Barefeats benchmarks of iMac G5

[Diatribe]

Barefeats posted the benchmarks for the new iMacs.

Barefeats

Although it is faster than its predecessor it's astonishing to me that it is almost not a lot faster in most of the tests than the 1.5GHz Powerbook, even slower in some. There's to all the people complaining about the G4. Either it isn't all that bad or the G5 isn't all that good.
I'd say I'd be happy with a 2Ghz G4 with an upped frontside bus. This would be an awesome notebook and should blow a G5 1,8 out of the water. Sadly.

 

[TyleRomeo]

Barefeats posted the benchmarks for the new iMacs.

Barefeats

Although it is faster than its predecessor it's astonishing to me that it is almost not a lot faster in most of the tests than the 1.5GHz Powerbook, even slower in some. There's to all the people complaining about the G4. Either it isn't all that bad or the G5 isn't all that good.
I'd say I'd be happy with a 2Ghz G4 with an upped frontside bus. This would be an awesome notebook and should blow a G5 1,8 out of the water. Sadly.

hey maybe the motion playback without render graph is a little off unless the 1.5 powerbook is now faster than a dual 1.8 G5. Ohh and the Halo and unreal tests are really test on the graphics card and not solely the CPU. goodnight

Tyler

 

[Zaty]

Barefeats posted the benchmarks for the new iMacs.

Barefeats

Although it is faster than its predecessor it's astonishing to me that it is almost not a lot faster in most of the tests than the 1.5GHz Powerbook, even slower in some. There's to all the people complaining about the G4. Either it isn't all that bad or the G5 isn't all that good.
I'd say I'd be happy with a 2Ghz G4 with an upped frontside bus. This would be an awesome notebook and should blow a G5 1,8 out of the water. Sadly.

The only major flaw of the G4 architecture is that it only supports bus speed up to 167 MHz. If Freescale were able to change that, the G4 could keep us happy as a mobile CPU for the months and years to come.

 

[Mord]

5200 ultra Vs mobility 9700, hmm i wander which is faster.

 

[cluthz]

The graphic intensive tests clearly shows that apple shouldn't use the fx 5200 anymore.
The iMac is showing good performance in the cpu intensive test.
The only thing that i fould a bit weird is that the eMac 1.25ghz is par with the iMac 1.6mhz in ut2004 ??

Cinebench (cpu redering) really shows what the imac G5 (1.8) is capable of, its twice as fast as the 1.25 iMac

 

[dongmin]

I'm suspicious about their Motion test: I've never seen it before and it seems wildly inconsistent. A dual 2.0ghz PM gets a score of 18 while a dual 1.8ghz PM gets a whopping 66??? I'd discount this test altogether.

And the graphics tests are obviously determined by the graphics card which we all know stinks in the iMac. But it's comparable to the PB's 9700 mobility (presumably 64MB).

So basically that leaves us with three meaning tests (iMovie render, Filemaker script, and Cinebench) in which the iMac G5 1.8 outperforms the PowerBook 1.5 and the dual 1.4 G4 Power Mac. Not bad.

 

[Elan0204]

I'd say the benchmarks look really good. One of the strange Motion results have been removed. The other Motion test result look weird too, though.

Th only problem is the graphics card, you can already see how the new iMac will struggle with games played on high settings. Luckily, I don't play many games (if any) and I don't require super high settings. By the way, does anyone know if the PowerBook had 64MB or 128MB of VRAM?

Overall, I'm quite pleased with the iMac's scores. Too bad Barefeats didn't compare it to a single 1.6GHz and 1.8Ghz G5 PowerMac so we could see the effects of the reduced bus speed on the iMac. I bet the effect is insignificant, and the results would help reduce peoples complaining.

 

[Timelessblur]

I might like to point out that those bench marks in Halo and UT2k4 and Halo do point out that Apple is massicly fudging it results in adveratsing how much fast the G5 imacs are in those games. some how I dont see the 189% and 212% speed boost apple is adversting. Welcome to again of apple fudging the results of there test.

It more like 31% better on both acconts. that is a fudging of over 600%. Just though I like to point that out.

It does show that the card is being push to it limits already...... what does that say for it live over the next 3 years.....

 

[Mord]

I'm suspicious about their Motion test: I've never seen it before and it seems wildly inconsistent. A dual 2.0ghz PM gets a score of 18 while a dual 1.8ghz PM gets a whopping 66??? I'd discount this test altogether.

And the graphics tests are obviously determined by the graphics card which we all know stinks in the iMac. But it's comparable to the PB's 9700 mobility (presumably 64MB).

So basically that leaves us with three meaning tests (iMovie render, Filemaker script, and Cinebench) in which the iMac G5 1.8 outperforms the PowerBook 1.5 and the dual 1.4 G4 Power Mac. Not bad.

not in a million years is a 5200 compareable to a 9700 mobility it's more like a 5650 or a 9600pro

 

[Diatribe]

By the way, does anyone know if the PowerBook had 64MB or 128MB of VRAM?

