Nehalem Mac Pros Arrive: Unboxing and Benchmarks

[Tesselator]

Think about it.

No don't.

Yes.

No.

?

.

LOL!!!

RED = I Disagree
GREEN = I agree.
YELLOW = I dunno.

Some people says the new MAC PROs are too expensive.
Some others says it's a huge step foward.
Some that it's a huge step backward.
Some says that the single chip edition sucks because of ram limit (8gb).
Some says that you won't need more that 8gb of ram in the next 2 years.
Some that the single 2.93 chip is the only one that gives you "bang for your bucks".
Some that only the most expensive octo 2.93 is the one that is really enough powerful to be taken in consideration.
Some says to wait another year for the 6 cores edition because this is a useless update and the next one will be much better. Not useless tho.. Just a rip off $$$
Some that we will see the real advantage only when snow leopard comes.
Some that we can already see huge advantage on the programs that uses multi-core processing.
But some says that there are no program that really uses multi-core advantage. Some others don't.
Some says that hyperthreading and ecc-ram is super good. Meaning only together with each other?
Some that hyperthreading can be an obstacle that lower the performances and ecc-ram is useless because you don't really need it.
Some say dell are much better than apple.
Some says that apple is loosing touch.
Some other give kudos to apple for having released one of the world's most powerful computer. Kinda...
​

 

[Tesselator]

It won't fix the serious drop in single-threaded performance, which is what many posters are complaining about.

Then they're daft. There is no drop in the performance of a single threaded app running on a multi core processor. It runs at the speed of a single core. Same as on a Duo, a quad, or an octad.

In fact the new procs are faster per clock then anything previous. This is both common sense and it's been proven. The new 2.66 runs single threaded apps from between 10% slower to 10% faster then the old 2.8 did.

The new 2.66 runs all single threaded apps faster than any previous 2.66 did. The new chips are inherently faster clock for clock.

Additionally if Apple did have a single core computer that ran at 2.66 it would not run a single threaded app faster than their current duals, quads, or octads do at the same 2.66.

So...(i) I'm ready to buy, and (ii) have read through this fascinating thread. My basic question is this (and some posts have touched on this): am I better off with the DP octo 2.26, or the SP quad 2.66 (or SP quad 2.93 for that matter)?

(One thing I DO know is that whichever one I get will blow my current PowerMac G5 DP 2.5GHz right out of the water - but that's besides the point.)

A little background: I do audio work on my current G5 (Logic Studio, samples, plug-ins, etc.), and to a lesser extent, some video work using FCE. I don't know whether Logic is already or will be optimized for multi-cores and/or multi-processors. I just want something that will last me a good 6-7 years.

So boys and girls: more cores or higher clock speed?? Any suggestions or comments are greatly appreciated!

If it were me I would ask myself how much work I would be doing in those areas. If I had boatloads piling up on me then maybe a multi-core system would be the way to go. 6 or 7 years is at least the 2.66 Octad. But an older 2008 2.8GHz will do too!

If I was working at a more casual pace (where it wasn't all about meeting strict and impossible deadlines everyday) then I would go for the Mac Mini (with two nice monitors!!!) for now and get a MacPro when the prices they are currently asking are justifiable. I guess that's 2010?

You're coming from a G5 so that's not too bad - but if you look at Apple's procession and where they are now it's kinda bad in a cost performance way. We can also maybe see that 2010 is the magic year??

1998 Apple Releases a 1 core G3 266MHz (AV) $2500 ... Speed increase from previous = more than 2X
2000 Apple Releases a 2 core G4 450MHz $2500 ... Speed increase from previous = more than 2X
2002 Apple Releases a 2 core G4 1.00GHz $2500 ... Speed increase from previous = more than 2X
2004 Apple Releases a 2 core G5 2.00 GHz $2500 ... Speed increase from previous = more than 2X
2006 Apple Releases a 4 core 2.66 GHz $2500 ... Speed increase from previous = more than 2X
2008 Apple Releases a 8 core 2.8 GHz $2800 ... Speed increase from previous = more than 2X

See this is about how it's supposed to go. Since about 1980 this has been pretty much the case with every vendor. Every two years we get 2X the speed for about the same price. But what about this year?

2009 Apple releases a 8 core (w/HT) 2.66 GHz $4700 ... Speed difference = between 0.9X ~ 1.68X

Yeah, between slower and only about 70% increase when heavily multitheading - but almost TWO THOUSAND dollars more???

