"This driver provides optimizations for the upcoming Star Wars: Battlefront Beta that we will be seeing next week and for the Fable Legends DX12 benchmark that we saw last week"
Does this mean you will revisit the Fable Legends article, or are PR articles not subject to revisits and preferred to be ran unoptimized for the party that is meant to be smeared?
If you read the article on Fable Legends you would have seen that they specifically noted that they did not test with the latest AMD driver (This one just coming out, which AMD had sent them), but did have the most recent nVidia drivers.
They only had the latest WHQL drivers from nvidia which werent put out specifically for the fable legends bench like the amd beta one was. They wanted it to be fair and unoptimized vs unoptimized.
It was an early preview! Who was getting smeared? They ran the tests with the drivers available. Now if they re-run with the new drivers, I of course like everyone else would like to see the diff/improvements. Still its just a scenery benchmark and while it may be indicative of performance, I'll still wait till an actual game benchmark comes out to pass judgment. however it's not going to affect my buying as I've already purchase GPU's earlier this year and will not upgrade until GPU's w/ HBM2/8GB frame buffer are available. Even then it will be based on bang for the buck, not brand.
My assumption is that as a long time nvidia sponsored company, the engine behind that benchmark is well optimized for nvidia hardware, and the whole thing is just a PR stunt to show that "nvidia doesn't suck in DX12". Thus my inquiry whether the PR article will be revisited once amd optimize their drivers, and possibly diminish the intent of the PR article.
Your assumption? Did you read the article? The way DX12 handles draw calls will benefit both AMD and Nvidia HW. If and that's a big "IF" this benchmark was made to prop up Nvidia then it didn't do a good job. The performance gap between the respective cards is par for the course. There wasn't any kind of runaway performance gap indicative of a benchmark created soley for the benefit of Nvidia. This test was done by Anandtech and they have been pretty fair to both companies. The tests did indicate that cards from the respective companies may bottleneck in different areas and that is something to explore further. I know a lot of people want to believe that DX12 will be the savior of AMD. The only savior of AMD will be AMD. They will either put out a competitive product or they wont. If they can't match on performance, they need to match with an appropriate price. I picked up two 290X 8GB cards in early March for just under $600. For me a much better value/performance than a single 980 at the time. It was a hell of a deal and I'm happy with my choice but there's no denying, Nvidia has the faster hardware. There's no PR conspiracy here just fanboyism...
"There wasn't any kind of runaway performance gap"
And how exactly did you figure that one out? It is not a given that a PR stunt will take two equally performing products and elevate one above the other, it may well take an inferior product and prop it up to look comparable to a superior one.
No fanboy here, I am running nvidia for graphics and radeo for compute, since under Linux radeons suck as much as nvidia sucks for DP compute, I merely asked whether the benchmark article will or will not be revisited now that AMD have optimized their driver for it, which logically implies the previous driver was not optimized for it.
Go look at other bench marks and games that are actually out. Look at the performance delta. No change in the preview benchmark really. You said "Does this mean you will revisit the Fable Legends article, or are PR articles not subject to revisits and preferred to be ran unoptimized for the party that is meant to be smeared?" You're the one calling it a PR article in Nvidia's favor LOL. And which cards are the equal performing ones? Only a couple of the midrange cards stack up equally. For the most part AMD cards trail 5-15% compared to their Nvidia counterparts.
Of course, the announcement that AMD released a driver optimized for it implies the previous driver was NOT, which means AT published an article, benchmarking an nvidia sponsored engine against an UNoptimized amd driver. This is as far from objective as it gets, and does sit well within the PR stunt range.
I buy whatever best suits my need, and run Linux, so I couldn't care less about amd, nvidia or DX12, but I care about objectivity. And don't pretend you are not a fanboy by calling others that. It is just lame.
So now I'm the fanboy? LOL Seriously? You're the one on about it being a PR piece for Nvidia! The game which the benchmark is based off of is an XBONE/Win 10 exclusive! Why would MS the publisher allow the game run worse on AMD hardware that powers their console? Doesn't make sense. From the preview benchmark article "Fable Legends is an Xbox One/Windows 10 exclusive free to play title built by Lionhead Studios in Unreal Engine 4. ". It also states as another poster mentioned in the paragraph labeled "The Test: The software provided to us is a prerelease version of Fable Legends, with early drivers, so ultimately the performance at this point is most likely not representative of the game at launch and should improve before release. What we will see here is more of a broad picture painting how different GPUs will scale when DX12 features are thrown into the mix. In fact, AMD sent us a note that there is a new driver available specifically for this benchmark which should improve the scores on the Fury X, although it arrived too late for this pre-release look at Fable Legends (Ryan did the testing but is covering Samsung’s 950 Pro launch in Korea at this time). It can underscore just how early in the game and driver development cycle DirectX 12 is for all players. But as with most important titles, we expect drivers and software updates to continue to drive performance forward as developers and engineers come to understand how the new version of DirectX works."
