iMac 1.8 Ghz G5 Benchmarks

[ki-goi]

What I have a problem with is the fact that the design was intentionally made to look like an ipod. In order to achieve that, chipset cooling might have been sacrificed.

i think this is wrong. if you think of the imac as an xserve stuck to the back of a flat monitor, then there's nothing about the "pretty" requirement that forces the machine be slow. the screen is smaller than the innards; the screen wants to be a certain height, for user comfort; the machine can't be 8 feet tall; so, there's a big wide space under the screen. also, the logo can be bigger, as per marketing fashion.

ever since the real 2nd generation imac - "slot-loading" - was released with its convection cooling setup, a big focus of the imac's design has been making it quiet. something that can sit in a public place and not intrude. this is not a new design requirement, so there's no reason to expect that this imac is faster or slower than it was hoped to be, unless somebody said so, on or off the record...

 

[Converted2Truth]

Oh my God. So you're saying that I ordered a 20" iMac by mistake, without realising that the graphics chip contained therein wasn't up to snuff for running the games I want to run? Oh my god! I'd better get on to my supplier and cancel the order straight away! Thanks so much for pointing out my mistake to me!

Newsflash for you. Not everybody buys computers to play games. Not everybody who buys computers has children. If I want to play games, I have a Gamecube which does perfectly fine, thank you very much. In any case, the games that I happen to enjoy don't particularly benefit from a 3D graphics accelerator; I'm talking about games like Starcraft, Diablo, Railroad Tycoon, Pikmin, FIFA 2004, Eternal Darkness, that sort of thing. Not first person shooters, as a general rule. Yes, ok, most games these days take advantage of a 3D accelerator, because it's assumed hardware, but apart from FPS, how much benefit is to be had by buying a faster accelerator? Not much (if any). I pretty much guarantee: 99% of non-FPS games will run just fine with the built-in GPU on the iMac.

No, I'm very happy with my purchase. It will do exactly what I want it to do, and very capably thank you very much. The CPU grunt will be adequate, as will the RAM (once I've upgraded it from stock levels), and I don't need a massive hard drive. I don't even particularly care if the LCD is of the same or lesser quality as the ACD 20", as long as it is clear enough -- I'm not looking for a screen capable of doing photographic quality work. If I were, I'd be buying a PowerMac and hooking it up to a high quality CRT. Hell, I don't even care if the pixel refresh is sub or super 20 ms, as long as it's not of the order of 50 ms. And this computer isn't really an "investment" in the traditional sense of the word. It's a tool. It does what it has to, and it does it well. That's all that matters to me, provided it's priced reasonably (and the iMac definitely is.)

Look at the capabilities of the GPU here. It's streets ahead of what most non-gaming people need. If you need more GPU than that, you're either a gamer (in which case, you're probably better off with a PC anyway, to be brutally honest), or somebody who does serious design work (in which case, you're better off with a PowerMac). Get a bit of perspective here, get a grip, and quit bitching about problems that, when you get right down to it, aren't all that serious in the end anyway. Anything you can buy is obsolete anyway -- it's a simple fact of life.

If apple were to release the equivilant of a 486 33sx with monocrome video out, you'd be the type of guppy they'd need to make a profit. Just say "MOOO" and follow the heard...

 

[Converted2Truth]

This is so silly !! Video card selection and video performance are NOT the ONLY reasons to buy a new iMac G5. If you are dissatisfied with the 5200 video card, then buy a PM G5 and pick your own video card or buy a Compaq, Sony, or HP "all in one". There is some real innovative design, engineering, and manufacturing (what a joke) !! I guess most of the complainers are gamers only.

Put yourself in Apple's position where performance and price are BOTH factors in the roll out of this machine.

Consumers who would like to play the occasion network game of Halo on a mac SHOULD NOT BE FORCED TO SPEND $2500 ON A POWERMAC! Put myself in apple's position? **** YA! SPEND AN EXTRA $10 TO PUT THE 9600PRO, AND GAIN THOUSANDS OF MORE SALES! You have no idea what you're talking about. Gamers don't demand the fastest, but they do demand 'playable'. this new imac sucks for any recent game, and will NOT play new ones. Sims2? forget it (when it's ported)!

