Intel Shows Off Conroe (Desktop) Processors [Archive]

MacRumors

Mar 7, 2006, 04:06 PM

http://www.macrumors.com/images/macrumorsthreadlogo.gif (http://www.macrumors.com)

AnandTech provides coverage (http://www.anandtech.com/tradeshows/showdoc.aspx?i=2711) for Intel's presentations at the Intel Spring 2006 Developer Forum today.

Intel discussed their new "Core" microarchitecture which utilizes Intels' 65-nm manufacturing process. The Core Duo and Core Solo processors used in the iMac and Mac mini are already based on this new architecture. By mid-2007, Intel plans to offer a 45nm manufacturing process which promises even faster and lower power processors. In the meanwhile, however, Intel is expected to offer a full line of 65-nm chips (Merom, Conroe and Woodcrest) in the second half of 2006.

These slides (http://www.anandtech.com/tradeshows/showdoc.aspx?i=2711&p=4) show what kind of performance is expected from Intel's upcoming processor lines:

- Merom (mobile) is said to offer 20% more performance than the Core Duo at the same battery life.
- Conroe (desktop) is said to offer 40% more performance at 40% less power than the Pentium D 950.
- Woodcrest (servers) is said to offer 80% more performance at 35% less power than the Xeon.

Intel performed some benchmarks (http://www.anandtech.com/tradeshows/showdoc.aspx?i=2712) of Conroe compared to the Pentium D 950. And Anandtech offers (http://www.anandtech.com/tradeshows/showdoc.aspx?i=2713) some benchmarks comparing gaming performance and media encoding on the two platforms.

Intel's Conroe processor is speculated to be the heart of the upcoming Intel PowerMacs (http://guides.macrumors.com/Intel_PowerMac_%28Rumored%29) and should be released in the second half of this year. Apple's announcement that WWDC will take place in August (http://www.macrumors.com/pages/2006/03/20060307132013.shtml) of this year would provide a convenient launching pad for Intel PowerMacs which were rumored/speculated (http://www.macrumors.com/pages/2006/02/20060208225709.shtml) to be delivered in September.

rye9

Mar 7, 2006, 04:10 PM

Doesn't this put PowerMacs at the same "power" as everyday PC's now if they use the same processor?

EDIT: Oh wait, the PowerMacs will still have Mac OS X. :D

~Shard~

Mar 7, 2006, 04:12 PM

Great news, it's nice to see some of this made "official" by Intel, as all I had heard up to this point was unconfirmed #s.

Conroe PowerMacs are going to be great, and it will be nice to see the Xserve get Woodcrest. :cool:

MacMyDay

Mar 7, 2006, 04:15 PM

Those specs are pretty impressive. Intel is looking more and more promising, but then we all knew that in the first place.

rye9

Mar 7, 2006, 04:15 PM

Conroe PowerMacs are going to be great, and it will be nice to see the Xserve get Woodcrest. :cool:

I dont know why, but I think the Merom chips will get the most attention. :cool:

treblah

Mar 7, 2006, 04:16 PM

If for nothing other than AnandTech's coverage of Intel, I am glad Apple switched. :)

ravenvii

Mar 7, 2006, 04:16 PM

Meron, or Merom? ;)

FireArse

Mar 7, 2006, 04:25 PM

Meron, or Merom? ;)

Is Conroe 64-bit?

NewbieNerd

Mar 7, 2006, 04:27 PM

Yay Merom! Even though I'll be getting a used 17'' pbook from a fellow macrumors.com guy, I'm just excited at the prospect of what the future holds for them. I'll be really interested to see what appears as a new form factor for the powerbook/mbp and powermacs in the probably not-too-distant future. Something that will make the current forms look old and lousy... :D

QCassidy352

Mar 7, 2006, 04:29 PM

Merom sounds like a nice bump, but hardly earth-shattering... the others sound more impressive to me.

Fiveos22

Mar 7, 2006, 04:30 PM

I dont know why, but I think the Merom chips will get the most attention. :cool:

They're getting all of my attention (a merom pb will hopefully be my first laptop).

VanMac

Mar 7, 2006, 04:31 PM

Good to see.

I'm sure we will continue to see continual performance gains and less power consumption with this move to Intel. Bravo.

runninmac

Mar 7, 2006, 04:35 PM

Thats great that they compare it to the pentium D and all but I havnt heard great things from it, can anyone post some benchmarks from it?

rockthecasbah

Mar 7, 2006, 04:37 PM

jabooth

Mar 7, 2006, 04:38 PM

Wow, conroes shaping up nicely...:D

Question is, will we see a Conroe or a Merom in rev b. iMac?

Are people thinking that apple will get the merom and conroes straight into Macs, or will Mr. Jobs make us wait....:(

any thoughts?

Compatiblepoker

Mar 7, 2006, 04:41 PM

nagromme

Mar 7, 2006, 04:43 PM

My quad-Conroe Mac tower gets closer to reality :)

PS, how is "Merom" properly pronounced? MARE-um? Mer-OAM?

~Shard~

Mar 7, 2006, 04:46 PM

I dont know why, but I think the Merom chips will get the most attention.

For mobile solutions at least, yes indeed. The question will be, will the new iMacs in 2007 receive Conroe or Merom chips? An argument could be made for both...

Merom sounds like a nice bump, but hardly earth-shattering... the others sound more impressive to me.

Merom is what Conroe and Woodcrest are based off, so I wouldn't slam it too badly. :p ;) :cool:

nagromme

Mar 7, 2006, 04:46 PM

Are people thinking that apple will get the merom and conroes straight into Macs, or will Mr. Jobs make us wait....:(
Core Duo and Core Solo went straight to Mac the instant they were available--and those were the FIRST of the big transition.

I see no reason to expect we'd have to wait when Conroe and Merom appear. Sometimes a new model isn't ready for other reasons besides the processor, but there's no reason to assume the worst.

runninmac

Mar 7, 2006, 04:46 PM

My quad-Conroe Mac tower gets closer to reality :)

PS, how is "Merom" properly pronounced? MARE-um? Mer-OAM?

Oh I just relized I have been saying it wrong this whole time, I always thought it was Mem-rom... me and my dyslexia

macorama

Mar 7, 2006, 04:47 PM

So Woodcrest for the new xServes then? It's been a looooooong time since there's been some xServe news. But will it be Conroe or Woodcrest for the Powermacs?

~Shard~

Mar 7, 2006, 04:48 PM

PS, how is "Merom" properly pronounced? MARE-um? Mer-OAM?

I pronounce it "MARE-om", but what do I know.... ;)

realityisterror

Mar 7, 2006, 04:48 PM

Oh I just relized I have been saying it wrong this whole time, I always thought it was Mem-rom... me and my dyslexia

I thought it was just pronounced "ME-rom"... ?

Let's keep in mind that these numbers are often optimistic :cool:

maxterpiece

Mar 7, 2006, 04:49 PM

it's nice that there is such a performance boost with Merom (20% at the same battery life?) but who here would rather see a laptop that's a smidge LESS powerful to get a lot more battery life? :rolleyes:

Yeah, really... Wasn't a big part of the rationale to switch to intel that we would be getting higher performance/watt? Well yeah, the core duo is more powerful than the G4 by a lot, but apple had to put a bigger battery in there just to give it close to equal battery life to the last gen of powerbooks (they still get a little less life, but I'll write that off to the brighter screens).

How about a chip that runs just at G4 speed but uses like 1/4 the power? That would get me excited. Oh, and make it cheap too.

~Shard~

Mar 7, 2006, 04:51 PM

So Woodcrest for the new xServes then? It's been a looooooong time since there's been some xServe news. But will it be Conroe or Woodcrest for the Powermacs?

Woodcrest would be amazing in the PowerMacs, but I think Conroe is more of a realistic guess. That being said, there is nothing saying that Apple would not do something along the lines of the Mac mini and (presumably) the new iBook, where they have one chip (Solo) in the low end model and another chip (Duo) in the high end model. Of course, in that case, you're simply talking about different flavors of Yonah - I think it might be a bit extreme to mix Conroe and Woodcrest in the same product line...

Wags

Mar 7, 2006, 04:52 PM

So, The Core Duo in the MacBook Pro is not 65mm as in iMac and Mini??

~Shard~

Mar 7, 2006, 04:53 PM

How about a chip that runs just at G4 speed but uses like 1/4 the power? That would get me excited. Oh, and make it cheap too.

I think you'd be surprised to see just how many people would indeed be quite content with this scenario. ;)

In the battle between more power and better battery life, many laptop owners would choose the latter, not the former.

nagromme

Mar 7, 2006, 04:53 PM

Merom sounds like a nice bump, but hardly earth-shattering... the others sound more impressive to me.
20% more performance (REAL performance, not Ghz) in seven months isn't earth-shattering? Are you new to Macs? :p

Plus they'll be 64-bit like the G5.

I like those gaming benchmarks against AMD. However, when they say 180 average fps in Quake 4, what they don't tell you is that it probably dips down as far as 120 ;)

jabooth

Mar 7, 2006, 04:54 PM

Core Duo and Core Solo went straight to Mac the instant they were available--and those were the FIRST of the big transition.

I see no reason to expect we'd have to wait when Conroe and Merom appear. Sometimes a new model isn't ready for other reasons besides the processor, but there's no reason to assume the worst.

True. We already know Merom is pin compatible with current Core Duos, so assuming Merom is used in the iMac it shouldn't present too many problems...

Still, could we see a Conroe powered iMac?

Personally, I doubt it... on a side note does anyone know if Conroe will use a different socket to Merom?

strider42

Mar 7, 2006, 04:55 PM

Yeah, really... Wasn't a big part of the rationale to switch to intel that we would be getting higher performance/watt?

Performance per watt means higher performance at the same level of power usage. You know, kind of like 20% better performance at the same battery life. So this is delivering EXACTLY what steve jobs said it would. It wasn't about better battery life, it was about more power without frying the computer, and more power being available in a laptop. battery performance is obviously important to, but these will smoke the G4's at the same power usage. That's a GOOD thing.

amateurmacfreak

Mar 7, 2006, 04:56 PM

Oh I just relized I have been saying it wrong this whole time, I always thought it was Mem-rom... me and my dyslexia
Eh, I'm worst. I've been thinking mem-or-rame. Don't ask how I got that. :rolleyes:

anastasis

Mar 7, 2006, 04:58 PM

And this is precisely why Apple jumped ship from IBM. They have nothing to compete with this... And can you imagine a dual quad-core Mac?? :eek:

Legacy

Mar 7, 2006, 05:00 PM

boncellis

Mar 7, 2006, 05:01 PM

...In the battle between more power and better battery life, many laptop owners would choose the latter, not the former.

I think you're right, Shard, as usual. The question is, now that the performance (at least according to these numbers) is much improved, will Intel do something to enhance the battery life? Apple portables were stuck with relatively little innovation for so long that perhaps more power is a play to compensate for any perceived repuation they might have acquired.

Platform

Mar 7, 2006, 05:02 PM

I wonder, by the time I will be in the market for a new Mac, they will be something like 2000% faster :eek: :D

d.perel

Mar 7, 2006, 05:07 PM

cool stuff... laaaaame names;)

Whistleway

Mar 7, 2006, 05:11 PM

so why did apple put a mobile chip on the imac?

Forevercoolin

Mar 7, 2006, 05:15 PM

Would'nt Woodcrest as a replacement for the Xeon be the only proc that coluld be used in a dual proc configeration. Therefore it would be the logical replacement for the high-end Power Macs. I think most people who buy Power Macs are more concerned with speed not power consumption or price point.

kahos

Mar 7, 2006, 05:17 PM

cool stuff... laaaaame names;)

Well those are the intel "code name" for the chip

I dont know about you but i dont think a "7447" G4 sounds any better then a "merom" core duo

Anyways, these conroe benchmark agains the athlon 64 x2 are quite impressive

ChrisA

Mar 7, 2006, 05:17 PM

If you are building a notebook, or a small desktop like a Mini or an iMac then "power matters." You can't get the heat out of such a small enclosure without a huge noisy fan. So use these new chips when compute power per watt matters

What I don't understand is way you'd want them in a Power Mac replacement If you want a Power Mac replacement you can make a nice one today. No waiting. Simply build one of these
http://tinyurl.com/zsaxg
but put it in a pretty box and load Mac OSX on it The specs are just about right for a PoerMac. It's the only thing I've seen that would seriously outperform a G5 Quad Core Powermac It that's not enough Sun has four chip. eight core Opertoon boxes that have been sellig for some time now.

I just don't see Intel catching up to AMD at the high end.

Mikael

Mar 7, 2006, 05:17 PM

Hmm...well the G4 was dead on the mobile mac platform and mini...but how will Conroe really compare to the Quad G5? Given that the Pentium M architecture performs at around clock-for-clock the same as a G5 equivalent, we would need two Conroes at at least 2.8Ghz to match or beat the Quad G5..right?
Say again? If what you say about Pentium-M vs G5 is true, then two Conroes at 2.1GHz would match the current G5 Quad.

I just don't see Intel catching up to AMD at the high end.
Check out Anand's benchmarks of Conroe that were just published:

http://anandtech.com/tradeshows/showdoc.aspx?i=2713

They benchmark it against an overclocked Athlon64 FX-60 (@ 2.8GHz). The Conroe CPU runs at 2.66GHz and thoroughly smashes the Athlon64 performance wise. It should also be a good deal cooler.

This sure doesn't look good for AMD.

EDIT: Might be worth noting that Intel supplied both systems and most of the benchmarks. Anand couldn't find anything fishy about the systems when he inspected them, though.

Flash3441

Mar 7, 2006, 05:17 PM

Hi everyone! I'm a first time macrumors poster but long time reader.

I'm curious to know why some of you automatically assume that Conroe is going into the PowerMacs? Maybe you all know something I don't, but Woodcrest is suppose to be replacing Xeon and Xeon processors have found there way into not just servers but workstations in the PC market. I would think that Woodcrest would be the preferred processor in PowerMacs rather than Conroe if the price is right.

Electro Funk

Mar 7, 2006, 05:17 PM

Mord

Mar 7, 2006, 05:21 PM

Electro Funk

Mar 7, 2006, 05:22 PM

I thought it was just pronounced "ME-rom"... ?

lol... so did i... funny how we all perceived the pronunciation so differently...:D

Peace

Mar 7, 2006, 05:27 PM

These things will PUMP YOU UP ! :D

Lets hope OS X improves along with them..

ksz

Mar 7, 2006, 05:30 PM

ChrisA

Mar 7, 2006, 05:30 PM

... these conroe benchmark agains the athlon 64 x2 are quite impressive

Yes, but you are comparing a not yet released for sale Intel chip against a not top of the line AMD. AMD's top performer is the Operon.

dashiel

Mar 7, 2006, 05:32 PM

I just don't see Intel catching up to AMD at the high end.

i don't quite agree with you there. i used to think the same thing about PPC when they had a significant lead over intel around the G3 era. coupled with the RISC/CISC marketing it seemed intel was on the way out.

while i do believe they will catch and surpass AMD for a while, the very fact that intel has legitimate competition is a good thing for apple. neither ibm or motorola cared about the desktop platform or apple's small orders; at least not beyond the free marketing they got by piggy backing on the jobsian RDF field. motorola is almost completely focused on the embedded market, while ibm is laughing themselves silly with the next gen gaming consoles and to a lesser extent their server market. intel meanwhile is probably a bit worried about microsoft's "defection" with the xbox 360 and looking for a high profile client to get/keep them in the living room.

