Developer Transition Kit Benchmarks

[PixelFactory]

Here is one thing we can all agree on. It won't matter anymore once the processor is replaced. Even if the last PowerPC Macs prove to have more performance then the Intel replacements, it won't matter. The Mac will be the same as any other PC you can go out and buy. This is probably the biggest advantage of going with Intel. Even if we had a crystal ball and we could see something GREAT happen in the future of the PowerPC, it won't matter because the Mac will be equal to every other PC out there. They will never sell a Mac on performance again so it is pointless. It will be sold on OS and a trendy design only. Performance won't matter because it will be the same as all the x86 boxes out there. That is the biggest advantage I can see.

Performance will still be an issue. But now we are talking software performance vs hardware. It will be interesting to see identical machines, one running OS X, one running Windows, and having benchmarks that show how fast OS X can be compared to an equally equipped Windows machine.

 

[shidoshi]

IBM will deliver in 2 years time because they could fall apart if they don't deliver to microsoft and sony also nintendo

Even though the relevance of this statement has already been noted as non-existant, there's another reason that this statement makes no sense. IBM will be delivering to Sony, Microsoft, and Nintendo chips of a pre-determined spec for the entire run of the hardware.

Would you like IBM to come up with a CPU of a certain power and speed for Apple, and then use that and only that for the next five or so years? Didn't think so.

 

[illumin8]

Their 64 bit chips run two separate 32 bit registers. Their 32 bit chips are really 8 bit chips interlaced to 32 bits. Pentium chips employ long pipelines and CISC processing. Because of the CISC processing they have to run at higher processing cycles, hence the higher GHZ. They generate a helluva lot of heat and the latest interation, Prescott, pulls 100 watts of power. It's all smoke and mirrors to say the 3.6 GZH chip is faster than a 2.7 GHZ G5. Jobs complained that IBM had only increased processing power from 2 GHZ to 2.7 in two years. However, during that same time period Intel has only increased the speed of the P4 from 3.4 to 3.6 GHZ. Late last year they issued a press release admitting they had hit the wall with the current P4 technology in terms of squeezing more speed out of that line of chips. Perhaps Apple is banking on a new line they are going to develop for them that will eliminate the limitations and bottlenecks of the current P4's, but that's down the road - perhaps two years. In the meantime, I wonder where IBM and Freescale will be with their chips?

Does anybody find it strange that Microsoft and Sony dumped Intel in their gaming machines citing similar complaints about Intel as Jobs did with IBM?

Next question. Does Apple really think they're going to shoehorn the P4 into a slim 1 inch Powerbook case when no pc maker has been able to do this? All laptops running the P4 are over 2 inches thick. Personally, I sure would have loved to have seen the 3 GHZ dual core G4 in the next generation Powerbook.

The only smoke and mirrors is your claim that a 2.7 ghz. G5 chip is faster than a 3.6ghz. P4. The benchmarks on Apple's site are obviously only showing you the part of the picture they want you to see. Sure, the 2.7 ghz. G5 is faster at things using Altivec like rendering video, Photoshop plugins, and other things that we always knew the PPC was better at, due to integrated Altivec, but we always knew that was the case.

Also, you're an idiot if you think Apple is going to try and shoehorn a desktop P4 processor into a PowerBook. They will use the Pentium-M or a descendent of the Pentium-M (otherwise known as Centrino), which has already been proven to perform much better at lower clock speeds and with lower power and heat requirements. In fact, the benchmarks I've seen say that a 1.6 ghz. Pentium-M processor performs about equivelant to a 2.8 ghz. P4. That's pretty impressive for mobile performance, and you can still get 3-4 hours of battery life from it.

Amazing. I thought the Apple zealots would quit all of this PPC r0x0rz, Intel blowz chunkz propaganda when Apple switched platforms, but they're still spouting the same FUD that Apple's been brainwashing them with for years.

And before you knock me as an Anti-Apple person, I own a 15" PowerBook G4 1.25 and a 30G 3G iPod. I also own a Dell SC420 and a Dell Dimension 8300. I'm not religious about my choice in computers... One Dell is a Linux server, the other Dell is for gaming, the Mac is for work and everything else.

 

[rubberband]

Does anybody find it strange that Microsoft and Sony dumped Intel in their gaming machines citing similar complaints about Intel as Jobs did with IBM?

