rMBP Performance in Unigine Heaven tested!

[Zer0n1nja]

Hey so I was able to get some benchmarks knocked out at my local bestbuy which had the new models out on display. They were originally locked from being able to install anything but the apple rep was there who let me install the new software on the macbooks and he was very patient!

Anyways, here's a run down of the performance I got of each model. I only tested unigine Heaven 4.0 as I didn't want to waste more of his time and have all the macbooks in use.

So here's the run down, I almost attached photos of each. All tests were done at Medium default settings in OpenGL at 1440x900 except on the 13 inch retina which wouldn't go past 1280x800

Macbook Air 13 Base model. (had to stop test about 70% of the way for a customer)
15.6 fps

rMBP 13 with i5 2.6 (1799 model)
21.2 fps 2 Things of note, the processor gpu is at 1.2 instead of 1.1ghz on the i5 which is about 10% more performance, and it would only run at 1280x800. I would best the comparable performance would be closer to 19 fps if all things were equal

rMBP 15 Base model
27.5 fps

rMBP 15 with 750M
28.5 FPS

My gaming desktop with i5-3570k running at 4.3ghz and stock Nvidia 760 gtx
82.9 fps just for comparison sake.


Temps for all the models were looked like this.
13 inch air, 78c
13 rMBP, 72c
15 rMBP base, 69c
15 rMBP 750m, 79c

Hope this helps in your decision making, I know it isn't a thorough gaming test but it's the best I could muster in the amount of time and patience I was working with. I included photos of each test except the macbook air.

Here's a Link to photos I took of all the tests. Here's a Link to photos I took of all the tests.

 

[johnnylarue]

Hey so I was able to get some benchmarks knocked out at my local bestbuy which had the new models out on display. They were originally locked from being able to install anything but the apple rep was there who let me install the new software on the macbooks and he was very patient!

Anyways, here's a run down of the performance I got of each model. I only tested unigine Heaven 4.0 as I didn't want to waste more of his time and have all the macbooks in use.

So here's the run down, I almost attached photos of each. All tests were done at Medium default settings in OpenGL at 1440x900 except on the 13 inch retina which wouldn't go past 1280x800

Macbook Air 13 Base model. (had to stop test about 70% of the way for a customer)
15.6 fps

rMBP 13 with i5 2.6 (1799 model)
21.2 fps 2 Things of note, the processor gpu is at 1.2 instead of 1.1ghz on the i5 which is about 10% more performance, and it would only run at 1280x800. I would best the comparable performance would be closer to 19 fps if all things were equal

rMBP 15 Base model
27.5 fps

rMBP 15 with 750M
28.5 FPS

My gaming desktop with i5-3570k running at 4.3ghz and stock Nvidia 760 gtx
82.9 fps just for comparison sake.


Temps for all the models were looked like this.
13 inch air, 78c
13 rMBP, 72c
15 rMBP base, 69c
15 rMBP 750m, 79c

Hope this helps in your decision making, I know it isn't a thorough gaming test but it's the best I could muster in the amount of time and patience I was working with. I included photos of each test except the macbook air.

Here's a Link to photos I took of all the tests. Here's a Link to photos I took of all the tests.

Wow--the 750M, running 10 degrees hotter, gets one whole FPS more than the Iris Pro?

Can someone tell me what's wrong with this picture?

 

[mattcube64]

Hey so I was able to get some benchmarks knocked out at my local bestbuy which had the new models out on display. They were originally locked from being able to install anything but the apple rep was there who let me install the new software on the macbooks and he was very patient!

Anyways, here's a run down of the performance I got of each model. I only tested unigine Heaven 4.0 as I didn't want to waste more of his time and have all the macbooks in use.

So here's the run down, I almost attached photos of each. All tests were done at Medium default settings in OpenGL at 1440x900 except on the 13 inch retina which wouldn't go past 1280x800

Macbook Air 13 Base model. (had to stop test about 70% of the way for a customer)
15.6 fps

rMBP 13 with i5 2.6 (1799 model)
21.2 fps 2 Things of note, the processor gpu is at 1.2 instead of 1.1ghz on the i5 which is about 10% more performance, and it would only run at 1280x800. I would best the comparable performance would be closer to 19 fps if all things were equal

rMBP 15 Base model
27.5 fps

rMBP 15 with 750M
28.5 FPS

My gaming desktop with i5-3570k running at 4.3ghz and stock Nvidia 760 gtx
82.9 fps just for comparison sake.