Forevercoolin

Mar 7, 2006, 05:35 PM

The problem with putting a Conroe in a power mac is that the Conroe would not be capable of a dual proc config. Only the woodcrest would be. Unless you know something we dont.

QCassidy352

Mar 7, 2006, 05:40 PM

Merom is what Conroe and Woodcrest are based off, so I wouldn't slam it too badly. :p ;) :cool:

oh, i'm not slamming it; I'm excited for it. But as someone else posted, I'd like to see an increase in battery life over an increase in processor power. More power at the same consumption is always good, but how 'bout a laptop that really lasts 7 hours?

ChrisA

Mar 7, 2006, 05:42 PM

Woodcrest is meant for the server market so it's possible Apple will use it in the new XServes. The PowerMac's replacement may well use Conroe, which fits the workstation category.

Why does Apple ned a high end CPU in an XServe? What do those run that needs it? They are mostly used as file servers. or maybe directory services and light duty web hosting. You don't see high CPU usage for these tasks

On the other hand people sometimes use PowerMacs to render hD video wich is typically 100% CPU bound.

arn

Mar 7, 2006, 05:44 PM

Analog Kid

Mar 7, 2006, 05:50 PM

I still can't get myself excited about watching the Intel roadmap... Feels a little like rooting for the pace car, or watching the rabbit at the dog track. There's nothing to bite your nails over-- whatever they do is vanilla by definition and everyone else is compared against it.

I guess this is what we're in for though-- 10% here, 20% there... Just enough to keep the quarter over quarter sales at Intel on a steady curve.

Gone are the days of realizing over the course of a year or two that what you have is probably good enough to last you for a while and then having a rabbit come out of the hat with a 200% performance increase before the years of rumors and false hopes start over again. I'll miss the mood swings and the careful analysis of whether the PPC is slightly better or slightly worse than, well... Intel.

boncellis

Mar 7, 2006, 05:51 PM

...intel meanwhile is probably a bit worried about microsoft's "defection" with the xbox 360 and looking for a high profile client to get/keep them in the living room.

As much as I would like to believe you, I just don't see Intel getting too worried. I don't mean to imply that they are complacent, but I am impressed with the roadmap and the offerings they have heretofore provided.

You may be proven correct in the not to distant future--the more I read about the VIIV architecture the more I'm impressed with the possibilities. Could Intel come up with some fusion of gaming and home entertainment (music, video downloads) in one box for the living room?

I apologize in advance if my speculation is too far off topic for this thread.

nagromme

Mar 7, 2006, 05:56 PM

The problem with putting a Conroe in a power mac is that the Conroe would not be capable of a dual proc config. Only the woodcrest would be. Unless you know something we dont.
That's an important point, and one I hadn't run across before this thread.

Is that something Intel has stated? Or just something people are assuming?

I do think dual-duals are to be expected in Intel PowerMacs--at least in high-end models when it becomes possible. Is that known to be impossible with Conroe?

Analog Kid

Mar 7, 2006, 05:56 PM

Performance per watt means higher performance at the same level of power usage. You know, kind of like 20% better performance at the same battery life. So this is delivering EXACTLY what steve jobs said it would. It wasn't about better battery life, it was about more power without frying the computer, and more power being available in a laptop. battery performance is obviously important to, but these will smoke the G4's at the same power usage. That's a GOOD thing.
Performance per watt also means the same performance with lower power usage. Ratios are funny that way... ;)

I think Intel usually sells "low voltage" and "ultra low voltage" versions of their chips too, don't they? There's probably going to be a version optimized for low power, it's just not a glamorous. Merom can probably get the same performance as Yonah with 30-40% less power, I'd guess. It's those last MIPS that cost the most juice...

ZorPrime

Mar 7, 2006, 06:01 PM

i don't quite agree with you there. i used to think the same thing about PPC when they had a significant lead over intel around the G3 era. coupled with the RISC/CISC marketing it seemed intel was on the way out.

while i do believe they will catch and surpass AMD for a while, the very fact that intel has legitimate competition is a good thing for apple. neither ibm or motorola cared about the desktop platform or apple's small orders; at least not beyond the free marketing they got by piggy backing on the jobsian RDF field. motorola is almost completely focused on the embedded market, while ibm is laughing themselves silly with the next gen gaming consoles and to a lesser extent their server market. intel meanwhile is probably a bit worried about microsoft's "defection" with the xbox 360 and looking for a high profile client to get/keep them in the living room.

good points but let's not also forget IBM's making a grip of $$$ from it's many R&D contracts developing BlueGene, then there's cell (http://www.cellsupercomputer.com/). :cool:

shawnce

Mar 7, 2006, 06:03 PM

so why did apple put a mobile chip on the imac?

...because the Core Duos are available now and they perform rather well and run rather cool compared to the G5 they replaced. They also are a better chip then any other currently shipping desktop chip from Intel with the heat profile needed for the iMac case.

blackfox

Mar 7, 2006, 06:04 PM

I do think dual-duals are to be expected in Intel PowerMacs--at least in high-end models when it becomes possible. Is that known to be impossible with Conroe?
Depending on timing, you may see quad-core chips finding their way into PMs or Xserves. Intels quad-core (clovertown iirc) are expected Q1 2007, making them a potential for a revB PM/Xserve, or even a late revA (with shipping delays).

danielwsmithee

Mar 7, 2006, 06:04 PM

I do think dual-duals are to be expected in Intel PowerMacs--at least in high-end models when it becomes possible. Is that known to be impossible with Conroe?

I think PowerMacs will be Woodcrest based for this reason. The Conroe will be used in the iMac and maybe lowest end PowerMac. Merom in MacBook Pro and CoreDuo in the iBook and mini.

I've got the CCard ready for a Conroe based iMac!!!

chaos86

Mar 7, 2006, 06:06 PM

shawnce

Mar 7, 2006, 06:11 PM

That's an important point, and one I hadn't run across before this thread.

Is that something Intel has stated? Or just something people are assuming?

I do think dual-duals are to be expected in Intel PowerMacs--at least in high-end models when it becomes possible. Is that known to be impossible with Conroe?

Maybe Kentsfield is the desktop class answer to that question...

Desktop Platform
Desktops can deliver greater compute performance as well as ultra-quiet, sleek and low-power designs.

Intel is developing a desktop-optimized, dual-core processor based on the new, state of the art, Intel Core microarchitecture, codenamed Conroe. The Conroe processor will work within the 2006 Digital Home platform codenamed Bridge Creek, and the 2006 Digital Office platform, codnamed Averill. Conroe is targeted for introduction in the third quarter of 2006.

Intel will also deliver a quad-core (4 full execution cores) processor to the high-end desktop based upon this new microarchitecture, codenamed Kentsfield. Kentsfield is targeted for introduction in the first quarter of 2007.

The Kentsfield appears to two Conroe dies (or very close relatives of) in a single package.

boncellis

Mar 7, 2006, 06:14 PM

...I've got the CCard ready for a Conroe based iMac!!!

When you say "CCard" do you mean CableCard? jk

DeathChill

Mar 7, 2006, 06:19 PM

no conroe cpu will ever be used in a powermac, ever woodcrest is whats going in powermacs, conroe is not an SMP chip, apple is not giveing up on dual physical cpu workstations, and will at least have quads, if not an all quad lineup.
Uh, I don't see why Apple wouldn't use Conroe since it will be available first. They will be a dual CPU workstation, without the hassle of two separate chips. :)

DVK916

Mar 7, 2006, 06:30 PM

True. We already know Merom is pin compatible with current Core Duos, so assuming Merom is used in the iMac it shouldn't present too many problems...

Still, could we see a Conroe powered iMac?

Personally, I doubt it... on a side note does anyone know if Conroe will use a different socket to Merom?

It is far more likely that we will see a Conroe iMac and a Merom iMac. There is no reason to go with Merom for an iMac.

Those wondering how to pronounce Merom, it is a hebrew word so just ask someone who speaks hebrew how you would say it.

shawnce

Mar 7, 2006, 06:30 PM

Uh, I don't see why Apple wouldn't use Conroe since it will be available first. They will be a dual CPU workstation, without the hassle of two separate chips. :)

I also see Apple using the Conroe in the PowerMac replacement since a dual core Conroe will outperform any dual core PowerMac and outperform a quad core PowerMac in all but thread heavy work loads (likely hold its own even then).

I think we may see quad core Intel "PowerMac" systems in early 2007 with dual core coming in late 2006.

DVK916

Mar 7, 2006, 06:33 PM

I think PowerMacs will be Woodcrest based for this reason. The Conroe will be used in the iMac and maybe lowest end PowerMac. Merom in MacBook Pro and CoreDuo in the iBook and mini.

I've got the CCard ready for a Conroe based iMac!!!

This is what I am thinking as well.

Woodcrest for PowerMac
Conroe for iMac
Merom for MacBook Pro

and Core Duo for iBook and mini

runninmac

Mar 7, 2006, 06:33 PM

runninmac

Mar 7, 2006, 06:35 PM

This is what I am thinking as well.

Woodcrest for PowerMac
Conroe for iMac
Merom for MacBook Pro

and Core Duo for iBook and mini

Is there any reason why a Server chip shouldnt be in a desktop? Are they just better overall except for the power consumption? What about in games?

shawnce

Mar 7, 2006, 06:35 PM

It is far more likely that we will see a Conroe iMac and a Merom iMac. There is no reason to go with Merom for an iMac. I doubt Apple will use higher wattage chip in the iMac given the iMacs form factor. Apple may stay with using high-end versions of laptop chips. In the end it likely comes down to cost to Apple (chip price)... between using a low-end desktop or high-end laptop chips.

DVK916

Mar 7, 2006, 06:41 PM

I doubt Apple will use higher wattage chip in the iMac given the iMacs form factor. Apple may stay with using high-end versions of laptop chips. In the end it likely comes down to cost to Apple (chip price)... between using a low-end desktop or high-end laptop chips.

iMac form factor should still be able to work with a Conroe. Remember the iMac housed G5s in them and we know what power hogs those were. Per clock Merom will cost more than Conroe. So based on cost Apple will go with Conroe.

shawnce

Mar 7, 2006, 06:47 PM

iMac form factor should still be able to work with a Conroe. Remember the iMac housed G5s in them and we know what power hogs those were. Per clock Merom will cost more than Conroe. So based on cost Apple will go with Conroe.

The G5s used in the iMac ran in the 30-50 watt range which is still below the Conroe (depending on which clock rate of Conroe you use). Also the mother board in the iMac Core Duo can support 1st generation Merom but not the Conroe, that would be a cost savings.

Anonymous Freak

Mar 7, 2006, 06:58 PM

Is there any reason why a Server chip shouldnt be in a desktop? Are they just better overall except for the power consumption? What about in games?

Well, with the new 'Core' lineup, the processors are at their core all the same. It's just the extra functions that differ. The server and laptop chips will probably have a slower front side bus than the desktop chip (for power reasons on the laptop; for 'stability' reasons on the server,) and the server chip MAY include extra server-related functionality that the others don't. (Virtualization would fall in this category, but even the Core Duo has it, so it's not a great example.)

For example, currently, the Core Duo is only on a 667 MHz bus, and only just moved to that; while the desktop Pentium 4 and Pentium D are on an 800 MHz bus, and have been there for a few years. Server Xeons JUST moved to an 800 MHz bus, after being on 667 or even 533 MHz, depending on the use, for many years. And the ultra-high-end desktop Pentium Extreme Editions use a 1066 MHz bus. And while Core Duo is a different 'core' than Pentium 4/D or Xeon, P4/D and Xeon all share the same basic core. So there are many variations only in bus speed in that family; much less in other areas of design.

In games, they should perform equally to an equal-spec desktop or mobile chip. So a current Xeon 1MB L2 cache 3.2 GHz on an 800 MHz bus should perform identically to a Pentium 4 1 MB L2 cache 3.2 GHz on an 800 MHz bus. (Or even a Pentium 4-M of same spec, the direct mobile equivalent of the Pentium 4; although Intel is positioning the Core Duo to replace that as the 'high end' mobile chip at present, in preparation for the replacement of the whole Pentium 4 line with the Core line later this year.)

DavidCar

Mar 7, 2006, 07:03 PM

My quad-Conroe Mac tower gets closer to reality :)
...
My question is whether a four core Conroe will be pin compatible with a dual core Conroe, allowing a processor upgrade to double the number of cores.

ksz

Mar 7, 2006, 07:07 PM

Why does Apple ned a high end CPU in an XServe? What do those run that needs it? They are mostly used as file servers. or maybe directory services and light duty web hosting. You don't see high CPU usage for these tasks

On the other hand people sometimes use PowerMacs to render hD video wich is typically 100% CPU bound.
Price is an important factor. Xeons are very expensive, even in volume purchases. The Xeon dual core processor (2.8 GHz) sells for about $1100 to $1200 (retail).

XServes are used in high-performance clusters for industrial and scientific purposes. Virginia Tech isn't the only institution that has built a supercomputer out of XServes; we have as well (but I cannot say who 'we' are). Servers are used for many other purposes than web hosting.

For the desktop Macintosh, price is important. Apple will most likely price the new machines at the SAME levels as existing PowerMacs. Woodcrest will be too expensive, I think, to achieve those price points. And it's not necessary.

I haven't seen a reference from Intel that rules out SMP on Conroe (but then I haven't looked), but it's understandable given that Pentium does not support SMP, but Xeon does.

So if Conroe==Pentium and Woodcrest==Xeon, it's possible that only the top end PowerMac replacement will get Woodcrested.

DavidCar

Mar 7, 2006, 07:10 PM

I also see Apple using the Conroe in the PowerMac replacement since a dual core Conroe will outperform any dual core PowerMac and outperform a quad core PowerMac in all but thread heavy work loads (likely hold its own even then).

I think we may see quad core Intel "PowerMac" systems in early 2007 with dual core coming in late 2006.

I would think that Apple would want the first Intel PowerMac to be a big improvement over the quad, and not just "hold its own."

Mechcozmo

Mar 7, 2006, 07:31 PM

Oh I just relized I have been saying it wrong this whole time, I always thought it was Mem-rom... me and my dyslexia

I didn't notice I was wrong until I saw your post. Oops.

I'm still waiting for the "Same speed, 2x battery life" processor. :rolleyes:

capone2

Mar 7, 2006, 07:49 PM

I pronounce it "MARE-om", but what do I know.... ;)

.."more-um" merom

cant wait for the 17" merom 2.5-3.0ghz 64bit.....with leopard of course

shawmanus

Mar 7, 2006, 08:07 PM

Price is an important factor. Xeons are very expensive, even in volume purchases. The Xeon dual core processor (2.8 GHz) sells for about $1100 to $1200 (retail).

