Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
I always benchmark at a baseline of all third party apps terminated. No background Dropbox, MagicPrefs, Paste, Bartender, Droplr, 1Password Mini or other utilities. This is usually what professional testers do to be consistent and establish a baseline for comparison. I also wait at least a day on a new machine so Spotlight can stop indexing, iCloud Photos can sync, Time machine can backup, etc. I also unplug drives and external displays. I benchmark after a reboot and wait about 5-10 minutes for everything to settle down. Might explain why mine is faster, but it’s also not a fake result even though you’d be running these apps while working because it removes variables between user configurations when testing. Many people don’t think about this and could be running tests with all kinds of things going on in the background and who knows, maybe even half the Adobe Creative Cloud running, lol. I also check activity monitor before testing to make sure nothing weird is going on and that the CPU is close to 0%. I’d say my results are more likely to be consistent with professional benchmark results.

Yes I had Steam running in the background for instance last time. Will try it with everything else closed down to get more accurate numbers.
 
So I had some time to run the benchmarks again, on Windows and MacOS this time.

All benchmarks where ran with nothing else in the background and after waiting a few minutes following startup. I got my iMac since Tuesday so I don't have any iCloud downloading or indexing in the background by now.

Extreme Preset, Open GL, MacOS :
da2bmTf.jpg


Extreme Preset, Open GL, Windows :
NkKN896.jpg


Extreme Preset, DirectX11, Windows :
8ojnwXu.jpg


Bonus picture, it's from the Radeon Pro settings panel, it lists some numbers about the RX580X :
vQFva5R.jpg


Important numbers :
Memory Clock : 1695MHz
Processor Clock : 1200MHz
6.78Gbps
Memory Bus : 216GBytes/s
 
Last edited:
So I had some time to run the benchmarks again, on Windows and MacOS this time.

All benchmarks where ran with nothing else in the background and after waiting a few minutes following startup. I got my iMac since Tuesday so I don't have any iCloud downloading or indexing in the background by now.

Extreme Preset, Open GL, MacOS :
da2bmTf.jpg


Extreme Preset, Open GL, Windows :
NkKN896.jpg


Extreme Preset, DirectX11, Windows :
8ojnwXu.jpg


Bonus picture, it's from the Radeon Pro settings panel, it lists some numbers about the RX580X :
vQFva5R.jpg


Important numbers :
Memory Clock : 1695MHz
Processor Clock : 1200MHz
6.78Gbps
Memory Bus : 216GBytes/s

Thanks for doing all this, it's really apreciated.

Is ther eno direct3D11 for MAC OS?
[doublepost=1554649355][/doublepost]
Wow, that benchmark depends a lot on the software/driver used! Thanks for taking the time.

Makes the Vega nearly 50% faster on MacOS, that's considerable.
Where are you getting the 50% faster on MacOS, I can't see but one image for MacOS
 
Where are you getting the 50% faster on MacOS, I can't see but one image for MacOS

Sorry for not being clear-what I meant was: The Vega is nearly 50% faster than the 580X in this benchmark. For the Vega we got a value of approx. 1700 (see e.g. previous page: 1713), here we have 1156, so the Vega is 48% faster if we take the 580X as the baseline. Of course, benchmarks will vary from run to run, but still, this is a very noticeable advantage for the Vega IMHO.
 
How well will the 580x handle 4K in a program such as iMovie - I’m not talking major projects but 10-15 min videos tops. I’m a hobbyist and shoot 4K on my phone now and do a lot of photo editing too.
 
Sorry for not being clear-what I meant was: The Vega is nearly 50% faster than the 580X in this benchmark. For the Vega we got a value of approx. 1700 (see e.g. previous page: 1713), here we have 1156, so the Vega is 48% faster if we take the 580X as the baseline. Of course, benchmarks will vary from run to run, but still, this is a very noticeable advantage for the Vega IMHO.

I do think that's a mistake of mixing different OS and openGL/direct3D11. I Think the difference is more a 20-30% I would actually need better vega 48 results, as everything is really scattered. I'm trying to compile many of these results, and once I get them I'll create a new post. Kaelis has been a great support.
 
  • Like
Reactions: Kaelis
How well will the 580x handle 4K in a program such as iMovie - I’m not talking major projects but 10-15 min videos tops. I’m a hobbyist and shoot 4K on my phone now and do a lot of photo editing too.

Hey,

If anyone knows where I can get some 4K raw footage to download I could try that. My Nikon D810 shoot up to 1080p but no more.

