AMD Omega Update is Now God/ Experiences with Omega

[soaringchicken]

Holy balls, the omega update has made my experience so much better. My fps in heavily modded skyrim (which before was nearly unplayable) is now considerably better and stuttering is eliminated (I have a 6800k and a 280x). I thought they were stretching when they said there was a massive performance boost, however it looks like (in my case) that they weren't lying at all.

 

Anyone else have an amazing performance boost in games? 

 

So happy

 

soaringchicken

 

 

EDIT: Alright I had my first crash in BF4

[Katsu-K1L]

Not me.

i got artifacts and crashs.

[Deletive]

First thread i'd seen where it was 20% post and 50% chin.

[soaringchicken]

Not me.

i got artifacts and crashs.

[shadowblitz3]

fuck that chin

[cec4]

well it works. i've had NO problems. i like it 

[KemoKa]

This whole Omega Driver thing has been bugging me lately, so I'll tell you what I literally just did. I benchmarked the drivers.

System Specs:

Core i7-2600K @ 4.5GHz

8GB DDR3-1600
AMD Radeon 7970 Asus DirectCU II @ 1000MHz core, 1500MHz memory

 

So basically, we have probably the most bog-standard all-purpose gaming rig in my opinion. I'm benchmarking these drivers using Unigine Valley.

Settings:
DirectX 11

1680x1050
Ultra
8x AA

I ran the standard benchmark on Unigine Valley, and because I like to use the scientific method when benchmarking, I ran it three times for each driver and averaged the values. So here are the results.

Catalyst 14.9

Avg. High: 91.9fps
Avg. Low: 25.7fps
Avg. Avg: 48.6fps

so far so good, right?

Catalyst 14.12 (Omega)

Avg. High: 90.9fps
Avg. Low: 26.8fps
Avg. Avg: 48.4fps

Let's be clear, I wasn't really expecting it to have that much of an effect on graphical performance, and perhaps I missed something, or perhaps people are going to chastise me saying that "ooh, that's not what they meant", but you know what, "massive increases in performance" are the kind of claims that I have been hearing from both AMD and testers of Omega. Okay, sure, maybe it has some better features and maybe there are some improvements in other applications or APIs like Mantle, but with the bog-standard DX11 on a bog-standard rig, we saw the average highs go down one fps, and the average lows go up one fps, and the average averages stay almost exactly where they were. the differences in results were well within margin of error, with variances in the individual highs from the Omega testing actually being greater than those in the 14.9 testing by about 0.3fps. These results are so unspectacular that even the margin of error itself is within margin of error! *mind explodes*

So in terms of overall raw graphical performance, we're not really seeing anything here. Nothing really happened. The Omega drivers are most likely being developed for stability(?) purposes and more fancy features. I'm sure that somewhere else someone will find some other benchmark, game or application where the new drivers seem to give a noticeable performance boost. It's just that with bog-standard rendering, these drivers are no different from the old ones.

have something to add? Please, feel free, in fact, I want to hear from people.

[Jade]

I've seen more BSoDs since updating, that's about it.

[SkilledRebuilds]

FC4 is a few fps on average faster, say 50-55fps now pushing 55-62fps

AC:U is a bigger difference yet! Enough of the extra performance to then use 3200x1800 and keep it enabled having 31-35fps avg or leave it at 1080p for 55-65fps average

Some smalltime gains in other games I play.

Pretty happy with that..

[KemoKa]

AC:U is a bigger difference yet! Enough of the extra performance to then use 3200x1800 and keep it enabled having 31-35fps avg or leave it at 1080p for 55-65fps average

Wow, so AssCreed: Unplayable is actually playable now? nice!

[MeltyMoon]

This whole Omega Driver thing has been bugging me lately, so I'll tell you what I literally just did. I benchmarked the drivers.

Let's be clear, I wasn't really expecting it to have that much of an effect on graphical performance, and perhaps I missed something, or perhaps people are going to chastise me saying that "ooh, that's not what they meant", but you know what, "massive increases in performance" are the kind of claims that I have been hearing from both AMD and testers of Omega. Okay, sure, maybe it has some better features and maybe there are some improvements in other applications or APIs like Mantle, but with the bog-standard DX11 on a bog-standard rig, we saw the average highs go down one fps, and the average lows go up one fps, and the average averages stay almost exactly where they were. the differences in results were well within margin of error, with variances in the individual highs from the Omega testing actually being greater than those in the 14.9 testing by about 0.3fps. These results are so unspectacular that even the margin of error itself is within margin of error! *mind explodes*

So in terms of overall raw graphical performance, we're not really seeing anything here. Nothing really happened. The Omega drivers are most likely being developed for stability(?) purposes and more fancy features. I'm sure that somewhere else someone will find some other benchmark, game or application where the new drivers seem to give a noticeable performance boost. It's just that with bog-standard rendering, these drivers are no different from the old ones.

have something to add? Please, feel free, in fact, I want to hear from people.

To be fair the driver is designed to improve performance in game, so the code in the driver is written to be more efficient in certain titles!

Hence the lack of improvement in valley

[Frix]

Haven't noticed an improvement as I don't run Fraps, its probably improved FPS and I know it definitely hasn't decreased my FPS.

[MoonSpot]

--snip--

FYI

A year old now, but contains results from multiple sources/games unlike your exclusive use of valley

 

and more recently the same person touched base with the same topic again.

 

And with that said the OP declared an actual perceived difference where their skyrim is now playable, you can't just disregard them because your test results do not reflect their statement(s).

 

 

 

P.S. Nvidia, if you're out there, I would love it if y'all could make modded skyrim playable for GT 130m via a driver update...

