Sidebar Sidebar
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

BareFeats Tests show Retina MB Pro best for games

A

apolloa

Suspended
Original poster
Oct 21, 2008
12,318
7,802
Time, because it rules EVERYTHING!
So far, in Starcraft 2 and Portal 2 and WOW, T=the new top spec Retina MB Pro is the best gaming laptop from Apple!

http://www.barefeats.com/mbp12gx.html

They are continuing to test it further....

QUOTE:

INSIGHTS and ANALYSIS
The 'mid 2012' Retina MacBook Pro is clearly a "cut above" when it comes to extreme gaming. However, if you opt for the entry level 15" 'mid 2012' MacBook Pro 2.3GHz, it gets beat by the best 'late 2011' MacBook Pro in two out of three tests above. (Psst. I read Apple is having a closeout sale on the 'late 2011' 17" MacBook Pro.)

You might say, "but the 2.3GHz MacBook Pro only had 4GB of main memory." True. But when we monitored the total wired and active real memory usage, it never exceeded 3GB in all three cases.

"What about the fact that the 2.3GHz MacBook Pro has half the video memory of the other two?" That could be a factor. Using OpenGL Driver Monitor, we determined that Portal 2 used 680MB of video memory during our run. Cutting the resolution down to 1280x720 dropped it to 536MB. Going from "BEST" quality to "DEFAULT" brought the video memory use down to 325MB.

As for World of Warcraft and Starcraft II, they used 335MB and 367MB respectively in our test runs.
 
It's not any faster than the high-end 15" MBP. Barefeats only tested the low-end 15" which has an inferior GPU.
 
Hellhammer said:
It's not any faster than the high-end 15" MBP. Barefeats only tested the low-end 15" which has an inferior GPU.
Click to expand...

Ah, good observation mate. Will be interesting to see how the tests develop but it's also interesting to see the Retina machine can handle games at high resolutions well, but not the screens native res or 1440x900?? The new base MB Pro has the same 650m but with half the video memory.
 
apolloa said:
Ah, good observation mate. Will be interesting to see how the tests develop but it's also interesting to see the Retina machine can handle games at high resolutions well, but not the screens native res or 1440x900?? The new base MB Pro has the same 650m but with half the video memory.
Click to expand...

Here are some early gaming tests as well: http://www.anandtech.com/show/5998/macbook-pro-retina-display-analysis

Some games run at the native res but obviously the performance is not all that great. Future drivers may help with that, though.
 
Hellhammer said:
It's not any faster than the high-end 15" MBP. Barefeats only tested the low-end 15" which has an inferior GPU.
Click to expand...

? It says they tested the absolute highest build of the 2011 15". 6770m 1GB, 2.5 i7.

edit: Gotcha, the MBP vs RMBP. That's obvious, though, I wouldn't expect any difference between the 2 models this year. A different screen shouldn't affect performance.
 
" However, if you opt for the entry level 15" 'mid 2012' MacBook Pro 2.3GHz, it gets beat by the best 'late 2011' MacBook Pro in two out of three tests above."

Strange they'd even say this, when those framerates are clearly the result of being vram constrained. They even mention further down that Portal 2 used quite a bit more than 512mb RAM at those settings.

I'd like to see the FPS difference between this years 2.3 and 2.6 models with all other specs being equal.

I'm also curious about GPU temps. This is promising, though. If the temps warrant it, the performance under Windows with a silght OC could make this a damn nice gaming laptop :) All I ask is for good quality/performance with Diablo 3 and Guild Wars 2.
 
Hellhammer said:
2011, yes, but the 2012 MBP they tested was low-end (2.3GHz, 650M with 512MB).
Click to expand...

Can't wait for the AnandTech reviews. I wish Anand also has the non-Retina high-end 2012 Pro for comparison.

Best regards to your team.
 
Hellhammer said:
It's not any faster than the high-end 15" MBP. Barefeats only tested the low-end 15" which has an inferior GPU.
Click to expand...

Am I the only one who thinks it's weird that they chose 512MB and 1GB for vram on a machine in that price range midway through 2012?
 
Computationally there is litte between them Retina or not
new-macbook-pro-2012-geekbench1.jpg


A 2012 2.6 MBP with or without Retina will be so close that it will be imperceptible, I doubt that you would even see a significant difference with the Late 2011`s
 
thekev said:
Am I the only one who thinks it's weird that they chose 512MB and 1GB for vram on a machine in that price range midway through 2012?
Click to expand...