See what I mean? What the future holds I dunno. Hopefully an Apple remedy! But this kind of jump in price for little or no benefit... Shrug! I dunno what to make of it. Other vendors have not and likely will not be doing anything like this. So it's only Apple? If so then I think it's time to get OS X running on non-Apple Hardware where we will be getting our 2X speed increase for the same $2500 to $2800 and not $4700. Either that or wait till 2010 and see what happens.


.

 

[eeboarder]

Then they're daft. There is no drop in the performance of a single threaded app running on a multi core processor. It runs at the speed of a single core. Same as on a Duo, a quad, or an octad.

In fact the new procs are faster per clock then anything previous. This is both common sense and it's been proven. The new 2.66 runs single threaded apps from between 10% slower to 10% faster then the old 2.8 did.

The new 2.66 runs all single threaded apps faster than any previous 2.66 did. The new chips are inherently faster clock for clock.

Additionally if Apple did have a single core computer that ran at 2.66 it would not run a single threaded app faster than their current duals, quads, or octads do at the same 2.66.





If it were me I would ask myself how much work I would be doing in those areas. If I had boatloads piling up on me then maybe a multi-core system would be the way to go. 6 or 7 years is at least the 2.66 Octad. But an older 2008 2.8GHz will do too!

If I was working at a more casual pace (where it wasn't all about meeting strict and impossible deadlines everyday) then I would go for the Mac Mini (with two nice monitors!!!) for now and get a MacPro when the prices they are currently asking are justifiable. I guess that's 2010?

You're coming from a G5 so that's not too bad - but if you look at Apple's procession and where they are now it's kinda bad in a cost performance way. We can also maybe see that 2010 is the magic year??

1998 Apple Releases a 1 core G3 266MHz (AV) $2500 ... Speed increase from previous = more than 2X
2000 Apple Releases a 2 core G4 450MHz $2500 ... Speed increase from previous = more than 2X
2002 Apple Releases a 2 core G4 1.00GHz $2500 ... Speed increase from previous = more than 2X
2004 Apple Releases a 2 core G5 2.00 GHz $2500 ... Speed increase from previous = more than 2X
2006 Apple Releases a 4 core 2.66 GHz $2500 ... Speed increase from previous = more than 2X
2008 Apple Releases a 8 core 2.8 GHz $2800 ... Speed increase from previous = more than 2X

See this is about how it's supposed to go. Since about 1980 this has been pretty much the case with every vendor. Every two years we get 2X the speed for about the same price. But what about this year?

2009 Apple releases a 8 core (w/HT) 2.66 GHz $4700 ... Speed difference = between 0.9X ~ 1.68X

Yeah, between slower and only about 70% increase when heavily multitheading - but almost TWO THOUSAND dollars more???

See what I mean? What the future holds I dunno. Hopefully an Apple remedy! But this kind of jump in price for little or no benefit... Shrug! I dunno what to make of it. Other vendors have not and likely will not be doing anything like this. So it's only Apple? If so then I think it's time to get OS X running on non-Apple Hardware where we will be getting our 2X speed increase for the same $2500 to $2800 and not $4700. Either that or wait till 2010 and see what happens.
.

Why are you so dedicated to troll this thread? I bookmarked this thread to read about people getting their new MacPros as I look forward to receiving mine. Yes, it is more expensive. It's been said by multiple people multiple times. We get it. You're numbers are skewed heavily towards the negative side with every post. I don't know what your problem is. Maybe you're upset about your older system or you just like spreading your negativity. I don't know. I'm not a psychiatrist.

There are lots of positives here:

1) Impressive redesigned interior.
2) Faster cheaper RAM(which is usually a bottleneck in most systems).
3) Improved graphics.
4) SATA optical drives.
5) 45nm process with lower power consumption.
6) Excellent multi-threading performance.

It's a decent improvement on the previous generation. Not shocking, but decent. Yes, the next generation will be better, but why wait? I'm going to get another year of production out of this machine than the next generation. I would image that it's rare for anyone to use up all power available in these MacPros at most times, at any rate. There may not be lots of applications than can utilize all 8-cores at all times, but guess what? I don't use one application at a time. I use 5-10. So do most people.