Next question. Does Apple really think they're going to shoehorn the P4 into a slim 1 inch Powerbook case when no pc maker has been able to do this? All laptops running the P4 are over 2 inches thick. Personally, I sure would have loved to have seen the 3 GHZ dual core G4 in the next generation Powerbook.

First: MS and Sony dumped Intel b/c they can't deliever technology. They'r behind in the industry and they know; diference b/t and xbox chip and the ones intel can provide now, does not even give the bases for a new xbox: keep the one you got ppl making add on will make more $$ than any of them with a new box and a intel chip. IBM and in AMD are the ones leading the field but IBM is the one capable of delievering custom chips 'faster' and with pretty scary (good) specs. Apple is very stupid for dumping the PPC, but at least go with AMD-- yes they emulate Intel chips, but they are better at that than Intel at themselves!

Second:Great point! The road map shows off future chips with lower watt consumption but then you get the lower productivity and GHZ. But to the problem with IBM, you can't fit the processor in the case but at least is a processor not a quadriplegic thing. Hey, 2in is about the depth of a iMac, oh sorry, Apple only makes 1in PowerBooks even though the power part of the name is not granted with intel chips. Basically, Intel will in the next 2 years work backwards: unloading chip from its power for better power=heat; and IBM will innovate the world with powerful and smart chip
Let me here a ra, ra, ra for the coming intel macs

 

[chibianh]

A post a few pages back pointed to

http://www.xlr8yourmac.com/

Doesn't seem like many really paid attention so I bring it up again.

First, the thing is fast. Native apps readily beat a single 2.7 G5, and sometimes beat duals. Really.
(I asked about real-world apps - if any were already available in native code-Mike)
All the iLife apps other than iTunes, plus all the other apps that come with the OS are already universal binaries....

There's more information as well from xlr8yourmac. Good read.

Also, all the cell people and the AMD people need to be quiet. Apple evaluated both. AMD has the same, if not worse, supply problems as IBM. Their roadmap is fine, but the production capacity is not.

The tested Cell as well. That processor is NOT intended for PC applications. (it was designed for game systems, not as a general use CPU) The lack of out of order execution and ILP control logic creates very poor performance with existing software. Having developers rewrite for cell would have been MUCH more work than reworking for Intel. And that's what this is, you rework your codebase in ALL cases, not rewrite it.

 

[sord]

If I had to choose between buying a mac mini with a Celeron 3.06 GHz or a G4 1.42 GHz at the same price, I'd go for the Celeron one, MHz myth or no MHz myth.

My girlfriend's parents have a 2GHz celeron - it runs like a 400mhz pentium 2 (even after a clean install of Windows)

 

[DakotaGuy]

Do not leave out innovation.

Innovation is copying everyone else? I mean I always hated it when the PC's and Windows copyed Apple, now it looks like Apple is copying the PC industry. That is not all bad, however, I just don't see the incentive for Apple to go above and beyond the rest of the industry in performance now. They have nothing to prove anymore. I will buy an Intel Mac in the future once the kinks are worked out, but I just don't see Apple putting a lot of money into the guts of the case when most of it will already be done by Intel and other vendors. They want to save cost with this change and that will be a huge savings to use industry standard hardware.

 

[seashellz]

>>Does Apple really think they're going to shoehorn the P4 into a slim 1 inch Powerbook case...?
---
No-but they WILL be able to shoehorn the coming Pentium M, however, as part of the plan...

 

[ailleur]

First: MS and Sony dumped Intel b/c they can't deliever technology. They'r behind in the industry and they know; diference b/t and xbox chip and the ones intel can provide now, does not even give the bases for a new xbox: keep the one you got ppl making add on will make more $$ than any of them with a new box and a intel chip. IBM and in AMD are the ones leading the field but IBM is the one capable of delievering custom chips 'faster' and with pretty scary (good) specs. Apple is very stupid for dumping the PPC, but at least go with AMD-- yes they emulate Intel chips, but they are better at that than Intel at themselves!

Second:Great point! The road map shows off future chips with lower watt consumption but then you get the lower productivity and GHZ. But to the problem with IBM, you can't fit the processor in the case but at least is a processor not a quadriplegic thing. Hey, 2in is about the depth of a iMac, oh sorry, Apple only makes 1in PowerBooks even though the power part of the name is not granted with intel chips

Please stop fueling stupid comments by your own false analysis to the situation.
The cpu in the xbox is a celeron 700, comparing it to a p4 3.6ghz is, to be gentle, completly moronic.