XServes are used in high-performance clusters for industrial and scientific purposes. Virginia Tech isn't the only institution that has built a supercomputer out of XServes; we have as well (but I cannot say who 'we' are). Servers are used for many other purposes than web hosting.

For the desktop Macintosh, price is important. Apple will most likely price the new machines at the SAME levels as existing PowerMacs. Woodcrest will be too expensive, I think, to achieve those price points. And it's not necessary.

I haven't seen a reference from Intel that rules out SMP on Conroe (but then I haven't looked), but it's understandable given that Pentium does not support SMP, but Xeon does.

So if Conroe==Pentium and Woodcrest==Xeon, it's possible that only the top end PowerMac replacement will get Woodcrested.

Intel has priced woodcrest very competitively.

http://www.theinquirer.net/?article=29510

Intel Woodcrest at 3.0 GhZ at $850 is not bad. I think Apple can put woodcrest in powermac or mac pro.

nagromme

Mar 7, 2006, 08:09 PM

The Kentsfield appears to two Conroe dies (or very close relatives of) in a single package.
Yes, but meanwhile, can two separate Conroes be used together, much like the G5 quad has dual duals?

I've seen some posts here suggesting Conroe can't do that, and I'm wondering if people are just ASSUMING that, or if it's something Intel has actually indicated?

Is there any reason why a Server chip shouldnt be in a desktop? Are they just better overall except for the power consumption? What about in games?
Good question. If it comes down to cost, how much more do server chips cost? (EDIT: thanks shawmanus! Now how much less, in theory, would Conroe be at the same Ghz?)

Those wondering how to pronounce Merom, it is a hebrew word so just ask someone who speaks hebrew how you would say it.
Yeah, I was wondering what it would really be. My friend who speaks Hebrew says it's pronounced much like the English "this side." :confused:

runninmac

Mar 7, 2006, 08:16 PM

I just thought of this also, since the clock speed starts out great thats good, but then will they hit a wall once they hit the 3.__(you fill in number)Ghz? Or have they planed around that?

~Shard~

Mar 7, 2006, 08:25 PM

oh, i'm not slamming it; I'm excited for it. But as someone else posted, I'd like to see an increase in battery life over an increase in processor power. More power at the same consumption is always good, but how 'bout a laptop that really lasts 7 hours?

Yep, I completely agree, and know where you're coming from. Sometimes more processing power isn't always better if it means no reduction in power consumption itself. :cool:

Marble

Mar 7, 2006, 09:06 PM

I'll be very interested to see if the iMac gets Conroe or Merom. I think this G4 TiBook still has a year in it and I'd like to get a desktop class processor that has as much life in it as this baby has.

Anonymous Freak

Mar 7, 2006, 09:09 PM

Yeah, I was wondering what it would really be. My friend who speaks Hebrew says it's pronounced much like the English "this side." :confused:

I have a feeling he was pulling your leg.

But, I (and two Intel employees I know) all pronounce it 'MARE-om' (Mare like a horse, om rhyming with pond.) But every once in a while, I hear a 'MARE-um'.

Conroe is 'CON-row'.

boncellis

Mar 7, 2006, 09:19 PM

I have a feeling he was pulling your leg.

But, I (and two Intel employees I know) all pronounce it 'MARE-om' (Mare like a horse, om rhyming with pond.) But every once in a while, I hear a 'MARE-um'.

Conroe is 'CON-row'.

Actually, nagromme is pulling our leg. I personally can't get enough of any and all things having to do with those wonderful chinese characters...

I laughed, nagromme.

Analog Kid

Mar 7, 2006, 09:37 PM

Have you seen the tests (http://www.anandtech.com/tradeshows/showdoc.aspx?i=2713) against AMD's top processor thats been overclocked? I bet not. It even should be getting faster than that by the time its ready for retail and it will also be offered in a 3.0 extreme edition. Read/look that before you start pooping on everyone parade.
:D Wow... Silly me, thinking it appropriate to express an opinion in a "forum". You take a swing at everyone you overhear saying something you disagree with?

No parade pooping going on here, just realizing that times are changing. Not as exciting, but hopefully more consistent.

Before you swipe at the next guy though, you might try putting your tongue in your cheek while you read the post and see if it feels right there...

hadlock

Mar 7, 2006, 09:38 PM

wyatt23

Mar 7, 2006, 09:42 PM

does anyone know what socket size the new processors are?

it'd be nice to know that if say i bought a low end intel iMac now, i could swap in a Conroe / Meron later.

austinbond

Mar 7, 2006, 09:44 PM

When do ya'll think they will release an iBook rev. B...History shows that iBooks are usually announced in April or May, and around October or November. I don't want to buy an iBook this summer for college and have a significant upgrade coming months after. You don't think they will get conroe as well?

Thanks guys!

AidenShaw

Mar 7, 2006, 09:44 PM

I'm curious to know why some of you automatically assume that Conroe is going into the PowerMacs? Maybe you all know something I don't, but Woodcrest is suppose to be replacing Xeon and Xeon processors have found there way into not just servers but workstations in the PC market. I would think that Woodcrest would be the preferred processor in PowerMacs rather than Conroe if the price is right.
Most readers here probably aren't familiar with the way that Intel has traditionally drawn a line between single-socket chips and dual-socket capable chips - even when the basic design is the same.

A Xeon (DP) and a Pentium 4 are almost the same chip, but only a Xeon runs in a dual-socket config.

Many readers here will say "dual Conroes" without realizing that's a marketing impossibility - you'll need "dual Woodcrests".
_______________________________________________________

However, I predict that Apple will in fact introduce a Conroe minitower - something to fill the gap between the tragically constrained MiniMacIntel and the tragically huge PowerMac G5. A Conroe (dual-core, single socket) in a typical minitower form factor (or even one of the SFF like a Dell Optiplex- see picture below) would be a good fit for someone who wants more expandability than a Mini, doesn't want the forced monitor of the iMac, but doesn't need a huge PowerMac.

http://img.dell.com/images/global/products/optix/gx_4_chassis_180x110.jpg

AidenShaw

Mar 7, 2006, 09:49 PM

does it feel weird to anyone else here having intel chip announcements on the MR front page?

well i guess it's about to become a regular occurance since they tend to update their lines every time they sneeze.
The IDF is Intel's equivalent to the WWDC - so it's obvious why it rates coverage for anyone interested in the x86/x64 platform.

AidenShaw

Mar 7, 2006, 10:00 PM

Is there any reason why a Server chip shouldnt be in a desktop? Are they just better overall except for the power consumption? What about in games?
Intel typically prices DP-capable server chips (dual socket workstation and low-end server) quite a bit higher than single-socket only chips.

MP-capable server chips are *much* more expensive - but often have much larger caches as well (current Xeon MPs have up to 8 MiB on-chip caches).

backdraft

Mar 7, 2006, 10:00 PM

Not impressed by Intel... Same garbage X86 chips with an inferior design...

boncellis

Mar 7, 2006, 10:03 PM

...However, I predict that Apple will in fact introduce a Conroe minitower - something to fill the gap between the tragically constrained MiniMacIntel and the tragically huge PowerMac G5. A Conroe (dual-core, single socket) in a typical minitower form factor (or even one of the SFF like a Dell Optiplex- see picture below) would be a good fit for someone who wants more expandability than a Mini, doesn't want the forced monitor of the iMac, but doesn't need a huge PowerMac.

Thank you, thank you, thank you, AidenShaw (for the validation). I was saying something similar in the WWDC thread and got hammered by different posters including one who defensively suggested that I be content with not buying a G5 rather than expressing an opinion that didn't conform with his own. Not exactly helpful, or responsive, for that matter.

I'm glad to see I'm not the only one who thinks Apple could introduce an intermediary machine--I would be in the market for it.

Donm

Mar 7, 2006, 10:10 PM

Donm

Mar 7, 2006, 10:15 PM

Yes, but you are comparing a not yet released for sale Intel chip against a not top of the line AMD. AMD's top performer is the Operon.

but its still desktop processor to desktop processor. When they test Woodcrest to optinon, what are you going to complain about then?

Chundles

Mar 7, 2006, 10:17 PM

I saw this asked before, but nobody responded -

is the Conroe processor 64 bit?

If it is, we could see the entire mac lineup switch entirely to Merom/Conroe processors, which would be interesting, since they'll be announcing the release of 10.5 - which could be the first fully 64 bit version of OSX.

64 bit finder in cocoa... mmm.

People who bought the gimpy 32 bit intel powerbooks are going to be complaining mightily come August :D

Yes, it is 64 bit. The Merom processor from which Conroe and Woodcrest are based is a faster, 64 bit version of the current core duo processor we know as Yonah with a few more bells and whistles. The Core Duo debuted at a minimum of 1.67GHz, Merom will debut at 2.33GHz - is fast yes? Yes indeed.....

duffman9000

Mar 7, 2006, 10:27 PM

Not impressed by Intel... Same garbage X86 chips with an inferior design...

Please die... or better yet, choke on your G3.

Finally... seems that Intel wasn't just blowing smoke but instead came out swinging. Stevie boy must have seen this coming. Performance + supply = a helluva of lotta happy Mac users.

duffman9000

Mar 7, 2006, 10:31 PM

However, I predict that Apple will in fact introduce a Conroe minitower - something to fill the gap between the tragically constrained MiniMacIntel and the tragically huge PowerMac G5. A Conroe (dual-core, single socket) in a typical minitower form factor (or even one of the SFF like a Dell Optiplex- see picture below) would be a good fit for someone who wants more expandability than a Mini, doesn't want the forced monitor of the iMac, but doesn't need a huge PowerMac.

http://img.dell.com/images/global/products/optix/gx_4_chassis_180x110.jpg

QFT... good to see the Intel haters quickly disappearing.

jared_kipe

Mar 7, 2006, 10:50 PM

Hey guys, who cares if Conroe is dual processor config ready. There are Conroe chips with 4 cores on them already
http://anandtech.com/tradeshows/showdoc.aspx?i=2713
Top chip.

Also, if it isn't a trick, it looks like the chips are hovering a bit. So I think we can assume these aren't LGA775.

Multimedia

Mar 7, 2006, 11:26 PM

danielwsmithee

Mar 7, 2006, 11:28 PM

I'm glad to see I'm not the only one who thinks Apple could introduce an intermediary machine--I would be in the market for it.

That would be my ideal market as well!

nagromme

Mar 7, 2006, 11:30 PM

Not impressed by Intel... Same garbage X86 chips with an inferior design...
Actually, Intel has as much as admitted the marketing-driven, brute force bad direction the Pentium 4 went in. Mobiles forced them to.

That's why Pentium M scrapped that and started fresh building on the Pentium 3 in a different and better direction. That's why Core Duo and all future Intel chips are moving ahead with THAT direction instead of the Pentium 4.

They aren't the same old chips.

Multimedia

Mar 7, 2006, 11:47 PM

Yes, but meanwhile, can two separate Conroes be used together, much like the G5 quad has dual duals?

I've seen some posts here suggesting Conroe can't do that, and I'm wondering if people are just ASSUMING that, or if it's something Intel has actually indicated?Doesn't matter to Quad PPC Owners. We'll Just Wait For Dual Kentsfield Quad Cores. :p

fiercetiger224

Mar 8, 2006, 12:18 AM

Multimedia

Mar 8, 2006, 12:19 AM

And this is precisely why Apple jumped ship from IBM. They have nothing to compete with this... And can you imagine a dual quad-core Mac?? :eek:I not only imagine it. I EXPECT it. It's called a Dual Kentsfield. And it will be shipping by Spring 2007 FOR SURE. That wil be my first Mac Pro. :p :D

Multimedia

Mar 8, 2006, 12:37 AM

Agreed... My credit card will take another punishing when me sees a merom macbook pro;)I'm with you on that. I'll wait for a 17" Merom MBP with Expresscard 54. :) That will be the first 64-bit portable Mac.

shawnce

Mar 8, 2006, 12:48 AM

I would think that Apple would want the first Intel PowerMac to be a big improvement over the quad, and not just "hold its own."

Oh a Conroe (dual core) will out perform a quad PowerMac in many work flows, only in heavily threaded work flows (optimized for many cores) would the quad PowerMac have a chance of holding its own.

(and I am a happy Quad PowerMac owner)

DavidCar

Mar 8, 2006, 12:48 AM

Doesn't matter to Quad PPC Owners. We'll Just Wait For Dual Kentsfield Quad Cores. :p
What about Quad owner wannabes who don't want to wait until 2007?

DVK916

Mar 8, 2006, 01:10 AM

Also people who keep saying Merom will work in current Intel Macs, are wrong. Pin to Pin compatible doesn't mean it will work. The chip set isn't compatible and that is what matters. The current Napa chip set don't support 64 bit processor, but Napa64 will.

Multimedia

Mar 8, 2006, 01:12 AM

What about Quad owner wannabes who don't want to wait until 2007?Should be Intel Quads with two Dual Core Conroes first. Or you can buy a PPC Quad refurb now for only $2799.

Someone mentioned perhaps Dual Core Conroe's can't be paired into a Quad. I don't know. Anyone here know? :confused:

DVK916

Mar 8, 2006, 01:21 AM

Should be Intel Quads with two Dual Core Conroes first. Or you can buy a PPC Quad refurb now for only $2799.

Someone mentioned perhaps Dual Core Conroe's can't be paired into a Quad. I don't know. Anyone here know? :confused:

Conroe doens't support MP configurations, but woodcrest will which is why I think the PowerMacs will use woodcrest.

mongoos150

Mar 8, 2006, 01:23 AM

I'm waiting for someone to pop a Conroe in their iMac Core Duo...so I can follow suit!:D

Multimedia

Mar 8, 2006, 01:34 AM

I'm waiting for someone to pop a Conroe in their iMac Core Duo...so I can follow suit!:DThis can't happen because the 32-bit Yonah Chipset Won't Support the 64-bit Conroe. Sorry. :(

nataku

Mar 8, 2006, 01:44 AM

wow the performance is amazing. AMD better start pouring money into the R&D department. I never thought I'd see the day Intel regain the top spot. :) GOod news for Mac geeks!!! :D

ksz

Mar 8, 2006, 01:47 AM

This is what I think will happen. Apple will have two lines of desktop Macs, or THREE if you include include the Mac mini, or FOUR if you include the iMac (but the iMac is an all-in-one solution, so it doesn't fall into this category).

It will be Mac mini, Mac (yes, simply called the Mac), and Mac Pro. The Mac will just include Conroe processors, and the Mac Pros will include the Woodcrest processors. Maybe this will happen? I don't know, but I'm assuming this will happen, just because I've seen them split the Macbook into two lines. Or rather, they just changed the names of the iBook and PowerBook. Oh well, they've never really had a "mid-level" desktop line.

Also, maybe they'll include the Woodcrest processors in the Xserve lines...Obviously this is just wishful thinking! ^_^
I think you, AidenShaw, and others may be onto something here...

a. Mac mini (no name change) with Yonah then Merom
b. Mac (mini-tower form factor) with Conroe
c. Mac Pro (full tower) with Woodcrest

The 'Mac' as defined in (b) may take significant sales away from the iMac, splitting that demographic, but also luring more buyers because of greater flexibility in configuring a system with their own choice of LCD, graphics card, etc.

