NASA Benchmarks the New G5 Powermac 751
sockit2me9000 writes "Well NASA's Langley Research Center recently benchmarked the new G5 dual 2ghz Powermac against a dual 1ghz Xserve, a dual 1.25 ghz Powermac, a Pentium4 2 ghz, and a Pentium4 2.66 ghz. To make things fair, the second processor in the G5 was switched off, as well as the other dual sysytems. Then, they all ran Jet3d. Even with un-optimized code and one processor, the G5 performance is impressive."
Summary (Score:3, Insightful)
"Benchmarks from the scalar version of Jet3D are shown in Figure 1 (MFLOPS) and Figure 2 (MFLOPS normalized by MHz). In terms of raw MFLOPS, the 2GHz G5 is about 32% faster than the 2GHz P4, 97% faster than the 1.25GHz G4, 142% faster than the 1GHz G4, and within 1 MFLOP of the 2.66GHz P4."
Translation: Slower than the P4 for anyone who didn't look at the grid. And M stands for million. Not one.
Re:NASA Verifies Apple Benchmarks? (Score:4, Insightful)
First, 498 vs 255 is 95.3% (instead of 195.3%) advantage.
Second, why compare dual fastest G5 vs single mid-range P4? Singe 2GHz G5 scored 254 MFLOPS, it quite fast, but then again equally fast 2.66GHz P4 is available at $188.
It would be much more interesting to compare dual 2GHz G5 against dual Opterons and Xeons.
Only benchmarks that matter are.... (Score:5, Insightful)
Costs (Score:3, Insightful)
I'd read some thread a while back on another board saying that "Macs are cheaper than PCs". I still can't believe anyone would make that argument. Doesn't being really good in a few areas satisfy the mac people? Do they have to try to spin higher costs as 'lower' (craziest thing I'd ever heard...)
The G5 (Score:3, Insightful)
Their scheme for OS X is the equivalent of Microsoft charging $100+ for a service pack, I just don't understand it.
I've used OS X, and it blows everything else out of the water in elequency and it seems the perfect balance between productive and 'cool factor'
But until I win the lottery, I'll stick with my cheap x86 machines
Re:MFLOPS/MHz? No AMD, Old P4, Old Redhat. (Score:3, Insightful)
Why no Athlon? They probably didn't have one and with the P4 at 3ghz and climbing, the old althon is becomming less and less significant for these pure number crunching apps. Plus maybe they've done previous tests that said that P4 they uesd was faster than an equivelent athlon, so it didn't need to be tested.
Why no Opteron? They probably didn't have one. This is the most valid question you ask.
Why RH 7.1? That's probably what they use. They are benchmarking PURE CPU so the OS doesn't matter too much for this kind of thing.
Re:The G5 (Score:5, Insightful)
Microsoft Windows XP Pro Upgrapde: $199
http://shop.microsoft.com/Referral/Productinfo.as
MacOS X 10.3/2/1 Full price: $129
http://www.apple.com/macosx/
Microsoft Windows XP Pro (5 Users): $1315.60
MacOS X 10.3/2/1 (5 Users): $199
If you bought Windows XP ($299), and then can upgrapde to Longhorn for $199, you paid $498. If you bought MacOS X 10.1, 10.2, 10.3, and 10.4, you paid $516. Pretty similar, and that's assuming you only have to pay $199 for Longhorn. In the meantime, Apple users enjoy continued advance, while Windows stagnates for 4+ years.
Do the same with a family licence of 5. Buy Windows XP for $1315.60, then upgrade for $875.60: $2191.20 (over 4 years, for 5 people: $109.56/user/year).
Buy MacOS 10.1, 10.2, 10.3, 10.4 (5 User Licence): $796 (over 4 years, for 5 people: $39.80/user/year).
Using http://shopper.cnet.com I found a copy of Windows XP Pro for $207, and an upgrade for Windows XP Pro for $177. I found a copy of MacOS X 10.2 for $98.
If these prices hold over to the newer Operating Systems these companies release, then Windows would cost $384 (23% savings), and MacOS X would cost $196 (24% savings). If you bought every point upgrade Apple released it would cost $392.
Dak
Wrong. (Score:4, Insightful)
The major OS version updates are when new features are added, etc. That is the equivelant of upgrading 98 to XP. The cost of buying a Mac is high. The cost of maint. is probably less than a Windows box.
MFLOPS/Mhz. - Useless Metric (Score:4, Insightful)
One might be tempted to design a chip that does more in one cycle and then clock it as fast as a chip that does less in one cycle. Unfortunately, while reality is a little more complex than this, the basic reason is that the more a chip does per cycle, the more heat it generates per cycle. If you try to squeeze too many cycles through it in a second it will fry.