2nd point. Lower productivity and ghz? WTF are you talking about. I thought you apple zealots were the one claming the mhz myth was crap. The lower watt consumption is a derivative of a p3/p4 design that is MEANT to be run at slower speed yet get the same work done a p4 would do at 4ghz.

All in all, you people will scare the less knowledgeable people in here by claming you know something you dont.
And if i didnt understand your post correctly, blame it on the syntax.

 

[LaMerVipere]

A supposed pic of the Pentium 4 3.6GHz devel box (ArsTechnica):

ooh la la! that looks interesting

 

[DakotaGuy]

Performance will still be an issue. But now we are talking software performance vs hardware. It will be interesting to see identical machines, one running OS X, one running Windows, and having benchmarks that show how fast OS X can be compared to an equally equipped Windows machine.

I agree completely with your statement. They better make sure it outperforms Windows...or they will have another issue to address.

 

[rubberband]

Also, all the cell people and the AMD people need to be quiet. Apple evaluated both. AMD has the same, if not worse, supply problems as IBM. Their roadmap is fine, but the production capacity is not.

The tested Cell as well. That processor is NOT intended for PC applications. (it was designed for game systems, not as a general use CPU) The lack of out of order execution and ILP control logic creates very poor performance with existing software. Having developers rewrite for cell would have been MUCH more work than reworking for Intel. And that's what this is, you rework your codebase in ALL cases, not rewrite it.

where you get this? is so no true the CELL desing for raw power 8 cores and central processing to start then cripled for the PS3. This chip only down side is the fact that is too powerful for ANYONE to get its full potential right now and the fact that IBM produces it so numbers are low. Besides, that a mac with a single cell processor yieling 250gigaflops is like a mac with 6 intels!! if they were capable of working together mind you, and nitrogen cooling was a feasible option

 

[Yebot]

Say it with me everyone. A @#$(* console chip is NOT a desktop chip!

Here, here.

The console chips are PowerPC-BASED and run at 3.2mhz.

They play a whole different game than G3/G4/G5 PowerPC's.

 

[ailleur]

where you get this? is so no true the CELL desing for raw power 8 cores and central processing to start then cripled for the PS3. This chip only down size is the fact that is to powerful for ANYONE to get its full potential right now and the fact that IBM produces it so number are low. Besides, that a mac with a single cell processor yieling 250gigaflops which is like 6 intels!! if they were capable of working together mind you.

STOP POSTING!!!!!!!!

 

[apple_intel]

Here is one thing we can all agree on. It won't matter anymore once the processor is replaced. Even if the last PowerPC Macs prove to have more performance then the Intel replacements, it won't matter. The Mac will be the same as any other PC you can go out and buy. This is probably the biggest advantage of going with Intel. Even if we had a crystal ball and we could see something GREAT happen in the future of the PowerPC, it won't matter because the Mac will be equal to every other PC out there. They will never sell a Mac on performance again so it is pointless. It will be sold on OS and a trendy design only. Performance won't matter because it will be the same as all the x86 boxes out there. That is the biggest advantage I can see.

Actually, this isn't true at all. The PC box builders all compete with one another, they all benchmark differently, etc.. There is much more to a computer than the processor and I'm pretty certain Apple will put together some impressive systems.

 

[DakotaGuy]

Actually, this isn't true at all. The PC box builders all compete with one another, they all benchmark differently, etc.. There is much more to a computer than the processor and I'm pretty certain Apple will put together some impressive systems.

You do have a point there. I am not sure what will happen.

I guess the last question I have is with the death of the PowerPC in desktop computing does this mean the RISC processor is dead replaced by the advanced CISC processor? Years ago I remember reading where RISC processors, like the PowerPC were supposed to be better then the CISC, but now it seems the opposite is true. I never really understood all of this.

 

[Porshuh944turbo]

Is it just me, or are there others who hate the fact that we can't use the term "Wintel" in the same sarcastic manner we used to.

Now we have Macintels!! [sigh]

 

[onlysublime]

Their 64 bit chips run two separate 32 bit registers. Their 32 bit chips are really 8 bit chips interlaced to 32 bits.

Please, before you speak this garbage, you really should know your facts. No wonder Apple only has 1.5% of the market with people like you. I bet you still believe everything Apple says.

