• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

***The DigitalFoundry Face Off Thread***

Farcry primal 970 vs 390


Well done AMD, although I did see at least one discrepancy in the test. They seem to have been carried out at a different time of day (in-game), and the Nvidia card is having to reproduce more shadows (on grass, leaves, and ground) than the AMD because of this. You can see an example of this by pausing at 1:26 and at 2:17-18 in the video. I'm not saying the AMD still wouldn't win, as this may not be an issue for most of the video, only in some parts, but it's there.
 
Last edited:
Well done AMD, although I did see at least one discrepancy in the test. They seem to have been carried out at a different time of day (in-game), and the Nvidia card is having to reproduce more shadows (on grass, leaves, and ground) than the AMD because of this. You can see an example of this by pausing at 1:26 and at 2:17-18 in the video. I'm not saying the AMD still wouldn't win, as this may not be an issue for most of the video, only in some parts, but it's there.

Having said that, the results are in line with other sites result.
 
Something different this time. CPU test with 6700K, vs 5820K vs 5960X.

Results clearly shows that Skylake with it's improved IPC is the winner in DX11 tittles even under SLI. Still waiting to see proper DX12 benches.
 
Something different this time. CPU test with 6700K, vs 5820K vs 5960X.

Results clearly shows that Skylake with it's improved IPC is the winner in DX11 tittles even under SLI. Still waiting to see proper DX12 benches.

Its not that much of a win though for the most part - a lot of the time there is barely a whisker in it aside from FC4 where the overclocked Skylake steps ahead - also be wary comparing the framerates at the top right hand corner as they are at times slightly out of sync with what scene each CPU is displaying - if you use the graph as a guide to when they are showing the same thing the difference in performance at some points is much less than the difference in the numbers at the top.
 
Its not that much of a win though for the most part - a lot of the time there is barely a whisker in it aside from FC4 where the overclocked Skylake steps ahead - also be wary comparing the framerates at the top right hand corner as they are at times slightly out of sync with what scene each CPU is displaying - if you use the graph as a guide to when they are showing the same thing the difference in performance at some points is much less than the difference in the numbers at the top.

Getting even those differencies is hard when your gpu bound in most scenarios. But even then, newer arch can fill those buffers faster when gpu is asking for more , to get that small improvement in performance. Even 6770K is fastest in general in test, I would take that 5820K anyday instead of it. Already seen in ashes that 3770k gets bottlenecked, and we're heading towards games that will be even more and more multithreaded (not just d3d12 rendering part), and they will give bigger difference between 4/6/8/x+ core cpu's.

About test:

I'm just disappointed in general. If you wanna do cpu test , why in the earth would you be choosing nvidia card for it, when AMD is known of driver overhead. Would have made a lot more difference in graphs aswell. Especially between overclocked and stocks.
 
I'd be worried if the 390 wasn't winning.

They might be around the same price but the 390 is a newer and technically more capable card.

LOL, you can't be serious.

AMD got ripped to shreds by you and other posters here for rebranding Hawaii (290 series) as Grenada (390 series). They are both in fact Hawaii, identical silicon. The 390 series simply has double the VRAM, 8GB, slightly faster clock speeds etc.

Maxwell is still a much newer architecture than the 390 series.

I'm quite a fan of NVIDIA since witnessing what my Asus Skylake 980M ROG laptop can do with gsync, also most likely getting a pascal card for my desktop PC, but I have to defend AMD when people like you make statements like that.
 
Wow, another nice win for AMD there! Hopefully it's just a driver issue with the nVidia lineup :)

Tbh I would say all users stay well away.
Even though amd is ahead shoddy performance is still there. The game for some reason on 60hz is locked at 50fps lol

I tell you what Microsoft is killing pc gaming with this horrible windows store junk.

I will say though excellent work from amd release drivers etc showing again there leading the way in DirectX12.
 
While the 390 is the faster card running that game - there is a problem there with VSync where the 970 is just missing the multiplier and the 390 is just hitting it - resulting in a massive but artificial performance differential...

GG Microsoft.

EDIT: To go into a bit more detail:

providing an objective frame rate of 45 FPS for frame rates 59 FPS to 45, 30 FPS for frame rates 44 FPS to 30 FPS, and so on.

This means that the 390 could be capable of say 45fps and the 970 44fps but one will output 45fps and the other 30fps - notice how in the video the framerates are very consistently sitting around 30fps for the 970 and 45fps for the 390 - with a little drift due to the fact that the framerate readout is sampled from an average over 1 (or more depending on the particular implementation) seconds. Which is due to the horrid MS store implementation not an AMD or nVidia thing.

Though the crashing drivers is entirely on nVidia.

EDIT2: Oh and there is a 50fps cap in the game wow! not in a position to watch the video in its entirety atm so some of that might be covered in the video.
 
Last edited:
While the 390 is the faster card running that game - there is a problem there with VSync where the 970 is just missing the multiplier and the 390 is just hitting it - resulting in a massive but artificial performance differential...

GG Microsoft.

EDIT: To go into a bit more detail:



This means that the 390 could be capable of say 45fps and the 970 44fps but one will output 45fps and the other 30fps - notice how in the video the framerates are very consistently sitting around 30fps for the 970 and 45fps for the 390 - with a little drift due to the fact that the framerate readout is sampled from an average over 1 (or more depending on the particular implementation) seconds. Which is due to the horrid MS store implementation not an AMD or nVidia thing.

Though the crashing drivers is entirely on nVidia.

EDIT2: Oh and there is a 50fps cap in the game wow! not in a position to watch the video in its entirety atm so some of that might be covered in the video.

There is more to it than that. I get your point though but the gtx970 is struggling to hold 30fps at points never mind 45.
 
Last edited:
Yeah even in the video the 970 was in high 20s.. With Nvidia driver crashing I would say this game need another driver from nvidia to boost performance.
50fps cap at 60hz from what I told if you using 120hz or 144hz you can go higher but the inconstant frame rate is horrible
 
There is more to it than that. I get your point though but the gtx970 is struggling to hold 30fps at points never mind 45.

Its fairly understandable why with the horrid VSync implementation forced by MS UWP - you only need one tiny miss when you are already down and you momentarily blip to 15 or 20fps or whichever the next multiplier down is with that implementation when you could have been at upto 44fps otherwise - its pretty much impossible to gauge the true performance differential - especially with the framerate cap further making things complicated. (Even with all that its obvious the 390 has a framerate advantage there but its impossible to tell whether that is a 1% advantage or a 200% advantage).

The game is broken, MS is broken and nVidia drivers are broken lol.
 
Last edited:
Back
Top Bottom