Jump to content


3D Game benchmarks for the MacBook Pro 2.6 "2008"


  • Please log in to reply
86 replies to this topic

#1 rob_ART

rob_ART

    Bare Feats

  • Members
  • PipPipPipPipPip
  • 792 posts
  • Location:Portland, OR

Posted 08 March 2008 - 10:05 PM

Apologies to the 20 or so of you who read this posting yesterday. My comparison of the two models of MacBook Pro was invalid. For one thing, the results from the older MacBook Pro were from tests performed in June 2007 using OS X Tiger. I decided to re-run all the games using our 17" MacBook Pro 2.4GHz* now that it has OS X Leopard (10.5.2) installed. The results were much faster.

So I've deleted the results from this posting and listed the latest results in the subsequent posting.



*NOTE: We no longer have the 15" MacBook Pro in the lab. If one of you has a 15" MacBook Pro 2.4GHz or 2.6GHz (2007 version) running Leopard, please contact me so you can run some of the same tests with your machine.

Edited by rob_ART, 09 March 2008 - 06:13 PM.

rob-ART morgan
mad scientist
BareFeats.com

#2 rob_ART

rob_ART

    Bare Feats

  • Members
  • PipPipPipPipPip
  • 792 posts
  • Location:Portland, OR

Posted 09 March 2008 - 06:37 PM

Here are the results after re-testing the 17" MacBook Pro 2.4GHz "mid 2007" (now running Leopard) and the 15" MacBook Pro 2.6GHz "early 2008." All tests were run at High Quality 1280x800:

Doom 3
2007 17" = 84 fps
2008 15" = 82 fps

Quake 4 (v 1.42)
2007 17" = 84 fps
2008 15" = 81 fps

Prey
2007 17" = 65 fps
2008 15" = 64 fps

Halo UB
2007 17" = 108 fps
2008 15" = 106 fps

UT2004 Inferno Flyby
2007 17" = 132 fps
2008 15" = 119 fps

Historically, the 15" GPU has been clocked lower than the 17" GPU due to overheating issues from the 15 inch's smaller heatsink. That's why before we draw any hard conclusions, we need to see the results for a 15" MacBook Pro "early or late 2007" model running at either 2.4 or 2.6GHz. If you have one of these models, please contact me. If I can borrow one, I'll run the tests on it under Leopard.
rob-ART morgan
mad scientist
BareFeats.com

#3 Quicksilver

Quicksilver

    Verbal Windbag

  • IMG Writers
  • 4227 posts
  • Location:Chicago Illinois
  • Pro Member:Yes

Posted 09 March 2008 - 06:55 PM

I can do some benchmarking for you (see my system specs).  Feel free to drop me an email--just click on my name in the signature field.  Why didn't you run your tests at 1440x900?  1280x800 doesn't make any sense.
Former Senior Hardware Editor
InsideMacGames.com

#4 QuantaCat

QuantaCat

    Master Blaster

  • Members
  • PipPipPipPipPipPip
  • 1995 posts
  • Location:Vienna, Austria

Posted 10 March 2008 - 02:12 AM

...wait wait wait. does this mean that the x1600 is actually faster than the 8600m?


Oh and I find your site a bit hard to navigate. Not because of the ads, but because it's basically just a list of stuff, with apparently *some* order, I have no clue which.. Was this test a recent one?
QC.


avatar courtesy of James Grimlee.

#5 rob_ART

rob_ART

    Bare Feats

  • Members
  • PipPipPipPipPip
  • 792 posts
  • Location:Portland, OR

Posted 10 March 2008 - 05:34 AM

View PostQuantaCat, on March 10th 2008, 01:12 AM, said:

...wait wait wait. does this mean that the x1600 is actually faster than the 8600m?
Oh and I find your site a bit hard to navigate. Not because of the ads, but because it's basically just a list of stuff, with apparently *some* order, I have no clue which.. Was this test a recent one?

No. The 2007 has an 8600M just like the 2008. Identical GPUs. Only difference is the amount of video memory.

As for the site, I have not posted anything on this comparison. I was giving you guys first look at the data.

I hope to get the data for the 2007 15" MacBook Pro before I publish the article since the 15" GPU is clocked lower than the 17" -- comparing the 17" to the 15" is not a "fair fight."
rob-ART morgan
mad scientist
BareFeats.com

#6 QuantaCat

QuantaCat

    Master Blaster

  • Members
  • PipPipPipPipPipPip
  • 1995 posts
  • Location:Vienna, Austria

Posted 10 March 2008 - 06:20 AM

if you want, I could throw in a bench run on my system.. I have a first gen MBP, 17" & X1600. (late 2006)
QC.


avatar courtesy of James Grimlee.