You want to compare benchmarks? The 8-core 2.26GHz is comparable to the last 3.2GHz for multi-threaded applications. As more and more applications and OS generations move to multi-core optimization, this machine will prove to be more powerful and useful than the previous generation. So will the generation proceeding this one. That's the way technology works. No need to get your panties in a bundle.

 

[kabunaru]

There are lots of positives here:

Let's have an analogy:
PowerMac G3 ---->PowerMac G5 upgrade as 2006 Mac Pro to 2009 Nehalem Mac Pro

PowerMac G5 Dual Processor ------> PowerMac G5 Quad Core as 2008 Mac Pro to 2009 Mac Pro upgrade.

 

[Tesselator]

@eeboarder

WTH???

Show me one skewed number. Just one.

You can't so I fully expect you to apologize for everything you just said!

 

[eeboarder]

@eeboarder

WTH???

Show me one skewed number. Just one.

You can't so I fully expect you to apologize for everything you just said!

kabunaru provided a excellent example. You're providing giant CPU leaps over two year periods and comparing them to the 2008 MP to the 2009 MP.

EDIT: And, you are trolling.

Not to mention pricing. Did you include RAM upgrades? Who uses their MP with 2GB RAM?

 

[Tesselator]

kabunaru provided a excellent example. You're providing giant CPU leaps over two year periods and comparing them to the 2008 MP to the 2009 MP.

His examples were all the at the same price point. The one's I'm exemplifying are at around a $2,000 dollar increase. Massive difference! If the prices were the same like in his examples or even in the same ballpark I wouldn't be making this case - which I only did BTW because it was asked. But you'd know that had you bothered read the thread and not just come here like a total troll blazing for trouble and wanting to fight with anyone that wasn't willing to lie about how awful these new machines are at this new price point.

 

[eeboarder]

Yes, it is more expensive. It's been said by multiple people, multiple times. We get it.

Again.

 

[mjteix]

In Transit: 2.26GHz 8-Core MacPro, ATI 4870HD, 9GB RAM, 3TB, 24" Samsung LCD

How are you going to achieve that?

 

[gnasher729]

The real hard issues, however, remain. The scheduling is dynamic, and changes millisecond by millisecond. One millisecond you may have 8 threads running on 8 cores, next millisecond 16 threads running on 8 cores, and the next millisecond 8 threads running on 4 cores. Since moving a thread can be expensive, trying to keep it all "optimal" millisecond by millisecond by moving threads around can make it slower.

gnasher729's scenario, while reasonable, is a static analysis that misses the issues that dynamic thread scheduling exposes.

You are very much exaggerating the problems.

When hyperthreading was introduced, it ran into operating systems and applications that had never heard of it, so unfortunate things happened, like a machine with two hyperthreaded cores using the two virtual threads on one core only, halving performance because they didn't know better.

However, your scenario seems very non-typical. It may be a problem for servers. Say a server with 32 users connected, who all use a bit of CPU time at unpredictable intervals. But the people buying MacPros most likely have workflows that use the CPUs for a long time. Like a video rendering, using all the CPUs it can get for a long time. These apps get hold of a CPU and never let go for a long time. And switching from one CPU to another is fast, as long as a thread stays on the same processor. All that needs moving is 512 KB from L2 cache to L3 cache on the same processor.

 

[Mendoman]

mac pro speed tests

For those of us graphics people who do not speak benchmark tests it would be nice to see a comparison of machines doing a Photoshop thing, like rotating a 500MB image 1 degree, something we can all relate to and have the ability to do on any computer (that will handle a 500MB file, so let's include a 100MB file for the "older" machines). My HP XW4200 workstation, 3Ghz, 3GB ram, XP Pro, for example took 5 seconds for the 100MB file and 2 minutes for the 500MB file. It would be quite valuable to know how much a Mac Pro, old or new, would improve this (say with Photoshop CS4 or 3). Specs such as installed ram and hard drive size and config would be important. Say stock out of the box configuration. Thanks.

 

[Tesselator]

For those of us graphics people who do not speak benchmark tests it would be nice to see a comparison of machines doing a Photoshop thing, like rotating a 500MB image 1 degree, something we can all relate to and have the ability to do on any computer (that will handle a 500MB file, so let's include a 100MB file for the "older" machines). My HP XW4200 workstation, 3Ghz, 3GB ram, XP Pro, for example took 5 seconds for the 100MB file and 2 minutes for the 500MB file. It would be quite valuable to know how much a Mac Pro, old or new, would improve this (say with Photoshop CS4 or 3). Specs such as installed ram and hard drive size and config would be important. Say stock out of the box configuration. Thanks.