So showing that the G5 has better performance per clock cycles is no more useful than showing that an AMD chip has better performance per clock cycle than an Intel chip. All that matters is how much performance you can get from a chip before it cannot be clocked any faster without requiring unreasonable cooling methods.
All this paper shows is that, while the G5 is designed to do more in a clock cycle than a P4 is, the chip tested is ultimately not any faster than the P4 they benchmarked it against. It remains to be seen how the G5 will do at higher clock speeds. With this in mind, it would be *far* more useful to see heat dissipation stats on the G5 since that might give us some idea how close to it's design limits. If it is cranking out high-end P4 performance and running cool *then* I will be impressed.
Why No Cray? (Score:5, Insightful)
Re:Linux is Dying (Score:2, Insightful)
Re:Costs (Score:1, Insightful)
Re:Costs (Score:5, Insightful)
The difference in speed mHz, RAM etc. is irrelevant for a person just getting used to computers and the net. When she is having weekly problems with the Siemens / Windows machine, it will be lost money in time. While the Mac is cheaper in usage, because of less "frustration time" and less hassle.
This is an argument I would strongly disagree with, if you asked me two years ago. But since then, I have come to the conclusion that the Mac simply work better for the lay people. It does the work, and faster because there are less frustrations and less hassle.
Re:G5 is really a full-blown workstation (Score:1, Insightful)
That you're talking out of your ass.
Re:Costs - correction (Score:2, Insightful)
How much for the dual Xeon system they were comparing that too? Yeah, you can build a P4 for $900, but not a dual Xeon.
However, yes, the $1999 low-end G5 is definitely more expensive than a P4 of similar performance, if the only consideration is raw speed. Macs are cheaper than PCs at the high end, not the low end.
Re:NASA Verifies Apple Benchmarks? (Score:5, Insightful)
Vector Performance (Score:5, Insightful)
This would seem to be one of the more interesting points made, actually. Prior to the announcement of the G5s, speculation on the PPC 970 suggested that it would be stellar with FP & so-so with integer; the real question surrounded how well IBM would implement SIMD. Many were pessimistic. Given that it seems like they've managed to add it efficiently a scaled-down POWER4 core, future refinements could make this series of chips (PPC 9X0s) real monsters.
But the future viability of that roadmap (given how ruthless the company as a whole tends to be when faced with departmental money losses) depends as much upon the success of IBM's Linux strategy as it does on its success in the PowerMac line.
[With apologies to BadAndy of the Ars Technica boards; thanks for sharing your insights.]
In all fairness... (Score:5, Insightful)
Re:The G5 (Score:1, Insightful)
all updates (let's say service packs for OS X) are free, it's the serious upgrades that cost. and nobody forces you to buy them anyway.
for me the mac is still the cheapest pc around.
This typed on a perfectly sensible orange iBook.
5177 MFLOPS 288 MFLOPS (Score:5, Insightful)
"The vector version of Jet3D runs an order of magnitude faster than the scalar version (speedups of 10X-13X are typical)." The dual 2GHz G5 was benchmarked at 5177 MFLOPS (a 1040% increase over the scalar test) and 1.29 MFLOPS/MHz."
5177 MFLOPS when running a Velocity Engine optimized version of Jet3D.
Now, how much does an P4 extrapolated to 3.2 GHz get? Like 288 MFLOPS?
Someone please explain to me how 5177 MFLOPS and ~300 MFLOPS are even comparable.
As the Mathematica guy said, the competition is no longer high-end PCs, it's now $10,000 UNIX workstations...and the G5 is still faster than any of them.
No wonder the G5s smoke the dual Xeon in the Photoshop, Mathematica, Logic, and Luxology app bake-off. All these apps would have been optimized to use the Velocity Engine.
If I were a scientist doing lots of image processing and vector calculations, I'd need a cluster of about 18 or so 3.2 GHz P4 machines to keep up with the dual 2 GHz G5 PowerMac running a typical Velocity Engine optimized app.
That's a sweet 5177 MFLOPS for you - evidence the G5s rock as hard as Apple has been indicating.
64 bit. (Score:2, Insightful)
Has anyone seen any such relivant benchmarks?
Re:The G5 (Score:1, Insightful)
WinXP Pro Family of five for four years: $0
MacOS-X Family of five for four years: $796
Savings with Windows: $796
Re:fortran compiler (Score:2, Insightful)
"Note that the higher level of optimization (-O2) and SSE/SSE2 options in the Portland compiler degraded Jet3D performance on the P4 system, and were therefore not used."