 

[sbarton]

First: MS and Sony dumped Intel b/c they can't deliever technology. They'r behind in the industry and they know; diference b/t and xbox chip and the ones intel can provide now, does not even give the bases for a new xbox: keep the one you got ppl making add on will make more $$ than any of them with a new box and a intel chip. IBM and in AMD are the ones leading the field but IBM is the one capable of delievering custom chips 'faster' and with pretty scary (good) specs. Apple is very stupid for dumping the PPC, but at least go with AMD-- yes they emulate Intel chips, but they are better at that than Intel at themselves!

Second:Great point! The road map shows off future chips with lower watt consumption but then you get the lower productivity and GHZ. But to the problem with IBM, you can't fit the processor in the case but at least is a processor not a quadriplegic thing. Hey, 2in is about the depth of a iMac, oh sorry, Apple only makes 1in PowerBooks even though the power part of the name is not granted with intel chips

I predict that IBM will be out of the desktop/server CPU buisness in 10 - 15 yrs.

Everybody is fawning over the fact that they are now making GAME CONSOLE cpu's for MS, Sony, Nintendo..whatever. The point is that all that R&D is now going down a path towards embeded appliances, game consoles, low power devices, etc.

The only desktop or enterprise product that will be left on PPC in 2yrs will be IBM's RISC based systems. Do you honestly think that they will be able to maintain the PPC design and compete in the future with INTEL on desktop and server class processors??

There is no question that they could dump money into R&D and keep it competivite for awhile, but tell me...who are they going to sell them to now that Apple is gone and thier RISC based server market is dying by the day?

They foolishly bet the farm on game systems with MS, and Steve shafted them and MS in the same stroke **BRILLIANT**. Only one, perhaps two of those game consoles are going to be left standing in the end, and unless that whole "living room media center" thing really pays off, they are going to be in the same postion with the winner as they were with Apple.

Trust me, this is a "New Coke" class F'up by IBM (if anyone around here is old enough to know what I'm talking about).

 

[chibianh]

where you get this? is so no true the CELL desing for raw power 8 cores and central processing to start then cripled for the PS3. This chip only down side is the fact that is too powerful for ANYONE to get its full potential right now and the fact that IBM produces it so numbers are low. Besides, that a mac with a single cell processor yieling 250gigaflops is like a mac with 6 intels!! if they were capable of working together mind you, and nitrogen cooling was a feasible option

CELL is a very specialized processor and i'm sure it'll kick ass at whatever it specialiises in. For general purpose computing.. i don't think so. Maybe IBM can (or have) developed a derviative for general computing, but the bottom line still stands. IBM makes cool processors, but if they can't deliver them in quantity, what's the point? Same with AMD. If IBM can't deliver chips in quantity to a 'tiny' customer like Apple, good luck supplying chips for MS, Nintendo, AND Sony at the same time.

 

[mt4design]

My first post - Couldn't the Apple/Intel alliance bring along an Altivec solution?

I've read these forums for years but never registered until today.

My question is this...

Could/would it be possible for Apple/Intel to actually create an Altivec friendly version of the x86 based processor? After all, IBM had to do this with the G5, right?

Couldn't Intel and Apple be working on such a solution that would make the ease of the transition easier and the processor that much more powerful?

Thanks!

 

[seashellz]

>>IBM and in AMD are the ones leading the field but IBM is the one capable of delievering custom chips 'faster' and with pretty scary (good) specs. Apple is very stupid for dumping the PPC, but at least go with AMD-- yes they emulate Intel chips, but they are better at that than Intel at themselves!
---


WAKE UP! IBM are putting all their efforts into SONY and MS-they dont have time for APPLE and their 3% anymore...
They will INOVATE for SONY and MS-they dont have time for APPLE and their 3% anymore...
They will deliver ON TIME-or as close as possible to MS and SONY because they dont have time for APPLE anymore-
These are chip specs peculiar to the gameboxes-
APPLE is a mere annoying gnat to IBM: Big Blue is now playing with the BIG BOYS...


and doesnt have time for APPLE anymore.

The handwrittiing was on the wall, when last week the Prez of Intel stated something like: "If you want security-go with the Mac"

I think maybe this guy has a hard-on for APPLE-and a new market to grow-APPLE could well control 25% of the market in 5 years-stay with IBM and maybe own 5%...

