I tested with the newer driver but i get much lower performance 30-50 fps lower fps, What is wrong?

Ok, I get that Nvidia GPUs are now faster with updated driver, but why AMD cards FPS drop by like 30% in updated chart ??? For example @4k: 390x 51->35 Fury X 50->33

Because shadow cache is disabled for amd as well. RX 480 doing around 128 fps@ultra in 1080p with on option. Some sites are reporting 122-116 or 128 and so on. Pretty much giving the similar performance as you have with 980Ti.

Ok, I get that Nvidia GPUs are now faster with updated driver, but why AMD cards FPS drop by like 30% in updated chart ??? For example @4k: 390x 51->35 Fury X 50->33

Shadow cache improves AMD Performance a lot but cuts NVIDIA Performance, thry disabled Shadow cache to "update" The review

Shadow cache improves AMD Performance a lot but cuts NVIDIA Performance, thry disabled Shadow cache to "update" The review

Did you see what shadow cache on/off does to Fury? http://www.guru3d.com/articles_pages/resident_evil_7_pc_graphics_performance_benchmark_review,8.html

I think it is useless to benchmark this game at te moment, the framerate every time different at the same place. I think it have to do with video memory management, cause wen my video memory runs full the framerate drops down.

Oh god, there he is, the worlds biggest AMD fanboy. Should I remove the new driver results for NV as well then? I mean the older drivers are much faster ? BTW Shadow Cache on creates big problems with ALL cards up-to 4GB, learn how to read the damn content before you thrown around idiot comments.

Original graph for 4k fury had higher FPS than on new one. I am just very much confused there because this does not make sense. If original had Shadow cache enabled and new one is identical but with Shadow cache off and what you said about all 4k cards, then id does not explain huge drop of fps on fury, it should be opposite. 😕 Maybe can you add new graph just for few AMD cards comparing Shadow cache OFF/ON effect ?

I honestly cannot spend even more time on this article as I need to move on towards next weeks articles. I tested all cards 3x each time with different results. The last batch is considered final and as good as it gets. The initial fury results you mention being higher FPS in Ultra HD might have had an error from my side. I was playing around with interlace modus and it might have stuck in Ultra HD. However I am not even 100% sure that could have caused the offset. That result set however was re-done and replaced in v2 of the benchmark session. The game is complicated for testing with so my differentials and anomalies per card. We test with one setting for all graphics card to offer an objective result set. However I recommend you guys play the game with cache off up-to 4 GB cards. This will offer a much better gameplay experience for say R9 Fury cards and for NV the GTX 980 series. So the current (revision 3) tests as they are online since yesterday are the closest match to performance. It however is a mixed bag of results, but there is nothing I can do about it as it is game-engine related. It is a tricky to test title. As stated in a few weeks we'll have had driver updates and game patches, likely I'll revisit this review then.

Thanks for explanation Hilbert , just wanted to understand those little bit erratic results.

I also love how no one cared when the preliminary Nvidia results got removed that were all significantly higher. I guess that doesn't fit the crybaby narrative so it's simply ignored.

Some have reported lower than some higher performance with the new driver, which one is it? I saw first HH benchmarks and indeed performance seemed better with the previous one but how can a "game ready" driver mess up the performance that bad? It sure isnt game or HH fault but nvidia.

Edit: Delete please

You seem to be forgetting revision 1:

I think the problem is that the picture isn't linked in the article. Most people are going to go to the article and not to the forum thread where they can see that. Perhaps just add a section in the article under the new results called "Cache On" or something with that graph(s) so people can see that turning it off impacts performance for AMD.

