Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Personally, I like using Galaxies 32k as a video card benchmark.

Here's what my Radeon 7970 from MacVidCards gives me:

3c7UGmG.png
 
The benchmark is heavily CPU bound, AMD drivers have less CPU overhead compared to Nvidia, so your 5870 will outperform every single Nvidia card you can buy, simple as that.

Is this still true with the new web drivers from NVIDIA?
 
Judging by benchmarks with heavy CPU usage (e.g. Cinebench or CS:GO) it still is, although the Web Drivers are definitely an improvement over OS X stock drivers (at least for Kepler, Fermi didn't change much).
I don't have any Maxwell card, so I can't test those.

I made some benchmarks in my 3,1 some months ago and posted results in some thread here, maybe I can find it.
 
Did you read the thread? Cinebench is completely meaningless.

The benchmark is heavily CPU bound, AMD drivers have less CPU overhead compared to Nvidia, so your 5870 will outperform every single Nvidia card you can buy, simple as that.

Have a look here: http://www.tonymacx86.com/graphics/177227-graphics-testing-benchmarking-chart-4.html#post1158102

Scores are perfectly proportional to CPU clock speed when OC'ing but doesn't change notably when upgrading GPU.


The thing you missed is that this completely irrelevant to other applications/games. A GTX 970 will easily outperform your 5870 in gaming or GPGPU applications, just not in that stupid benchmark.
The 9 series performs worst than the 7 or 6 series in other tests and apps at Barefeats. It's just not a Maxwell optimised driver so even though it has more compute units for GPGPU, it has rather ****** Open GL performance compared to what the same GPU could do with a proper driver support.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.