Im sure Intel takes no pleasure in being associated with the horrible likes of Microsoft anymore-despite the cash flow-and MS is never going to write Windows to work on a PPC chip-it would likely explode.

remember folks-the big bet is on the as of YET unreleased Pentium M chip-so quit sqwuakinig until you see the REAL specs.

 

[rubberband]

Please stop fueling stupid comments by your own false analysis to the situation.
The cpu in the xbox is a celeron 700, comparing it to a p4 3.6ghz is, to be gentle, completly moronic.

2nd point. Lower productivity and ghz? WTF are you talking about. I thought you apple zealots were the one claming the mhz myth was crap. The lower watt consumption is a derivative of a p3/p4 design that is MEANT to be run at slower speed yet get the same work done a p4 would do at 4ghz.

All in all, you people will scare the less knowledgeable people in here by claming you know something you dont.
And if i didnt understand your post correctly, blame it on the syntax.

Processor speed is a myth yes! but in the case of intel it comes down to the size of load that it can handleacket size. Altivec was design to increase the size to (say) large amounts while intel developed HT to compensate. Doesn't matter who came first, the point is that intel never unticipated a road block with with pentiums and design them for small packets and fast cores, and ibm design ppc for speed independence through packet size. Hence, ghz myth busted!

NOw, if you criple a intel chip you must increase packet size to compensate but they have not being able to so, reason for which, no new intel chips have being seen and rather stupid speed bumps.( celerons and so on are pentium's derivatives with same fate). So the next year chips promise this but that is like re-inventing yourselve or flagship. I dont have fate in intel pulling it through.

Now moron, if you dont understand this. I am sorry but i can't teach english to on a chat session.

 

[ailleur]

Processor speed is a myth yes! but in the case of intel it comes down to the size of load that it can handleacket size. Altivec was design to increase the size to (say) large amounts while intel developed HT to compensate. Doesn't matter who came first, the point is that intel never unticipated a road block with with pentiums and design them for small packets and fast cores, and ibm design ppc for speed independence through packet size. Hence, ghz myth busted!

NOw, if you criple a intel chip you must increase packet size to compensate but they have not being able to so, reason for which, no new intel chips have being seen and rather stupid speed bumps.( celerons and so on are pentium's derivatives with same fate). So the next year chips promise this but that is like re-inventing yourselve or flagship. I dont have fate in intel pulling it through.
The pentium m is faster than both pentium EE and athlon fx, which both run several laps around a g5.

Now moron, if you dont understand this. I am sorry but i can't teach english to on a chat session.

The next chip ALREADY IS OUT genius, and its called a pentium m. It has been out for, oh, 2 years. Yeah im really scared intel wont be able to come up with a chip next year, that has been out for 2 years today. Whew, what a suspense. You know nothing of the intel product like, so please PLEASE stop talking about it and scaring people who come here to learn something. Altivec is nothing magical and its nothing sse1-2-3 cant mimic. An instruction set is an instruction set.

Oh and about the english remark, i was refering to your syntax, not mine. Try reading some of the things you post before hiting submit.

 

[madmaxmedia]

You do have a point there. I am not sure what will happen.

I guess the last question I have is with the death of the PowerPC in desktop computing does this mean the RISC processor is dead replaced by the advanced CISC processor? Years ago I remember reading where RISC processors, like the PowerPC were supposed to be better then the CISC, but now it seems the opposite is true. I never really understood all of this.

The majority of today's processors can’t rightfully be called completely RISC or completely CISC. The two textbook architectures have evolved towards each other to such an extent that there’s no longer a clear distinction between their respective approaches to increasing performance and efficiency. To be specific, chips that implement the x86 CISC ISA have come to look a lot like chips that implement various RISC ISA’s; the instruction set architecture is the same, but under the hood it’s a whole different ball game. But this hasn't been a one-way trend. Rather, the same goes for today’s so-called RISC CPUs. They’ve added more instructions and more complexity to the point where they’re every bit as complex as their CISC counterparts. Thus the "RISC vs. CISC" debate really exists only in the minds of marketing departments and platform advocates whose purpose in creating and perpetuating this fictitious conflict is to promote their pet product by means of name-calling and sloganeering.


I'm so smart, huh? Actually, it came from here-

http://arstechnica.com/cpu/4q99/risc-cisc/rvc-1.html