Thus they are comparing the FP performance of the old x87 style, stack-based interface on the P4 - not to mention they are just using minimal optimization. When they start talking about how great vector peformance is, they need to be using SSE2, which is both modern and far more efficient than x87, plus SSE2 is essentially an implementation of a short-length vector processor, somewhat like altivec is on the G4/G5 chips.
This "report" sure smacks of the same bogosity that the original Apple benchmarks did - essentially hamstringing the competition before making a comparison. Since you can't *buy* a dual G5 system today, one must assume that Apple gave it to NASA and you can bet that early access comes with an NDA. So, Apple must have either approved or at least been involved in the generation of these results. Essentially making them just as biased as Apple's own.
Re:Interesting choice of processors (Score:2, Insightful)
really, do you think NASA's in on some conspiracy to hype Apple hardware? Good grief. I couldn't imagine any business, but a benchmarking business, approving an order for a bunch of new hardware solely for the purpose of conducting some mythical-the-truth-is-out-there, guaranteed, unbiased benchmark. There's nothing interesting about the choice of processors. It's what was lying around when the demo came in, fool.
Re:NASA Verifies Apple Benchmarks? The Future (Score:4, Insightful)
If, as widely reported, the PPC 970 goes from 2GHz to 3GHz in the next 12 months it will definitely be more than competative.
It has been along time since I've seen performance increase by near 50% in a year. Takes me back to the 486DX2-66 days.
Re:Costs (Score:3, Insightful)
Are you honestly claming that 129 is a higher number then 255? (fastest g4 to the fastest p4) Yeah, the g4 had a slightly higher flop/cycle score (that would have been much lower using the intel fortran compiler) but the p4's over all score was much higher.
When comparing the price of a 32bit system to that of a 64bit, it's a bit pointless...as you're getting so much more with the 64bit chip.
Um, no, you're not. For one thing these benchmarks were all done with floating point numbers, the Intel architecture has supported 64 bit floats since the 287. A 64 bit CPU won't have much of an advantage over a 32 bit CPU at all. Secondly, a 64 bit CPU can't do anything that a 32 bit CPU can't. It just needs to take more cycles. But that only happens in the rare case when you have to deal with a non floating point number which is larger then 2^32, or ~4.3 billion. That doesn't happen very often.
Re:Damn Dude, RTFA (Score:5, Insightful)
1. They weren't using a compiler optimized for the G5, and expect performance to increase when they have that opportunity.
2. Dual G5s appear to scale better than dual P4s. They're getting close to 2x performance with the dual G5s, much better than most folks are used to with SMP systems.
Re:Single Processor Mode (Score:3, Insightful)
If you can think of a good reason to turn off 50% of your processing, why not save a lot of money and go for one of the two mid-range single cpu G5 configurations that Apple will happily sell you?
By testing in this ludicrous 'half a machine' mode, we don't get any of the potentially really interesting information about the efficiency of apple's dual cpu hardware and OS support, and how much difference their 1Ghz bus makes compared to intel/amd offerings. It's this area which will make or break the claims of 'fastest PC'.
Re:Single Processor Mode (Score:3, Insightful)
I also remember a few games having issues on dual processor computers, a while back. Though most of these issues have been resolved now, it'd be interesting to see if the problems cleared up if you disabled a cpu.
Re:Costs (Score:2, Insightful)
Dual 2GHz PowerPC G5
2GB DDR400 SDRAM (PC3200) - 2x1GB
250GB Serial ATA - 7200rpm
ATI Radeon 9800 Pro
Apple Cinema Display (20" flat panel)
SuperDrive (DVD-R/CD-RW)
Apple Keyboard & Apple Mouse - U.S. English
Mac OS X - U.S. English
APP for Power Mac (w/ or w/o display) - Enrollment Kit
$5693.00
versus
Dell Dimension XPS Series
Pentium® 4 Processor at 3.2GHz with 800MHz front side busPrice: $4,268.00
SAVE $190! (Savings included in price) 2GB DDR SDRAM at 400MHz
Dell ® Quietkey ® Keyboard
SAVE $160! (savings included in price) 20.1 in Digital Flat Panel Display
New 128MB DDR ATI RADEON(TM) 9800 Graphics Card
200GB Ultra ATA/100 Hard Drive (7200 RPM) with DataBurst Cache(TM)
3.5 in Floppy Drive
Microsoft® Windows® XP Professional
Dell(TM) Optical USB Mouse
New Dell Gigabit Ethernet
16 Max DVD-ROM Drive
Sound Blaster Audigy 2(TM) sound card with DVD Audio
SAVE $60 (system price shown before rebate) 4 Yr Ltd Warr plus 4 Yr At-Home
Dell Movie Studio Plus with Roxio VideoWave Movie Creator(TM)
FREE UPGRADE! New 4x DVD+RW/+R Drive w/CD-RW
If we first add back in all of the "free" stuff DELL is giving by ordering online....$190 + $220 +$170 = $580 +$4268 = $4848 which is a more realistic total considering that I doubt NASA does their ordering online.