There's this thread: https://forums.macrumors.com/threads/200558/

Problem is... Trying to find a decent PS test that is distributable and that scales across multiple cores is about as difficult as trying to stuff a pound of peanut-butter down a pole-cat's throat in a telephone booth. For one there's only a few filters in PS that multi-thread at all. etc.

 

[drsmithy]

Early versions of OSX were rather weak on SMP systems, even though those have been around for decades.

Well, there is a significant difference between relatively simple modifications to scheduler algorithms and refactoring large amounts of code to make it more scalable.

I would not expect that Apple invested a lot of effort in making hyper-threading aware schedulers before now - when hyperthreaded chips are actually here.

Didn't some of the early x86 dev boxes have HT P4s in them ?

 

[AidenShaw]

All that needs moving is 512 KB from L2 cache to L3 cache on the same processor.

You don't understand caches, do you?

I don't believe it's worth the effort to continue this discussion about scheduling - you think it's simple, and Cambridge University considers it worthy of doctoral dissertations.

Well, there is a significant difference between relatively simple modifications to scheduler algorithms and refactoring large amounts of code to make it more scalable. ?

I could argue about the simplicity of the changes, but the argument that I was making was that software teams seldom make changes until they are required and necessary. Remember that 10.5 slipped because people were moved to work on the Iphone OS - what engineering manager would assign people to work on improved SMT scheduling when there was no apparent need?

Didn't some of the early x86 dev boxes have HT P4s in them ?

They all did.

 

[eeboarder]

How are you going to achieve that?

Two sets of three. 3x2GB from Newegg and 3x1GB from Apple.

 

[Eric S.]

what engineering manager would assign people to work on improved SMT scheduling when there was no apparent need?

Most likely none. In my experience, companies are always scrambling to get the next product out the door, or to fix bugs in existing products. That's even more true in recent years. In the past a large company might have devoted some software resources to "research," but it's unlikely today.

 

[Kiwi Mac]

For Aperture

OK - I have to replace my current machine, a dual 2.7 G5 with 7Gb of RAM and the OE graphics card.

My main use is professional photography using Aperture (although may switch to Adobe Light Room soon unless Apple fix the bugs in Aperture).

I am currently torn between the 24" 3.06 iMac and the cheaper of the 2 options for the Mac Pro.

I need a larger monitor as well - only have 20" ACD.

Ideally I would get the Mac Pro and the 30" ACD - a total of around $10,000 here in NZ, about twice the cost of the iMac option.

Will I see any significant benefit from the extra $5,000 (apart from the screen real estate increase) or do you think the iMac will do for another 3 years or so?

 

[avkills]

OK - I have to replace my current machine, a dual 2.7 G5 with 7Gb of RAM and the OE graphics card.

My main use is professional photography using Aperture (although may switch to Adobe Light Room soon unless Apple fix the bugs in Aperture).

I am currently torn between the 24" 3.06 iMac and the cheaper of the 2 options for the Mac Pro.

I need a larger monitor as well - only have 20" ACD.

Ideally I would get the Mac Pro and the 30" ACD - a total of around $10,000 here in NZ, about twice the cost of the iMac option.

Will I see any significant benefit from the extra $5,000 (apart from the screen real estate increase) or do you think the iMac will do for another 3 years or so?

Personally, I'd use the G5 instead of the iMac. The Mac Pro is a better option long road, since your internal upgrade options are a lot better. I am basically in the same boat, I have a dual 2Ghz G5, although I am just wondering if I should pull the trigger or wait. I would never buy an iMac. The iMac places too many limits for a pro in my opinion.

-mark

 

[LillDrutten]

Music studio computer?

Steinberg have made a official statement about their views of the new mac pro.
(Steinberg makes the music/audio software nuendo and cubase and are used in many
big studios and homes)

http://knowledgebase.steinberg.de/71_1.html

The same applies to the new Intel i7 ("Nehalem") processors that also feature Hyper-Threading. We are currently investigating using an Intel i7 on both PC systems and the latest Mac Pro models with HT enabled in conjunction with our sequencers. As long as our tests are still ongoing, we recommend turning off Hyper-Threading in the BIOS on PC systems. As this is not possible on the latest Mac Pro systems, we cannot recommend using those systems until further notice.