Apple has only 2 form factors today for headless Macs: really tiny and really large. They seem to have forgotten all about Goldilocks.

Woodcrest-based Mac Pros could just be priced a notch higher -- even at $850 for Woodcrest in large quantities, putting two of those puppies into one machine means $1700 in costs for the processor alone. Factor in costs for a dual-socket motherboard, other component costs, manufacturing costs, warranty costs, blah blah blah, and you've got a COGS of about $3000. Add in wholesale/retail commissions and leave a gross margin of 20%, and you've got an MSRP around $3700 with a good (but not great) graphics card and 1GB memory. This is not too far from today's quad-core PM, and could be really enticing for those who need the power. (P.S. These figures are all made up. Bash them with better data.)

jabooth

Mar 8, 2006, 01:57 AM

So... if conroe is released around September time, would we see a Conroe iMac shortly thereafter? If thats the case, it might just be enough to prevent me from buying a core duo iMac very soon.... :p

ksz

Mar 8, 2006, 02:04 AM

So... if conroe is released around September time, would we see a Conroe iMac shortly thereafter? If thats the case, it might just be enough to prevent me from buying a core duo iMac very soon.... :p
Conroe has 40% less power consumption than a Pentium D 950. While that is good, it may not be enough to meet cooling and silence requirements of the iMac. Merom, as the successor to Core Duo, is the logical choice for the next iMac.

jabooth

Mar 8, 2006, 02:30 AM

Conroe has 40% less power consumption than a Pentium D 950. While that is good, it may not be enough to meet cooling and silence requirements of the iMac. Merom, as the successor to Core Duo, is the logical choice for the next iMac.

Hmmm... is it worth waiting out for a Merom iMac? Personally, I don't think the 20% performance increase warrants a wait of 6 months, although waiting for a 64-bit chip could be very beneficial...

I'm obviously talking here of people who are planning on buying an iMac now, yet have seen Conroe/Merom could be out in half a year... :confused:

nataku

Mar 8, 2006, 04:50 AM

So... if conroe is released around September time, would we see a Conroe iMac shortly thereafter? If thats the case, it might just be enough to prevent me from buying a core duo iMac very soon.... :p

This shouldnt stop you from buying an iMac. This is a growing problem amongst us Mac users. We shouldn't be a victim of expecting too much/overhyping. Look at what happened at the Mac Mini announcement. People got really pissed that they didnt get what they expected, a Media Center Mac Mini, because of all the expectations that they built up. I don't think Apple would immediately switch every iMac to Conroe as soon as it gets released. I think this would be reserved for a High-end/BTO iMac or for PowerMacs. It would be safe to say that PowerMacs would definitely get this processor.

Conroe has 40% less power consumption than a Pentium D 950. While that is good, it may not be enough to meet cooling and silence requirements of the iMac. Merom, as the successor to Core Duo, is the logical choice for the next iMac.

Hmmm... another mobile processor for a desktop machine?:confused:

Evangelion

Mar 8, 2006, 05:28 AM

Check out Anand's benchmarks of Conroe that were just published:

http://anandtech.com/tradeshows/showdoc.aspx?i=2713

They benchmark it against an overclocked Athlon64 FX-60 (@ 2.8GHz). The Conroe CPU runs at 2.66GHz and thoroughly smashes the Athlon64 performance wise. It should also be a good deal cooler.

This sure doesn't look good for AMD.

Take those benchmarks with HUGE grain of salt. As you mentioned, the test-systems were provided by Intel. There are TONS of things they could do to cripple the AMD-system. They can say that "this has an overclocked A64-CPU in it", but there's simply no knowing how they systems were REALLY configured. Even the apps they were benchmarking were provided by Intel, so they could have simply picked and chosen the benchmarks where they will be faster than the AMD-system.

And it's worth noting that the AMD-CPU they were comparing to is available today (although they did overclock it), whereas Conroe is not. So they are basically telling people that their FUTURE CPU will be faster than their competitors CURRENT CPU. nBy the time Conroe ships, AMD will have new Athlons available as well.

Evangelion

Mar 8, 2006, 05:31 AM

Not impressed by Intel... Same garbage X86 chips with an inferior design...

Please explain. If they mop the floor with PowerPC, how exactly do they have "inferior design"?

jabooth

Mar 8, 2006, 05:34 AM

This shouldnt stop you from buying an iMac. This is a growing problem amongst us Mac users. We shouldn't be a victim of expecting too much/overhyping. Look at what happened at the Mac Mini announcement. People got really pissed that they didnt get what they expected, a Media Center Mac Mini, because of all the expectations that they built up. I don't think Apple would immediately switch every iMac to Conroe as soon as it gets released. I think this would be reserved for a High-end/BTO iMac or for PowerMacs. It would be safe to say that PowerMacs would definitely get this processor.

You're right nataku. I have told myself that I won't play the waiting game, but I'm about to buy my first Mac, I don't want it to be outdated in a few months!

I'm not bothered about the increased speed - its the 32/64bit issue that worries me more... :(

BRLawyer

Mar 8, 2006, 05:37 AM

wow the performance is amazing. AMD better start pouring money into the R&D department. I never thought I'd see the day Intel regain the top spot. :) GOod news for Mac geeks!!! :D

This was obvious since day one, and the main reason behind Apple's choice...even though AMD chips were seemingly "faster" in the beginning, the company has no scale to compete with Intel; it just got an edge because of its much more streamlined and less bureaucratic structure...

Just remember the Mac clone industry...they had far fewer computers to manufacture and could have quicker access to the latest technology, unlike Apple.

And now that Intel's R&D is completed, AMD will surely suffer a downward spiral run. It is gonna be in the beleaguered league pretty soon, trust me...AMD is history.

janstett

Mar 8, 2006, 05:50 AM

so why did apple put a mobile chip on the imac?

Same reason they did with the Mini. Keep it small and tight, re-use laptop components.

ScottB

Mar 8, 2006, 05:54 AM

I think we will see the Merom going into MacBook Pro's almost as soon as they have been released.

janstett

Mar 8, 2006, 05:57 AM

does it feel weird to anyone else here having intel chip announcements on the MR front page?

I thought that too! Mac announcements used to be fighting for attention AGAINST Intel events like IDF, now they are intimately joined.

nataku

Mar 8, 2006, 06:03 AM

You're right nataku. I have told myself that I won't play the waiting game, but I'm about to buy my first Mac, I don't want it to be outdated in a few months!

I'm not bothered about the increased speed - its the 32/64bit issue that worries me more... :(

Don't worry jabooth. A Core-Duo iMac is a very smart investment. Since this is your first mac, you don't have to worry about being outdated very very fast, unlike in the PC world. I'm a switcher myself. I had my powerbook since november last year. It is still going strong and I still like the performance. Don't worry about the 64bit thing. Mac OS X is a 32 bit os running on a 32 bit kernel. A 64bit Mac OS X is still out of the question since Apple won't force its customers to transition 2 times in such a short time (PowerPC to Intel and 32bit to 64bit). That would be very unApple :) Developers are also just starting to master the 32bit Intel code. so I suggest you get the Core-Duo Intel iMac. If you have the extra cash, by all means get the Conroe version (if any) later this year. If you don't have the extra cash, dont you worry about buying one right now. You will know what I mean when you get one.:)

jabooth

Mar 8, 2006, 06:04 AM

so why did apple put a mobile chip on the imac?

People forget that an iMac really is of laptop form factor...

Yes, it is slightly larger than a laptop, (and does incorperate a much bigger screen) but it is still only a few inches deep. The heatsink and fan on my A64 3500 is deeper than the whole iMac!

And its the form factor that makes me want an iMac soooooo bad.... :D

nataku

Mar 8, 2006, 06:04 AM

I think we will see the Merom going into MacBook Pro's almost as soon as they have been released.

I think so to. I'm guessing the 17" MacBook Pros. It's just a guess anyway :D

jabooth

Mar 8, 2006, 06:08 AM

Don't worry jabooth...

You know what nataku?

...You're right. :rolleyes: It really is stupid to play waiting game!

I'm just waiting untill April 1st just to check the MacBook isn't anything to change my mind, then I'll make a purchase. Thanks again mate :D

Glen Quagmire

Mar 8, 2006, 06:29 AM

so why did apple put a mobile chip on the imac?

Because the only alternative at the moment is the Pentium D, which would melt the iMac and set fire to your house.

janstett

Mar 8, 2006, 06:34 AM

64 bit finder in cocoa... mmm.

64-bit just for the sake of being 64-bit doesn't make it faster. It can actually be slower.

Mikael

Mar 8, 2006, 06:47 AM

Yes, but you are comparing a not yet released for sale Intel chip against a not top of the line AMD. AMD's top performer is the Operon.
A few relevant points:

- Conroe will hardly get slower and it will be released at even higher frequencies than 2.66GHz

- Opteron and Athlon64 are identical. Where does all these fantasies about Opteron being something special come from? There is no Opteron that beats the Athlon64 FX-60. There are however Opterons that perform the same and those are the 185, 285 and 885.

The test therefore compares AMD's absolutely fastest CPU, overclocked, to an upcoming Intel CPU in the middle of the field. Sure AMD will have a faster CPU by then, but it's just going to be the 2.8GHz FX-62 and the test just showed us how that might turn out for AMD...

This was obvious since day one, and the main reason behind Apple's choice...even though AMD chips were seemingly "faster" in the beginning, the company has no scale to compete with Intel; it just got an edge because of its much more streamlined and less bureaucratic structure...

Just remember the Mac clone industry...they had far fewer computers to manufacture and could have quicker access to the latest technology, unlike Apple.

And now that Intel's R&D is completed, AMD will surely suffer a downward spiral run. It is gonna be in the beleaguered league pretty soon, trust me...AMD is history.
I hope you're being sarcastic. AMD's answer to Conroe seems to be scheduled for a mid 2007 launch. This includes an all new mobile architecture.

Take those benchmarks with HUGE grain of salt. As you mentioned, the test-systems were provided by Intel. There are TONS of things they could do to cripple the AMD-system. They can say that "this has an overclocked A64-CPU in it", but there's simply no knowing how they systems were REALLY configured. Even the apps they were benchmarking were provided by Intel, so they could have simply picked and chosen the benchmarks where they will be faster than the AMD-system.

And it's worth noting that the AMD-CPU they were comparing to is available today (although they did overclock it), whereas Conroe is not. So they are basically telling people that their FUTURE CPU will be faster than their competitors CURRENT CPU. nBy the time Conroe ships, AMD will have new Athlons available as well.
Well, AMD's only response to Conroe at launch will be the rev F CPUs. These are said to be 5-10% faster than current rev E CPUs at most. Also, Conroe will hardly become slower as the launch date draws closer. Not that it's likely to get much faster either.

Not impressed by Intel... Same garbage X86 chips with an inferior design...
Agreed. It's hard to be impressed by a chip that seriously beats the G5 clock for clock, while reaching much higher frequencies. Garbage is what it is. :rolleyes:

Evangelion

Mar 8, 2006, 06:53 AM

- Opteron and Athlon64 are identical.

No they are not. They are built on different tolerances, Opterons support registered RAM and the 2xx and 8xx-series have additional HyperTransport-links. The difference between 1xx-series and A64 is a bit vague though.

The test therefore compares AMD's absolutely fastest CPU, overclocked, to an upcoming Intel CPU in the middle of the field

What makes you think the Conroe we saw was "Middle of the field"? It was propably the fastest Conroe Intel could come up with at this point. There might be faster Conroes available at launch, but there will propably be faster-running A64's available as well. Instead of doing relatively modest 200Mhz clock-speed increases, AMD could bumb the clock-speeds up by a considerable amount.

Well, AMD's only response to Conroe at launch will be the rev F CPUs. These are said to be 5-10% faster than current rev E CPUs at most. Also, Conroe will hardly become slower as the launch date draws closer. Not that it's likely to get much faster either.

Well, AMD will be having new socket with improved mem-bandwidth (double the bandwidth). And I have heard that there are revisions in the works that have additional FP-unit and more cache than the current CPU's do.

janstett

Mar 8, 2006, 07:00 AM

And now that Intel's R&D is completed, AMD will surely suffer a downward spiral run. It is gonna be in the beleaguered league pretty soon, trust me...AMD is history.

I wonder if people are forgetting AMD's history. Back in the early 90s a number of companies started making X86 clones. AMD, Cyrix, IBM, TransMeta, and a few others I am forgetting.

AMD is the only survivor, but they have an up-and-down history versus Intel. Several times they have been on the brink and manage to save themselves by one-upping Intel. My first AMD was a K6-2 laptop. Then they were almost crushed under the Pentium 4 express only to come back with the Athlon. Once again they were pushed to the edge and came up with their 64-bit chips.

However, I agree that they are in trouble. There are only so many times they can have their backs to the wall and pull a rabbit out of their hats, to mix metaphors.

Hattig

Mar 8, 2006, 07:02 AM

Good stuff, for Apple.

Not so good for AMD, although inertia in the marketplace, and possibly sensible pricing, will keep them safe for a while after these Intel chips launch, maybe until they get K8L out of the door in 2007.

Also high-speed dual-core benchmarks with memory intensive applications are going to choke on the old dual-channel DDR400 memory controllers on S939. The move to DDR2 in Socket AM2 will help those applications a lot.

But Conroe will probably still be faster than the fastest AMD processor.

I think AMD got complacent and expected another Intel farce.

Evangelion

Mar 8, 2006, 07:28 AM

AMD is the only survivor, but they have an up-and-down history versus Intel. Several times they have been on the brink and manage to save themselves by one-upping Intel. My first AMD was a K6-2 laptop. Then they were almost crushed under the Pentium 4 express only to come back with the Athlon. Once again they were pushed to the edge and came up with their 64-bit chips.

Huh? After K6-2 AMD introduced K6-III and Athlon. Athlon competed in the beginning mostly against Pentium III, not Pentium 4 (what is P4 Express BTW?). And Intel had serious problems competing with the Athlon, even with the P4 (which sucked in the beginning). Wikipedia sez that Athlon was the fastest x86-CPU from august 1999 to january 2002. And even after that, Athlon was competetive, although not necessarily the fastest. And in April 2003, AMD introduced the Opteron.

However, I agree that they are in trouble. There are only so many times they can have their backs to the wall and pull a rabbit out of their hats, to mix metaphors.

Well, as things are right now, AMD has never been better. Their sales are going up, their market-share is going up (in all segments, servers, laptops and desktops), they have solid technology and roadmap, superior multicore-tech when compared to Intel and they have the best interconnect-tech in the industry. What might be happening now is that instead of seeing AMD humiliate Intel in just about every benchmark since the launch of the Opteron, things are getting more equal. I wouldn't say that Intel getting more competetive with AMD means that AMD is doomed. AMD has just been doing exceptionally well for the past several years.

nataku

Mar 8, 2006, 07:30 AM

You know what nataku?