Now if we compare the 2 prices ($4848.00 DELL and $5693.00 APPLE), unless I am doing math poorly these days, we see a 15% savings by purchasing the DELL system.
HOWEVER....
What always seems to be forgotten is the cost of SUPPORTING these machines. The support costs is what makes the Mac shine when it comes to TCO. I cannot tell you how many times I have seen IT departments in large corporations that have 10+ Windows support people in Department X that are ALWAYS busy running around fixing Windows machines, compared to their Mac support group (generally about 1/2 or less the people) with an equal size department to support.
The support costs are where the Mac just totals Windows boxes over the useful life of the machine....
What most people FAIL to realize is that the upfront costs, != the total cost
Re:Single Processor Mode (Score:5, Insightful)
Given that the 2 processors are (presumably) sharing 1 video card, one set of RAM chips, one I/O system, and one disk drive, surely there must be *some* kind of performance hit from sharing the rest of the machine with a second processor? Or some performance drain from putting all the housekeeping talks onto just one of the 2 processors?
I think this adds weight to my original point - that benchmarking a dual G5 against dual amd or intel systems is the only way we'll settle this 'fastest PC' question, and that arbitrarily shutting half the sysem off isn;t a sensible way to test things.
Re:Interesting Thought (Score:2, Insightful)
Frankly, us Mac "fanboys" could care less if someone could pick up a $499 4 Ghz P4. The people touting these systems are driving rusted out Ford Escorts with the bumpers held on by duct tape (or drooling over the Ford Escorts while looking out of the windows of the loser-cruiser)
Seriously now, Mac/OSX fanboys like their computers. They don't care if it is the absolute fastest or the most/least expensive. The hardware has some style and the OS simply rocks.
Re:Costs (system, not processor...) (Score:2, Insightful)
160GB Serial ATA, SuperDrive (CD-RW, DVD-R), Firewire 800, USB 2, ATI Radeon 9600 Pro...
Plus all the standard things that apple don't actually mention in their specs any more of course. 16 bit sound, gigabit ethernet, and so on.
And that's without the free software. iMovie, iCal, iPhoto, iDvd, etc., (which you may or may not want, so adjust value of system accordingly).
The thing with your comparision, is that you are comparing the G5 processor with the P4 processor. Now try comparing the G5 system to a PC system with a P4 with equivilant specs. Hard drive, DVD writer, latest firewire, gigabit ethernet, etc, and you soon see the PC cost more.
And that's without touting the advantages of Mac OS X and all the other points that have already been made in this thread...
This is Apples advantage and disadvantage. Because they won't sell you anything less than top quality in their pro line... And that costs!
People seem to miss the big picture some times!
Re:fortran compiler (Score:5, Insightful)
Re:Wha? (Score:5, Insightful)
By a whopping 0.4%, and with one of the G5's processors disabled. You can spin it any way you want, but the clear fact is that with the G5 Macs are competitive in CPU performance again. I don't see why this disturbs you so; competition is good.
Re:Single Processor Mode (Score:4, Insightful)
My take.
Since everything inside the G5 is "point to point" then unless both CPUs need to talk to exactly the same thing at exactly the same time then the performance hit will be almost impossible to detect. (I wonder how RAIDing drives could make this overhead even more improbable?)
Since there is a lot of Mach-ness in the OS X kernel I would also think that there aren't too many critical regions there that force the use of only one CPU. In other words, I think the situation is better than that kernel (was it an early Solaris?) that had one mutex around the whole kernel.
As usual, anyone willing to pay me to benchmark one of these puppies for them? :-) The machine is suitable payment.
Re:NASA Verifies Apple Benchmarks? (Score:2, Insightful)
If you read the article the benchmarker wanted to know the performance levels of the different systems (note: not CPUs).
He did say that Pentium were probably more cost effective.
PS: Where can I buy a complete system with a 2.66GHz P4 for $188? is that U.S. or some other type of dollar?
