Intel Merom (64-bit Mobile Processor) Due in September?

[jabooth]

Yes, yes a thousand times yes! I had an AMD Athlon 64 processor and loading internet exploror in 64 bit vs 32 bit was probably 3-4 times faster. Milage might vary but if the program is designed to take advantage of it then it will be. So before you post saying its only good for more than 8 gigs of ram really research it. It doesn't really make the chip faster it just means that it can do certain calculations in one step that might have taken a few before.

You sure mate? I experimented in 64bit linux for a while and 64bit didn't mean programs open faster (they actually might take slightly longer due to the fact they take up slightly more memory than 32 bit versions) but yes, once a program is running you do see a speed benefit.

Sorry I can't cite you a source, can anyone confirm this?

Or am I just barking up the wrong tree

 

[Hunabku]

You can talk about all you need is a 32-Bit and that's understandable, however in the future apps will start to be written/optimized for 64-Bit. It is already the present as the G5 has been out for some years now. Some pro's do use their portables to do high-end stuff and if not will start to do so. It just gives another added dimension to computing. Not everyone will need this power of course, but many Pros will.

The Core Duo is a major stop gap chip and in the technology world things change quickly, however the CD's life is going to be amazingly short. If it was a 64-Bit chip then Intel would just be building on it more and rolling out more cores for 2-3 years. The CD most likely will not see any of that once Merom comes out as that's the chip that originally was wanted by Apple to be first in their lines. Intel is classic for rolling out stop gap chips. Pentium D's..etc..

So yeah some people will be bothered by this and some will say something new always comes out, however with Merom I think some will feel that they got the technology that is truly going to receive the development and attention. Or the one that was meant to be. So yeah they will add 8-Cores to Merom and better than the 2-Core Merom, but at least they still have a Merom.

Conroe is one that I'm really curious about and wanting to see how it will best the G5.

Here, here. As programs are written for 64 bit we can expect relatively decent speed increases in certain areas. If there are enough of these increases then the value of your 32 bit computer will drop accordingly.

With more & more 64 bit processors coming online, the demand for software to support them is increasing. In coming versions, we will see Photoshop optimized for 64 bit - even photoshop cs2 needs Windows XP 64 to access more than 2gb RAM.

It would be good to know when Adobe will pull this off. I don't expect it will be in their next major release of photoshop UB-Mac & PC. But I could be wrong - insights any one?

 

[iPie]

There will definitely be a rev of the iMac and mini this year. Has anyone else noticed that the new Mac mini and the iMac are labeled "early 2006" (as in edition) all over apple.com?

Start here: http://www.apple.com/support/macmini/

That's a good point! Anybody care to speculate as to why Apple needs to qualify their NEW mac mini with (early 2006)?

Do the conroe and merom chips give the current intel macs an 'nearing end of lifecycle' status? or is this just a standard and regular upgrade of the mac?

For many people there is no 'last moment' when buying a new mac. It is more a question of increasing hassle with what they already have. If these chips are such a significant change, maybe it is worth delayed gratification.

 

[DVK916]

Both the next Intel iMac and Powermacs should get conroe, with merom being reserved for the intel ibook, and mac book pro. So I wonder how apple will make the new mac book pro significantly better than the imac given that it is clear both will use conroe.

 

[sunfast]

13.3" Widescreen MBP w/ a Merom proc and OS X 10.5 Leopard pre-installed

My dream machine!!

 

[combatcolin]

New PowerMacs with 10.5 please.

And Apple can have my £1600 order.

 

[jocknerd]

Will it replace the MacBook Pro or be another line?

I can see Steve offering the MacBook Ultimate or something like that. Sell it for $3299. That would really suck.

 

[ScubaDuc]

Both the next Intel iMac and Powermacs should get conroe, with merom being reserved for the intel ibook, and mac book pro. So I wonder how apple will make the new mac book pro significantly better than the imac given that it is clear both will use conroe.

