Catalyst 10.9 is out

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
10.9 is a huge success for me. Kudos to AMD!

AVP FPS went up by 25%

Mafia 2 FPS increase 20%. (They fixed a issue)

10% more frames on Metro 2033

FPS up by 25 frames in BC2 when Direct X11 is enabled. (I usually disabled it because I get more frames in DX9, but now I get more frames in DX11 then Dx9!)
 
YAY Praise the....uhmmm people at ATI ! Proper 2D idle clocks! (157/300 and not 400/OC mem) Only took 5 months to fix!

I'll be comparing performance of the 10.9's against the 10.4's in various benchmarks and report back with the results.

Oh and installation went fine, no driver sweeper, no uninstall, just downloaded, ran it straight over 10.4, reboot and jobs a good'en.
Overscan worked perfectly, all I had to do was right click desktop and open CCC and it automaticly overscanned : )
 

Damn due, optimizing for BC2 makes sense, but really? Source??? Can't you already play CSS/HL,etc at like 100fps?!?!?
 
AMD Releases ATI Catalyst 10.9a Hotfix for Radeon HD 4800 X2 Series Graphics Cards
Less than a day after releasing the ATI Catalyst 10.9 WHQL driver suite, AMD rolled out the Catalyst 10.9a Hotfix driver for users of the ATI Radeon HD 4870 X2 and Radeon HD 4850 X2 graphics accelerators. Applicable for Windows Vista and Windows 7, Catalyst 10.9b Hotfix improves performance of the said graphics cards, eliminating a bug that lets the OS make use of only one of the two GPUs.
http://www.techpowerup.com/131114/AMD-Releases-ATI-Catalyst-10.9a-Hotfix-for-Radeon-HD-4800-X2-Series-Graphics-Cards.html
 

:lol: :lol: :lol: :lol: :lol: :lol: :lol: :lol: :lol: :lol: :lol: :lol: :lol: :lol:

The CS:S bench gives me 233.....and thats at 600/450 clocks. At default 750/900, I get......233.3. It's CPU limited!
 
Unfortunately, no time to test 10.9 out as I'm setting up my SSD but I did install it without issue... I should be gaming before the weekend tho so hopefully I'll have some feedback.

omgitzfatal, looking forward to seeing the benchs 😀
 


Yea, I guess there are a few out there that really don't care if a new driver is faulty or not since they can always revert to the previous driver and problem solved, basically what I do every time a new one is released. I guess I got used having good drivers for the most part since I have mostly been using NV cards in the past 15 years. The thing that bothers me is the fact that the 5xxx series is capable of so much more with better/stable/optimized drivers 🙁
 
Ok so got a few results for you guys.

All results done with a HD5850 @ 1000 Core 1200 mem

BFBC2 maxed (except only 1xAA) 1920 x 1080 - 1 minute of Heart Of darkness Level - recorded FPS with FRAPS

10.4 results: MinFPS = 63 AvgFPS = 76.883 MaxFPS = 136

10.9 results: MinFPS = 37(WTF?) AvgFPS = 86.402 MaxFPS = 136

Heaven Benchmark 2.1 1920 x 1080, Tesselation = normal, Shaders = High, Anisotropy = 4, AA = None Full Screen.

10.4 results: MinFPS = 17.5 AvgFPS = 39.7 MaxFPS = 104.6

10.9 results: MinFPS = 15.9 AvgFPS = 38.6 MaxFPS = 103.3

STALKER CoP Benchmark Maxed out 1920X1080 NOAA DX11 HBAO Enabled SSAO Quality: High Tesselation = Enabled

10.4 results:

Day: Min: 49.3FPS Avg: 195.6FPS Max: 317.7FPS
Night: Min: 43.5FPS Avg: 223.3FPS Max: 412.4FPS
Rain: Min: 49.1FPS Avg: 255.6FPS Max: 423.5FPS
Sun Shafts: Min: 40.5FPS Avg: 130 Max: 227.7FPS

10.9 results

Day: Min: 51.4FPS Avg: 202.1FPS Max: 314.8FPS
Night: Min: 46.2FPS Avg: 227.5FPS Max: 411.6FPS
Rain: Min: 52.4FPS Avg: 254.3FPS Max: 425.9FPS
Sun Shafts: Min: 66.3FPS Avg: 131.1 Max: 234.7FPS

So abit of a mixed bag so far, BFBC2's min FPS taking a huge hit, but the average FPS went up by 10FPS.... Weird.
Heaven results were dissapointing, seeing a loss across the board.

Stalker's results were better, good to see the minimum FPS getting a healthy boost.

More to come shortly.
 

What gets me is that for something that is issued every month and has been for the last few years they don't appear to be solid, reliable and improved each time. If the wages department at your place of work got your money wrong each month I'm sure that you'd want to know why people who are doing the same thing each month can't seem to get it right.
 
Lol I'm not trying to do anything underhanded, I'm surprised at the results too.

stalkerdx11ultratessela.png
 


You must be the luckiest guy on earth, especially when there were no improvements whatsoever for those games you listed above.

In my last 15 years of playing FPS games I have never seen a driver give the user an extra 25 frame boost, man you got lucky!! :lol:

 

The Stalker benchmark can be a tad odd sometimes, I recall Jennyh getting strange results with CF 5770's a while back.
 


Yeah I'll just take it as the benchmark being buggy, People can still see the difference between my 10.4 results and 10.9 results. 😉
 
I never leave anything at stock cuz I r harcor3! (Jokes) but seriously I don't know, but I can find out for you tomorrow, I have FC2, HawX (demo) and RE5, along with others like Dirt 2

I won't be doing anymore comparisons today, have to wait until tomorrow I'm afraid.
 
trial and error...installing and uninstalling are getting pretty tiresome.

They should just release it when they know for sure they have an improved driver...instead of rushing it out to meet monthly release.
 
Ok then tomorrow plz run dirt2,re5(variable benchmark).At stock setting of 5850 and ur cpu.

I'm not resetting my rig back to stock man, this is just a driver comparison, I'm happy to put my HD5850 at stock for BFBC2, and I'll run the benchmarks you asked for, but I only have 10.4 results for dirt 2 and re5 at 1000/1200 and CPU @ 4Ghz.
 
Yes I have them installed, and I already have the 10.4 results for those games (noted them down awhile ago)

What I'm not prepared to do is revert back to 10.4 and put everything to stock, run the benchmarks again, then reinstall 10.9, re-run the benchmarks then set my OC's back.... Its just alot easier if I compare the drivers with my card and CPU/RAM at there normal speeds.

This isn't a Overclock Vs Stock Vs 10.4 Vs 10.9...thing.