When it comes to Apple, nothing is weird :p Apple has never been too lenient when it comes to VRAM. I think the high-end MBPR should have had 2GB due to the resolution. Don't let the amount of VRAM fool you, though. Some cheap PCs may also have 1GB but that's usually slow DDR3, not GDDR5.


Queen6 said:
Computationally there is litte between them Retina or not
Image

A 2012 2.6 MBP with or without Retina will be so close that it will be imperceptible, I doubt that you would even see a significant difference with the Late 2011`s
Click to expand...

The difference is so small that it doesn't really count - it could be just a measuring error. If the tests were repeated 10 times, I'm sure the average would be about the same. Keep in mind that GeekBench only measures CPU and RAM performance, so it's not really a good measure of overall performance.
 
Being as new as it is, I would imagine the 650M is still constrained by driver quality (NVIDIA still hasn't release a Windows driver for it). So, we're likely to see even better results as time goes by.
 
corvus32 said:
Being as new as it is, I would imagine the 650M is still constrained by driver quality (NVIDIA has yet to release a Windows driver for it). So, we're likely to see even better results as time goes by.
Click to expand...

aren't portal and sc2 cpu limited? It's kind of funny that they didn't at least use the same cpu speeds to take that out of the equation.
 
Hellhammer said:
The difference is so small that it doesn't really count - it could be just a measuring error. If the tests were repeated 10 times, I'm sure the average would be about the same. Keep in mind that GeekBench only measures CPU and RAM performance, so it's not really a good measure of overall performance.
Click to expand...

Yep i know, it`s just an early indication of system performance, and this is my point that there is little or no difference, an increase yes but nothing noteworthy performance wise. As more benchmarks are published we will see, likely splitting hairs over a few FPS
 
Hellhammer said:
When it comes to Apple, nothing is weird :p Apple has never been too lenient when it comes to VRAM. I think the high-end MBPR should have had 2GB due to the resolution. Don't let the amount of VRAM fool you, though. Some cheap PCs may also have 1GB but that's usually slow DDR3, not GDDR5.
Click to expand...

I'm aware of that. I didn't mean $500 laptop cheap. I was thinking more in the $1300-1500 range.

Note the $1799 model. A lot of graphic design and photography students probably go with this. Adobe requires a minimum of 512GB for mercury engine/OpenGL functionality to run. These are right on the edge. Many games require a minimum of 512. I thought it was a little skimpy there much like the early 2011s that started with 256MB. I've always felt Apple should maintain a really strong base configuration when they go with a high cost of entry. Anyway I wanted to make it clear that I was talking about more than just games.
 
Queen6 said:
Yep i know, it`s just an early indication of system performance, and this is my point that there is little or no difference, an increase yes but nothing noteworthy performance wise. As more benchmarks are published we will see, likely splitting hairs over a few FPS
Click to expand...

You do know that the table you posted was for CPU and RAM performance when this thread is more on gaming and GPU, right?
 
arctic said:
You do know that the table you posted was for CPU and RAM performance when this thread is more on gaming and GPU, right?
Click to expand...

As said it`s just an initial indication, 2012 MBP Retina or not share the same CPU/GPU so why expect miracles, and in no ways am i deriding the Retina MBP as i am looking at one myself, only I dont anticipate a massive jump in performance over my Late 2011 !5" 2.4
corvus32 said:
@Queen6:While I suppose most games are still 32-bit, the 64-bit geekbench results are about 1400 points apart.
Click to expand...

It is what it is, i am by no means an expert, we will see once people start to publish FPS etc for the Retina`s I personally dont expect much more.
 
arctic said:
You do know that the table you posted was for CPU and RAM performance when this thread is more on gaming and GPU, right?
Click to expand...

That's right. I made this thread because I am interested to see how the retina machines games because of the resolution of that screen.
In all other tasks a top end Retina and non retina MB Pro will be the same so long as the non Retina MB Pro has an SSD :)

It was interesting to read that Anandtech brief glimpse on the gaming performance.
 
Fynd said:
Huge price jump for a theoretical ~12% performance increase. You do get a bump in storage as well, I guess.
Click to expand...

A 512GB Samsung 830 SSD will cost you about $700 from Amazon or Newegg, and it will not have Apple's custom firmware.
 
You must log in or register to reply here.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.
Top Bottom
Back