I've just cleaned up this thread and said goodbye to one more user. I'll repost what I stated earlier. There have been an unfortunate three revisions of the benchmarks with results all over the place. The last batch is the most valid. The Shadow Cache settings disabled offers the best game-play for the majority of cards hence it is the preferred and recommended setting, for 6GB and upwards shadow cache enabled can offer better results, use it at your own peril. In-between revision 1 and 3 of the review a lot has happened, new drivers 5 minutes after I posted the article, totally weird anomalies but also an error on my side that boosted perf on the earlier tests. I think this was the cause of having interlaced modus activated (mistakenly) on some of the cards as I have been goofing around with that on day 1. Totally my bad, but I am not even 100% sure either that was issue as the game remains difficult to measure. A mentioned several times now, I will retest the game in a few weeks when the drivers on both sides have mature and when the game has had a patch or two. The initial results up-to revision 3 have been far from ideal, but the rev3 results do match what you guys should get gameplay / FPS wise closest. I have also retested an rx480 with both the shadow cache on and disabled. The perf differential is not far away from each other really. However on the opposite side, going to a 4GB card with shadow cache enabled brings in issues, stutters and game perf differences as shown in the FCAT results. That said, the results as they stand ever since yesterday evening with rev3 stand and are sound. Last thing, I am getting very tired of the attacks on my personal account here. I spend hours offering you guys un-biased results. Commenting is absolutely fine as mistakes do happen. But the verbal attacks lately are beyond what I deem is acceptable. The near sociopath behavior that some of you guys have been able to witness from the two users is just unacceptable. It partly comes with this job I know, but please do have some respect as I certainly do not deserve the incredibly crap that has been outted here. Thanks guys.

Last thing, I am getting very tired of the attacks on my personal account here. I spend hours offering you guys un-biased results. Commenting is absolutely fine as mistakes do happen. But the verbal attacks lately are beyond what I deem is acceptable. It partly comes with this job I know, but please do have some respect as I certainly do not deserve the incredibly crap that has been outted here. Thanks guys.

The "you guys" are only a couple trolls HH 😀. The hours you spend working on site are appreciated by vast majority here.

Last thing, I am getting very tired of the attacks on my personal account here. I spend hours offering you guys un-biased results. Commenting is absolutely fine as mistakes do happen. But the verbal attacks lately are beyond what I deem is acceptable. The near sociopath behavior that some of you guys have been able to witness from the two users is just unacceptable. It partly comes with this job I know, but please do have some respect as I certainly do not deserve the incredibly crap that has been outted here. Thanks guys.

Dont take it personal HH, most of us appreciate everything you are doing here and lets hope you'll continue to bring us the benchmarks of latest games despite everything that happened. Like you said, it comes with the job. 😉

Oh obviously we'll keep on making perf reviews, I love doing them. Example: the initial results didn't sit well, something had to be wrong. Then in the middle of the night I wake up at 3:30 and had the eureka moment. 04:00 in that night yesterday I started working on the shadow cache findings. That is the kind of guy that I am and that is how passionate I am on these articles. It's just that it is so nasty to remain uplifted with vicious personal attacks. I make these articles out of passion and dump so much time in them. Anyway, nuff said there. Moving on!

There is also such a thing as the released version of the game being different to the version the latest driver/s are optimised for. This is a drawback of being in the online-age where the game can be worked-on right up to the last minute and even then we see time and again day 1 patches etc. RE7 is not an established game on a well-known engine; it's brand new and we should expect bugs and problems. This is also a Capcom game. Expect the unexpected. Remember SF V? It should be obvious to everyone who's tested this game that it's not stable and due to this everyone will have a hard time benchmarking it with repeatable and consistant results.

What it did seems to do to 4GB gpu's .... Look like shadow cache increase highly the Vram usage.. This said, i suspect the way it do it is a bit buggy.

Yes, but you missed the context. I responded to someone who framed it as an Nvidia vs AMD thing rather than the 4gb issue that was affecting both camps.

The initial fury results you mention being higher FPS in Ultra HD might have had an error from my side. I was playing around with interlace modus and it might have stuck in Ultra HD.

Ah that's why they were so high at first.. Makes more sense now 🤓