I have some XS Max 4K footage but I’m not sure if it’s as heavy as some real camera footage.
[doublepost=1554706946][/doublepost]
Sorry for not being clear-what I meant was: The Vega is nearly 50% faster than the 580X in this benchmark. For the Vega we got a value of approx. 1700 (see e.g. previous page: 1713), here we have 1156, so the Vega is 48% faster if we take the 580X as the baseline. Of course, benchmarks will vary from run to run, but still, this is a very noticeable advantage for the Vega IMHO.

Hey Bohemien, the score you got for the Vega 48 is the good one but the screenshot on the previous page does mention DirectX11 in the graphic APU information.

From what I gathered so far the Vega 48 is about 25% faster than the RX580X in this benchmark when the same graphic API are used.
 
The 580x is more than sufficient to handle and edit 4K videos. There is a lot of talk here about the Vega being a lot faster but cost wise the performance leap may not be worth it if you are planning on editing 4K. Remember, GPU's geared for gaming do not necessarily make great video editing GPU's. I mean, they will work and will probably speed up rendering but by in large, the 580x will perform just fine. Actually the CPU has more of an impact on rendering time so get the fastest CPU with the most cores you can afford.
 
Hey,

If anyone knows where I can get some 4K raw footage to download I could try that. My Nikon D810 shoot up to 1080p but no more.

I have some XS Max 4K footage but I’m not sure if it’s as heavy as some real camera footage.
[doublepost=1554706946][/doublepost]

Hey Bohemien, the score you got for the Vega 48 is the good one but the screenshot on the previous page does mention DirectX11 in the graphic APU information.

From what I gathered so far the Vega 48 is about 25% faster than the RX580X in this benchmark when the same graphic API are used.

Hi Kaelis, Is there no direct3D11 for MAC OS result? from the set of screenshots you supplied , thanks a lot for the effort, I could not see that one.
 
Direct 3D 11 is exclusive to Windows as part of Direct X, so that is not something that can be tested with macOS.
Ah cool, that makes sense. I don't know why I Though it was independent to Direct X. Then we have all of them for Karlis model. I'm now procuring some from another forum user for the i5 vega 48
 
My 580X didn't go beyond 70 degrees during Unigine benchmarking, compared to 84 degs on my old 2017 580.

Also here's my 2019 i9/580X cinebench results
 

Attachments

  • Screen Shot 2019-04-12 at 7.49.27 PM.png
    Screen Shot 2019-04-12 at 7.49.27 PM.png
    225.3 KB · Views: 510
  • Like
Reactions: pixelshaders
can somebody here please use Unigine Superposition benchmark on windows? One at "4k optimized" preset and one with "1080p medium" preset. It would be very useful if we have Vega benchmarks as well with this.
Unigine Heaven is quite outdated and lack a lot of modern features currently used on modern 3d software, Superposition at 4k helps a lot understanding how slow this machines are with stuff as Unreal Engine, Substance, Mari or Maya running at working desktop resolution.
Those scores I saw on Heaven worry me a lot, the 580X seems almost twice as slow as a mobile rtx2060 from the razer blade and performance will likely tank at working resolution. But with vega gpus it could be a different story.
Thank you.
 
My 580X didn't go beyond 70 degrees during Unigine benchmarking, compared to 84 degs on my old 2017 580.

Also here's my 2019 i9/580X cinebench results


I am really want to get the i9/580x config, really not need the extra performance on the vega48, except if it is much cooler more efficient to run.
 
I am really want to get the i9/580x config, really not need the extra performance on the vega48, except if it is much cooler more efficient to run.
I'm in the same boat. I'm programmer and want maximum CPU performance with average GPU. I was afraid that 580x will be real heat source to the system. I understand that basically Vega 48 and Radeon 580x has some process 14nm, so it should not be much difference in heating. Many people writing here and on reddit that Vega 48 MUCH cooler, but I don't really believe in this. It's looks like they trying to justify $450 premium for themselves :)

So, I take a risk and order iMac 27/i9/8GB/512GB/580x. And here's my results:
Please take a look on GPU temperature. I'm not saw it more than 85C.
Also please take a look to the second image. When I run 16th consoles with 'yes > /dev/null' and load CPU, fan speed goes up and GPU temperature drops less than 75C.

I'm very happy with my purchase. I hope this information helps somebody.

https://imgur.com/a/gov3d7c

Thanks
 
Last edited:
In Valley, again, if CPU is not loaded, fans not going up and GPU is heating up to 88C.
 

Attachments

  • Screen Shot 2019-07-10 at 11.30.53 PM.png
    Screen Shot 2019-07-10 at 11.30.53 PM.png
    3.3 MB · Views: 410
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.