I think it's likely that also i-Macs wil get the 64 bit Merom chip because it seems to be pin-to-pin compatible with Yonah, as was evidenced in the "upgrading Mac Mini" thread. This way Apple will just drop in the new chips in the MacBook pros and iMacs and bingo, they have "new" products with no significant re-designing

The MacMinis and new iBooks (macbooks) will remain Yonah for a while so that there is a clear segmentation between the pro and the consumer lines.

I would like to see benchmarks of the new Mini though: where are they?

 

[nataku]

wow this is really cool. I want to see Merom MacBooks. But I can not wait for September for a next-gen Mac laptop! I want my MacBook now!!!

 

[MacQuest]

Great news for Mac Mini Core Solo to Core Duo upgrade?

There will definitely be a rev of the iMac and mini this year. Has anyone else noticed that the new Mac mini and the iMac are labeled "early 2006" (as in edition) all over apple.com?

Start here: http://www.apple.com/support/macmini/

That's a good point! Anybody care to speculate as to why Apple needs to qualify their NEW mac mini with (early 2006)?

Do the conroe and merom chips give the current intel macs an 'nearing end of lifecycle' status? or is this just a standard and regular upgrade of the mac?

Assuming that the 64 bit Merom Mac Mini replaces the current 32 bit Yonah Mac Mini [7-10 months] and therefore drastically reduces the value of the 32 bit Yonah Mac Mini, wouldn't that mean that buying a current [socketed] Mac Mini Core Solo that we now know is upgradeable to [at least for now] the 2.16 Core Duo is a REALLY good investment?

Think about how much and how quickly those [currently] $400 Yonah Core Duo chips will drop in price once the Merom's are introduced!

BETTER YET, does anyone know if the Merom chips will be pin-compatabile with the Yonah chips? Now that would be a serious upgrade. Going from a 32 bit Yonah 1.5Ghz Core Solo Mac Mini to a 64 bit Merom 2.33Ghz Core Duo Mac Mini.

H0LEE ****!!!

As far as the heat issue argument, if the Core Duo 2.16 Mac Mini seems to be holding up okay, then the lower wattage Merom should hold up as well, correct?

Please advise, as I will buy a Mac Mini Core Solo immediately if this is the case.

Thanks!

 

[MacQuest]

BETTER YET, does anyone know if the Merom chips will be pin-compatabile with the Yonah chips? Now that would be a serious upgrade. Going from a 32 bit Yonah 1.5Ghz Core Solo Mac Mini to a 64 bit Merom 2.33Ghz Core Duo Mac Mini.

I think it's likely that also i-Macs wil get the 64 bit Merom chip because it seems to be pin-to-pin compatible with Yonah, as was evidenced in the "upgrading Mac Mini" thread.

OH, SWEET CHEEZITS LET THIS BE TRUE!!!

Can anyone else verify pin-compatability between Merom and Yonah, and maybe give me the slightest glimmer of hope that a 32-bit Yonah 1.5Ghz Core Solo Mac Mini could possibly be upgraded to a 64-bit Merom 2.33Ghz Core Duo Mac Mini?

Not that an upgrade to a 2.16Ghz 32-bit Yonah Core Duo [from a 1.5Ghz Core Solo] at a [hopefully drastically] reduced price would be anything to sneeze at. Just the possibility of upgrading to a 64-bit 2.33Ghz Merom makes the investment all that much sweeter though.

Thanks for the feedback!

 

[Kelmon]

...and the wait continues...

I had a play with a MacBook Pro in the Sheffield Apple Store while my wife was left to rampage through the clothing shops and I was pretty impressed. However, I have been planning for some time to wait for Merom (oddly the guys in the store hadn't heard of it) as I had always seen Yonah, once Merom was announced, as being an interim solution and I'm looking for something that will last me for some time. Interestingly the linked article suggested that the Santa Rosa chipset for Merom would be both delayed and not deliver what it was initially expected to do. I had been planning to wait for that but there's now the possibility that I might as well buy in September, which is pretty much when there will be sufficient cash in the bank for a balls-out 17" version.