...You're right. :rolleyes: It really is stupid to play waiting game!

I'm just waiting untill April 1st just to check the MacBook isn't anything to change my mind, then I'll make a purchase. Thanks again mate :D

Glad to help:)

BRLawyer

Mar 8, 2006, 07:36 AM

I hope you're being sarcastic. AMD's answer to Conroe seems to be scheduled for a mid 2007 launch. This includes an all new mobile architecture.

Hejsan, Mikael, hur är det? ;)

Nope, I am not being sarcastic...you may have been fooled by the apparently "bad" results by Intel for this last quarter, or by AMD's marketshare increase in the server market...the fact of the matter is: when AMD launches its "answer" to Conroe and the other Intel chips it's gonna be already too late for them...

Right now AMD is making some money exactly because of this performance/watt gap that Intel has...unfortunately for AMD, the gap is closing sooner than they had realized...therefore, AMD will NOT have the same performance advantage by 2007, let alone the market power...trust me, AMD is going the way of the dodo in no time...and its previous balance sheets show that the company is not really a shining star in the business either...

AidenShaw

Mar 8, 2006, 07:41 AM

Hmmm... is it worth waiting out for a Merom iMac? Personally, I don't think the 20% performance increase warrants a wait of 6 months, although waiting for a 64-bit chip could be very beneficial...

I'm obviously talking here of people who are planning on buying an iMac now, yet have seen Conroe/Merom could be out in half a year... :confused:
If you can wait half a year, wait. If you need a faster Apple now, buy it.

The only risks to the 32-bit system are:

Resale value of the 32-bit systems will fall much faster than has been traditional for PPC Apples, so if you plan to eBay the Yonah to get a Merom you might get sticker shock
Sooner or later there will be some application or device or O/S (Leopard + 1?) that has something that is only available with a 64-bit CPU.

A 32-bit MacIntel will probably be good for 18 months or so before the second situation occurs - so ask yourself if about $100/month for a computer is OK. If yes, then buy it and any resale value is simply a bonus.

Mikael

Mar 8, 2006, 08:14 AM

No they are not. They are built on different tolerances,
That's something you've made up yourself. AMD may choose to use the best chips from each wafer for Opteron products, but so far I've seen zero evidence that they're manufactured on separate lines.

Opterons support registered RAM and the 2xx and 8xx-series have additional HyperTransport-links.
Those are features that are most likely enabled and disabled after manufacturing of the chips.

What I was trying to say is that Opteron and Athlon64 are identical from an architechtural standpoint. Sure, there might be a few "shallow" differences in features depending on what market they're aimed at, but check under the hood and the architecture is the same.

The real point of the response was however to explain that there are no performance differences between Opteron and Athlon64. The registered memory for the S940 platform might slow down those Opterons a percent or so, so Opteron certainly isn't any faster.

I just hate this kind of hairsplitting...

What makes you think the Conroe we saw was "Middle of the field"? It was propably the fastest Conroe Intel could come up with at this point. There might be faster Conroes available at launch, but there will propably be faster-running A64's available as well. Instead of doing relatively modest 200Mhz clock-speed increases, AMD could bumb the clock-speeds up by a considerable amount.
Well, unless something goes terribly wrong, Intel is set to release Conroe CPUs at 2.93GHz and probably an even faster EE edition at launch. AMD might have something faster by then, but there are a few things speaking against that. First of all they will have just released their 2.8GHz FX-62 CPU. Second, 3GHz will certainly be pushing it on their 90nm process. It was my understanding that Conroe was supposed to be launched this summer, which puts it very close to AMD's AM2 launch and the FX-62.

Exactly how do you think AMD would be able to increase clock frequency that much on their 90nm process? If these initial performance tests prove to be true, then AMD will need a 3.4GHz X2 just to match the 2.66GHz Conroe and that's certainly not something AMD could accomplish on 90nm.

Well, AMD will be having new socket with improved mem-bandwidth (double the bandwidth). And I have heard that there are revisions in the works that have additional FP-unit and more cache than the current CPU's do.
Yes, I already mentioned AMD's next-gen/updated CPUs in my last post. As I said in that post, they will not be here when Conroe arrives. They will probably not surface until H2 2007, which gives Intel a good amount of time to reign supreme. I also mentioned the new rev F CPUs in my last post. Yes, they will have more bandwidth, but even AMD told us to expect at most a 10% increase in performance. This is likely to be in very memory intensive applications. There doesn't seem to be much in the way of core enhancements in rev F, so computational performance in non memory intensive apps looks to remain very similar.

Hejsan, Mikael, hur är det?
Hej på dig! Det är bara bra. :)

Nope, I am not being sarcastic...you may have been fooled by the apparently "bad" results by Intel for this last quarter, or by AMD's marketshare increase in the server market...the fact of the matter is: when AMD launches its "answer" to Conroe and the other Intel chips it's gonna be already too late for them...

Right now AMD is making some money exactly because of this performance/watt gap that Intel has...unfortunately for AMD, the gap is closing sooner than they had realized...therefore, AMD will NOT have the same performance advantage by 2007, let alone the market power...trust me, AMD is going the way of the dodo in no time...and its previous balance sheets show that the company is not really a shining star in the business either...
No, I'm not drawing any conclusions based on recent market share improvements. I'm just curious as to how anyone can think that a 37 year old company is going to lay down and die, simply because of a normal shift in performance leadership based on a new product generation. This is how it's always been and I honestly fail to see how it's all that different this time.

Evangelion

Mar 8, 2006, 08:32 AM

That's something you've made up yourself.

Um, no, I didn't

AMD may choose to use the best chips from each wafer for Opteron products, but so far I've seen zero evidence that they're manufactured on separate lines.

No, they are not made on separate lines, and I never claimed anything of the sort. They do have different packaging, IIRC (ceramic vs. organic).

Those are features that are most likely enabled and disabled after manufacturing of the chips.

So? Fact remains that that is a feature that Opterons have, while A64 do not have.

What I was trying to say is that Opteron and Athlon64 are identical from an architechtural standpoint. Sure, there might be a few "shallow" differences in features depending on what market they're aimed at, but check under the hood and the architecture is the same.

No **** Sherlock! I heard that Xeon and P4 are also 98% identical! And in any case, your claim that "Opteron and A64 are identical" is plain FALSE.

The real point of the response was however to explain that there are no performance differences between Opteron and Athlon64. The registered memory for the S940 platform might slow down those Opterons a percent or so, so Opteron certainly isn't any faster.

And if you need registered RAM, that difference does not matter. And if you need multi-CPU system, A64 does not cut it.

Well, unless something goes terribly wrong, Intel is set to release Conroe CPUs at 2.93GHz and probably an even faster EE edition at launch. AMD might have something faster by then, but there are a few things speaking against that. First of all they will have just released their 2.8GHz FX-62 CPU. Second, 3GHz will certainly be pushing it on their 90nm process.

You know that for a fact?

It was my understanding that Conroe was supposed to be launched this summer, which puts it very close to AMD's AM2 launch and the FX-62.

Conroe should be launched this summer, whereas AM2 should be out in Q2. So that propably means that AM2 will be here first. The new socket/RAM propably gives around 10% speed-increase, and AMD could get another 10-15% from tweaking the core. Add to that increased clock-speeds (lets say 200-400Mhz), and you have a solid competitor right there. Even if the disregard any tweaks to the core, it would still be competetive. It wouldn't wipe the floor with Intel like it does today though.

And let's not forget that we have so far seen exactly ONE benchmark of Conroe, and it was supplied by Intel.

Exactly how do you think AMD would be able to increase clock frequency that much on their 90nm process? If these initial performance tests prove to be true, then AMD will need a 3.4GHz X2 just to match the 2.66GHz Conroe and that's certainly not something AMD could accomplish on 90nm.

And that's a mighty big "if".... They tested systems that were provided by Intel. The apps they benchmarked were provided by Intel. I would take those benchmarks with huge grain of salt.

Mikael

Mar 8, 2006, 09:06 AM

No, they are not made on separate lines, and I never claimed anything of the sort. They do have different packaging, IIRC (ceramic vs. organic).
I already said that I meant that the cores are the same and that both cores perform identical. We are talking about different things. I understand that, why don't you? Yes, I may have been a bit "fuzzy" with the details in my initial post, but the point still holds: Opteron doesn't perform better than Athlon64. Can't we just leave it at that?

BTW, I thought that all Opterons and Athlon64 were in organic packages these days. I know that the first Opterons were in cermaic packages, though. My bad if I was wrong about this.

You know that for a fact?
What part? That Intel has a 2.93GHz Conroe planned for launch? Well, that's what I've seen on roadmaps posted.

Or did you mean that 3GHz would be pushing it? 3GHz could certainly be possible (although hot running), but the important part is that far more than 3GHz may be required.

Conroe should be launched this summer, whereas AM2 should be out in Q2. So that propably means that AM2 will be here first. The new socket/RAM propably gives around 10% speed-increase, and AMD could get another 10-15% from tweaking the core. Add to that increased clock-speeds (lets say 200-400Mhz), and you have a solid competitor right there. Even if the disregard any tweaks to the core, it would still be competetive. It wouldn't wipe the floor with Intel like it does today though.
Unless AMD is holding some information back regarding rev F, there will not be a 10-15% increase in performance because of core enhancements. They said that rev F/AM2 would bring up to 10% increased performance, that's it.

Current rev F samples also seem to lack any core enhancements and those should be in place this close to launch. Everything at the moment points towards the extra bandwidth providing the extra performance. That means that many apps won't see a speed-up at all.

And let's not forget that we have so far seen exactly ONE benchmark of Conroe, and it was supplied by Intel.
Yes, I know. This whole discussion is based on the assumption that the benchmarks are somewhat correct.

And that's a mighty big "if".... They tested systems that were provided by Intel. The apps they benchmarked were provided by Intel. I would take those benchmarks with huge grain of salt.
As I said, I'm assuming that they're correct for this discussion. If they aren't, then congrats to AMD. The real difference might very well prove much smaller when we get some actual reviews.

EDIT: And don't get me wrong. I have used AMD chips the past six years and been very happy with them. I want AMD to continue to be strong, but at the same time I don't want the stagnant CPU market we've seen the past few years. Best for everybody (well, except maybe for Intel) would be if Conroe turns out excellent, but AMD manages to get out a very competitive chip very fast to counter Conroe.

jared_kipe

Mar 8, 2006, 10:01 AM

EDIT: And don't get me wrong. I have used AMD chips the past six years and been very happy with them. I want AMD to continue to be strong, but at the same time I don't want the stagnant CPU market we've seen the past few years. Best for everybody (well, except maybe for Intel) would be if Conroe turns out excellent, but AMD manages to get out a very competitive chip very fast to counter Conroe.
Thank you. You manage to keep it civil, in the face of such comments. I love it when the arguing people save it so it doesn't escalate.

ksz

Mar 8, 2006, 10:53 AM

If you can wait half a year, wait. If you need a faster Apple now, buy it.

The only risks to the 32-bit system are:

Resale value of the 32-bit systems will fall much faster than has been traditional for PPC Apples, so if you plan to eBay the Yonah to get a Merom you might get sticker shock
Sooner or later there will be some application or device or O/S (Leopard + 1?) that has something that is only available with a 64-bit CPU.

A 32-bit MacIntel will probably be good for 18 months or so before the second situation occurs - so ask yourself if about $100/month for a computer is OK. If yes, then buy it and any resale value is simply a bonus.
I think some intriguing questions are:

1. Does Yonah already have 64-bit extensions? Someone raised this earlier, but I haven't seen a clear yes or no.

2. The new Mac mini and iMac have socketed (not soldered) CPUs. It's already been demonstrated that the processors can be replaced in the field. What is not clear, however, is whether we'll be able to replace Yonah with its pin-for-pin compatible successor, Merom, without needing to replace the chipset.

If (2) is doable, then (1) is a moot point. Just buy an Intel iMac or Mac mini today and replace the processor at the end of the year when retail prices are reduced.

jabooth

Mar 8, 2006, 11:38 AM

If option 2 was available, that would be perfect!

Still, do people genuinely think you could find 64bit exclusive software in just 18 months?

As I understood it, 64bit excels in some areas, yet is pointless (and possibly even slower) in others. Is this correct? If so, why would companies limit market by making software 64 bit exclusive?

Is their anyone who has a good understanding of 64bit who can give us all a good lesson into its +/- and how soon it will be implemented in software? :confused:

Cheers

itguy06

Mar 8, 2006, 11:41 AM

Well, unless something goes terribly wrong, Intel is set to release Conroe CPUs at 2.93GHz and probably an even faster EE edition at launch. AMD might have something faster by then, but there are a few things speaking against that. First of all they will have just released their 2.8GHz FX-62 CPU. Second, 3GHz will certainly be pushing it on their 90nm process. It was my understanding that Conroe was supposed to be launched this summer, which puts it very close to AMD's AM2 launch and the FX-62.

Let's just say this:
Intel has a long history of paper launches (P4 EE and P3 1.x Ghz), so if they get it out in time, expect long lines to get one. And don't think Apple will get them before anyone else. Do you really think Intel is going to PO Dell and HP, who make up almost half of the PC market by giving Apple a heads up on their fastest chips?

Exactly how do you think AMD would be able to increase clock frequency that much on their 90nm process? If these initial performance tests prove to be true, then AMD will need a 3.4GHz X2 just to match the 2.66GHz Conroe and that's certainly not something AMD could accomplish on 90nm.

Or they could increase the cache or do some other tweaks to get the CPU to where it needs to be.

Look at it this way:

This is Intel saying our future chips will be faster than our competitor's current chips. Well, DUH, of couse they will be faster. Intel isn't going to release another chip like the P4 where it is noticeably slower than anything else on the market.

they will have more bandwidth, but even AMD told us to expect at most a 10% increase in performance.

So, if Intel is claiming 20% speed increase (didn't look like that in Anand's tests, but I didn't bring out the calculator) than AMD's current chips, and AMD's new chips will be 10% faster, the gap is only 10%. IOW, you won't notice it.

No, I'm not drawing any conclusions based on recent market share improvements. I'm just curious as to how anyone can think that a 37 year old company is going to lay down and die, simply because of a normal shift in performance leadership based on a new product generation. This is how it's always been and I honestly fail to see how it's all that different this time.

And a 36 year old company won't either. AMD is not some newcomer to the game. They have been around just as long as Intel and have a product portfolio that is almost as large. They have even produced CPU's for Intel when they could not meet demand (8088, 286 and 386, IIRC).

You also fail to explain away the increased sales AMD has been having for a long time now. People with new computers are not going to go out and buy a new computer this year because it's 10% faster than what they have. They'll buy a new one in 2-3 years.

We'll have to wait and see what AMD comes up with. I have a feeling it will be very competitive. Maybe not the fastest - it's about time Intel has something worth buying, but AMD is not going anywhere.

I just wish Apple would not have went the Intel route.... Doesn't make sense from so many angles.