So it's obvious you didn't bother to actually read the article. You're popping off about it being a PR hit piece but I'm the fanboy? Objectivity? Bullshit. For which brand am I a supposed shill? I the owner of two 290X cards already stated that Nvidia as the test shows outperforms the AMD cards. The performance Delta is still within what has been already benchmarked with other games. There is no conspiracy here, no significant advantage was given to anyone. Just to you maybe. Good Lord...
- you have no problem with benchmarking nvidia sponsored engine on an amd gpu with a driver not even optimized for the title - you have a big problem with revisiting the benchmark to utilize the now available optimized driver
... and you still claim to not be a fanboy? And the proof for that is your claimed ownership of amd GPUs, which I can bet every undercover nvidia fanboy will claim as well...
If I believed in conspiracy theories I would think that you were acting like this because you were paid by Nvidia to act like an obnoxious AMD fanboy, to turn people away from AMD's brand. Thankfully I don't subscribe to conspiracy theories.
This is one area I think ddriver has a point. I don't think its a PR stunt, but without proper drivers established for both sides the results are pretty much non conclusive, and DOES make AT look like nvidia pr pushers. The engine is nvidia focused, and when the bench was ran the article made it known that there were no optimized AMD drivers for it.
Now... revisiting an article based on a benchmark of an unfinished game? Whats the point? What type of performance increase would you seriously expect from an optimized driver on a benchmark that's probably going to go through changes up to the point of the game being released, making the current state of optimized drivers irrelevant correct?
Its a benchmark of an unfinished game... and its a benchmark. Which in the end really means nothing. If the game was out and a review was done with unoptimized AMD drivers vs optimized nvidia drivers that would most certainly call for a revisit once AMD released new drivers.
In reality, AT just jumped the gun. They were given a new toy and wanted to get out preliminary results to its reader base. Should they have waited for AMD drivers? No. Should they revisit the previous article? No, whats the point...? We are after all talking about a benchmark of a game that has not yet been released.
It's a better result for NV than Ashes, but Fable might actually be the better result for AMD in context. AMD and NV walked into a UE4 game, and for once the game didn't dramatically underperform on AMD hardware.
That's a good sign for AMD, that DX12 won't just sucker punch them on engines that aren't in their wheelhouse (actually a good sign for both, because neither really wants huge swingy performance gaps between Gameworks and NV having the market share that a big game or two strongly favoring AMD would likely hurt them).
Yeah UE usually do favor Nvidia HW. Being that this is an XBONE/WIN 10 exclusive, I'm sure a lot of work was done to make it run well on AMD's GCN HW. Hoping what they learned from this will help with drivers for other UE based games. Also since AMD has a lock on console HW I think we will see optimizations that benefit AMD in UE games overall. UE does power a lot of games regardless of platform. Still, AMD needs a solid win in the GPU space. Optimizations can only close the gap so much.
I installed it and played call of duty advanced . It was horrible my fps went down 40 % . I usually get 91 fps but since that new beta driver I get 39 - 45 fps. I do have a amd r9 fury x . When I installed back to the driver 15.8 everything is back to 90 fps. Also my vram of card is no back to 2300 mb when I have installed the newest driver my vram usage was 3900 mb constantly . ? Any suggestions ? Thanks in advance
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
28 Comments
Back to Article
ddriver - Tuesday, September 29, 2015 - link
"This driver provides optimizations for the upcoming Star Wars: Battlefront Beta that we will be seeing next week and for the Fable Legends DX12 benchmark that we saw last week"Does this mean you will revisit the Fable Legends article, or are PR articles not subject to revisits and preferred to be ran unoptimized for the party that is meant to be smeared?
squngy - Tuesday, September 29, 2015 - link
Why not wait for both companies to do optimizations for one arbitrary benchmark and revisit then?Or you know, wait for actual games before making any decisions.