 

[Converted2Truth]

...Sure I'd prefer a better card, but since a 19 month old G4 iMac could do everything I wanted it to I'm pretty sure the brand new G5 iMac will too...

This is where you're wrong. If you plan on playing any FUTURE games, this card will leave you wanting. Then again, if you don't plan on playing ANY future games, then go ahead and spend a ton of money on something that you don't need. You said it yourself that you're old imac does everything you want!...

 

[nospleen]

Here are the results for my new 20" imac with 1.25 gb of ram. I am not sure if these numbers are good or not. But, I have owned lots of macs and this thing runs great. I am more impressed than I thought I would be!



Results 154.36
System Info
Xbench Version 1.1.3
System Version 10.3.5 (7P35)
Physical RAM 1280 MB
Model PowerMac8,1
Processor PowerPC G5 @ 1.80 GHz
L1 Cache 64K (instruction), 32K (data)
L2 Cache 512K @ 1.15 GHz
Bus Frequency 600 MHz
Video Card GeForce FX 5200
Drive Type ST3160023AS
CPU Test 159.28
GCD Loop 96.86 3.78 Mops/sec
Floating Point Basic 284.16 1.03 Gflop/sec
AltiVec Basic 123.59 3.59 Gflop/sec
vecLib FFT 151.05 2.34 Gflop/sec
Floating Point Library 352.73 14.12 Mops/sec
Thread Test 102.45
Computation 67.84 915.77 Kops/sec, 4 threads
Lock Contention 209.16 2.63 Mlocks/sec, 4 threads
Memory Test 217.71
System 245.50
Allocate 659.36 430.10 Kalloc/sec
Fill 204.18 1625.25 MB/sec
Copy 172.25 861.25 MB/sec
Stream 195.56
Copy 171.80 1255.89 MB/sec [G5]
Scale 169.55 1251.27 MB/sec [G5]
Add 222.47 1423.78 MB/sec [G5]
Triad 235.84 1441.01 MB/sec [G5]
Quartz Graphics Test 203.94
Line 190.80 4.86 Klines/sec [50% alpha]
Rectangle 183.07 12.88 Krects/sec [50% alpha]
Circle 199.15 4.59 Kcircles/sec [50% alpha]
Bezier 186.36 2.03 Kbeziers/sec [50% alpha]
Text 291.85 4.76 Kchars/sec
OpenGL Graphics Test 207.52
Spinning Squares 207.52 145.22 frames/sec
User Interface Test 239.21
Elements 239.21 76.94 refresh/sec
Disk Test 92.47
Sequential 87.38
Uncached Write 87.90 36.64 MB/sec [4K blocks]
Uncached Write 69.96 28.65 MB/sec [256K blocks]
Uncached Read 81.14 12.84 MB/sec [4K blocks]
Uncached Read 128.48 51.91 MB/sec [256K blocks]
Random 98.20
Uncached Write 88.01 1.32 MB/sec [4K blocks]
Uncached Write 100.44 22.65 MB/sec [256K blocks]
Uncached Read 94.31 0.62 MB/sec [4K blocks]
Uncached Read 113.47 23.35 MB/sec [256K blocks]

 

[tex210]

The future Conan?

This is where you're wrong. If you plan on playing any FUTURE games, this card will leave you wanting. Then again, if you don't plan on playing ANY future games, then go ahead and spend a ton of money on something that you don't need. You said it yourself that you're old imac does everything you want!...

Well if that's the argument...
you really shouldn't get anything other than a dual core/dual processor machine, as anything less will be unable to run the artificially intelligent O.S. 11.
p.s. A ton of money, really?

 

[Telomar]

L2 Cache 512K @ 1.15 GHz
Spinning Squares 207.52 145.22 frames/sec

Two things I've noticed in Xbench. First the L2 Cache on the new iMacs doesn't appear to show as running at processor speed, which it should be.

Second there seems to be a peak on OGL scores at around 150 fps. Whether that's a constraint based on Apple's implementation or something else it's interesting but would certainly explain the poor OGL scores in Cinebench for Macs. I'd be very interested to see somebody's OGL scores using either a Radeon 9800 or a Geforce 6800, when it comes out.

