AMD's Omega Driver Decreased 3Dmark performance and increase Game FPS

[Hatsune Miku 「 」]

 http://www.techpowerup.com/reviews/AMD/Catalyst_14.12_Performance/4.html

 

As listed in the source, the new Omega driver apparently lowers 3Dmark 11 performance.

Offering a slight performance drop around 100 points from the score.

This happens to be the case with all 3Dmark benchmarks and varying resolutions.

 

Only the sky diver benchmark had a slight increase in score.

Performance is lowered in 1080, 1440, 2160p resolutions.

 

GAME TESTS HERE: http://www.techpowerup.com/reviews/AMD/Catalyst_14.12_Performance/5.html

 

Considerable improvements in many games in 1080, 1440 and 2160p resolutions. Test setup was a r9 290 gpu.

[Enderman]

lol I knew these new revolutionary "omega" AMD drivers were still going to have issues.

Soon we will probably start to see more black screen issues, or other performance degradation, or just outright not-working systems because of these AMD drivers.

It was too good to be true.

[Deli]

So much hate on AMD in this forum!

[Daegun]

Isn't it still in Beta? I mean, has it actually even been officially anounced yet, or still just that leak that let us know what it was?

[Glenwing]

Unfinished driver has flaws.  Shocking.

 

I'm not even sure that 100 points in 3DM even really counts as that.  Does anyone care about it if game performance isn't affected?  And even if game performance mirrored that loss it would be a loss of what, 1fps?  Downright nightmare.

[Kevinkt]

But did game preformance go up?

from the gaming tests on that page, two fps.

[Picklenuttz]

I buy a GPU for synthetic benchmarks...oh wait a sec.

[Kukielka]

I buy a GPU for synthetic benchmarks...oh wait a sec.

I do

[WereCat]

I buy a GPU for synthetic benchmarks...oh wait a sec.

I run synthetics benchmark 8 hours a day on my PC so that when friends come to visit me they can be like .. OH.. WAIT?! YOU GOT 5600 points in Fire Strike? Awesome!

Then when they leave my home I feel a big satisfaction and go on the forums and try to throw it into peoples face.

 

Thats how it should be right?

[Just The Tips]

Unfinished driver has flaws.  Shocking.

 

I'm not even sure that 100 points in 3DM even really counts as that.  Does anyone care about it if game performance isn't affected?  And even if game performance mirrored that loss it would be a loss of what, 1fps?  Downright nightmare.

1 fps to 0 fps is a big difference mannnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn

[EChondo]

Oooh, TPU is reporting smoother experience due to better frame pacing?

 

I'll have to try this. Can't wait to see/read about PCPer's experience with this!

 

Too bad I might be going green this Christmas, but I'll hang onto my 7870 for playing around with.

[Techhog]

You people are seriously complaining about a small decrease in a benchmark and don't care about the increase in game performance.

 

You guys are really stupid sometimes.

[Dive5885]

You people are seriously complaining about a small decrease in a benchmark and don't care about the increase in game performance.

 

You guys are really stupid sometimes.

+1

[TechFan@ic]

No one's really complaining, synthetics went slightly down, gaming performance went up considerably 15-20% in a number of titles including Ryse Son of Rome, Bioshock Infinite, Batman Arkham Origins, Rome 2 Total War, Grid 2, CoD Ghosts and Sniper Elite 3.

[Hatsune Miku 「 」]

You people are seriously complaining about a small decrease in a benchmark and don't care about the increase in game performance.

 

You guys are really stupid sometimes.

But benchmarks are for enthusiasts who want the best...

[Humbug]

big gains on Ryse!

Don't care about synthetic benchmarks..

[xAcid9]

This is bullshit, all those hypes and i cant even get a performance increase in my favorite game Firestrike.

[Lord Baal]

No one's really complaining, synthetics went slightly down, gaming performance went up considerably 15-20% in a number of titles including Ryse Son of Rome, Bioshock Infinite, Batman Arkham Origins, Rome 2 Total War, Grid 2, CoD Ghosts and Sniper Elite 3.

The only one of these i have is Grid 2 which is incredibly easy to run anyway even my AMD HTPC with a 4gb 270x run it maxed out at 1080p with over 100fps at all times haha. I care more about a stability increase as I was running beta drivers then I do any drop in benchmarks or mild increase in games that were already easy to run. No noticeable performance gains in Far Cry 4(only thing I've run since updating) Also the 3d mark scores for my all AMD HTPC seem to be up about 500 points over the scores 2-3 days ago.

[Dive5885]

Has anyone with crossfire played Battlefield 4 after the update ?

[SpaghettiCarbonara]

You people are seriously complaining about a small decrease in a benchmark and don't care about the increase in game performance.

 

You guys are really stupid sometimes.

 

You say "you people" but I only saw one person legitimately complain about this.

I know what you mean though, the shitposting/hate-mongering is strong in most threads

[Master Disaster]

I run synthetics benchmark 8 hours a day on my PC so that when friends come to visit me they can be like .. OH.. WAIT?! YOU GOT 5600 points in Fire Strike? Awesome!

Then when they leave my home I feel a big satisfaction and go on the forums and try to throw it into peoples face.

 

Thats how it should be right?

In the spirit on your post...

I throw your 5,600 FS score back at you and replace it with my 12,000 FS score

Seriously though who cares if 3D Mark scores less if game performance goes up?

[Khvarrioiren]

I've actually had quite a noticeable difference in image smoothness on some games. No complaints here.

[Master Disaster]

I don't know who did these benches but my experience is totally different.

 

I kept my I7 4770K at stock, RAM is on XMP with manual timings (as for some reason it defaults to 2T instead of 1T), also running 2 x MSI R9 280X Gaming 3Gs in a 16x/16x Xfire config.

 

Catalyst 14.11.2 BETA.

 

For this test i had my GPUs OCed from 1020Mhz Core, 1500Mhz RAM to 1125 Mhz Core, 1600Mhz RAM.

 

http://www.3dmark.com/fs/3237996 - Result: 12168

 

Catalyst 14.12 Omega

 

For this test my GPU OC reset or didn't apply so was actually running at stock.

 

http://www.3dmark.com/fs/3422883 - Result: 12335

 

Thats right folks, i got 200 more points in 3D Mark with a lower clock spped on both the GPU and vRAM. Am going to figure out whats wrong with my OC then re run the test with Omega and OCed GPUs.

 

EDIT

 

LOL, got less points with my GPU OC working?

 

http://www.3dmark.com/fs/3422955 - Result: 12289

 

Now onto AC Unity

[ComradeHX]

But benchmarks are for enthusiasts who want the best...

At the moment they won't be spending money on AMD anyway...

[harrynowl]

But benchmarks are for enthusiasts who want the best...

And enthusiasts who want the best should really know that a 100pt difference in a 3D Mark Firestrike score isn't really basis to choose 1 GPU over another.

 

This may matter to benchmarkers and overclockers who are competing for epeen on hwbot, not gamers.

 

If 290X was 100 points faster than GTX970, and they were both the same price. Almost everyone would buy 970 still-why? Because of the other reasons to choose a GPU.