I'm going to continue following news of the Santa Rosa chipset but I can't really wait much longer as my 1GHz Ti is getting long in the tooth. If it won't deliver much more than Napa64 and won't arrive until mid 2007 then I'll buy a Merom-based system once they are released by Apple. Given the competitive market that Apple is now in with their systems running the same hardware as the likes of Dell and HP I'm expecting new MacBook Pros to arrive with Merom soon after the official launch so as not to be left behind the "competition".

 

[ScottB]

I'm quite sure it has been said before on this thread but I think these will clearly be in the PowerMac or whatever they rename it as. Because aren't the intel chips Apple are using now laptop ones?

 

[AidenShaw]

ask Yahoo! "yonah merom pin-compatible"

BETTER YET, does anyone know if the Merom chips will be pin-compatabile with the Yonah chips? Now that would be a serious upgrade. Going from a 32 bit Yonah 1.5Ghz Core Solo Mac Mini to a 64 bit Merom 2.33Ghz Core Duo Mac Mini.

http://www.xbitlabs.com/news/cpu/display/20051019183430.html

Intel Merom to Be Pin-to-Pin Compatible with Yonah – Intel Exec
Intel Discloses Some Specs of Next-Gen Mobile Product

Category: CPU

by Anton Shilov

[ 10/19/2005 | 06:35 PM ]


"Intel Corp. disclosed certain specifications of its code-named Merom processors, which is slated to come in the second half of 2006 and utilize fully new architecture. An important point is that Merom will be compatible with mobile platforms designed for Intel Pentium M processor based on the Yonah core.
...

Additionally, a being dual-core processor made on a single piece of silicon, Merom will sport L1-to-L1 cache transfer as well as, possibly, unified L2 cache (up to 4MB) for better performance in applications that heavily rely on threading."

 

[AidenShaw]

there's progress, and there's desperation

Intel is classic for rolling out stop gap chips. Pentium D's..etc.

Aren't all computers just stop gap solutions? Further down the pipeline is always the holy grail of fast that you wait for, and then once it's come, it's just a stopgap to the next thing

But the Pentium D runs existing software, and is compatible with motherboards that support Pentium 4 and Celeron D chips. While it is fair to call the Pentium D a "stopgap", it is almost completely painless for computer makers, software makers, and users. It is a cheap and effective way to get the benefits of dual-core to users earlier than the unified multi-core design cycle would allow.

Apple's "stopgap 32-bit machines", however, require that the OS and all applications be ported to the new hardware if you want native performance.

When Merom comes around, the OS and all applications will need to be ported again to 64-bit if you want the best performance. (Wait for WWDC to hear the presentations about how OSx64 will be a true 64-bit OS stem-to-stern (like XP 64-bit), and unlike the lame 64-bit stuff tacked onto Tiger.)

The "9 months of Yonah" will be a speed bump that will cost Apple, vendors and users for years down the road. "Even fatter" binaries with x86, x64 and PPC code. Testing, QA and support for three versions of each product. Early "retirement" for 32-bit Intel support, as companies realize that they won't make enough money from 32-bit MacIntels to cover the costs of the extra fat. Rapidly dropping resale value for Yonah as soon as Merom appears, and especially when OSx64 shows up.

 

[gnasher729]

So besides 64-bit based software and more possible RAM, does 64-bit capabilities make the chip any faster, or does that depend just on clock speed/cache speed/FSB speed?

Generally, 64 bit doesn't make the chip faster, except for some very specialised applications (it helps with encryption, for example).

The Intel/AMD chips are a bit unusual: In 64 bit mode, not only do they have 64-bit registers, they also have twice as many integer and floating point registers as in 32 bit mode, and that usually can make code run 10% to 20% faster.

So x86 chips run faster in 64 bit mode, but not because of the 64 bit.

 

[ieani]

Looks like my ideal laptop will not be out until right before I am due at college. So I guess I will get the Imac now and then sell it later on for this one. With hopefully not too much money lost due to Apples poor matching of my wants/needs.

 

[AidenShaw]

The Intel/AMD chips are a bit unusual: In 64 bit mode, not only do they have 64-bit registers, they also have twice as many integer and floating point registers as in 32 bit mode, and that usually can make code run 10% to 20% faster.

