Wednesday, May 20th 2015

It's Now Been Over 160 Days Since a Catalyst WHQL Release

As of today (20/05/2015), it has been over 160 days since AMD released a WHQL-signed Catalyst driver update, in what is a clear sign of decay in the company's after-sales support for the consumer graphics market. Once tuned to a near-monthly release of its Catalyst Software suite, which added optimzations for new games, improved upon support for existing ones; CrossFire multi-GPU support profiles; even if not adding support for new GPUs; AMD slipped into quarterly WHQL release cycle in 2013-14. It now seems to have deviated from even that.

The company's last WHQL-signed Catalyst release was Catalyst 14.12 Omega WHQL, which released on 09/12/2014, 161 days ago. The company has since only released two "Beta" drivers, notably Catalyst 15.4 Beta, with optimization for Grand Theft Auto V, and AMD FreeSync support. In contrast, NVIDIA adopted a faster driver update cycle than its previous monthly GeForce WHQL driver releases, under its "Game Ready" driver program. New WHQL-signed releases predate almost every AAA PC game release. There's still no word on a Catalyst WHQL update, and with launch of new graphics cards slated for the third week of June, it's unlikely that the company will release one interim. By then, it will have been 196 days since a Catalyst WHQL driver release. Such a slow driver update cycle would do little to inspire confidence in buying the next-generation Radeon product, even if it establishes a performance lead over GeForce.
Add your own comment

161 Comments on It's Now Been Over 160 Days Since a Catalyst WHQL Release

#126
Aquinus
Resident Wat-man
rtwjunkieExcept, this "fair and value oriented pricing" is not good business practice. When you eternally sell things for cheap, people expect it. You then didg yourself into even more debt. I think putting out top-notch products and higher pricing will actually bring alot of people over FROM Nvidia, because the PERCEPTION is the products are now competitive and top-notch. Yes, many benchmarks have shown that already for a long time. Most users however, do not check benchmarks. They buy on reputation and PERCEPTION. Perception, sadly, is reality in this world.
I remember when Athlon FX cpus costed as much as EE Intel CPUs.
Posted on Reply
#127
Rei86
AquinusI remember when Athlon FX cpus costed as much as EE Intel CPUs.
You mean during the FX-50 to FX-60ish days when it was worth picking AMD over Intel? Yup.

I still have my FX-53 Clawhammer in a box sitting on my bookcase.
Posted on Reply
#128
Kaleid
WHQL driver releases are absolutely no guarantee that the driver works perfectly. I've had crap WHQL from both Nvidia and AMD. Currently my 280x works fine.
Posted on Reply
#129
R-T-B
KaleidWHQL driver releases are absolutely no guarantee that the driver works perfectly. I've had crap WHQL from both Nvidia and AMD. Currently my 280x works fine.
While this is true, it only makes the lack of this relatively cheap and easy to pass cert from AMD even more disturbing...
Posted on Reply
#130
Haytch
I can't complain about my AMD drivers. I am running 3x 290X's in crossfire without any issues. All games I play work perfectly including most of the latest AAA games. In-fact, I couldn't ask for more performance in games such as Project Cars and GTA V.

I was wondering what happened to the monthly driver update but was not fussed about it.

I generally find that when new AAA games are released and they constantly crash because the GPU needs some software update, is usually because the game was poorly coded (Not saying that GTA V and Project Cars are greatly coded), the game simply sucks or the game would never interest me in a million years (like Bitcher 3).

I have more issues running my 2 Titans in SLi than the 290x's in Crossfire on w7, w8.1 and w10. Especially on w8.1!

WHQL is a meaningless label for me, it just means Microsoft had a chance to look at something that already works perfectly and made some money in the process.

I look forward to the next AMD series. Dx12 + HBM, who wouldn't.

Note :- If it is a human writing the article, there will almost always be a human perspective on the matter and hence an opinion. I enjoy and appreciate the facts that TPU brings forth. I guess what everyone here needs is the news articles to be specific to facts and perhaps a reply to the editor's own thread by the editor with his/her opinion, separate to the news article itself.

In a nutshell, thanks TPU for bringing it to our attention that AMD have not released a WHQL driver in roughly 5 months, but don't worry, they are releasing beta drivers. No issue's or complaints from me regarding their products performance or support.
Posted on Reply
#131
HumanSmoke
AquinusI remember when Athlon FX cpus costed as much as EE Intel CPUs.
The difference is that AMD were a much different company then than they are now. Back then, the company walked the walk as well as the talking/PPS presentation part. The company was still recognised as an extension of Jerry Sanders straight shootin' persona as Hector Ruiz had yet to imprint the company as a dysfunctional basket case, they were in the process of wrenching a large (~ 25% ) market share away from Intel in the budding x86 server sector with the debut of Opteron (the 100 series also proved to be a great overclockers and fairly cheap), and went toe-to-toe with Intel pretty much across the board - especially when AMD paired up their offerings with Nvidia's nForce2 chipset for the masses.

That landscape no longer exists, neither in performance actuality, nor brand perception.
Posted on Reply
#132
Aquinus
Resident Wat-man
HumanSmokeThe difference is that AMD were a much different company then than they are now. Back then, the company walked the walk as well as the talking/PPS presentation part. The company was still recognised as an extension of Jerry Sanders straight shootin' persona as Hector Ruiz had yet to imprint the company as a dysfunctional basket case, they were in the process of wrenching a large (~ 25% ) market share away from Intel in the budding x86 server sector with the debut of Opteron (the 100 series also proved to be a great overclockers and fairly cheap), and went toe-to-toe with Intel pretty much across the board - especially when AMD paired up their offerings with Nvidia's nForce2 chipset for the masses.