#7 teflon

teflon

    Bastard of the Popeye Analogy

  • Members
  • PipPipPipPipPipPipPipPipPipPipPipPip
  • 9589 posts
  • Location:London, UK

Posted 10 March 2008 - 07:02 AM

like QS, I have that same MBP and all those games but Doom 3.
Polytetrafluoroethylene to my friends.

Macbook Pro - C2D 2.4Ghz / 4GB RAM / Samsung 830 256GB SSD / Geforce 8600M GT 256Mb / 15.4"
Cube - G4 1.7Ghz 7448 / 1.5GB RAM / Samsung Spinpoint 250GB / Geforce 6200 256Mb
Self-built PC - C2Q Q8300 2.5Ghz / 4GB RAM / Samsung 830 256GB SSD / Radeon 7850 OC 1GB / W7 x64
and a beautiful HP LP2475w 24" H-IPS monitor

#8 Quicksilver

Quicksilver

    Verbal Windbag

  • IMG Writers
  • 4227 posts
  • Location:Chicago Illinois
  • Pro Member:Yes

Posted 10 March 2008 - 12:34 PM

FYI, I'm not going to have this MacBook Pro for more than a day or two longer.
Former Senior Hardware Editor
InsideMacGames.com

#9 teflon

teflon

    Bastard of the Popeye Analogy

  • Members
  • PipPipPipPipPipPipPipPipPipPipPipPip
  • 9589 posts
  • Location:London, UK

Posted 10 March 2008 - 01:02 PM

oh really? how so?? you selling it on?
Polytetrafluoroethylene to my friends.

Macbook Pro - C2D 2.4Ghz / 4GB RAM / Samsung 830 256GB SSD / Geforce 8600M GT 256Mb / 15.4"
Cube - G4 1.7Ghz 7448 / 1.5GB RAM / Samsung Spinpoint 250GB / Geforce 6200 256Mb
Self-built PC - C2Q Q8300 2.5Ghz / 4GB RAM / Samsung 830 256GB SSD / Radeon 7850 OC 1GB / W7 x64
and a beautiful HP LP2475w 24" H-IPS monitor

#10 Janichsan

Janichsan

    Jugger Bugger

  • Forum Moderators
  • PipPipPipPipPipPipPipPipPipPipPipPip
  • 8076 posts
  • Steam Name:Janichsan
  • Location:over there

Posted 10 March 2008 - 01:40 PM

View PostQuicksilver, on March 10th 2008, 01:55 AM, said:

Why didn't you run your tests at 1440x900?  1280x800 doesn't make any sense.
A wild guess: comparability to the 13" Macbooks?

"We do what we must, because we can."
"Gaming on a Mac is like women on the internet." — "Highly common and totally awesome?"


#11 Quicksilver

Quicksilver

    Verbal Windbag

  • IMG Writers
  • 4227 posts
  • Location:Chicago Illinois
  • Pro Member:Yes

Posted 10 March 2008 - 01:56 PM

View PostJanichsan, on March 10th 2008, 02:40 PM, said:

A wild guess: comparability to the 13" Macbooks?

Hence my question.  Since the MacBook can't really run any of those games, why bother?
Former Senior Hardware Editor
InsideMacGames.com

#12 Janichsan

Janichsan

    Jugger Bugger

  • Forum Moderators
  • PipPipPipPipPipPipPipPipPipPipPipPip
  • 8076 posts
  • Steam Name:Janichsan
  • Location:over there

Posted 10 March 2008 - 04:36 PM

View PostQuicksilver, on March 10th 2008, 08:56 PM, said:

Since the MacBook can't really run any of those games, why bother?
To show how crap the MacBook does run this games?

"We do what we must, because we can."
"Gaming on a Mac is like women on the internet." — "Highly common and totally awesome?"


#13 Quicksilver

Quicksilver

    Verbal Windbag

  • IMG Writers
  • 4227 posts
  • Location:Chicago Illinois
  • Pro Member:Yes

Posted 10 March 2008 - 04:41 PM

Wouldn't that be tantamount to writing an article about poo being stinky?   :blink:
Former Senior Hardware Editor
InsideMacGames.com

#14 Janichsan

Janichsan

    Jugger Bugger

  • Forum Moderators
  • PipPipPipPipPipPipPipPipPipPipPipPip
  • 8076 posts
  • Steam Name:Janichsan
  • Location:over there

Posted 10 March 2008 - 04:43 PM

View PostQuicksilver, on March 10th 2008, 11:41 PM, said:

Wouldn't that be tantamount to writing an article about poo being stinky?   :blink:
Well, someone has to say it... ;)

"We do what we must, because we can."
"Gaming on a Mac is like women on the internet." — "Highly common and totally awesome?"


