Is Apple cooking their gaming performance numbers?

Posted:
in Current Mac Hardware edited January 2014
http://www.barefeats.com/imacg5b.html



Quote:

My posting last week of HALO and UNREAL TOURNAMENT 2004 results created such a controversy that I decided to devote a special page to iMac G5 3D Game results. We chose those two apps since they are the newest and best examples of first person 3D accelerated games.





Not having attended Mac Expo in Paris and not having paid close attention to Apple's iMac G5 web pages, I didn't realize that Apple had compared the iMac G5 and iMac G4 using the same two 3D games that I chose. Their results showed the iMac G5 to be as much as 212% faster than the iMac G4. My results showed it to be only 12% faster at most.



My report obviously got Apple's attention because they contacted me to see what I was doing differently. I sent them all my information on hardware and software and settings. I expected them to reciprocate but, so far, the only bit of information I could get from them is that they used the lowest quality settings. I had used high to highest settings.



So I made two 120 mile round trips during the last two days to remote sites so I could retest the fastest iMac G5 and fastest iMac G4 with the lowest quality settings for both Halo and UT2004. Here are the results:



CONFIG settings for Halo: Resolution 1024x768, Vertex Shaders ON, Detail Objects OFF, Model Reflections OFF, FSAA OFF, Lens Flare OFF, Model Detail LOW, Specular OFF, Shadows OFF, Decals OFF, Particles LOW, Texture Quality LOW, Sound ON.



CONFIG settings for UT2004 Full Version (patch 3323): When using Santa LCDBench, we ran Antalus Botmatch, specifying 1024x768 and "Minimum Settings" which translates to: 12 bots, Colordepth 16bit, Texture Detail LOWEST, Character Detail LOWEST, World Detail LOW, Physics Detail LOW, Dynamic Mesh LOD LOW, Decal Stay OFF, Character Shadows NONE, Decals OFF, Dynamic Lighting OFF, Coronas OFF, Trilinear Filtering OFF, Projectors OFF, Foliage OFF, Weather Effects OFF, Fog Distance Bar MINIMUM, Sound Detail OFF.



With the latest "lowest quality" settings, the iMac G5 was 10% faster running Halo and 50% faster running UT2004. That compares to iMac G5 advantage of 189% and 212% respectively reported on Apple's GRAPHICS page for the iMac G5.



There are two other theories put forth by readers as to why I can't match Apple's numbers:

1. Apple used more memory -- like dual 512MB or dual 1024MB matching 400MHz DIMMs in the iMac G5. We used dual 256MB matching 400MHz DIMMs in our test machine. Apple hasn't responded to our request for their hardware configuration. We aren't at liberty to add more memory to the iMac G5 demo units we were allowed to test, but based on tests we performed on a G5/2.0GHz MP Power Mac, two 256MB matched DIMMs deliver the same game frame rates as two 512MB or two 1024MB matching 400MHz DIMMs. So this argument is not a strong one.

2. Apple used an iMac G4 with GeForce4 MX instead of GeForceFX 5200. This theory intrigues me. I've asked Apple to confirm this. Meanwhile, I'll try to find a recent model iMac with the GeForce4 MX to see if this theory flys.





I understand Apple was trying to put the iMac G5's best foot forward, but serious gamers seek a balance between quality of graphics and performance. I believe my original graphs reflect "real world" game settings used most often by gamers. So I've keeping them posted here for those of you interested:



CONFIG settings for Halo: Resolution 1024x768, Vertex Shaders, Detail Objects ON, Model Reflections ON, FSAA OFF, Lens Flare OFF, Model Detail HIGH, Specular OFF, Shadows OFF, Decals ON, Particles HIGH, Texture Quality HIGH, Sound ON.



CONFIG settings for UT2004 Full Version (patch 3323): When using Santa LCDBench, we ran Antalus Botmatch, specifying 1024x768 and "Maximum Settings" which translates to: 12 bots, Colordepth 32bit, Texture Detail HIGHEST, Character Detail HIGHEST, World Detail HIGH, Physics Detail HIGH, Dynamic Mesh LOD HIGHEST, Decal Stay HIGH, Character Shadows FULL, Decals ON, Dynamic Lighting ON, Coronas ON, Trilinear Filtering ON, Projectors ON, Foliage ON, Weather Effects ON, Fog Distance Bar MAX, Sound Detail LOW.





With my "real world" settings, the advantage of the iMac G5 increases from 10% to 12% with Halo and shrinks from 50% to 11% with UT2004 (compared to the "lowest quality" runs).





CONCLUSION

I'm sure you haven't heard the end of this puzzling iMac G5 speed issue. I'll continue to probe the mysteries of Apple's phenomenal 3D Game tests results. For now, it's clear that the new iMac is faster than the old one. But we're convinced it's not as fast as Apple thinks it is -- or would have you believe.





FOR MORE TEST RESULTS ON THE IMAC G5, SEE PAGE ONE.



GRAPH LEGEND

Pmac G5/2.0MP = G5/2.0GHz MP Power Mac with Radeon 9800 Pro SE

Pmac G5/1.8MP = G5/1.8GHz MP Power Mac with GeForceFX 5200

iMac G5/1.8 = iMac G5/1.8GHz with GeForceFX 5200

iPbook G4/1.5 = G4/1.5GHz PowerBook 17" with Radeon 9700 mobility

Pmac G4/1.42MP = G4/1.42GHz MP Power Mac with Radeon 9700 Pro

iMac G4/1.25 = iMac G4/1.25GHz 20" with GeForceFX 5200



Is this a blatant case of false advertising? Why are so many people willing to give them a pass on this stuff? Even if you don't like Rob Art, there must certainly be some independent verification of Apple's outrageous results. What's the deal here?