AidenShaw

Mar 8, 2006, 11:58 AM

What I don't understand is way you'd want them in a Power Mac replacement If you want a Power Mac replacement you can make a nice one today. No waiting. Simply build one of these
http://tinyurl.com/zsaxg
but put it in a pretty box and load Mac OSX on it
But OSx86 won't load on an Opteron, and Apple is very unlikely to make that possible (TPM, for example).

It that's not enough Sun has four chip. eight core Opertoon boxes that have been sellig for some time now.
So does HP, and several other vendors. Sun is one of the least interesting AMD vendors - they're focussed on trying to get the mob of people abandoning SPARC to chose Sun's Intel architecture machines instead of any others.

Sun is also on very dangerous ground with their hyping of the power consumption advantages of the Sun boxes. Not only are they basically dishonest, but dual-core Xeons and new architecture chips turn the tables on them.

I just don't see Intel catching up to AMD at the high end.
Look closer at the news from IDF...

itguy06

Mar 8, 2006, 12:21 PM

But OSx86 won't load on an Opteron, and Apple is very unlikely to make that possible (TPM, for example).
I thought someone already did. Apparently all it needs is SSE3, which the Opterons and X64's have.

Mord

Mar 8, 2006, 12:29 PM

But OSx86 won't load on an Opteron, and Apple is very unlikely to make that possible (TPM, for example)..

no reason why it cant, new opterons support sse3.

lexfuzo

Mar 8, 2006, 12:42 PM

AMD is going the way of the dodo...
Well, it won't. Intel needs AMD as a counterpart on the market. Anti-trust anyone? As long as AMD isn't a serious threat (and at 20% vs. 75+% isn't) they will let it live.

Mord

Mar 8, 2006, 12:49 PM

Well, it won't. Intel needs AMD as a counterpart on the market. Anti-trust anyone? As long as AMD isn't a serious threat (and at 20% vs. 75+% isn't) they will let it live.

AMD and intel have some intresting agreements, if intel wanted to they could legally start makeing athlon 64's just as AMD can make P4's if they wanted to.

tutubibi

Mar 8, 2006, 02:31 PM

Interesting to see so many of you celebrating Intel and questioning even AMD survival. What's with you people? Looks like everybody was pulled in Steve's distortion field :D

First of all, Intel is just closing a gap and certainly not overtaking AMD. For more realistic view, please check http://arstechnica.com/news.ars/post/20060307-6334.html

backdraft

Mar 8, 2006, 02:35 PM

Agreed. It's hard to be impressed by a chip that seriously beats the G5 clock for clock, while reaching much higher frequencies. Garbage is what it is.

INTEL Duo does not beat the single core G5 clock for clock, hell the Intel duo has trouble vs the single core G4. Not to mention the G4 & G5 are at 90nm... Look at the floating point performance the PowerPC wins period.

I would rather see the Dual Core G4 in the Powerbooks vs INTEL Duo...

Higher frequency does equate to better performance...

http://macdailynews.com/index.php/weblog/comments/8271/

INTEL chips are not better than PowerPC chips, Apple is forcing them upon Mac users and everyone is just jumping on the bandwagon blindly. If Apple "must" use INTEL chips they should leave rinky dink X86 chips for their consumer lines (iMac, iBook, Mac mini) and PowerPC chips for the pro's (Powerbook and PowerMac)

ksz

Mar 8, 2006, 03:01 PM

I think some intriguing questions are:

1. Does Yonah already have 64-bit extensions? Someone raised this earlier, but I haven't seen a clear yes or no.

2. The new Mac mini and iMac have socketed (not soldered) CPUs. It's already been demonstrated that the processors can be replaced in the field. What is not clear, however, is whether we'll be able to replace Yonah with its pin-for-pin compatible successor, Merom, without needing to replace the chipset.

If (2) is doable, then (1) is a moot point. Just buy an Intel iMac or Mac mini today and replace the processor at the end of the year when retail prices are reduced.
On point 2, here's very encouraging information from AnandTech (http://www.anandtech.com/tradeshows/showdoc.aspx?i=2715):

"Merom will make its debut in the second half of this year on a platform Intel is referring to as the Napa Refresh. The new platform is essentially Napa, but with a Merom processor instead of Yonah. The other beauty of Merom is that it is 100% compatible with existing Yonah designs, meaning that all Core Duo notebooks today should be able to accept a Merom processor with at most a BIOS update."

Sadly, MacBook Pros have soldered CPUs which (in practical terms) rules this out, but new Mac mini and iMac owners have something to cheer about!

Mikael

Mar 8, 2006, 03:06 PM

Do you really think Intel is going to PO Dell and HP, who make up almost half of the PC market by giving Apple a heads up on their fastest chips?
Where did I say that? I'll buy a PC with Conroe this fall. Don't care when it ends up in a Mac. I'll look at an Intel based MacBook Pro the next time I buy a laptop, but that's pretty far into the future.

Look at it this way:

This is Intel saying our future chips will be faster than our competitor's current chips. Well, DUH, of couse they will be faster. Intel isn't going to release another chip like the P4 where it is noticeably slower than anything else on the market.
Well, I haven't disputed that fact, have I?

So, if Intel is claiming 20% speed increase (didn't look like that in Anand's tests, but I didn't bring out the calculator) than AMD's current chips, and AMD's new chips will be 10% faster, the gap is only 10%. IOW, you won't notice it.
You're right, Conroe was shown to be faster than 20% in several cases, despite some 5% lower clock frequency. Even if Conroe's advantage turns out to be just 10% and a lower power consumption, it will still be enough to get back most of those lost marketshares. Well, unless AMD dumps their prices like in the good old AthlonXP days. We all know how good that was for business, though...

Conroe performance, especially in games, does look a bit too good, though. There's not much doubt that it will be faster than the A64 clock for clock, but a few of those gaming test do look a little too good... :o If they are true, the new Core microarch really is quite an accomplishment, though.

You also fail to explain away the increased sales AMD has been having for a long time now. People with new computers are not going to go out and buy a new computer this year because it's 10% faster than what they have. They'll buy a new one in 2-3 years.

We'll have to wait and see what AMD comes up with. I have a feeling it will be very competitive. Maybe not the fastest - it's about time Intel has something worth buying, but AMD is not going anywhere.
I'm not the one that said AMD was going anywhere. I said the opposite. Seems you got us mixed up or something...

INTEL Duo does not beat the single core G5 clock for clock, hell the Intel duo has trouble vs the single core G4. Not to mention the G4 & G5 are at 90nm... Look at the floating point performance the PowerPC wins period.
Yes, the G5 is likely faster at floating point. It is however probably slower on integer math. It's also worth noting that floating point performance without AltiVec isn't all that impressive. From the few tests I've seen, it can't really match an Athlon64 clock-for-clock.

The thing is that we're not talking about Yonah here. Conroe is a pretty different design compared to Yonah. My guess is that G5 isn't going to seem very impressive when Conroe arrives...

I would rather see the Dual Core G4 in the Powerbooks vs INTEL Duo...
Why? Barefeats just tested a 2GHz G4 versus a 2GHz Core Solo and although there were only a few tests that were totally CPU dependant, the G4 didn't look very nice in comparison. This is not the first test showing that the G4 generally performs quite a bit worse than the Core arch.

What's with this cult around old PPC chips anyway? The G4 is still a pretty nice CPU, but just like a Pentium III or AthlonXP it does show its ages these days.

dragula53

Mar 8, 2006, 03:07 PM

INTEL Duo does not beat the single core G5
.
.
(iMac, iBook, Mac mini) and PowerPC chips for the pro's (Powerbook and PowerMac)

You don't know what you are talking about.

Back when the G5 actually showed up on spec.org it got trounced by the Pentium-M (Dothan). Something like 1300 vs 900 integer and 700 vs 900 (the other way) for floating point. Yonah is 2 generations newer than Dothan, is faster, has more cache, and also has 2 cores.

The Single G5 has NEVER been faster than a comparable pentium 4. It may have squeaked by in floating point, but the integer gap has always been humongous.

In all actuality, even the G3 which supposedly trounced the equivalent pentium 2's, did not. Read some old arstechnica articles, back before they were pitching softball to apple.

The G5 brought apple close to parity with Intel/AMD. And I think there is considerable doubt that PowerPC was ever in the lead.

That being said, faster is always better, and I have a G5 on my desk, because I like their hardware and operating system. And a dual processor G5 is nothing to sneeze at.

I am still up in the air about buying a new intel mac pro/macbook pro, etc. not because I doubt the viability of the chips, but because I don't see enough progress on the software side of things yet-- I'm waiting to pull the trigger.

Regards

Anonymous Freak

Mar 8, 2006, 03:36 PM

What's with this cult around old PPC chips anyway? The G4 is still a pretty nice CPU, but just like a Pentium III or AthlonXP it does show its ages these days.

It stems from the fact that on PPC's introduction, and even through the G5, it is a vastly superior architecture, it's just that the implemented chips haven't lived up to the hype.

Yes, A 64-bit PPC chip with AltiVec is a kick-ass product. Yes, if operated at equal clock frequencies to a NetBurst-core Intel, it would blow the Intel away. The problem has been Mot/Freescale and IBM's manufacturing. The day the G5 came out, it was the best processor with the best support chipset available. Unfortunately, Intel's manufacturing dominance won. Intel has scaled their clock speeds significantly higher than IBM did.

While Intel has hit a wall, they realized that switching to a more efficient core would gain them processing power improvements even at lower clock speeds. The 'Core' architecture is at least an equal to the G5, clock-speed-for-clock-speed. Yes, the G5 may win in a few minor artificial tests, just as the Core appears to win by huge margins in a few minor artificial tests.

Unless IBM pulls a rabbit out of their hat, and releases a 45nm process dual-core G5 at 3.5+ GHz, the G5 doesn't stand a chance against Core.

This actually makes me sad. PPC's big selling point early on was that it was a much simpler architecture that could then scale to obscenely fast clock rates compared to the more complex x86. Unfortunately, this hasn't been the case. And it's mostly due to manufacturing issues. If Intel was to take the design for the PPC970MP, and re-jigger it for their manufacturing, they could probably get it to fly at 3.5+GHz. I would actually have rather seen that than a switch to x86.

I'm not saying Intel is perfect... Just look at Itanium. Another architecture that looked great on paper, but thoroughly failed in the market. And in the interest of disclosure, once upon a time I was an Intel employee; a technician/engineer in the server dept. I had no impact on the design of processors or chipsets, and our department even released product with competitor's chipsets, eschewing Intel's own chipsets.

Anonymous Freak

Mar 8, 2006, 03:40 PM

1. Does Yonah already have 64-bit extensions? Someone raised this earlier, but I haven't seen a clear yes or no.

No.

Hell no.

This was indeed clearly answered. The rumor came from a misreading of some Intel specs. That is that Yonah is the base for a server chip codenamed 'Sossaman' that will be sold under the label 'Xeon LV' and aimed for high-density blade servers, so it needs to be high-processing-power, low-electrical/thermal-power. Someone got confused and assumed that since it would be a 'Xeon', that it would be 64-bit; and then that got re-told and re-told until someone came to the conclusion that Yonah was 64-bit.

The reality is that Sossaman is 32-bit, just like Yonah.

There are three kinds of lies: Lies, damned lies, and rumors.

KNGPaul

Mar 8, 2006, 04:11 PM

For those who had questions regarding Merom working on Yonah machines

From Anandtech (http://www.anandtech.com/tradeshows/showdoc.aspx?i=2715)

Merom will make its debut in the second half of this year on a platform Intel is referring to as the Napa Refresh. The new platform is essentially Napa, but with a Merom processor instead of Yonah. The other beauty of Merom is that it is 100% compatible with existing Yonah designs, meaning that all Core Duo notebooks today should be able to accept a Merom processor with at most a BIOS update.

Bonte

Mar 8, 2006, 04:20 PM

INTEL chips are not better than PowerPC chips, Apple is forcing them upon Mac users and everyone is just jumping on the bandwagon blindly. If Apple "must" use INTEL chips they should leave rinky dink X86 chips for their consumer lines (iMac, iBook, Mac mini) and PowerPC chips for the pro's (Powerbook and PowerMac)

Apple must use Intel if they want to sell a Windows compatible machine, it can easily create a 2 digit market-share with it and a stock happy company like Apple can't drop that profit for a few Macheads like us. x86 and Windows is just the only possible way to keep growing because trying to make everybody switch clearly didn't work. As long as Apple keeps developing osX i'm happy, Windows compatibility is just an extra bonus for who needs it and lots a lots of cash for Apple. i'm keeping my stock firmly in hand :)

If PPC becomes promising i'm certain Apple somehow will keep using it, they have a choice now. Having choices is a good thing and Apple is in a unique position in that regard.

Mikael

Mar 8, 2006, 04:22 PM

If Intel was to take the design for the PPC970MP, and re-jigger it for their manufacturing, they could probably get it to fly at 3.5+GHz.
That is most likely not the case. The reason that Intel manages to get their Netburst chips to those high frequencies is mostly because of the very long pipeline. I don't know how familiar you are with pipelining, but, basically, the longer the pipeline you have the shorter the signal delay between each stage and the higher you can go with the fequency. Just look at the differences in max frequency between Yonah and Dothan compared to the P4 and Pentium-D. The former use a pipeline length similar to the 970 and they are also able to run at about the same frequencies.

Anonymous Freak

Mar 8, 2006, 05:22 PM

That is most likely not the case. The reason that Intel manages to get their Netburst chips to those high frequencies is mostly because of the very long pipeline. I don't know how familiar you are with pipelining, but, basically, the longer the pipeline you have the shorter the signal delay between each stage and the higher you can go with the fequency. Just look at the differences in max frequency between Yonah and Dothan compared to the P4 and Pentium-D. The former use a pipeline length similar to the 970 and they are also able to run at about the same frequencies.

I am familiar with the problems with pipelines.

I was referring mostly to the fact that Intel is indeed good at getting max frequency out of a given design, and the fact that they now have a production 65nm process.

ksz

Mar 8, 2006, 05:33 PM

No.

Hell no.

This was indeed clearly answered.
Thank you, but it's really a moot point.

shawnce

Mar 8, 2006, 06:44 PM

INTEL Duo does not beat the single core G5 clock for clock, hell the Intel duo has trouble vs the single core G4. Not to mention the G4 & G5 are at 90nm... Look at the floating point performance the PowerPC wins period.

http://macdailynews.com/index.php/weblog/comments/8271/

INTEL chips are not better than PowerPC chips, Apple is forcing them upon Mac users and everyone is just jumping on the bandwagon blindly. If Apple "must" use INTEL chips they should leave rinky dink X86 chips for their consumer lines (iMac, iBook, Mac mini) and PowerPC chips for the pro's (Powerbook and PowerMac)

Rather poorly written article MacDailyNews with bad interpretation of information (except for pointing out when Apple made various claims they did make sense at the _time_ they made them).

In a nut shell and at a high level...

- A single core in an Intel Core Duo will usually outperform a single core G5 clock for clock when doing integer scaler operations.

- A single core in an Intel Core Duo will usually be about the same as a single core G5 clock for clock when doing floating point scaler operations.

- A single core in an Intel Core Duo will usually underperform a single core G5 clock for clock when doing integer vector operations.