#15 rob_ART

rob_ART

    Bare Feats

  • Members
  • PipPipPipPipPip
  • 792 posts
  • Location:Portland, OR

Posted 11 March 2008 - 08:35 PM

I have some interesting Halo UB (2.0.2) results to share:

2007 MacBook Pro 2.4GHz (15") = 97 fps
2008 MacBook Pro 2.4GHz (15") = 101 fps
2008 MacBook Pro 2.6GHz (15") = 106 fps

This supports my theory: the new 45 nm processor uses less power and runs cooler. So the GPU, which shares the same heatsink, can be clocked higher.
rob-ART morgan
mad scientist
BareFeats.com

#16 QuantaCat

QuantaCat

    Master Blaster

  • Members
  • PipPipPipPipPipPip
  • 1995 posts
  • Location:Vienna, Austria

Posted 12 March 2008 - 03:31 AM

can, or is?

IE. are you talking about user clocking, or Apple clocking?

BTW, I'll get on with those tests as soon as I finish watching Terminator series.. (which should be today/tonight)
QC.


avatar courtesy of James Grimlee.

#17 Quicksilver

Quicksilver

    Verbal Windbag

  • IMG Writers
  • 4227 posts
  • Location:Chicago Illinois
  • Pro Member:Yes

Posted 12 March 2008 - 06:03 AM

Or, the higher-clocked CPU is feeding more data to the card.  Or some combination of both.
Former Senior Hardware Editor
InsideMacGames.com

#18 rob_ART

rob_ART

    Bare Feats

  • Members
  • PipPipPipPipPip
  • 792 posts
  • Location:Portland, OR

Posted 12 March 2008 - 06:07 AM

View PostQuantaCat, on March 12th 2008, 02:31 AM, said:

IE. are you talking about user clocking, or Apple clocking?

The core and memory clock speeds of the MacBook Pro's GPU are variable, depending on what you are doing. Though there is currently no Mac OS X utility to measure this, we were able to confirm the variable frequencies of the 15" 2007 MacBook Pro 2.4, for example, using ATITool under Windows XP Pro (Boot Camp partition):
CORE
Idle = 169MHz, Normal = 375MHz, Max (Gaming) = 470MHz
MEMORY
Idle = 100MHz, Normal = 502MHz, Max (Gaming) = 635MHz

A similar variable clocking goes on under Mac OS X, according to ATI engineers.

The 17" 2007 MacBook Pro 2.4 maxed at 520MHz Core and 650MHz Memory. That explains why the 17" is always faster than the 15", all other things being equal. I'm told its max is higher because it has a bigger, more efficient heatsink on the GPU/CPU.

As soon as we have Windows installed on our 2008 15" MacBook Pro, we'll sample the frequencies to see if they are higher than the 2007 model.
rob-ART morgan
mad scientist
BareFeats.com

#19 rob_ART

rob_ART

    Bare Feats

  • Members
  • PipPipPipPipPip
  • 792 posts
  • Location:Portland, OR

Posted 12 March 2008 - 06:23 AM

Here's Unreal Tournament 2004 Inferno Flyby results at 1280x800 for more MacBook Pro examples:

2006 MBP 2.3 (17") = 102 fps (Radeon X1600 - thanks to QuantaCat)
2007 MBP 2.4 (15") = 115 fps (thanks to teflon)
2008 MBP 2.4 (15") = 120 fps
2008 MBP 2.6 (15") = 120 fps
2007 MBP 2.4 (17") = 132 fps

(UT2004's Flyby is one of the few "pure GPU" tests you can run.)
rob-ART morgan
mad scientist
BareFeats.com

#20 Quicksilver

Quicksilver

    Verbal Windbag

  • IMG Writers
  • 4227 posts
  • Location:Chicago Illinois
  • Pro Member:Yes

Posted 12 March 2008 - 06:41 AM

The UT2004 tests clearly show that the GPU is being clocked higher, and weirdly enough, that the GPU is the limiting factor in game that's notoriously CPU-bound.  

I don't think that Apple is going to exceed the specifications that nVidia laid down for the 8600M GT's clock speeds, hence, the only way that you'll figure out where the performance difference is coming from is by charting clock speed against system load.  In that case, I'd predict that the newer MacBook Pro's curve would have a wider plateau but the same amplitude.  

Also, it's probably time to move beyond UT2004, or increase resolution to 1440x900 and add FSAA to (hopefully) simulate the higher VRAM allocation required by newer engines.  For instance, a Call of Duty 4 test in Windows XP with everything maxed-out should result in a huge win for the 2008 MacBook Pro with 512 MB of VRAM.
Former Senior Hardware Editor
InsideMacGames.com