Temps for all the models were looked like this.
13 inch air, 78c
13 rMBP, 72c
15 rMBP base, 69c
15 rMBP 750m, 79c

Hope this helps in your decision making, I know it isn't a thorough gaming test but it's the best I could muster in the amount of time and patience I was working with. I included photos of each test except the macbook air.

Here's a Link to photos I took of all the tests. Here's a Link to photos I took of all the tests.

Thank you very much for doing this and reporting back! Always eager to see the new benchmarks and whatnot when new Macs are released... always feel like things trickle out over a few weeks.

Wow--the 750M, running 10 degrees hotter, gets one whole FPS more than the Iris Pro?

Can someone tell me what's wrong with this picture?

It's a synthetic benchmark. It's a good way of getting a rough idea of performance between different computers. Either way, though, a synthetic benchmark is designed to run a component (in this case the CPU/GPU) at maximum threshold - so it's going to get as hot as possible. I think everyone was expecting the 750 to be hotter than the Iris Pro. But, doesn't mean the benchmark was showing the best of what the 750 can do best on.

I think it's a safe assumption to say in day-to-day tasks, the Iris Pro and 750m are going to exchange punches. But for games and GPU-heavy tasks, the 650M is probably about 25% fast than the Iris Pro, and the 750M will probably end up being about 10% faster than the 650M with some extra VRAM to boot.

 

[johnnylarue]

Thank you very much for doing this and reporting back! Always eager to see the new benchmarks and whatnot when new Macs are released... always feel like things trickle out over a few weeks.



It's a synthetic benchmark. It's a good way of getting a rough idea of performance between different computers. Either way, though, a synthetic benchmark is designed to run a component (in this case the CPU/GPU) at maximum threshold - so it's going to get as hot as possible. I think everyone was expecting the 750 to be hotter than the Iris Pro. But, doesn't mean the benchmark was showing the best of what the 750 can do best on.

I think it's a safe assumption to say in day-to-day tasks, the Iris Pro and 750m are going to exchange punches. But for games and GPU-heavy tasks, the 650M is probably about 25% fast than the Iris Pro, and the 750M will probably end up being about 10% faster than the 650M with some extra VRAM to boot.

Thanks for talking me down from the ledge, matt. I'd overlooked the fact that this was essentially a form of "stress test" rather than anecdotal evidence from an actual game (or whatever.)

 

[Zer0n1nja]

Tomorrow I am going back and I will attempt to run 1 or 2 more benchmarks related to gaming performance. Seeing as how these are public computers with limited time I can test this stuff, what would you guys be interested in seeing? Any small demo's I can quickly run?

 

[mattcube64]

Tomorrow I am going back and I will attempt to run 1 or 2 more benchmarks related to gaming performance. Seeing as how these are public computers with limited time I can test this stuff, what would you guys be interested in seeing? Any small demo's I can quickly run?

I think I'll have mine before you report back, buuuuut.... if you don't mind, go to TheVerge.com and Polygon.com and try to swiftly scroll through the page back and forth. I know it'll be a tad subjective, but is it choppy at all or smooth as butter?

Thanks!

 

[aiyaaabatt]

Are those GPU temps or CPU temps? And are they max or avg temps?

 

[Zer0n1nja]

To answer both your questions, scrolling on the verge is like butter so long as you don't do scaled resolution. The 15 and 13 stutter but not as bad as the previous models. Haven't been to polygon but I'll check it out. The stutter isn't as bad on the 15.

I believe that unigine heaven reports back the gpu temps on the test which is what I based it off of. The temps were pretty stable only fluctuating a degree or 2. One thing of note, the 13 rmbp felt the coolest near the end of the touch and made the least noise, the air and base model retina made the most while the nvidia was a little quieter. Maybe that explains the big temps difference?

I plan on buying the 13 inch retina with my next paycheck so I was curious about the performance difference between the 2 models and if it was worth the upgrade. For me I don't think so as I'm personally a fan of the 11 inch but as a photographer the screen is gorgeous.

 

[fotuwe]

The 11 inch MacBook Air Screen is gorgeous for a photographer... ? Or do I get smt wrong?

My gosh its the worst screen u can imagine for that purpose

 

[briand05]

I just ran this benchmark on my 15" base model Iris Pro with identical settings to what you had on the 13" Iris. This could help provide some additional prospective on how the Iris and Iris Pro compare.