• 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.

AMD Driver/Benchmark Review Thread

Caporegime
Joined
12 Jul 2007
Posts
44,521
Location
United Kingdom
AMD 12.9 Beta Driver/12.9 Caps Performance.

When new drivers come out i will update this thread with new results.

hmAkl.jpg


TWZND.jpg


Settings Used

vfac4.jpg



Official AMD Driver Reveiws & Performance Index

http://www.ocaholic.ch/xoops/html/modules/smartsection/item.php?itemid=883&page=11
 
Last edited:
May i remind everyone of the thread title.

I just ran a quick 10 minute benchmark with my i5 @ 3.7ghz stock and 4.7ghz overclocked.

I was convinced that i could feel no difference from overclocking my 2500k and i was correct.


Battlefield 3 - Bandar Desert - 64 man server
Low Quality Preset @1080p 2500k @3.7ghz
7970 -12.9 Beta+12.9 Caps @ 1125/1575

2012-10-16 13:00:14 - bf3
Frames: 40960 - Time: 300443ms - Avg: 124.331 - Min: 92 - Max: 176

Battlefield 3 - Bandar Desert - 64 man server
Low Quality Preset @1080p 2500k @4.7ghz
7970 -12.9 Beta+12.9 Caps @ 1125/1575


2012-10-16 12:50:58 - bf3
Frames: 43478 - Time: 300255ms - Avg: 144.804 - Min: 101 - Max: 193
 
Last edited:
I can't feel the difference on my x6 Thuban either, but i have seen people on stock or lightly clocked x4's bottleneck a 7950. especially if well clocked.

I think it's case of SB having the muscle to deal with it to start with.

Yeah, almost makes me think i should just run it at stock setting tbh, maybe 4ghz or something.

Can run 4ghz at 1.2v and under as opposed to 1.4 and under for 4.7.
 
matt, is that your every day 79 clocks?

If it is, then I'm feeling better that I don't have a lazy 79.;)

No Tommy, however those settings are enough for any game to run smoothly, including bf3 full ultra.

My 24/7 settings are 1200/1700. My max fully stable settings are 1225/1750. It can gi higher on the core, up to 1250 or so but once i hit 75c artifacts arrive, and i can't keep it cool enough.

Battlefield 3 - Bandar Desert - 64 man server
Ultra Quality Preset @1080p 2500k @3.7ghz
7970 -12.9 Beta+12.9 Caps @ 1200/1700

2012-10-16 13:32:30 - bf3
Frames: 26493 - Time: 360238ms - Avg: 73.543 - Min: 48 - Max: 159


Will run it again now with my i5 @4.7ghz.
 
Will run it again now with my i5 @4.7ghz.

Battlefield 3 - Bandar Desert - 64 man server
Ultra Quality Preset @1080p 2500k @4.7ghz
7970 -12.9 Beta+12.9 Caps @ 1200/1700


2012-10-16 13:57:32 - bf3
Frames: 27133 - Time: 353951ms - Avg: 76.658 - Min: 50 - Max: 177
 
Ok matt, back to being a lazy 79 then, the clocks you used are near enough my full whack.

Doubt I'd notice a difference anyway, I suppose it's just an average clocker rather than a poor clocker and at the end of the day 9 out of 10 I'm running@stock clocks as it eats everything up anyway.

It's your turn for a good one anyway.

:)

Yep not a lot in it tbh mate.

The AMD rep on another forum said that pretty much all 7970's will do 1100/1200 on the core, so at best id consider mine just above average.
 
Good news my fellow amd brethren. I decided to contact an AMD rep on another forum about bf3 performance. :cool:

LRLWD.jpg
 
Matt can you do a quick test for me. Do you get a big FPS drop on Metro when leaving the RU deployment and heading towards the cafe (A). Mine drops to around 45fps there but it's fine for the rest of the map.

Lol yes mate i know the point you talk of. Hang tight and ill get some numbers for you.
 
Matt can you do a quick test for me. Do you get a big FPS drop on Metro when leaving the RU deployment and heading towards the cafe (A). Mine drops to around 45fps there but it's fine for the rest of the map.

I did two runs, one recording video footage and one without video footage. I recorded video to show you which route i took and if i did it correctly.

Recording lost me a few fps.

Full ultra with vsync on

Normal Run

2012-10-17 11:50:05 - bf3
Frames: 1786 - Time: 30045ms - Avg: 59.444 - Min: 54 - Max: 61

Recording Run

2012-10-17 11:49:08 - bf3
Frames: 1912 - Time: 30047ms - Avg: 58.033 - Min: 50 - Max: 62


Video

 
Amd 7xxx Brethren,

I decided to contact the amd rep again, here's something to wet your beaks.

zB5ip.jpg


I don't know about you, but i can't wait for the 22nd of October. I hope im not disappointed. :D
 
AMD 12.9 Beta Driver/12.9 Caps Performance.

When new drivers come out i will update this thread with new results.

Battlefield 3 operation swordbreaker - played from start of level to where it ends on this video.

hmAkl.jpg


TWZND.jpg


Settings Used

vfac4.jpg
 
Last edited:
Hm, nothing like running your own benches - great fun and all that - but ocaholic.ch do actually have performance ratings for all the Catalysts back 11.12 in their 12.9Beta article:

Seems that after a shaky start, AMD's driver team have been working really hard. Question is can they squeeze any more out of GCN or is this it? New drivers are due tomorrow so we'll see soon.

Yes i did include that url originally in the OP.

However i decided to run some of my own tests as well. :)

The new drivers are indeed due tomorrow and if you read my post above the AMD rep indicates more gains on the way for battlefield 3.

Can't wait. :D
 
Ah, how did I miss that? :o

Thing is, I don't remember where I came across ocaholic.ch so it might even have been in your OP.

At least there is a legit reason to optimize for BF3 since it's so popular - but I hope the fact that 11.12 vs 12.9 gained 47% for synthetics vs 23% for games doesn't meant that AMD have been wasting time trying to look good in 3DMark!

I'd call it that a waste of resources although I guess since even Intel optimize for synthetics benchmarks, AMD cannot afford to ignore 3DMark.

BTW, Nvidia seem to be either resting on their laurels with Kepler drivers, or Kepler was so similar to Ferni that their drivers are already optimized and there isn't much they can do to improve performance.

I can't and won't speak for Nvidia but installing and checking for performance improvements from AMD is all part of the fun.

Can't wait to try out these new drivers and BF3 performance, i hope im not disappointed.

Its great to see how the drivers have improved though, its a shame they couldn't get this performance out earlier but at least it shows the potential of GCN where after all this time things are still improving steadily.
 
Yes mate as my 2600k @ 4.5ghz doesn't have any slowdown there, or anywhere tbh since at least 12.6 drivers, 60fps throughout with vsync.

If you describe the exact place, i'll go there and grap a screenshot with FRAPS with the FPS showing?

That would be great Dave, thanks.

Actually i remember uploading a video here, trouble is i can't quite remember where i posted it now. In that thread i did two runs, one recording and one not recording. Fps were worse recording naturally, but even on the not recording run the fps did drop below 60 to 57 i think at that part.

Anyway, i found this vid. Its the part just before you get to the steps leaving the russian base i think.


I think where it drops to 51-54 near the start is the part the guy meant. As i said though it only dropped this low as i was recording the footage.
 
Dave, did you forget?

Also a note about the bf3 benchmark i posted above. Forgot to mention at the time but rather than copying the time demo which we all used i did a complete run through from the start of the level up to the bit where the video ends.
 
Back
Top Bottom