WYP
News Guru
Let's hope that this is better than the game's atrocious standalone benchmark.
Read more about Final Fantasy XV's PC release.
Read more about Final Fantasy XV's PC release.
Final Fantasy XV: Windows Edition PC Performance Review
- Thread starter WYP
- Start date
We already know this game is heavily promoted by nVidia and they have helped alot in it's developement. So for me personally this test isn't that interesting with the very limited GPU selection available. RX580 4 and 8 GB, Vega 64 and 56 is a must, otherwise I see no point in even looking.
WYP
News Guru
We are working to get a Vega 56 on site for games test, big problem here is Denuvo, which will lock people out after a certain number of hardware changes, which is a real pain. This is why not many places do CPU testing as well as GPU testing.
The RX 580 is basically a higher clocked 480, the perf gap is relatively small there, as our RX 480 is a factory overclocked Strix model. As is the difference between the 4GB and 8GB versions, only VRM capacity and in some cases VRAM clocks are changes, so differences only occur when capacity limited or bandwidth limited.
Thanks for the input.
Dicehunter
Resident Newb
The game is pretty but from the few hours I've played so far I don't think it should command such a performance hit, A few optimisation patches are needed me thinks.
WYP
News Guru
In densely forested areas the game can look phenomenal, though I do agree that some areas could have looked a lot better.
RedSymphony
New member
On page 4 in the (High vs Highest) comparison, it says 'Average Preset' in the top left and 'High Preset' in the top right of the slider image. Had me confused for a bit.
RedSymphony
New member
I'm hugely surprised about your GameWorks performance hit analysis. When playing the demo on a R9 390, I saw my framerate tank enabling turfworks alone. I went from what I believe was around 32fps down to the low 20s.
WYP
News Guru
Thanks for the spot there.
I have fixed the error. The images and sliders should appear correctly now.
The change could easily be architecture based, there are a lot of changes that AMD made with Polaris and newer architectures. Turf FX does use tessellation, which is a factor which was upgraded with Polaris, so that could explain the performance discrepancy. There is also the increased CPU loads offered by TurfFX to consider
Then there is also the fact that we found the demo to perform somewhat differently than the final release (older build), which could also play a role here.
Last edited:
NeverBackDown
AMD Enthusiast
Aren't you under water?
Yes. And that is what shocks me most.
After some analysis I found that the fittings paint has flaked off and caused restriction in the flow. Basically they are blocking my cpu and gpu.
I had this happen as soon as I switched to EK cryofuel navy blue. Now, it could be might fault (somehow) or the coolant itself. I haven't seen others with the issue so I'll take blame. I find it odd to cause flaking of the fittings though. Can't see how I could have caused it either.
Last edited:
I have noticed during my own running of the game that I can run around 80 - 90 FPS most of the time at 1440P simply by reducing LOD - I found it whilst reading through some optimisations for the demo that LOD seems to applied incorrectly and frees up around 20 FPS simply lowering from high to average on my system:
Ryzen 1700@ 3.925ghz
Palit 1080 Gameock Premium
16gb 3200mhz Trident Z RGB
Custom Loop
Ryzen 1700@ 3.925ghz
Palit 1080 Gameock Premium
16gb 3200mhz Trident Z RGB
Custom Loop
I found this aswell! To the point I have stripped my loop back and GPU block, split it, cleaned it, swapped the inlet and outlet around and finally replaced the EKDDC pump/res combo with an Alphacool Light Tower with DDC - Temps dropped 10C from all this, still has my 1080 sitting at 48C though under my loop (360 + 240/60mm + 120mm)
At least the pump is a tonne quieter!