[winningsince1337]

Works fine for me. No complaints

[Briggsy]

I didn't really get much FPS improvements, but crossfire got smoother for Dragon Age Inquisition. I even tried 1440p VSR with AA completely disabled and my framerate barely dropped compared to max settings 1080p (my 3570k @ 4.7Ghz is a bit of a bottleneck i guess), but visual quality is better by a huge amount. too bad 4k downscaling is only available on the R9 285...

[Just Monika]

I'm getting occasional crashing and complete system freezes with Divinity Original Sin, but to be fair, I don't think I've played the game without the Omega drivers so I can't fault them immediately.

[KemoKa]

 

A year old now, but contains results from multiple sources/games unlike your exclusive use of valley

 

and more recently the same person touched base with the same topic again.

 

And with that said the OP declared an actual perceived difference where their skyrim is now playable, you can't just disregard them because your test results do not reflect their statement(s).

I was doing just a single standard test with standard software. It's called a benchmark for a reason. But that's not what annoys me about your post. In what twisted version of reality did I ever say that I disregarded the other results? Okay, you're right, my test was single-faceted, but I'm not so hardcore a gamer as to have Skyrim, Assassin's Creed Unity AND Far Cry 4 in my library of libraries to willy-nilly benchmark every new game that might benefit from a driver update. If you actually properly read my post, you would have noticed this

 

Let's be clear, I wasn't really expecting it to have that much of an effect on graphical performance, and perhaps I missed something, or perhaps people are going to chastise me saying that "ooh, that's not what they meant", but you know what, "massive increases in performance" are the kind of claims that I have been hearing from both AMD and testers of Omega.

 I'm sure that somewhere else someone will find some other benchmark, game or application where the new drivers seem to give a noticeable performance boost. It's just that with bog-standard rendering, these drivers are no different from the old ones.

have something to add? Please, feel free, in fact, I want to hear from people.

I swear, I feel like whenever people use the -snip- scissors, it addles their memory worse than tl;dr.

I never intended to disregard people or their results. I actually welcomed people for any additional information to make up for my pathetic ability to benchmark hardware. But you apparently took my skepticism the wrong way, even after me stating repeatedly that I may well be wrong about the performance increases. I was pretty abrasive about the drivers, but really, can you blame me for being skeptical, especially when literally everyone is having different experiences with the drivers?

[MoonSpot]

--snip--

(I snip because it is annoying for most on non-pc platforms to scroll past the text we addle-brained folk just finished reading, but whatever to each their own pet peeves)

 

Sorry I did phase that previous post quite poorly and you did deserve the courtesy of.a fully and more accurately articulated reply.  At least more than "you poo poo on OP?  You suck!"(to loosely paraphrase).  I was and am more frustrated with incomplete tests and skewed data than anything else, as I've seen way to many examples of single data sets repeatedly waved around as gospel by some.  I do actually applaud your efforts to put this (or anything) to a test.  I was primarily trying to point out that there is other another P.O.V to consider that does not get reflected in the data so easily, and that I think it is a more worthwhile metric to use.  Which is why I linked to them for you to examine and consider.  That and I simply disagreed with your findings, due to the method employed.

 

Though I don't think I was entirely off base to read into you posting your benchmarks while the OP was in the middle of his happy dance as more than just skepticism, and I found suspicious. 

Anyways this is has gotten a bit too sensitive for my liking, and I'd prefer to know what you think about Ryan's findings in frame pacing and whether you believe they're with merit to a significant degree or not.

[soaringchicken]

(I snip because it is annoying for most on non-pc platforms to scroll past the text we addle-brained folk just finished reading, but whatever to each their own pet peeves)

 

Sorry I did phase that previous post quite poorly and you did deserve the courtesy of.a fully and more accurately articulated reply.  At least more than "you poo poo on OP?  You suck!"(to loosely paraphrase).  I was and am more frustrated with incomplete tests and skewed data than anything else, as I've seen way to many examples of single data sets repeatedly waved around as gospel by some.  I do actually applaud your efforts to put this (or anything) to a test.  I was primarily trying to point out that there is other another P.O.V to consider that does not get reflected in the data so easily, and that I think it is a more worthwhile metric to use.  Which is why I linked to them for you to examine and consider.  That and I simply disagreed with your findings, due to the method employed.

 

Though I don't think I was entirely off base to read into you posting your benchmarks while the OP was in the middle of his happy dance as more than just skepticism, and I found suspicious. 

Anyways this is has gotten a bit too sensitive for my liking, and I'd prefer to know what you think about Ryan's findings in frame pacing and whether you believe they're with merit to a significant degree or not.

I did mention in the post that "at least mine works better" in brackets 

[-CLASSIFIED-]

Not me.

i got artifacts and crashs.

it made my crossfire between my r9 280x and hd 7950 not work anymore  

[soaringchicken]

it made my crossfire between my r9 280x and hd 7950 not work anymore t

You got that to work in the first place......?

[-CLASSIFIED-]

You got that to work in the first place......?

yea because the r9 280x is based on the same architecture as the hd 7900 cards   

[soaringchicken]

yea because the r9 280x is based on the same architecture as the hd 7900 cards   

Yes but is is based on the 7970, I didn't think that would work the same way...... Thats like saying because they are both in the 700 series, I can SLI a 780 and a 770? Lol I have no idea what I'm talking about

[-CLASSIFIED-]

Yes but is is based on the 7970, I didn't think that would work the same way...... Thats like saying because they are both in the 700 series, I can SLI a 780 and a 770? Lol I have no idea what I'm talking about

lol , worked fine for me , maybe i just got lucky and flukes it into working ahahah 

[Constantine]

lol , worked fine for me , maybe i just got lucky and flukes it into working ahahah 

 

Its supposed to work. Dunno why it broke it.