A couple of additional points

• x86/x64 usually dedicates 3 of the 8 integer registers for program control and linkages. This means that in x64 mode you have 13 usuable registers, vs 5 usable in x86 mode - more than 2½ times as many.
• x64 also has twice as many 128-bit SSE registers - 16 instead of 8
• x86/x64 do not have any floating point registers - the FP87 uses an FP stack
• It is recommended to use SSE instructions for scalar floating point (use vector arithmetic, but with a vector length of 1). When this is done, x64 has twice as many SSE registers for floating point use.

There are a couple of other minor architectural differences that can improve performance in some circumstances.

 

[milo]

Going back to 32bit seemed like a step backwards..

I think the key word here is SEEMED. On the iMacs nothing was lost by going to 32 bit. The only real advantage of 64 bit is more than 4 gigs of ram, and they didn't have the slots for that anyway. On the iMac, nothing was lost, and a ton was gained by going to a dual core CPU.

Not to mention that the current version of OSX doesn't really allow 64 bit applications yet, it will take an OS rewrite.

Cost: Resale value - after selling 32-bit MacIntels for 6-8 months, Apple upgrades the line to 64-bit. Wanna buy a used 32-bit MacBook? Didn't think so. Boat anchor.

Cost: Developer pain - after forcing the vendors to make fat binaries for 32-bit x86, in a few months Apple will introduce "even fatter binaries" with code for x86, x64 and PPC. (Hidden cost - how many developers will skip x86 and wait to do "even fatter binaries" with x64 and PPC only? How many will say "to hell with it!" after putting up with the OS9 -> OSX -> x86 -> x64 code porting exercises?)

Why would anyone value a 32 bit machine any less? What can't they do that a 64 bit machine can?

And 64 bit apps provide an advantage only on a very few high end things. Only a handful of developers will release 64 bit apps, most will probably wait years and will only support 64 when it's done automatically by the compiler. How much extra work is it to create an app that runs in 32 and 64 anyway? Is it much different than current apps that have G5 optimization but still run on G4 or G3?

The situation is similar to G4 vs G5. Were the G4's suddenly crippled because they couldn't run apps? Nope. And the G4's didn't drop off a cliff in value, they dropped a bit just because they were a little bit slower, not because of 32 vs 64.

I just don't buy it.

 

[dosers]

64-bit support from Apple

I am with you.
Look, I love to have a 64-bit Merom MacBook and so forth, and the additional cache and few hertz will help, but why are people SO excited about this ?

Keep this in mind (as one of my co-workers pointed out):
There is NO mention in Apple's developer documentation about the amd64 ABI.

All of Apple's documentation that discusses 64-bit applications and ABIs is PowerPC-centric.

Don't get me wrong; they obviously are working on 64-bit capable Intel boxes, but it would be nice if they let their developers know about it, and to date they haven't:

<http://developer.apple.com/documentation/MacOSX/Conceptual/universal_bi
nary/universal_binary_abi_a/chapter_9_section_1.html>
<http://developer.apple.com/documentation/DeveloperTools/Conceptual/LowL
evelABI/index.html>

Conversely, Apple is currently looking for a Senior System Bringup Engineer with EM64T experience:

> The Kernel Engineering team within Apple's CoreOS organization is looking > for an exceptional engineer to work on state-of-the art kernel technology > for Mac OS X.

[ ... ]

> This position involves low-level platform support, particularly for > Intel-based products. You will be involved in system bringup, design > and implementation of low-level kernel enhancements, and adopting > new technologies. Specific experience with the IA32 supervisor mode > programming model, including the MMU, IPC, EM64, and other technologies > will be particularly helpful.

My point is that I doubt you will see any advantages from 64-bit, unless you are working with specific apps or need more than 4GB of flat-address memory. Yeah, it looks all great on paper, but I don't see the 'new 64-bit Macbook' to be THAT revolutionary compared to the current Yonah models...