 

[aliasfox]

My rev A 867 MHz PowerBook G4 12" that is 20 months old has been able to play Halo and Unreal Tournament 2003- albeit at a medium-low quality. The iMac G5 will be fine for the occassional gamer- remember, based on rough specs, it's roughly equivalent to a 2.5 GHz P4 running a GeForce3 generation card- and most people don't complain about a machine with that setup.

Yes, Apple could have put a 9600 class card in the iMac. Would it really have boosted sales? Considering most people buy Dells with Intel Integrated graphics, I don't think the average Joe knows or cares too much. So they would have spent (for example) an extra $15 over (about) 800,000 machines a year, meaning an extra $12,000,000 outlay to put a better graphics card in. Now, if Apple makes an average of $400 off of selling an iMac (this is an average of the low end and the high end and direct and resale channels), Apple would have to sell an additional 30,000 units based on the fact that the iMac has (say) a 128 MB 9600 XT vs a 64 MB 5200 Ultra alone. Apple's accountants probably made the conservative judgement and decided not to. Case closed. If you don't want it, wait for someone to come out with a midboard modification, a la original iMac and Mezzanine slot.

 

[BigEvan23]

CINEBENCH COMPRO!

Contenders:
20 inch G5 iMac
1.8ghz
1GB Ram
160GB HD
NV 5200 Ultra 64mb
Airport Express

--vs--

15inch Powerbook G4
1.5Ghz
1GB Ram
80GB 5400rpm HD
Radeon 9700 128mb
Airport Express

the iMac destroys it!

Btw avg xbench score for the powerbook with 5400rpm drive is 135, avg for the 20 imac is 162.


I think the iMac is pleanty powerful for con and quasi-prosumer use!

 

[AmigoMac]

It seems to me, you don't get it. It is the same with iMovie, Dashboard, Photoshop, Mac OS X itself, maybe Safari, lot of games, whatever.

BTW: Motion is a prosumer App with it's low price and Final Cut Express runs even fine on an iMac G4, so what?

This graphics card is just outdated, but the rest of the iMac is a very powerful computer.

Just 60$ more for the end user is the BTO option for the Powermac Radeon 9600XT with 128.

Or even more simple: Don't solder the GPU on the mainboard... So later an upgrade board can be made. Already fine would be, if it's only a newer low end gpu next year and maybe later another one too.

Dude, you don't get it/me when I end a whole sentence with ...
until now I have seen 2 posts claiming that they could use the iMac for pro apps, even there is one claiming Cinema4D use... God! those posts are on the wrong side of the equation ... iMac = Home use ... period.

iMovie won't be a problem with the GPU if you have enough RAM... (Only here I agree with the 512 MB standard petition, is that hard steve?)

Dashboard... ... it will be a piece of cake for any GPU in actual apple computers ...

Photoshop (There are a lot of posts claiming a good photoshop behavior in G3 models) ... Those who think this is sh*t, should look for Powermac ...

Apple is not guilty because of the lack of money of some pro users... I don't think iBooks users complain that high because there is no ATI 9700 in there...

Safari won't give the GPU that load, Exposé wont be different as soon as Tiger comes out... I do think that we are within a big GPU myth because of some pro users experiences, the next revision may give 128 MB option but 64 will be ok for the 4-5 years, we won't see Mac OS 10.5 soon...(Apple statement), come on, there a lot of iMac G3 users out there ... with 16 MB VRAM...

My feeling : The iMac G5 is a kick ass machine and will kick an iMac G4 away, but that power is needed in less than 25% of the market target... actually I wanted to buy a G4 yesterday in ebay but was too late... so I will get the G5 , it will be with me for the next years until 10.5 comes, I had that target with my actual eMac but it's pretty loud at night and can't run continuously as wanted...

 

[AmigoMac]

Contenders:
20 inch G5 iMac
1.8ghz
1GB Ram
160GB HD
NV 5200 Ultra 64mb
Airport Express

--vs--

15inch Powerbook G4
1.5Ghz
1GB Ram
80GB 5400rpm HD
Radeon 9700 128mb
Airport Express

the iMac destroys it!