Stuka87 - Tuesday, September 29, 2015 - link
If you read the article on Fable Legends you would have seen that they specifically noted that they did not test with the latest AMD driver (This one just coming out, which AMD had sent them), but did have the most recent nVidia drivers.Asomething - Tuesday, September 29, 2015 - link
They only had the latest WHQL drivers from nvidia which werent put out specifically for the fable legends bench like the amd beta one was. They wanted it to be fair and unoptimized vs unoptimized.Manch - Tuesday, September 29, 2015 - link
It was an early preview! Who was getting smeared? They ran the tests with the drivers available. Now if they re-run with the new drivers, I of course like everyone else would like to see the diff/improvements. Still its just a scenery benchmark and while it may be indicative of performance, I'll still wait till an actual game benchmark comes out to pass judgment. however it's not going to affect my buying as I've already purchase GPU's earlier this year and will not upgrade until GPU's w/ HBM2/8GB frame buffer are available. Even then it will be based on bang for the buck, not brand.ddriver - Tuesday, September 29, 2015 - link
My assumption is that as a long time nvidia sponsored company, the engine behind that benchmark is well optimized for nvidia hardware, and the whole thing is just a PR stunt to show that "nvidia doesn't suck in DX12". Thus my inquiry whether the PR article will be revisited once amd optimize their drivers, and possibly diminish the intent of the PR article.Manch - Tuesday, September 29, 2015 - link
Your assumption? Did you read the article? The way DX12 handles draw calls will benefit both AMD and Nvidia HW. If and that's a big "IF" this benchmark was made to prop up Nvidia then it didn't do a good job. The performance gap between the respective cards is par for the course. There wasn't any kind of runaway performance gap indicative of a benchmark created soley for the benefit of Nvidia. This test was done by Anandtech and they have been pretty fair to both companies. The tests did indicate that cards from the respective companies may bottleneck in different areas and that is something to explore further. I know a lot of people want to believe that DX12 will be the savior of AMD. The only savior of AMD will be AMD. They will either put out a competitive product or they wont. If they can't match on performance, they need to match with an appropriate price. I picked up two 290X 8GB cards in early March for just under $600. For me a much better value/performance than a single 980 at the time. It was a hell of a deal and I'm happy with my choice but there's no denying, Nvidia has the faster hardware. There's no PR conspiracy here just fanboyism...ddriver - Tuesday, September 29, 2015 - link
"There wasn't any kind of runaway performance gap"And how exactly did you figure that one out? It is not a given that a PR stunt will take two equally performing products and elevate one above the other, it may well take an inferior product and prop it up to look comparable to a superior one.
No fanboy here, I am running nvidia for graphics and radeo for compute, since under Linux radeons suck as much as nvidia sucks for DP compute, I merely asked whether the benchmark article will or will not be revisited now that AMD have optimized their driver for it, which logically implies the previous driver was not optimized for it.
Manch - Tuesday, September 29, 2015 - link
Go look at other bench marks and games that are actually out. Look at the performance delta. No change in the preview benchmark really. You said "Does this mean you will revisit the Fable Legends article, or are PR articles not subject to revisits and preferred to be ran unoptimized for the party that is meant to be smeared?" You're the one calling it a PR article in Nvidia's favor LOL. And which cards are the equal performing ones? Only a couple of the midrange cards stack up equally. For the most part AMD cards trail 5-15% compared to their Nvidia counterparts.ddriver - Tuesday, September 29, 2015 - link
Of course, the announcement that AMD released a driver optimized for it implies the previous driver was NOT, which means AT published an article, benchmarking an nvidia sponsored engine against an UNoptimized amd driver. This is as far from objective as it gets, and does sit well within the PR stunt range.I buy whatever best suits my need, and run Linux, so I couldn't care less about amd, nvidia or DX12, but I care about objectivity. And don't pretend you are not a fanboy by calling others that. It is just lame.
Manch - Tuesday, September 29, 2015 - link
So now I'm the fanboy? LOL Seriously? You're the one on about it being a PR piece for Nvidia! The game which the benchmark is based off of is an XBONE/Win 10 exclusive! Why would MS the publisher allow the game run worse on AMD hardware that powers their console? Doesn't make sense. From the preview benchmark article "Fable Legends is an Xbox One/Windows 10 exclusive free to play title built by Lionhead Studios in Unreal Engine 4. ". It also states as another poster mentioned in the paragraph labeled "The Test: The software provided to us is a prerelease version of Fable Legends, with early drivers, so ultimately the performance at this point is most likely not representative of the game at launch and should improve before release. What we will see here is more of a broad picture painting how different GPUs will scale when DX12 features are thrown into the mix. In fact, AMD sent us a note that there is a new driver available specifically for this benchmark which should improve the scores on the Fury X, although it arrived too late for this pre-release look at Fable Legends (Ryan did the testing but is covering Samsung’s 950 Pro launch in Korea at this time). It can underscore just how early in the game and driver development cycle DirectX 12 is for all players. But as with most important titles, we expect drivers and software updates to continue to drive performance forward as developers and engineers come to understand how the new version of DirectX works."So it's obvious you didn't bother to actually read the article. You're popping off about it being a PR hit piece but I'm the fanboy? Objectivity? Bullshit. For which brand am I a supposed shill? I the owner of two 290X cards already stated that Nvidia as the test shows outperforms the AMD cards. The performance Delta is still within what has been already benchmarked with other games. There is no conspiracy here, no significant advantage was given to anyone. Just to you maybe. Good Lord...
ddriver - Wednesday, September 30, 2015 - link
So let me see if I got it right:- you have no problem with benchmarking nvidia sponsored engine on an amd gpu with a driver not even optimized for the title
- you have a big problem with revisiting the benchmark to utilize the now available optimized driver
... and you still claim to not be a fanboy? And the proof for that is your claimed ownership of amd GPUs, which I can bet every undercover nvidia fanboy will claim as well...