I think the key word here is SEEMED. On the iMacs nothing was lost by going to 32 bit. The only real advantage of 64 bit is more than 4 gigs of ram, and they didn't have the slots for that anyway. On the iMac, nothing was lost, and a ton was gained by going to a dual core CPU.

Not to mention that the current version of OSX doesn't really allow 64 bit applications yet, it will take an OS rewrite.




Why would anyone value a 32 bit machine any less? What can't they do that a 64 bit machine can?

And 64 bit apps provide an advantage only on a very few high end things. Only a handful of developers will release 64 bit apps, most will probably wait years and will only support 64 when it's done automatically by the compiler. How much extra work is it to create an app that runs in 32 and 64 anyway? Is it much different than current apps that have G5 optimization but still run on G4 or G3?

The situation is similar to G4 vs G5. Were the G4's suddenly crippled because they couldn't run apps? Nope. And the G4's didn't drop off a cliff in value, they dropped a bit just because they were a little bit slower, not because of 32 vs 64.

I just don't buy it.

 

[iris_failsafe]

Maybe Apple would wait and release the 17' MBP with Merom.
Wait about a year a swth the Imac and the 15' MBP to it


About Conroes is that the repalecement of the P4 or of the Xeon?
I thing the replacement for the g5 must be a workstation class processor...

 

[dosers]

20% better ?

According Intel's roadmap, it's really Conroe, which is the star (Spring IDF 2006).

According to Intel, Conroe will give 40% more performance while using 40% less power.

Merom, will be "at a constant battery life compared to the Core Duo T2600", while providing 20% more performance.
That means, according to Intel, battery life will not be any 'better' by itself than that of the current Yonah, while we'll see a moderate performance increase.

 

[DVK916]

According Intel's roadmap, it's really Conroe, which is the star (Spring IDF 2006).

According to Intel, Conroe will give 40% more performance while using 40% less power.

Merom, will be "at a constant battery life compared to the Core Duo T2600", while providing 20% more performance.
That means, according to Intel, battery life will not be any 'better' by itself than that of the current Yonah, while we'll see a moderate performance increase.

Conroe will use 40% less power than the current P4, it will still consume more power than Yonah, and way more than Merom.

 

[DVK916]

I think it's likely that also i-Macs wil get the 64 bit Merom chip because it seems to be pin-to-pin compatible with Yonah, as was evidenced in the "upgrading Mac Mini" thread. This way Apple will just drop in the new chips in the MacBook pros and iMacs and bingo, they have "new" products with no significant re-designing

The MacMinis and new iBooks (macbooks) will remain Yonah for a while so that there is a clear segmentation between the pro and the consumer lines.

I would like to see benchmarks of the new Mini though: where are they?

The problem with this though is Merom will be more expensive than the equivalently clocked Conroe chip. Meaning if the new PowerMac get the Conroe it will allow them to be cheaper to produce.

 

[MacQuest]

BETTER YET, does anyone know if the Merom chips will be pin-compatabile with the Yonah chips? Now that would be a serious upgrade. Going from a 32 bit Yonah 1.5Ghz Core Solo Mac Mini to a 64 bit Merom 2.33Ghz Core Duo Mac Mini.

http://www.xbitlabs.com/news/cpu/display/20051019183430.html

Intel Merom to Be Pin-to-Pin Compatible with Yonah – Intel Exec
Intel Discloses Some Specs of Next-Gen Mobile Product

Category: CPU

by Anton Shilov

[ 10/19/2005 | 06:35 PM ]


"Intel Corp. disclosed certain specifications of its code-named Merom processors, which is slated to come in the second half of 2006 and utilize fully new architecture. An important point is that Merom will be compatible with mobile platforms designed for Intel Pentium M processor based on the Yonah core.
...

Additionally, a being dual-core processor made on a single piece of silicon, Merom will sport L1-to-L1 cache transfer as well as, possibly, unified L2 cache (up to 4MB) for better performance in applications that heavily rely on threading."

Just to clarify then, would this be:

a- very likely
b- very unlikely
c- too early to decide

Obviously it looks like it may be very likely, but just asking in case there might be some "unknowns" to factor in.