Btw avg xbench score for the powerbook with 5400rpm drive is 135, avg for the 20 imac is 162.


I think the iMac is pleanty powerful for con and quasi-prosumer use!

I'm very happy and look forward to get one of those 17" babies soon

 

[the dairy giant]

Contenders:
20 inch G5 iMac
1.8ghz
1GB Ram
160GB HD
NV 5200 Ultra 64mb
Airport Express

--vs--

15inch Powerbook G4
1.5Ghz
1GB Ram
80GB 5400rpm HD
Radeon 9700 128mb
Airport Express

the iMac destroys it!

Btw avg xbench score for the powerbook with 5400rpm drive is 135, avg for the 20 imac is 162.


I think the iMac is pleanty powerful for con and quasi-prosumer use!

megahertz is 20% higher, test score is about 20% higher.

Is there no advantage to it being the next generation chip with much faster bus? Seems like a 1.8 G5 would be about that same as a 1.8 G4 if such a thing existed. Which seems a bit disappointing to me (in a theoretical way -- probably still going to get one to replace my 400 G3!)

 

[AmigoMac]

What we need is a comparison between the G5 and the last G4 ... they have the same GPU, under the same conditions (RAM, HD, Screen size), at that point we can see if it is the 30.55555% faster, if so is really fine, great, everything over that is nice! someone with those results already?

I would guess 50% faster... for that price...

 

[groovebuster]

Here are the results for my new 20" imac with 1.25 gb of ram. I am not sure if these numbers are good or not. But, I have owned lots of macs and this thing runs great. I am more impressed than I thought I would be!

Results 154.36

Actually I am not impressed at all by these numbers! Only 154? My old G4 Quicksilver with a 1.2GHz (single!) processor upgrade gets a 132 with totally outdated technology.

I considered the new iMac as a replacement for that Quicksilver in my home office, but just a new TFT will do now. I really thought the G5 iMac would wipe the floor with my Quicksilver. I am a little bit disappointed...

Results 132.65
System Info
Xbench Version 1.1.3
System Version 10.3 (7B85)
Physical RAM 1536 MB
Model PowerMac3,5
Processor PowerPC G4 @ 1.20 GHz
Version 7455 (Apollo) v3.3
L1 Cache 32K (instruction), 32K (data)
L2 Cache 256K @ 1.20 GHz
L3 Cache 2048K @ 4.11 GHz
Bus Frequency 134 MHz
Video Card GeForce2 MX
Drive Type Maxtor 6Y120L0
CPU Test 141.61
GCD Loop 137.39 5.37 Mops/sec
Floating Point Basic 131.96 477.21 Mflop/sec
AltiVec Basic 146.77 4.26 Gflop/sec
vecLib FFT 150.19 2.33 Gflop/sec
Floating Point Library 143.27 5.73 Mops/sec
Thread Test 103.70
Computation 75.57 1.02 Mops/sec, 4 threads
Lock Contention 165.19 2.07 Mlocks/sec, 4 threads
Memory Test 115.37
System 140.44
Allocate 756.91 493.73 Kalloc/sec
Fill 99.91 795.25 MB/sec
Copy 99.69 498.46 MB/sec
Stream 97.89
Copy 93.59 684.13 MB/sec [altivec]
Scale 95.43 704.26 MB/sec [altivec]
Add 100.36 642.33 MB/sec [altivec]
Triad 102.74 627.71 MB/sec [altivec]
Quartz Graphics Test 147.34
Line 126.89 3.23 Klines/sec [50% alpha]
Rectangle 115.79 8.15 Krects/sec [50% alpha]
Circle 140.43 3.24 Kcircles/sec [50% alpha]
Bezier 150.49 1.64 Kbeziers/sec [50% alpha]
Text 273.72 4.46 Kchars/sec
OpenGL Graphics Test 125.63
Spinning Squares 125.63 87.91 frames/sec
User Interface Test 231.55
Elements 231.55 74.48 refresh/sec
Disk Test 120.01
Sequential 123.63
Uncached Write 120.05 50.04 MB/sec [4K blocks]
Uncached Write 118.61 48.57 MB/sec [256K blocks]
Uncached Read 138.15 21.87 MB/sec [4K blocks]
Uncached Read 119.68 48.36 MB/sec [256K blocks]
Random 116.60
Uncached Write 152.52 2.29 MB/sec [4K blocks]
Uncached Write 122.85 27.71 MB/sec [256K blocks]
Uncached Read 94.56 0.62 MB/sec [4K blocks]
Uncached Read 110.71 22.78 MB/sec [256K blocks]