 

[macshark]

A rip-off is a rip-off...

Current Mac Pros are a very poor value. I was actually going to buy a Mac Pro as soon as Apple announced the Nehalem models, but I postponed my purchase as the current price/performance point does not make any sense. Apple is asking for 150% premium for the single processor model compared to similar hardware from other vendors.

First, the single processor, quad-core Mac Pro: Xeon adds no practical value for this system. It is essentially the same as any other desktop Core i7 system out there in terms of performance. Yes, you do get a better designed enclosure, and better software, but I don't understand how Apple can justify charging $2500 for this system when you can get a comparably configured system from Dell/Gateway (and now Lenovo) in the $1K range. And most of these vendors do not give you a retarded graphics card like the NVidia 120, on top of that, they usually have some kind of a promotion going on where you can get extra 3GB of memory for free (for a total of 6GB)...

When Apple introduced the first Intel Mac Pros, pricing was actually quite competitive against comparable products from vendors like Dell. This was partly because the Xeon processors and chipset that supported SMP configuration was very pricey and therefore only available in high end workstation class platforms.

I think it would be fair for Apple to claim some premium over the generic PC workstation given the much better enclosure design and excellent software, but a reasonable cap for this premium should be in the 50% range. If I can get a XPS 435 with a single 2.66GHz Core i7, 6GB, 750GB HD, ATI 4670 graphics for $1099 (sale price), a similarly configured Mac Pro should not cost more than $1599-$1699. $2499 is a ridiculous price in today's market. If Apple wanted to maintain $2499 as an entry price, they should have offered the 2.93GHz processor, 6GB memory, 1TB HD and Radeon 4670 as the base configuration.

As for the dual processor model, the entry level processor choice of the 2.26GHz Core i7 Xeons is ridiculous. This model clearly has lower performance than the dual 2.8GHz Xeon Mac Pro it is replacing in almost all applications despite the higher price tag. The new dual processor Mac Pro should have been priced at $2999 and should have included dual 2.66GHz quad-core Core i7 processors, 6GB RAM, 1TB HD and ATI 4870 graphics. It is shameful that Apple puts a 640GB HD in a $3299 system to save maybe $20 cost when you can get 1TB HDs for $90 at retail.

Intel is going to do a major price adjustment in May and July of this year and introduce faster Core i7 chips. It would be interesting to see if Apple updates the entry level single processor Mac Pro to 2.93GHz and entry level the dual processor Mac Pro to 2.66GHz at that time while maintaining the price point.

 

[LillDrutten]

Current Mac Pros are a very poor value. I was actually going to buy a Mac Pro as soon as Apple announced the Nehalem models, but I postponed my purchase as the current price/performance point does not make any sense. Apple is asking for 150% premium for the single processor model compared to similar hardware from other vendors.

First, the single processor, quad-core Mac Pro: Xeon adds no practical value for this system. It is essentially the same as any other desktop Core i7 system out there in terms of performance. Yes, you do get a better designed enclosure, and better software, but I don't understand how Apple can justify charging $2500 for this system when you can get a comparably configured system from Dell/Gateway (and now Lenovo) in the $1K range. And most of these vendors do not give you a retarded graphics card like the NVidia 120, on top of that, they usually have some kind of a promotion going on where you can get extra 3GB of memory for free (for a total of 6GB)...

When Apple introduced the first Intel Mac Pros, pricing was actually quite competitive against comparable products from vendors like Dell. This was partly because the Xeon processors and chipset that supported SMP configuration was very pricey and therefore only available in high end workstation class platforms.

I think it would be fair for Apple to claim some premium over the generic PC workstation given the much better enclosure design and excellent software, but a reasonable cap for this premium should be in the 50% range. If I can get a XPS 435 with a single 2.66GHz Core i7, 6GB, 750GB HD, ATI 4670 graphics for $1099 (sale price), a similarly configured Mac Pro should not cost more than $1599-$1699. $2499 is a ridiculous price in today's market. If Apple wanted to maintain $2499 as an entry price, they should have offered the 2.93GHz processor, 6GB memory, 1TB HD and Radeon 4670 as the base configuration.