Sure thing pal...
D. Lister - Thursday, October 1, 2015 - link
@ddriverIf I believed in conspiracy theories I would think that you were acting like this because you were paid by Nvidia to act like an obnoxious AMD fanboy, to turn people away from AMD's brand. Thankfully I don't subscribe to conspiracy theories.
wolfemane - Monday, October 5, 2015 - link
This is one area I think ddriver has a point. I don't think its a PR stunt, but without proper drivers established for both sides the results are pretty much non conclusive, and DOES make AT look like nvidia pr pushers. The engine is nvidia focused, and when the bench was ran the article made it known that there were no optimized AMD drivers for it.Now... revisiting an article based on a benchmark of an unfinished game? Whats the point? What type of performance increase would you seriously expect from an optimized driver on a benchmark that's probably going to go through changes up to the point of the game being released, making the current state of optimized drivers irrelevant correct?
Its a benchmark of an unfinished game... and its a benchmark. Which in the end really means nothing. If the game was out and a review was done with unoptimized AMD drivers vs optimized nvidia drivers that would most certainly call for a revisit once AMD released new drivers.
In reality, AT just jumped the gun. They were given a new toy and wanted to get out preliminary results to its reader base. Should they have waited for AMD drivers? No. Should they revisit the previous article? No, whats the point...? We are after all talking about a benchmark of a game that has not yet been released.
xthetenth - Tuesday, September 29, 2015 - link
It's a better result for NV than Ashes, but Fable might actually be the better result for AMD in context. AMD and NV walked into a UE4 game, and for once the game didn't dramatically underperform on AMD hardware.That's a good sign for AMD, that DX12 won't just sucker punch them on engines that aren't in their wheelhouse (actually a good sign for both, because neither really wants huge swingy performance gaps between Gameworks and NV having the market share that a big game or two strongly favoring AMD would likely hurt them).
Manch - Tuesday, September 29, 2015 - link
Yeah UE usually do favor Nvidia HW. Being that this is an XBONE/WIN 10 exclusive, I'm sure a lot of work was done to make it run well on AMD's GCN HW. Hoping what they learned from this will help with drivers for other UE based games. Also since AMD has a lock on console HW I think we will see optimizations that benefit AMD in UE games overall. UE does power a lot of games regardless of platform. Still, AMD needs a solid win in the GPU space. Optimizations can only close the gap so much.Oxford Guy - Tuesday, September 29, 2015 - link
When they get around to looking at Ashes they could revisit Fable.jamjakap - Tuesday, September 29, 2015 - link
I installed it and played call of duty advanced . It was horrible my fps went down 40 % . I usually get 91 fps but since that new beta driver I get 39 - 45 fps. I do have a amd r9 fury x . When I installed back to the driver 15.8 everything is back to 90 fps. Also my vram of card is no back to 2300 mb when I have installed the newest driver my vram usage was 3900 mb constantly . ? Any suggestions ? Thanks in advanceManch - Tuesday, September 29, 2015 - link
Any other games take a hit?ruthan - Tuesday, September 29, 2015 - link
AMD happens.xthetenth - Tuesday, September 29, 2015 - link
Drivers happen, more like. NV's released a stinker or two this generation too. One nearly got me to get rid of my 970, and I avoided another.Manch - Tuesday, September 29, 2015 - link
What CPU do you have?NXTwoThou - Tuesday, September 29, 2015 - link
I've had two blue screens so far with 13.9 beta. One just happened when I was at lunch(aka, no user interaction at all). I think this just a dud.hansmuff - Tuesday, September 29, 2015 - link
https://twitter.com/AMDCare/status/648865439225876...AMD themselves telling users to roll back to previous drivers. These beta ones are a little under cooked.
NXTwoThou - Tuesday, September 29, 2015 - link
Ha! And when I go to Apps and Features, AMD Catalyst Install Manager has Modify and Uninstall greyed out.Daniel Williams - Tuesday, September 29, 2015 - link
Thanks for bringing this up hansmuff, an update is on it's way.Daniel Williams - Tuesday, September 29, 2015 - link
* to this article that is :)D. Lister - Wednesday, September 30, 2015 - link
"AMD earlier this morning announced that they were aware of a memory leak..."Well, it's all peachy then, as long as they're aware. :p