 

[will]

A few observations

I don't think anyone else has mentioned this, but two of the PowerMac models come with NVIDIA GeForce FX 5200 Ultra cards (64MB of DDR SDRAM), only the high-end model defaults to a ATI Radeon 9600 XT (128MB of video SDRAM). Yes, you can upgrade them, but if Apple had made the iMac a 9600, it would have had to bump the specs on the PowerMac. The PM will hopefully get a major revision in the January 2005 timeframe (PCI Express graphics, anyone?), by which time Apple should have had more of a chance to overcome heat dissipation issues of higher specification graphics cards in the iMac. I'd expect the next iMac revision, released before Tiger, to have 128M graphics card. (All these revision timescales will be affected by how fast IBM can crank out faster G5s, but Apple have traditionally updated about every six months).

XBench is interesting software, but, and this is a big but, it is in no way a proxy for actual performance. All benchmarks have weaknesses, but XBench is a particularly weak proxy for actual performance, as it's an arbitrary conbination of synthetic benchmarks. Benchmarks based on actual applications are much more useful. For example, I use Photoshop CS on a 867MHz G4, it would be good to know how much faster it's likely to be on the iMac G5. http://www.barefeats.com provides useful information, using actual applications, on Mac performance.

 

[Raveny]

the result of cinebench is that the iMac rendering (CPU) is 85% faster than the 1,5 GHz G4 Powerbook. And that's pretty cool I think.. Even the open GL results are better than the ones from the powerbook. So the imac will have more fps in Unreal than the powerbook...

 

[nospleen]

Actually I am not impressed at all by these numbers! Only 154? My old G4 Quicksilver with a 1.2GHz (single!) processor upgrade gets a 132 with totally outdated technology.

I considered the new iMac as a replacement for that Quicksilver in my home office, but just a new TFT will do now. I really thought the G5 iMac would wipe the floor with my Quicksilver. I am a little bit disappointed...

I personally think these benchmarks are useless. I have owned a dual 1.42 and a dual 1.8. I can honestly say that this imac is just as quick as my old dual 1.42, regardless of what the numbers say. I am looking forward to the barefeats tests though. But, trust me, this mac would be a huge upgrade for you. I was not expecting it to be this quick.

 

[Synthdance]

Hello, this is my first post to this forum, but I hope it's not the last one

I am amazed how you are attacking this machine for its low graph specs. In a typical use (exept for high demanding FPP games) it's unbeatable. I was waiting a long time for iMac G5 and I already love this machine. My record label will switch to Mac early next year. As we are indie label, the DP G5 is out of our reach. iMac G5 is cheap, silent, compact and looks sexy.

NVidia 5200 Ultra is far enough to accelerate desktop in Tiger. With an addition of FW Audio Interface and USB midi interface iMac G5 will find place in many home and project recording studios. This is an ideal machine aimed at musicians on a budget (my point of view, of course). Also, it has digital audio output and a large LCD screen (if you've ever tried to record a guitar near CRT - it's a disaster). And GarageBand is included for a start.

Add some RAM and it will run Photoshop and Dreamweaver without any problems. Even Cinema 4D, if you have patience and not work on large projects. It's ideal for semi-pro multimedia applications, when the creator is on a budget.

We are running Cinema 4D 7 on 512 MB Athlon 1800+ (1400 MHz) and Radeon 8500 (64MB). We render cover artworks at 300 DPI on it without any problems. Just a little patience for a final render

So please, as a would be switcher - don't tell me that this machine's graphics card sucks, because it don't. A (reasonable) PC user like me can see it, so why can't you?