As for the dual processor model, the entry level processor choice of the 2.26GHz Core i7 Xeons is ridiculous. This model clearly has lower performance than the dual 2.8GHz Xeon Mac Pro it is replacing in almost all applications despite the higher price tag. The new dual processor Mac Pro should have been priced at $2999 and should have included dual 2.66GHz quad-core Core i7 processors, 6GB RAM, 1TB HD and ATI 4870 graphics. It is shameful that Apple puts a 640GB HD in a $3299 system to save maybe $20 cost when you can get 1TB HDs for $90 at retail.

Intel is going to do a major price adjustment in May and July of this year and introduce faster Core i7 chips. It would be interesting to see if Apple updates the entry level single processor Mac Pro to 2.93GHz and entry level the dual processor Mac Pro to 2.66GHz at that time while maintaining the price point.

I dont understand why they use xeons on the single cpu version. The core i7 is much faster with its usage of mush faster memorys. Apple is pricing it self out of the market, they will go the same way as the other major unix workstation companys.

 

[Anonymous Freak]

I dont understand why they use xeons on the single cpu version. The core i7 is much faster with its usage of mush faster memorys. Apple is pricing it self out of the market, they will go the same way as the other major unix workstation companys.

The single-socket "Xeon 3500-series" are Core i7's. They are Core i7 plus ECC RAM support. At the same clock speed they are the exact same speed as the equivalent Core i7. RAM speeds are the same, 1066 MHz for the lower-end parts, 1333 MHz for the high-end part. It's just that PC motherboard manufacturers allow you to select higher (Intel considers if overclocking,) RAM speeds, Apple doesn't. And, to my understanding (not breaking NDA here, I haven't touched any Xeon 3500-series stuff, so I only know what I've read on public sources,) the Xeons will be locked down to 1066/1333 anyway, no 'overclocked' 1600 or 1866 MHz RAM. (Intel's own X58/i7 motherboard, the DX58SO, supports up to 1866 MHz RAM as 'standard', even though the processor-side considers that overclocking.)

 

[jbl108]

MP or iMac?

Personally, I'd use the G5 instead of the iMac. The Mac Pro is a better option long road, since your internal upgrade options are a lot better. I am basically in the same boat, I have a dual 2Ghz G5, although I am just wondering if I should pull the trigger or wait. I would never buy an iMac. The iMac places too many limits for a pro in my opinion.

-mark

I am in a similar position, with similar concerns. I, too, have a 2GHZ G5, and boy am I ready for an Intel Mac! Ultimately, for better or for worse, many of us long-time Macophiles are probably more concerned with whether the MacPro is the right computer for what we want to do than with its pricing (or overpricing as the case appears to be). Perhaps some of you could help me (and others) with this decision. I will be doing some non-professional FCP editing of AVCHD footage, rendering, photoshop stuff, etc. Does the entry level MP really provide that much of an advantage over the top end iMac? How much faster for this would it really be? The benchmarks I have seen compare a iMac with 4GB of memory to MP's with 8-12. How much of part of the speed differences with rendering, etc is this? Would maxing out the iMac with 8 GB of RAM and the 4570 video card bring it close to the entry level MP for these things?

I certainly appreciate any input any of you might have.

 

[gnasher729]

Steinberg have made a official statement about their views of the new mac pro.
(Steinberg makes the music/audio software nuendo and cubase and are used in many
big studios and homes)

http://knowledgebase.steinberg.de/71_1.html

Maybe, just maybe, this is related to the fact that Apple bought Emagic and not Steinberg. So there will be a certain amount of sour grapes involved.

 

[gnasher729]

You don't understand caches, do you?

I don't believe it's worth the effort to continue this discussion about scheduling - you think it's simple, and Cambridge University considers it worthy of doctoral dissertations.

Well, instead of being sarcastic you can try to educate us all and tell us what else is involved in moving a thread from one one core to another core on the same processor. In your esteemed opinion. I am looking forward to it.

Of course Cambridge University will find this worth of dissertations. Creating dissertations is their business, the whole reason for their existence. This dissertation is a few years old, so a lot is common knowledge now that wasn't know back then. And Apple doesn't have to make hyperthreading work in any optimal way; all they have to do is make it work better than no hyperthreading at all. Since hyperthreading on new processors like Nehalem and POWER5 is much more efficient than on an old Pentium 4, that is not a difficult task.