Re:MFLOPS/MHz? No AMD, Old P4, Old Redhat. (Score:3, Insightful)
Puh-lease... The gripe with the SPEC benchmarks was that Apples numbers for the competition were WAY below the OFFICIAL numbers, ot that Apples numbers for their own equipment was crap.
Jeeze...let's at least wait till these things are SHIPPING.
Re:No (Score:3, Insightful)
the hell you can (Score:1, Insightful)
Re:NASA Verifies Apple Benchmarks? (Score:5, Insightful)
You wanted to say year ago? Month ago Intel released 3 GHz, 800 MGh FSB P4 and new chipset for dual DDR 400 motherboards. Comparing G5 against older 533 MGh FSB processors on old chipset is hypocrisy.
Even 2.6 GHz 800 MGh FSU processor with i875 chipset-based motherboard and dual DDR 400 memory behaves much better than that 2.66 GHz processor.
Re:They left out an important graph. (Score:2, Insightful)
Re:Single Processor Mode (Score:4, Insightful)
Re:Still not the same comparisons... (Score:3, Insightful)
He did not say the first 64-bit computer or 64-bit server but 64-bit personal computer.
Sheesh.
Alpha's and Sparcs are not cheap and are considered servers . You do not go to CompUSA and buy a digital Alpha server or Ultra 10x workstation. THey are certainly not designed for the average joe and the price reflects it.
Their is no software for them that is not specialized anyway. They are specialized workstations for running a particular set of apps. No photoshop or quakeIII for the kids.
FYI I am not a mac user.
Re:5177 MFLOPS 288 MFLOPS (Score:1, Insightful)
The dataset I ran the test on was about 500MB in size, thus easily fitting into the RAM.
The simulation does two major things: It gives a simple visualization that it outputs to high-resolution but graphically simple pictures, for easy analysis. The second is that it dumps all the stuff to a log-file, saving coordinates, motion curves, dynamics states etc, that can be imported into Maya for prettier visualization.
Thus it's fairly dependant upon HD too, which is one place where the speed problems could be, since they are using MSI motherboards(bleurgh).
Another thing is the compiler, GCC. I compiled for x86-64 with the switches -m64 -O2 -msse2 -mpreferred-stack-boundary=4(Which sets it to 16 bytes which SSE/SSE2 needs) -mno-push-args -maccumulate-outgoing-args -mtthreads, but I don't think GCC properly read the stack boundary size. The Opterons should be faster than the result I got.
For the G5, the most important switches I used were -mpowerpc -O2 -maltivec -mnew-mnemonics -mabi=altivec.
Re:Single Processor Mode (Score:2, Insightful)
Ummm, nice try butt chunk, but the G5 for $3k is DUAL PROC. Please try again, this time with a dual proc intel machine. The one you spec'd out would get trounced by the 2x2GHz G5. Give it another go.
And BTW - those pricewatch "lowest prices" things are usually a crock - they usually have like $20 shipping fees to make up for the lower prices. Please factor those in too. Thanks. Still want to do this exercise? I didn't think so...
Grow up! (Score:1, Insightful)
Re:Costs - correction (Score:2, Insightful)
Does that include most of the hardware features of the G5? Gigabit Ethernet, FireWire, optical 5.1 audio, CD-RW/DVD-R drive?
A friend of mine ran the numbers on Pricewatch last week and didn't have the same results you did. I don't recall what he found, but for similar specs it wasn't cheaper than the G5.
Straight forward (Score:2, Insightful)
- A single G5/970 top of the line is probably 20% slower than P4 top of the line, on unoptimized software. By that I mean regular integer and floating point (no AltiVec), and with unoptimized compilers. The compilters may or may not improve significantly over time, so that shouldn't be afactor on release date, but may pan out later.
- The G5/970 is not designed/optimized to ship as a single CPU, instead the bus, etc. are designed for SMP of between 2 and 8 CPUs (might technically work with more, I don't know). So, you'll see that, in a dual CPU comparison that the G5/970 will scale better than the P4 (xeon) duals, narrowing the performance gap, and in some applications, putting the G5/970 on top.
- With software that is recoded to work with AltiVec, compared to SSE(2) optimized software, the AltiVec performs noticably better, but that depends on the application. When comparing AltiVec to regular (non-vector optimized code), there can be as much as an order of magnitude increase in speed, again depending on t6he application.
So, expect benchmarks to vary by a tremendous amount, depending on the number of CPUs, and how optimized for AltiVec the software is, and how mature the compiler that was used. Since a lot of software can be vectorised, I would expect software on the G5 to initially lag P4 performance, but then to dramatically speed up in some areas.
Also, expect a lot of people who don't grasp this to be getting their panties in a knot!