Best Regards,

Maciej Repetowski
AXIS Records
www.axisrecords.net

 

[Ensoniq]

I've learned 3 things from reading this thread:

1 - Apparently, many people are still dumbfounded that Apple included the 5200 Ultra in the iMac G5, even though that's exactly what Apple includes in all but the top of the line PowerMac G5s. Please get over it already!

2 - Even if you believe Apple coulda/shoulda included a faster graphic chip in the iMac G5, please recognize that most likely the 5200 Ultra was chosen partially due to heat constraints. But again...please get over it already!

3 - xBench sucks. There...I said it. The fact that thousands of people use it trying to get some magic number for their machine doesn't really mean anything of value. If xBench really worked, there is no way that a two year old QuickSilver could be equivalent to an iMac G5 with significantly faster hard drive, RAM, bus speed, processor speed, and video. The only logical conclusion is that xBench just doesn't work.

Maybe if you compare individual scores for each section (memory, OpenGL, etc.) from one machine to another, there might be some comparitive value...assuming xBench is properly calibrated to exploit the specific nuances of the G3/G4/G5 properly at all times. But the "total score" nonsense is a completely unreliable and only has value to the same people who whine incessantly about the iMac G5's 5200.

Final point...the G5 iMac gets nearly 3 times the performance out of the 5200 Ultra as the G4 iMac did. So a lot of those PC kingdom charts placing the 5200 way below the "newer/better" cards don't take into account at all the benefits of running on an optimized platform instead of the hacked together hardware platform that is the Wintel world.

Maybe 5 people on these forums have actually used an iMac G5 in person. Let's wait until some real tests are done before we claim the sky is falling.

 

[myapplseedshurt]

i think this is wrong. if you think of the imac as an xserve stuck to the back of a flat monitor, then there's nothing about the "pretty" requirement that forces the machine be slow....

That is a good point. do you know what the cooling solution is in an xserve? I would think they wouldn't care about noise, and would have a stack cooling fan system to force air at a much higher rate than what the imac's little fans do. But I'm completely in the dark wrt/ xserve.

 

[Telomar]

Actually I am not impressed at all by these numbers! Only 154? My old G4 Quicksilver with a 1.2GHz (single!) processor upgrade gets a 132 with totally outdated technology.

I considered the new iMac as a replacement for that Quicksilver in my home office, but just a new TFT will do now. I really thought the G5 iMac would wipe the floor with my Quicksilver. I am a little bit disappointed...

I think you'd find those HD scores are the cause of your QS doing so well. They're surprisingly high and HD performance has a notable effect on the final Xbench score.

 

[H. Georgan]

That is a good point. do you know what the cooling solution is in an xserve? I would think they wouldn't care about noise, and would have a stack cooling fan system to force air at a much higher rate than what the imac's little fans do. But I'm completely in the dark wrt/ xserve.

I've heard that, noise-wise, standing beside an xserve is like standing beside a 747. Clearly unacceptable for an iMac.

 

[corbin_a2]

iMac G5 17" 1.8 w/512mb ENERGY SAVER ON AND OFF

I ran the Xbench test with the energy saver processor performance first on auto then on highest. What a difference!

PROCESSOR ON AUTO:

Results 89.19
System Info
Xbench Version 1.1.3
System Version 10.3.5 (7P35)
Physical RAM 512 MB
Model PowerMac8,1
Processor PowerPC G5 @ 1.80 GHz
L1 Cache 64K (instruction), 32K (data)
L2 Cache 512K @ 1.15 GHz
Bus Frequency 600 MHz
Video Card GeForce FX 5200
Drive Type ST380013AS
CPU Test 89.19
GCD Loop 61.45 2.40 Mops/sec
Floating Point Basic 126.67 458.08 Mflop/sec
AltiVec Basic 61.43 1.78 Gflop/sec
vecLib FFT 99.25 1.54 Gflop/sec
Floating Point Library 180.58 7.23 Mops/sec


PROCESSOR ON MAX:

Results 158.28
System Info
Xbench Version 1.1.3
System Version 10.3.5 (7P35)
Physical RAM 512 MB
Model PowerMac8,1
Processor PowerPC G5 @ 1.80 GHz
L1 Cache 64K (instruction), 32K (data)
L2 Cache 512K @ 1.15 GHz
Bus Frequency 600 MHz
Video Card GeForce FX 5200
Drive Type ST380013AS
CPU Test 158.28
GCD Loop 103.04 4.02 Mops/sec
Floating Point Basic 165.17 597.29 Mflop/sec
AltiVec Basic 123.47 3.59 Gflop/sec
vecLib FFT 202.39 3.14 Gflop/sec
Floating Point Library 358.38 14.35 Mops/sec

My question is as far as running my iMac on a normal day what setting should I set? Is there a reason to leave it on auto?

 

[stevesien]

iMac G5 1.8 -- 256 MB arrived. Have not yet upgeaded memory:

Results	134.71	
	System Info		
		Xbench Version		1.1.3
		System Version		10.3.5 (7P35)
		Physical RAM		256 MB
		Model		PowerMac8,1
		Processor		PowerPC G5 @ 1.80 GHz
			L1 Cache		64K (instruction), 32K (data)
			L2 Cache		512K @ 1.15 GHz
			Bus Frequency		600 MHz
		Video Card		GeForce FX 5200
		Drive Type		ST380013AS
	CPU Test	136.48	
		GCD Loop	91.90	3.59 Mops/sec
		Floating Point Basic	285.46	1.03 Gflop/sec
		AltiVec Basic	122.14	3.55 Gflop/sec
		vecLib FFT	100.86	1.57 Gflop/sec
		Floating Point Library	241.02	9.65 Mops/sec
	Thread Test	84.47	
		Computation	53.04	716.03 Kops/sec, 4 threads
		Lock Contention	207.38	2.60 Mlocks/sec, 4 threads
	Memory Test	210.75	
		System	229.53	
			Allocate	616.09	401.87 Kalloc/sec
			Fill	209.06	1664.14 MB/sec
			Copy	150.07	750.36 MB/sec
		Stream	194.81	
			Copy	166.59	1217.79 MB/sec [G5]
			Scale	169.44	1250.43 MB/sec [G5]
			Add	224.87	1439.19 MB/sec [G5]
			Triad	239.16	1461.27 MB/sec [G5]
	Quartz Graphics Test	188.34	
		Line	189.47	4.82 Klines/sec [50% alpha]
		Rectangle	156.19	10.99 Krects/sec [50% alpha]
		Circle	193.08	4.45 Kcircles/sec [50% alpha]
		Bezier	167.62	1.82 Kbeziers/sec [50% alpha]
		Text	268.61	4.38 Kchars/sec
	OpenGL Graphics Test	187.25	
		Spinning Squares	187.25	131.04 frames/sec
	User Interface Test	209.37	
		Elements	209.37	67.34 refresh/sec
	Disk Test	79.19	
		Sequential	67.00	
			Uncached Write	69.51	28.97 MB/sec [4K blocks]
			Uncached Write	43.58	17.85 MB/sec [256K blocks]
			Uncached Read	68.48	10.84 MB/sec [4K blocks]
			Uncached Read	128.78	52.03 MB/sec [256K blocks]
		Random	96.82	
			Uncached Write	94.15	1.41 MB/sec [4K blocks]
			Uncached Write	91.65	20.67 MB/sec [256K blocks]
			Uncached Read	92.34	0.61 MB/sec [4K blocks]
			Uncached Read	111.69	22.99 MB/sec [256K blocks]