Comments

  • Reply 1 of 11
    kickahakickaha Posts: 8,760member
    It's not outrageous, blatant or false until proven wrong.



    Very odd though. I'll be interested to see how it pans out.
  • Reply 2 of 11
    well that's advertising...



    apple could have set the old G4 to 1024 @ high detail and set the G5 @ 800x600 medium detail.



    it doesn't list the specs at which each machine was tested. Perhaps they left it even adn tested both at a much lower resoltuion like 640... or 800. With that graphics card, 1024 is a bit high. I would never try to play UT2k4 at that kinda resolution with a 64 megger.



    Or, resoultion and detail could have remained the same, choked the G4 with 256 megs of ram and stuff the G5 with a gig.



    There are alot of ways to control the output without randomly making up numbers -- it's twisted truth but that's what advertising is.



    :c
  • Reply 3 of 11
    placeboplacebo Posts: 5,767member
    I notice that Apple removed the mention of DooM 3 on their iMac Graphics page.
  • Reply 4 of 11
    Quote:

    Originally posted by Placebo

    I notice that Apple removed the mention of DooM 3 on their iMac Graphics page.



    They didn't - that reference has for some reason always been on the "design" page, not the graphics page.



    http://www.apple.com/imac/design.html
  • Reply 5 of 11
    kotatsukotatsu Posts: 1,010member
    Quote:

    Originally posted by fishdoc

    They didn't - that reference has for some reason always been on the "design" page, not the graphics page.



    http://www.apple.com/imac/design.html




    You can indeed play Doom 3. At around 12fps.



    Wooo!
  • Reply 6 of 11
    mac voyermac voyer Posts: 1,294member
    Quote:

    Originally posted by cubedcompanies

    well that's advertising...



    apple could have set the old G4 to 1024 @ high detail and set the G5 @ 800x600 medium detail.



    it doesn't list the specs at which each machine was tested. Perhaps they left it even adn tested both at a much lower resoltuion like 640... or 800. With that graphics card, 1024 is a bit high. I would never try to play UT2k4 at that kinda resolution with a 64 megger.



    Or, resoultion and detail could have remained the same, choked the G4 with 256 megs of ram and stuff the G5 with a gig.



    There are alot of ways to control the output without randomly making up numbers -- it's twisted truth but that's what advertising is.



    :c




    Sorry, that's not advertising. That's lying. It is not a comparison at all if they were not using like machines. By that, I mean a high end stock G4 iMac vs. a high end stock G5 iMac, or something like that. If they did something to one configuration without doing the same thing to the other, then the results are as good as random. If they starved one and fed the other or used the low end of one and the high end of the other or used low settings on one and high settings on the other, then they are misrepresenting the performance of the machines. Advertising would be to say that the new machine offers vastly improved gaming performance over the old. The "vastly" part is subjective. The "improved" part is empirical fact. If you actually believe that Apple did any of the above things you suggested, then you must admit that their tactics are at best, slimy, and at worst, intentional misrepresentations. Here's the kicker. The G5 iMac didn't need that sort of hype. It stands up well enough on its legitimate merits.
  • Reply 7 of 11
    Back when the G5s first came out I always thought the numbers for the PC benchmarks in Quake III looked a little low.



    Example:

    Quake III benchmarks from Apple

    vs.

    Quake III benchmarks from Intel



    Apple has a 3.0Ghz P4 w/ Radeon 9800 Pro doing 277fps at 640x480 16 bit color, while Intel with a similar processor that is only .4ghz faster managed 389fps at 640x480 32 bit color. I realize it isn't a fair comparison since the CPUs aren't identical.



    Saying Doom 3 is playable on an iMac is a joke, but gaming isn't the reason I am looking at getting an iMac. I'll still try to play Doom III on it, if I end up getting one.
  • Reply 8 of 11
    Maybe Apple will get sued for false advertising and they will upgrade their video cards to something nice.
  • Reply 9 of 11
    amorphamorph Posts: 7,112member
    Wow. Barefeats benchmarking vs. Apple benchmarking. I don't know whether to laugh or cry.



    I think I'll laugh.



    Barefeats' inconsistencies aside, it certainly is telling that Apple isn't releasing their test criteria. That's a big flashing red light that they aren't completely on the up-and-up.



    Ah well. Anything that requires them to tone down the impossibly upbeat copy on their product pages can't be a bad thing...
  • Reply 10 of 11
    kotatsukotatsu Posts: 1,010member
    Quote:

    Originally posted by Amorph





    Ah well. Anything that requires them to tone down the impossibly upbeat copy on their product pages can't be a bad thing...




    It does strike me as bizarre that they're pushing it so much as a gaming rig. By all means push the iLife stuff, Mac OS X, the style and size of the machine itself.. but they're really pushing it to keep hyping that crappy NVidia GPU for games.



    It's especially important to have a fast GPU in a machine with a LCD display as they don't look great when running lower resolutions than native.



    Come on Apple, give the machine the GPU it deserves.
  • Reply 11 of 11
    Quote:

    Originally posted by talksense101

    Maybe Apple will get sued for false advertising and they will upgrade their video cards to something nice.



    Thats exactly what i expect.
Sign In or Register to comment.