- A single core in an Intel Core Duo will usually underperform a single core G5 clock for clock when doing floating point vector operations.

- A single core in an Intel Core Duo will usually outperform a G4 when doing any operation.

Then looking at some real-world examples...

MacBook Pro 2.0GHz versus PowerBook 2.0GHz (!) (http://www.barefeats.com//pbcd.html)
iMac Intel Core Duo 2.0GHz versus Others (http://www.barefeats.com//imcd.html)
Intel Mac mini Core Duo versus PPC Mac mini (http://www.barefeats.com//mincd.html)
MacBook Pro Performance Analysis (http://www.craigtheguru.com/reports/MacBook_Pro_Performance_Analysis.php)
iMac 17" Core Duo (http://arstechnica.com/reviews/hardware/imac-coreduo.ars/5)

The merom & conroe look to have improvements to SSE which will improve vector abilities in comparison to AltiVec.

/me a PowerPC ISA lover

DeathChill

Mar 8, 2006, 08:46 PM

The problem with all these arguments for PowerPC is that they all rely on the "If this was done then PowerPC would be faster." The chips don't exist and currently Intel's x86 chips are handing PowerPC chips their...behind.

I'm not really for one side or the other, as I used to believe PowerPC chips were faster (ala Apple benchmarks, then I learned how fabricated those are) so I wondered why Apple was switching to Intel. However, even the EXTREMELY inefficient Pentium-4 line competes with the G5's, which is extremely sad.

Anyway, Conroe is looking to make sure that the G5's don't even stand a chance. Comparing the Core Duo and Core Solo to a G5 is also sad, simply because these chips are LAPTOP chips and aren't meant to be powerhouses. When a laptop chip is beating a desktop chip in any area you know something's not right.

DavidCar

Mar 8, 2006, 10:58 PM

I found a diagram for a dual Woodcrest workstation. Scroll down the page until you see Glidewell Workstation Platform. This is different from the Woodcrest server platform in that it includes PCI Express. It also looks like it will accept the next generation Woodcrest chips in 2007, which I assume means quad core. By that I mean socketed replaceable chips would be a possibility. Don't take this to mean I actually know anything.

http://www.2cpu.com/review.php?id=109&page=3

~Shard~

Mar 8, 2006, 11:06 PM

It also looks like it will accept the next generation Woodcrest chips in 2007, which I assume means quad core.

You mean Kentsfield?

DavidCar

Mar 8, 2006, 11:43 PM

You mean Kentsfield?
I had to go look up a Wikipedia article because I don't have these names memorized. According to that, Kentsfield will be the four core version of Conroe, while the four core version of Woodcrest will be called Cloverton, or Clovertown depending on which article you look at.

So I'm saying that since it appears Intel is developing the Glidewell platform for using two Woodcrest processors in a workstation, it sounds reasonable that it would be the foundation for a new PowerMac, or Mac Pro.

I was trying to find a source the $850 Woodcrest price but couldn't.

ksz

Mar 8, 2006, 11:54 PM

I was trying to find a source the $850 Woodcrest price but couldn't.
See post #80 in this thread.

http://www.theinquirer.net/?article=29510

DavidCar

Mar 9, 2006, 12:07 AM

See post #80 in this thread.

http://www.theinquirer.net/?article=29510
Thanks for the reminder. I remember seeing $850, but forgot it included a link. I agree with your point there that Apple would want the new PowerMac to be priced about the same as the current edition. Checking the link gives me hope for a lower Mhz version at a lower price.

Two 3.0 Ghz Woodcrest chips = $1700
Two 2.0 Ghz Woodcrest chips = $ 660

I could be happy with a 2.0 Ghz Woodcrest "Quad."

From the link:

"The 5160 clocks at 3GHz, the 5150 at 2.66GHz, the 5140 at 2.33GHz, the 5130 at 2GHz, the 5120 at 1.86GHz and the 5110 at 1.6GHz.

The price cashcade when you buy loads of these is $850, $700, $470, $330, $270 and $230."

Multimedia

Mar 9, 2006, 02:52 AM

I think we will see the Merom going into MacBook Pro's almost as soon as they have been released.I agree ScottB. And the 64-bit advantage will make waiting for Dual Core Merom in a 17" MBP logical to all but who must have faster portables NOW. I for one will wait for a Merom 17" MBP before I pull the trigger on my first MacIntel portable. I don't really want a 32-bit processor in a portable when I am driving four 64-bit processors on the desktop already.

Seems too retro when I know 64-bit is just a few more months away. WWDC SteveNote August 7 will tell us a lot about what's really up. I think my Quad will hold its own against dual Conroe or Woodcrest. And I'm thinking that until a pair of Kentsfield Quad Cores are inside a Mac (for a total of 8), I'm not gonna feel too weak to enjoy this platform a few more months - esp with a Universal Quad Enhanced, Oct Ready, Leopard and iLife '07 out in ten months max.

While I enjoy watching this unfoldment, I have to say this year is looking more and more like a wait and see time for me. I'm starting to think Leopard and iLife '07 are going to need to be the bait that will make me want to bite the Intel portable hook after January's MacWorld San Francisco 2007 SteveNote.

That's only 5 months after the August 7-11 WWDC this year - delayed two months this year for at least two reasons:

1. Leopard needs to be almost ready to ship so developers can adapt to it ASAP after WWDC in time for MacWorld SF Leopard enhanced & compatible versions.
2. Merom, Conroe and Woodcrest need to be near shipping dates so Steve can blow everyone's minds with one of the most amazing software and hardware unveilings he's ever made at once. :)

PiTT

Mar 9, 2006, 05:12 AM

Merom will be compatible with the current iMac and MacMini !!!!!!!!

During a separate briefing, Intel's Mooly Eden showed a benchmark pitting a Dell Core Duo system against the same system with a Merom processor (Eden literally swapped out the Core Duo CPU and stuck in a Merom processor, partly to showcase its backwards comptability).
http://www.anandtech.com/tradeshows/showdoc.aspx?i=2715&p=2

Multimedia

Mar 9, 2006, 05:42 AM

Merom will be compatible with the current iMac and MacMini !!!!!!!!
http://www.anandtech.com/tradeshows/showdoc.aspx?i=2715&p=2Have to admit this is pretty amazing. I think I posted earlier that Merom wouldn't work with Yonah chipsets. I was mistaken. Too bad they didn't put a socket inside the MacBook Pro. That really blows for all the MBP buyers. But the iMac and Mac mini buyers are good to go for an easy off the shelf processor upgrade. All they have to do is figure out how to get inside those difficult to open up models. By the looks of that chart though, I am more inclined than ever to wait for the 2007 support chips so I can have faster 802.11n wi-fi and graphics for the long term.

Good find PITT. ;) :cool:

Multimedia

Mar 9, 2006, 06:02 AM

Now to observe what Apple may be aiming at with their first MBP 17" Model:

Toshiba Core Duo 17" Laptop with HD-DVD Drive (http://www.dailytech.com/article.aspx?newsid=1153). The HD-DVD drive can write 320 GB to one HD-DVD platter. WUXGA is 1920 x 1200 :D :eek:

AidenShaw

Mar 9, 2006, 06:35 AM

I don't really want a 32-bit processor in a portable when I am driving four 64-bit processors on the desktop already.
The only initial advantage to Merom will be future-proofing - you'll be "driving" that Merom MBP in 32-bit mode at 32-bit speeds until

Apple comes out with a true 64-bit OSx64 operating system (Apple has to say something about this at WWDC - look for the "Leopard x64" announcements)
Applications are ported to 64-bit for the extra speed

But, since probably all the applications that you run on the quad are 32-bit, "future-proofing" will be good enough.

Intel says that Merom is 20% faster than Yonah. 20% is also a typical figure for the speedup going from 32-bit x86 to 64-bit x64....

I wonder if Intel is comparing 64-bit Merom to 32-bit Yonah - and the speed boost is only seen in x64 mode?

thunng8

Mar 9, 2006, 07:01 AM

You don't know what you are talking about.

Back when the G5 actually showed up on spec.org it got trounced by the Pentium-M (Dothan). Something like 1300 vs 900 integer and 700 vs 900 (the other way) for floating point. Yonah is 2 generations newer than Dothan, is faster, has more cache, and also has 2 cores.

The Single G5 has NEVER been faster than a comparable pentium 4. It may have squeaked by in floating point, but the integer gap has always been humongous.

Not exactly true. We do have a new datapoint with the release of the 2.7Ghz 970MP in the IBM JS21 Blade system.

http://www.spec.org/cpu2000/results/res2006q1/

2.7Ghz 970MP
SpecInt2000:1706
SpecFP2000:2259

2-way 2.7Ghz 970MP
SpecIntRate:38.5
SpecFPRate:43.9

4-way 2.5Ghz 970MP
SpecIntRate:67.5
SpecFPRate:58.8

As a comparison, the 2.16Ghz Core Duo (2-way):
SpecIntRate:34.9
SpecFPRate:27.3

And best current P4 system:

3.46Ghz P4EE (2-way):
SpecIntRate:41.1
SpecFPRate:36.7

However, the most remarkable figures are the Power5+ based p5-575 system (up to 16 core 2U rack system)

8-way 2.2Ghz
SpecIntRate:200
SpecFPRate:382

16-way 1.9Ghz
SpecIntRate:314
SpecFPRate:571

Truly amazing numbers - nothing at the moment comes remotely close to this performance (with same number of cores).

thunng8

Mar 9, 2006, 07:21 AM

That is most likely not the case. The reason that Intel manages to get their Netburst chips to those high frequencies is mostly because of the very long pipeline. I don't know how familiar you are with pipelining, but, basically, the longer the pipeline you have the shorter the signal delay between each stage and the higher you can go with the fequency. Just look at the differences in max frequency between Yonah and Dothan compared to the P4 and Pentium-D. The former use a pipeline length similar to the 970 and they are also able to run at about the same frequencies.

That is generally the case - however IBM has revealed that their next generation POWER6 chip has the same number of pipeline stages the POWER5 but will clock 2X higher.

http://news.zdnet.com/2100-9584_22-6039293.html

"We doubled the frequency but held the pipe depth the same as Power5."

Shipping frequency will be 4-5Ghz and they have the POWER6 chip running at 5.1Ghz in the labs with standard voltages (and up to 6Ghz with higher voltages).

thunng8

Mar 9, 2006, 07:53 AM

People forget that an iMac really is of laptop form factor...

Yes, it is slightly larger than a laptop, (and does incorperate a much bigger screen) but it is still only a few inches deep. The heatsink and fan on my A64 3500 is deeper than the whole iMac!

And its the form factor that makes me want an iMac soooooo bad.... :D

I think Apple will change to Conroe with the iMac when that chip is introduced.

The reason for putting a Core Duo now is that the dual core P4 is way too hot. With Conroe being much cooler, I don't see any more reason on putting a laptop chip in an iMac. The mac mini on the other hand will likley continue with laptop chips because it is indeed very small.

janstett

Mar 9, 2006, 07:59 AM

INTEL Duo does not beat the single core G5 clock for clock, hell the Intel duo has trouble vs the single core G4.

Sigh... It must be some strange anomaly, then, that my MacBook Pro D-E-S-T-R-O-Y-S my G4 Mac Mini by a factor of 5:1 on Handbrake. I mean, the Core Duo has trouble vs. a single core G4, right?

INTEL chips are not better than PowerPC chips, Apple is forcing them upon Mac users and everyone is just jumping on the bandwagon blindly. If Apple "must" use INTEL chips they should leave rinky dink X86 chips for their consumer lines (iMac, iBook, Mac mini) and PowerPC chips for the pro's (Powerbook and PowerMac)

Right, and you know this based on your personal use of an X86 Mac, whereas people like me who actually OWN a PPC and an X86 Mac, well, we're just cwazy.

AidenShaw

Mar 9, 2006, 08:03 AM

http://www.spec.org/cpu2000/results/res2006q1/
Truly amazing numbers - nothing at the moment comes remotely close to this performance (with same number of cores).
Check http://www.spec.org/cpu2000/results/ - you're looking only at the results submitted in the most recent quarter. If you look at "all results", there are Itanium II systems that are close to the POWER 5 numbers.

It would be more interesting to see SPECint/SPECfp (single stream) numbers for all the chips as well.

"SPECxx_rate" runs separate copies of the bench on each core, it gives a good idea of the memory system scales to support the additional cores (Opteron does rather poorly in this regard), but few single applications will see anything close to that much speedup. (Both numbers are useful)

DeathChill

Mar 9, 2006, 08:12 AM

Intel says that Merom is 20% faster than Yonah. 20% is also a typical figure for the speedup going from 32-bit x86 to 64-bit x64....

Uh, this has nothing to do with anything. 64-bit support wouldn't make a chip any faster, it just allows it to address more RAM (besides supporting 64-bit of course!).

~Shard~

Mar 9, 2006, 08:15 AM

I had to go look up a Wikipedia article because I don't have these names memorized. According to that, Kentsfield will be the four core version of Conroe, while the four core version of Woodcrest will be called Cloverton, or Clovertown depending on which article you look at.

So I'm saying that since it appears Intel is developing the Glidewell platform for using two Woodcrest processors in a workstation, it sounds reasonable that it would be the foundation for a new PowerMac, or Mac Pro.

I was trying to find a source the $850 Woodcrest price but couldn't.

Ah, I see. Cool, thanks for the info - makes sense. :cool:

AidenShaw

Mar 9, 2006, 08:45 AM

Uh, this has nothing to do with anything. 64-bit support wouldn't make a chip any faster, it just allows it to address more RAM (besides supporting 64-bit of course!).
True for PPC, not true for x64.

When running in 64-bit mode, the Intel architecture chip has more than twice as many usable integer/pointer registers, twice as many 128-bit SSE/FP registers, and a few other tweaks.

See http://en.wikipedia.org/wiki/X86-64

Benchmarks typically show an average of around 20% speed improvement for 64-bit applications/OS vs. 32-bit applications/OS running on the same system. This has been discussed many times in this forum - PPC slower in 64-bit mode, x64 faster in 64-bit mode.

For example, the following chart from BareFeats shows one comparison - the three PD lines show XP 32/64 bit with a 32/64 application:

http://www.barefeats.com/image06/dc-cin.gif
32/32 = Windows XP 32-bit with 32 bit application
64/32 = Windows XP 64-bit with 32 bit application
64/64 = Windows XP 64-bit with 64 bit application
PD = Pentium Dual Core (Pentium-D)

http://www.barefeats.com/dualcore.html

ksz

Mar 9, 2006, 11:02 AM

Intel says that Merom is 20% faster than Yonah. 20% is also a typical figure for the speedup going from 32-bit x86 to 64-bit x64....

I wonder if Intel is comparing 64-bit Merom to 32-bit Yonah - and the speed boost is only seen in x64 mode?
The new Core architecture in Merom has these additional enhancements (click here (http://www.extremetech.com/article2/0,1697,1935736,00.asp)):

1. Ability to issue and retire 4 instructions at once (4-wide architecture) compared to 3 for Yonah.

2. Macro-Op fusion in addition to micro-op fusion that is present in Yonah. Load and Compare operations, for example, are two macro-ops that can be combined and executed as if they were one op. Because Load/Compare occurs very often, macro-op fusion can provide improvements across a wide range of applications.