Original single G5 1.8 w Radeon 9800 SE, 1.5GB ram
Results 124.41
System Info
Xbench Version 1.1.3
System Version 10.3.5 (7M34)
Physical RAM 1536 MB
Model PowerMac7,2
Processor PowerPC 970 @ 1.80 GHz
L1 Cache 64K (instruction), 32K (data)
L2 Cache 512K @ 1.80 GHz
Bus Frequency 900 MHz
Video Card ATY,R350
Drive Type ST3160023AS
CPU Test 120.78
GCD Loop 76.95 3.00 Mops/sec
Floating Point Basic 173.62 627.89 Mflop/sec
AltiVec Basic 89.02 2.59 Gflop/sec
vecLib FFT 139.06 2.16 Gflop/sec
Floating Point Library 237.06 9.49 Mops/sec
Thread Test 81.74
Computation 56.76 766.22 Kops/sec, 4 threads
Lock Contention 146.00 1.83 Mlocks/sec, 4 threads
Memory Test 214.03
System 187.02
Allocate 343.28 223.92 Kalloc/sec
Fill 143.33 1140.93 MB/sec
Copy 162.57 812.84 MB/sec
Stream 250.17
Copy 200.43 1465.13 MB/sec [G5]
Scale 228.55 1686.70 MB/sec [G5]
Add 311.62 1994.34 MB/sec [G5]
Triad 292.79 1788.97 MB/sec [G5]
Quartz Graphics Test 156.14
Line 108.95 2.77 Klines/sec [50% alpha]
Rectangle 172.75 12.15 Krects/sec [50% alpha]
Circle 181.67 4.19 Kcircles/sec [50% alpha]
Bezier 161.15 1.75 Kbeziers/sec [50% alpha]
Text 187.07 3.05 Kchars/sec
OpenGL Graphics Test 131.86
Spinning Squares 131.86 92.27 frames/sec
User Interface Test 192.26
Elements 192.26 61.84 refresh/sec
Disk Test 84.10
Sequential 98.22
Uncached Write 123.57 51.51 MB/sec [4K blocks]
Uncached Write 114.14 46.74 MB/sec [256K blocks]
Uncached Read 65.57 10.38 MB/sec [4K blocks]
Uncached Read 116.01 46.87 MB/sec [256K blocks]
Random 73.54
Uncached Write 53.23 0.80 MB/sec [4K blocks]
Uncached Write 86.19 19.44 MB/sec [256K blocks]
Uncached Read 74.12 0.49 MB/sec [4K blocks]
Uncached Read 95.10 19.57 MB/sec [256K blocks]

 

[RedEric]

Spot On

I've learned 3 things from reading this thread:

1 - Apparently, many people are still dumbfounded that Apple included the 5200 Ultra in the iMac G5, even though that's exactly what Apple includes in all but the top of the line PowerMac G5s. Please get over it already!

2 - Even if you believe Apple coulda/shoulda included a faster graphic chip in the iMac G5, please recognize that most likely the 5200 Ultra was chosen partially due to heat constraints. But again...please get over it already!

3 - xBench sucks. There...I said it. The fact that thousands of people use it trying to get some magic number for their machine doesn't really mean anything of value. If xBench really worked, there is no way that a two year old QuickSilver could be equivalent to an iMac G5 with significantly faster hard drive, RAM, bus speed, processor speed, and video. The only logical conclusion is that xBench just doesn't work.

Maybe if you compare individual scores for each section (memory, OpenGL, etc.) from one machine to another, there might be some comparitive value...assuming xBench is properly calibrated to exploit the specific nuances of the G3/G4/G5 properly at all times. But the "total score" nonsense is a completely unreliable and only has value to the same people who whine incessantly about the iMac G5's 5200.

Final point...the G5 iMac gets nearly 3 times the performance out of the 5200 Ultra as the G4 iMac did. So a lot of those PC kingdom charts placing the 5200 way below the "newer/better" cards don't take into account at all the benefits of running on an optimized platform instead of the hacked together hardware platform that is the Wintel world.

Maybe 5 people on these forums have actually used an iMac G5 in person. Let's wait until some real tests are done before we claim the sky is falling.

What a well thought out, reasoned and sensible point.

Well done that man: though I doubt it will ever shut the naysayers up.

Many will be happy with their new machine, those that aren't will be those that shouldn't have bought it. It is what it is, no more, no less. You want more, buy more. You don't like the chin or GPU, don't buy it.

If you have no interest in owning one, shut up about it. We are all able to make our own informed choice when spending hard earned cash. If we want advice about something, we'll ask all you experts who haven't even seen one, let alone know how fast they go.

My 2p

RedEric