They used the 64MB version first and then tested the 128MB and found the latter to be only marginally faster, meaning 1-2 frames. So it does not relly matter.

 

[JFreak]

The only major flaw of the G4 architecture is that it only supports bus speed up to 167 MHz. If Freescale were able to change that, the G4 could keep us happy as a mobile CPU for the months and years to come.

there, you said it. and in audio work i have found that the G4 is exceptionally good chip - it's more efficient than the G5 clock-for-clock! if the G4 can have faster fsb, and if it can clock higher, too, then it might still live for a long time.

in audio work, the G4 is currently more stable than the G5, though partly because G5 has only been around for a little while and software is not yet optimized for it.

still... great to see G5 in a new imac

 

[slughead]

I'm suspicious about their Motion test: I've never seen it before and it seems wildly inconsistent. A dual 2.0ghz PM gets a score of 18 while a dual 1.8ghz PM gets a whopping 66??? I'd discount this test altogether.

Yes, this is because the 5200 sucks, and motion uses the video card for much of the rendering operations.

See how far the 5200 drags down a DP2.0? Well the iMac has a much slower processor.

I've been ranting about this ever since the specs were rumored. It's the only problem I ever saw with the G5 iMac, aside from the precariousness of having a 20" monitor on a 5" stand with 25 cords coming out the side of it :X

 

[MacsRgr8]

For $ 50 more Apple could have used the Radeon 9600.... (comparing the PM G5 prices)

Don't benchmark Motion if you have a GeForce 5200, it will get you depressed... Also don't benchmark games (anything else?).
Look at other Barefeats benchmarks concerning Motion.... Even a Dual G4 1.42 GHz PowerMac with Radeon 9700 is faster than a Dual PowerMac G5 1.8 GHz with that damn 5200....

GeForce 5200: game over....
Nuff said.

 

[takao]

2 whopping frames more than the G4 with the same graphics chip in UT2004 (on the 20" version)and still only 20 ..i wonder where apple got that 200% faster thing...perhaps they maxed out the imac g5 and put a only 64 mb ram into the old one ...

 

[quagmire]

I'm suspicious about their Motion test: I've never seen it before and it seems wildly inconsistent. A dual 2.0ghz PM gets a score of 18 while a dual 1.8ghz PM gets a whopping 66??? I'd discount this test altogether.

Read the top part of the graph. It says the shorter bar is faster.

 

[slughead]

2 whopping frames more than the G4 with the same graphics chip in UT2004 (on the 20" version)and still only 20 ..i wonder where apple got that 200% faster thing...perhaps they maxed out the imac g5 and put a only 64 mb ram into the old one ...

Maybe it was in software acceleration mode

 

[slughead]

Read the top part of the graph. It says the shorter bar is faster.

If it has to be disputed:

 

[tliptak]

Of those tests the first shows that motion tests are all about the graphics card.

 

[slughead]

Of those tests the first shows that motion tests are all about the graphics card.

Exactly. Nobody's saying the iMac's CPU is slow, we're talking about the GPU.

 

[tinydancer]

Is there any chance that Apple will significantly upgrade the GPU in their next revision of the G5 imac? It seems that there was such a drive for acheiving a particular architecture that there was little concern for GPU capabilities. I love the looks, and I think the price is really good, but if it really isn't much better than my G4 why spend the money.

BTW, out of curiosity..does Apple get the message from its user about dissatisfaction. I'm just wondering (having read numerous post dealing with the GPU concerns) if Apple is aware that they may have been better using a better card.

Dont' get me wrong...I'm not a hater...love my mac, but I think they shot a liitle low here.

 

[slughead]

Is there any chance that Apple will significantly upgrade the GPU in their next revision of the G5 imac?

It's possible, but if they just upgraded it to a 9600 and the next rev is a year from now then.. well actually nevermind, that'd still be better.

I don't want to go through how antiquated the 5200 really is, but the latter statement should give you an idea.

 

[wkw]

How does the 5200 compare to the ATI 9000 that is currently in my PM G4 dual 1.25? I have been considering switching to the 20" iMac and selling off the G4 tower and my 20" cinema display. Would this be a good move? I suck at FSP games and don't play them.

 

[cluthz]

How does the 5200 compare to the ATI 9000 that is currently in my PM G4 dual 1.25? I have been considering switching to the 20" iMac and selling off the G4 tower and my 20" cinema display. Would this be a good move? I suck at FSP games and don't play them.

The FX 5200 is faster.
Both in 2d and 3d. FX5200 isn't that bad, but you want more power when you pay much for a computer.
So if you don't game and don't use motion, then the fx5200 will suit you fine!
if you like some further reading.. http://graphics.tomshardware.com/graphic/20031229/index.html

 

[wkw]

I suck at FSP games and don't play them.

woops I meant FPS!

 

[dubbz]

not in a million years is a 5200 compareable to a 9700 mobility it's more like a 5650 or a 9600pro

Heh, I wish... the 5200 isn't even comparable to the 9600 Pro.

Putting the 5200 in any new computer, without having a better solution as an option is an insult to humanity. Well, ok, maybe not... but sometimes it feels that way..