AMD Catalyst 15.7 Driver is now available!

[Rvenger]

GCN 1.0 is 100% programmed at the shader level I believe.

 

[tential]

I thought I referenced a link on guru3d. If that's not allowed and it was deleted then oops, but if I just completely forgot because I'm deliriously hungry then I'll actually post it lol.

 

[dn7309]

I like the new driver. The frame rate control is a nice alternative to Vsync. Limits the card to 60fps so the there's not a lot of tearing and my FPS doesn't tank every time it drops below 60.

 

[VulgarDisplay]

Witcher 3 just seems a lot more smooth now. It was fine before, but now it's playing noticeably better.

No idea on framerates before and after.

 

[Udgnim]

FRTC does not seem to work for me in Witcher 2 with a R9 290

 

[burninatortech4]

FRTC does not seem to work for me in Witcher 2 with a R9 290

FRTC only works on 300 series cards.

 

[freeskier93]

FRTC only works on 300 series cards.

Guess the FRTC controls are just for show then with my 280... You should probably read the full release notes.

 

[biostud]

Still gets tons of flickering in Dying Light

 

[burninatortech4]

Guess the FRTC controls are just for show then with my 280... You should probably read the full release notes.

My bad. It was 300 exclusive on 15.15.

 

[MeldarthX]

FRTC only works on 300 series cards.

that's wrong FRTC will only work on DX 10 and above games; thought witcher 2 was still dx 9...but I could be wrong....

*DX 10 confirm came from AMDMatt off OCUK subs*

 

[wege12]

So has anyone re benchmarked the Fury X with these new drivers against the 980ti?

 

[railven]

So has anyone re benchmarked the Fury X with these new drivers against the 980ti?

Won't matter, these drivers bring no performance gains it seems:

http://www.hardwarecanucks.com/foru.../69792-amd-r9-fury-performance-review-20.html

So there you have it: exactly as AMD described, there isn’t anything to distinguish these two driver sets from one another on the performance front. While there are times when one or the other pushes a bit ahead, these new drivers are a literal carbon copy of the 15.15’s since all of the differences can easily be chalked up to margin of error.

Regardless what some early forum posts said, these drivers aren't performance drivers.

 

[wand3r3r]

Won't matter, these drivers bring no performance gains it seems:

http://www.hardwarecanucks.com/foru.../69792-amd-r9-fury-performance-review-20.html

Regardless what some early forum posts said, these drivers aren't performance drivers.

The release notes specify crossfire improvements.

Whether or not these are already included in the early Fury X driver is a question which would be interesting to know.

 

[96Firebird]

Anyone re-bench a 290X vs 390X at the same clockspeeds with these latest drivers?

 

[PG]

Averages might not have improved much, but many are reporting better minimums and smoother overall performance in a variety of games. A simple graph with averages doesn't really show that.

 

[freeskier93]

I like the new driver. The frame rate control is a nice alternative to Vsync. Limits the card to 60fps so the there's not a lot of tearing and my FPS doesn't tank every time it drops below 60.

FRTC is in no way shape or form even close to a V-Sync alternative.

 

[Gonad the Barbarian]

Won't matter, these drivers bring no performance gains it seems:

http://www.hardwarecanucks.com/foru.../69792-amd-r9-fury-performance-review-20.html

Regardless what some early forum posts said, these drivers aren't performance drivers.

They're using Windows 8. The big performance gains are in the Win10 drivers. Plenty of posts on the guru3d forums have been showing a ~40% improvement in draw calls over previous versions. Any review site should really have a better handle on this. Relevant Guru3d thread is here: http://forums.guru3d.com/showthread.php?t=398858&page=39

 

[xthetenth]

Is that just in DX12 synthetics or current games?

 

[Gonad the Barbarian]

The improvements show in DX11 in Futuremark's API overhead test and I have every expectation they would also show in current games where CPU overhead was/is an issue.

Edit: From this post:
http://forums.guru3d.com/showpost.php?p=5117743&postcount=955
15.6

15.7

 

[Madpacket]

Wow, great timing. I just bought a new 1440P freesync monitor and my 290X needs all the help it can to drive it. Hopefully this stops some of the dips in the framerates from going under 35FPS (my cut off point for freesync).

Will do Witcher 3 / BF4 comparisons sometime this weekend.

This driver overhead thing has been an issue for a LONG time. Better late than never, keep closing the gap AMD.

 

[RaulF]

I dont know whats going on, but i just put my 295X2 since im sending my Ti for repairs and wow, is running like shit on this 15.7 release.

Going to roll back to omega and see what happens.

 

[kawi6rr]

Saphire Trixx-OC 290, installed new drivers but can't find where I can limit frame rates in CCC even under the game I was wanting to limit. Other then that great drivers!!

 

[dn7309]

Saphire Trixx-OC 290, installed new drivers but can't find where I can limit frame rates in CCC even under the game I was wanting to limit. Other then that great drivers!!

under gaming tab when you can also see the AMD overdrive mode

 

[RaulF]

Might be something else wrong.

I take back I typed before, I notice I was running the game borderless and not full screen and that might be the problem.





Yup, when i switched to the Ti i went borderless, and i guess that kills AMD for some reason, it almost felt like i was running a single card and not Xfire. Can't tell anything about the whole freesync, i usually run fairly high frame rates.

The only time i have for sure notice tearing is the minibird chopper blades as a passenger, will report back when i get a chance to see them.

 

[Rvenger]

Anyone re-bench a 290X vs 390X at the same clockspeeds with these latest drivers?

I bet the clock for clock performance is identical now. I have a suspicion they withheld 290/x from the driver updates to paint the 390x in a better light. Sounds like funny business and I hope I am wrong.