3. 4MB L2 cache versus 2MB on Yonah.

4. "Substantial enhancements have been made to the SSE engine. SSE registers are 128 bits wide, but in the past, data paths within the CPU were 64 bits wide. That meant that SSE instructions would take two cycles. In Conroe, the entire SSE engine is 128 bits, not just the registers. So SSE instructions can now execute in a single cycle."

There is no difference between Conroe and Merom other than:

1. "Conroe shares the architecture with the upcoming mobile equivalent, codenamed Merom. The main differences between the two revolve around implementation of power management and subtle differences in the manufacturing process."

ksz

Mar 9, 2006, 11:14 AM

Thanks for the reminder. I remember seeing $850, but forgot it included a link. I agree with your point there that Apple would want the new PowerMac to be priced about the same as the current edition. Checking the link gives me hope for a lower Mhz version at a lower price.

Two 3.0 Ghz Woodcrest chips = $1700
Two 2.0 Ghz Woodcrest chips = $ 660

I could be happy with a 2.0 Ghz Woodcrest "Quad."

From the link:

"The 5160 clocks at 3GHz, the 5150 at 2.66GHz, the 5140 at 2.33GHz, the 5130 at 2GHz, the 5120 at 1.86GHz and the 5110 at 1.6GHz.

The price cashcade when you buy loads of these is $850, $700, $470, $330, $270 and $230."
Agreed. At those prices, why not! Intel claims an 80% performance improvement over the current 2.8GHz Dual-Core Xeon (2x2MB Cache). Apple never had choices like this before...

DavidCar

Mar 9, 2006, 11:41 AM

Agreed. At those prices, why not! Intel claims an 80% performance improvement over the current 2.8GHz Dual-Core Xeon (2x2MB Cache). Apple never had choices like this before...
I wonder if Apple could create one motherboard with socketed processors that could be sold with 2Ghz Woodcrest dual core chips, and upgraded by either Apple or the user to 3 Ghz chips, or Cloverton quad core chips. Those would be great choices.

All I need now for my dream system is for EyeTV 500 to go MacIntel, and an Apple Cinema Display with an HDMI/DRM connector for next generation DVDs.

DavidCar

Mar 9, 2006, 11:46 AM

Benchmarks typically show an average of around 20% speed improvement for 64-bit applications/OS vs. 32-bit applications/OS running on the same system. This has been discussed many times in this forum - PPC slower in 64-bit mode, x64 faster in 64-bit mode.

I didn't realize that. So universal applications will have to come in PPC, Intel 32, and Intel 64 flavors?

Anonymous Freak

Mar 9, 2006, 01:20 PM

I didn't realize that. So universal applications will have to come in PPC, Intel 32, and Intel 64 flavors?

Unfortunately, true.

thunng8

Mar 9, 2006, 02:36 PM

Check http://www.spec.org/cpu2000/results/ - you're looking only at the results submitted in the most recent quarter. If you look at "all results", there are Itanium II systems that are close to the POWER 5 numbers.

Not even close:

8-way 2.2Ghz Power5+
SpecIntRate:200
SpecFPRate:382

8-way 1.67Ghz Hitachi IT2 (best IT2 result)
SpecIntRate:134
SpecFPRate:222

It would be more interesting to see SPECint/SPECfp (single stream) numbers for all the chips as well.

"SPECxx_rate" runs separate copies of the bench on each core, it gives a good idea of the memory system scales to support the additional cores (Opteron does rather poorly in this regard), but few single applications will see anything close to that much speedup. (Both numbers are useful)

Opteron does pretty good, it is the Intel Xeons that do very poorly:

Current best 8 way results:

Opteron 2.6Ghz 8-way
SpecIntRate:147
SpecFPRate:150

Xeon 3.0Ghz 8-way (HP DL580)
SpecIntRate:104
SpecFPRate:48.2

Note the extremely bad scaling of FP in the Xeon. This does not bode well for high core count Intel Powermac systems. The upcoming Bensley platform will help the Xeons out with FB-DIMM, independent buses (this will help fp_rate a lot) and Woodcrest. Intel has quoted a 80% increase in performance in specINT_rate of Woodcrest over current Xeons (not sure about core count of that comparison though).

Also note that the Power5+ system is almost 8X faster at specfp_rate with the same number of cores over current Xeon for 8 core system.

AidenShaw

Mar 9, 2006, 03:25 PM

Not even close:
Since you said "not even remotely close" originally, you've changed your position a bit ;)

This was my bad - that one 2.2 GHz POWER5 line is a bit out-of-place - I missed it earlier.

Note the extremely bad scaling of FP in the Xeon.
I can't note "bad scaling" with a single data point.

IMO, it makes much more sense to look at SPECint_2000/SPECfp_2000 (base) than the peak rate numbers. This reflects more on what an application with a couple of active threads will see.

It doesn't seem that reasonable to compare enterprise systems in the $50,000 to $200,000 range - and then extrapolate to what that means for some future chips in a PowerMac....

thunng8

Mar 9, 2006, 04:57 PM

Since you said "not even remotely close" originally, you've changed your position a bit ;)

This was my bad - that one 2.2 GHz POWER5 line is a bit out-of-place - I missed it earlier.

I can't note "bad scaling" with a single data point.

OK .. here are the results of that particular xeon system (specfp_rate):

2 core: 24.5
4 core: 32.1
8 core: 48.1

Do you admit the extremely bad scaling of the Xeon system on fp?

IMO, it makes much more sense to look at SPECint_2000/SPECfp_2000 (base) than the peak rate numbers. This reflects more on what an application with a couple of active threads will see.

Going for the base results doesn't change performance that much (~5% for POWER5 and very little difference for Xeon)

It doesn't seem that reasonable to compare enterprise systems in the $50,000 to $200,000 range - and then extrapolate to what that means for some future chips in a PowerMac....

You are right, the system I quote is rather expensive, however there are much cheaper Power5 systems.

POWER5 systems start from $3,399 (1.65Ghz P5-505) and $4,485(1.9Ghz P5-510). At the moment no entry level systems use the 2.2Ghz chip.

BTW, here are the results of the P5-510 1.9Ghz (2 core):

SpecIntRate:39.9
SpecFPRate:67.1

AidenShaw

Mar 9, 2006, 09:29 PM

OK .. here are the results of that particular xeon system (specfp_rate):

2 core: 24.5
4 core: 32.1
8 core: 48.1

Do you admit the extremely bad scaling of the Xeon system on fp?
Do you honestly believe that SPECfp_rate measures floating point performance?

It doesn't, and nobody buys an MP Xeon system for what it does measure. Yes, it's obvious to everyone who really understands SPECfp_rate that large numbers of Xeons don't scale. Dell stopped selling 8-way Xeons - there's a reason.

Which is exactly why an 8-way PowerMacIntel will have two quad-core chips with a huge shared cache and a 1333 MHz FSB. Do you think that Intel isn't concerned about this issue, and isn't working on improvements for the next generation chips?

What is your point, anyway? That a quarter of a million dollar POWER 5+ system is faster than a $50K Xeon server? I'd hope so.... What about a 64-way SPARC system vs a Celeron?

You're looking at server benchmarks on high-end enterprise systems vs. low-end servers, and trying to make some statement about what the next generation of desktop chips will deliver on desktop apps? The next Apple isn't going to have a POWER 5+ chip, nor is it going to have a Netburst Xeon chip. Stop the mental masturbation about these SPEC rate scores.

ksz

Mar 9, 2006, 10:31 PM

I wonder if Apple could create one motherboard with socketed processors that could be sold with 2Ghz Woodcrest dual core chips, and upgraded by either Apple or the user to 3 Ghz chips, or Cloverton quad core chips. Those would be great choices.
I hope we're witnessing the birth of the Era of Replaceable Processors in desktop Macs and eventually in laptops as well.

All I need now for my dream system is for EyeTV 500 to go MacIntel, and an Apple Cinema Display with an HDMI/DRM connector for next generation DVDs.
It does not seem likely that a single next-gen DVD format will emerge. BluRay and HD-DVD are likely to coexist just as DVD+R and DVD-R. All I hope is for inexpensive (relatively speaking) dual-format drives to be available when the Mac Pros are released.

thunng8

Mar 9, 2006, 10:47 PM

Do you honestly believe that SPECfp_rate measures floating point performance?

What does it measure then?

It doesn't, and nobody buys an MP Xeon system for what it does measure. Yes, it's obvious to everyone who really understands SPECfp_rate that large numbers of Xeons don't scale. Dell stopped selling 8-way Xeons - there's a reason.

Which is exactly why an 8-way PowerMacIntel will have two quad-core chips with a huge shared cache and a 1333 MHz FSB. Do you think that Intel isn't concerned about this issue, and isn't working on improvements for the next generation chips?

What is your point, anyway? That a quarter of a million dollar POWER 5+ system is faster than a $50K Xeon server? I'd hope so.... What about a 64-way SPARC system vs a Celeron?

What are you talking about?, I've pointed out a $4-5k Power5+ system that beats the 50k Xeon system in specfp_rate. And the p5-575 does not cost a quarter of a million dollars .. it is closer to $100k.

You're looking at server benchmarks on high-end enterprise systems vs. low-end servers, and trying to make some statement about what the next generation of desktop chips will deliver on desktop apps? The next Apple isn't going to have a POWER 5+ chip, nor is it going to have a Netburst Xeon chip. Stop the mental masturbation about these SPEC rate scores.

It was just something I noted in the original post , and then you challenged me on it (firstly by saying an itanium 2 system would be close to Power5+ performance), not once, but twice (asking me to provide more data points for bad specfp scaling) - nothing more. I think this can stop now.

Mechcozmo

Mar 9, 2006, 11:34 PM

blahblahblahblahblahblahblahblahblahblahblahblah<funny>Stop the mental masturbation</funny>blahblahblahblahblahblahblahblahblahblahblahblah
Read the 'tar.

rhashem

Mar 11, 2006, 12:30 PM

Hmm...well the G4 was dead on the mobile mac platform and mini...but how will Conroe really compare to the Quad G5? Given that the Pentium M architecture performs at around clock-for-clock the same as a G5 equivalent, we would need two Conroes at at least 2.8Ghz to match or beat the Quad G5..right?

The P-M is considered clock-for-clock equivalent to the G5 because of poor floating-point performance. The Merom design fixes that, so the SPEC scores for a 3 GHz Conroe are 2800/2500, versus 1438/2076 for a 2.5 GHz 970MP. So a 2.5 GHz Conroe would match the 970MP in SPECfp and outdo it by a large margin in SPECint. Considering Woodcrest (the CPU you'll need for a Quad-like machine, because of SMP) will be available up to 3.3 GHz, Apple doesn't even have to ship Intel's fastest parts to outdo the old Quad.

rhashem

Mar 11, 2006, 12:45 PM

64-bit just for the sake of being 64-bit doesn't make it faster. It can actually be slower.

On x86, this generally isn't true. x64 gets rid of most of the performance limitations Intel and AMD haven't been able to exorcise out of x86 already. Generally, it leads to a 10-20% speedup on a processor that implements x64 natively (eg: the Opteron, and hopefull Conroe).

rhashem

Mar 11, 2006, 12:52 PM

The G5 brought apple close to parity with Intel/AMD. And I think there is considerable doubt that PowerPC was ever in the lead.

The G5's quite legitimately beat the Pentium 4's and Athlon's of the time. Apple's GCC-based benchmarks of the G5 versus the Pentium 4 were perhaps the most legitmate benchmarks they've ever conducted --- no theoretical mastrubation about programs compiled with super-compilers that nobody actually uses.

However, a few months later, the Opteron came out, and once that hit 2.0 GHz, x86 took the lead again.

rhashem

Mar 11, 2006, 01:01 PM

This actually makes me sad. PPC's big selling point early on was that it was a much simpler architecture that could then scale to obscenely fast clock rates compared to the more complex x86.

The thing is, both the 970 and the modern x86 chips do instruction translation (x86 or PPC -> native, undocumented ISA). That's because PowerPC isn't quite as simple as the CPU implementors wanted it (ie: it has some 3 operand or 2 output instructions). Once you're doing instruction translation anyway, your core is limited by your native ISA, not the legacy ISA exposed to software.

Unfortunately, this hasn't been the case. And it's mostly due to manufacturing issues. If Intel was to take the design for the PPC970MP, and re-jigger it for their manufacturing, they could probably get it to fly at 3.5+GHz.

Heck, IBM could do that, if they rejiggered it to fix the power density issues. The 970 was designed for 3 GHz operation at 90nm (looks like they finally made it too). Going by historical projections, that means 4 GHz at 65nm (the same process as the Core Duo). The problem is that even at that speed, it'd be about as fast as a 2.5 GHz Conroe in integer code. That's the thing really hurting the G5 in a lot of code --- it's integer performance isn't enough to keep up with its FPU performance. Careful coding can get around that limitation, but most code isn't that carefully written (and GCC can't optimize for such a picky processor anyway).

backdraft

Mar 11, 2006, 06:08 PM

http://www.theregister.co.uk/2006/03/10/intel_heat/

DeathChill

Mar 11, 2006, 06:44 PM

http://www.theregister.co.uk/2006/03/10/intel_heat/
Wait, you're complaining about Intel talking about the advantages of technologies they're now implementing? They never said they invented them, but certainly Intel is in the lead power per watt wise.

Also, you wouldn't think you'd hear complaining about 'lies' on an Apple messageboard. Everything Apple says benchmark or performance wise is always tilted in their favour..

thunng8

Mar 11, 2006, 11:26 PM

The P-M is considered clock-for-clock equivalent to the G5 because of poor floating-point performance. The Merom design fixes that, so the SPEC scores for a 3 GHz Conroe are 2800/2500, versus 1438/2076 for a 2.5 GHz 970MP. So a 2.5 GHz Conroe would match the 970MP in SPECfp and outdo it by a large margin in SPECint. Considering Woodcrest (the CPU you'll need for a Quad-like machine, because of SMP) will be available up to 3.3 GHz, Apple doesn't even have to ship Intel's fastest parts to outdo the old Quad.

Just some minor corrections:

IBM just submitted spec scores for 970MP in their JS21 blade:
2.5Ghz: 1587/2119
2.7Ghz: 1706/2259

Which is a bit higher than the previous numbers quoted.

Conroe will clock at 2.67GHz at lanuch, not 3Ghz.
Woodcrest is an up to 3GHz part (and that where the 2800/2500 figure comes from). Also mentioned is an 80% specIntrate increase of Woodcrest over Paxville (Netburst Xeon chip). This would leave Woodcrest (dual core) with a specIntRate of approx. 34*1.8~=60.

Conroe XE is rumoured to clock at up to 3.33Ghz although nothing was confirmed at IDF about this part.

But overall, Conroe/Woodcrest/Merom looks very good and will definitely beat current dual core G5s in SPEC benchmarks (especially big advantage in Int).