That landscape no longer exists, neither in performance actuality, nor brand perception.
Exactly, that's the point. The FX CPUs were expensive when they competed with the top. For the person I was quoting, the point was that it makes no difference which company is the top dog. Someone is going to pay out the nose for "the best" hardware. Cost only works when you have a selling point on the product. You drop prices to stay competitive or to get rid of stock. That's all I'm trying to get at.
Posted on Reply
#133
HumanSmoke
AquinusExactly, that's the point. The FX CPUs were expensive when they competed with the to-p. For the person I was quoting, the point was that it makes no difference which company is the top dog. Someone is going to pay out the nose for "the best" hardware. Cost only works when you have a selling point on the product. You drop prices to stay competitive or to get rid of stock. That's all I'm trying to get at.
I did see your valid point, but as far as rtwjunkie's point goes, you cant just pin a big price tag on parts - even if they are superb, and expect the customer base to quickly come back around to the brand, just as a single flop doesn't cause people to flee a brand that has top of the mind brand perception. Case in point is the Cougar Point flaw and board recall, or the much more publicized FDIV bug before it. Both barely caused a ripple in the public consciousness, Indeed, Intel's sales within months of issuing their mea culpa regarding Pentium actually increased.
AMD, like any company will need a track record of producing quality competitive product, and producing it on time.
Posted on Reply
#134
raghu78
btarunrOh you mean that site which, when the GTX 970 3.5GB drama was just on forums/reddit, was among the first to lunge forward with detailed articles flaming NVIDIA; with several follow-ups, until Jen-Hsun wrote a sorry-letter? Stop making yourself look like an ignorant jackass without reading enough of our content.
you guys did squat till the issue blew up on forums. you and the rest of the press were just trolled by Nvidia. For 4 months you guys knew nothing. The users did all the hardwork analyzing the problems they faced at high res and high settings and in multi-GPU when the last 0.5GB was used. So how about not bragging about something you should be ashamed of. That you got fooled because your testing did not catch the problem. Instead of writing useless editorials like this improve your testing to catch such situations.

btw did you guys do any testing to find out the ramifications and negative performance effects of the memory partition so as to better inform the consumer. Thats more important. No you did not. you guys must have been busy counting your kickbacks. Even though pcper is considered Nvidia biased they atleast did 2 articles and earned some reputation.

www.pcper.com/reviews/Graphics-Cards/Frame-Rating-Looking-GTX-970-Memory-Performance
www.pcper.com/reviews/Graphics-Cards/Frame-Rating-GTX-970-Memory-Issued-Tested-SLI

Your site on the contrary did not do a single review or test to address the problem and its ramifications. The last GTX 970 review on your site was mid-dec. after the gtx 970 controversy you guys moved on to more exciting reviews (or shall we say kickbacks) like GTX 960 and GTX Titan-X.

www.techpowerup.com/reviews/MSI/GTX_970_Gaming/32.html

"Jen-Hsun wrote a sorry letter"

blogs.nvidia.com/blog/2015/02/24/gtx-970/

"Instead of being excited that we invented a way to increase memory of the GTX 970 from 3GB to 4GB, some were disappointed that we didn’t better describe the segmented nature of the architecture for that last 1GB of memory."

Come on. are you kidding me. There was no sorry or even apology in his letter. It was about a new Maxwell feature. rofl. The guy wants us to be excited about something which Nvidia did not even think to disclose at launch. what a joke. If you believe thats a sorry I feel sorry for your readers.

oh btw just a sample of your cheerleading for Nvidia

www.techpowerup.com/reviews/NVIDIA/GeForce_GTX_980/31.html

"Oh, and AMD seems fucked."

We now know in hindsight that its Nvidia which f****d their loyal customers with the GTX 970 feature (as Jen Hsun calls it) which they did not disclose at launch.
Posted on Reply
#136
raghu78
R-T-BU mad bro?
u mad at me for calling out their paid FUD and pure drivel.
Posted on Reply
#137
R-T-B
They just called out nvidia on their clock lock a few minutes ago.

And no, I'm not mad. Life's too short to get mad about companies peddling wafers of silicon... But I do know I've been here a number of years, and the idea that this site is NVIDIA biased is quite frankly, hillarious.
Posted on Reply
#138
rtwjunkie
PC Gaming Enthusiast
HumanSmokeI did see your valid point, but as far as rtwjunkie's point goes, you cant just pin a big price tag on parts - even if they are superb, and expect the customer base to quickly come back around to the brand, just as a single flop doesn't cause people to flee a brand that has top of the mind brand perception. Case in point is the Cougar Point flaw and board recall, or the much more publicized FDIV bug before it. Both barely caused a ripple in the public consciousness, Indeed, Intel's sales within months of issuing their mea culpa regarding Pentium actually increased.
AMD, like any company will need a track record of producing quality competitive product, and producing it on time.
I agree with you, but if they are offering quality cards, then at some point (and hopefully not too late), AMD is going to have to raise prices and get the full value of the cards if they want to survive and eventually profit. But you're right, there will have to be some lead up into that stage.
Posted on Reply
#139
Nordic
I remember when people used to think tpu was amd biased. I only joined in 2011 too.
Posted on Reply
#140
Aquinus
Resident Wat-man
I remember when people just liked building computers and playing video games and didn't get so bent out of shape about hardware. I feel that there has been a shift to more confrontational attitudes towards "what people like," in general. Not even with just hardware.

All in all, I think people need to take a sedative, smoke a bowl, or drink a beer. Something just to get people to calm down. I'm here to embrace technology and give input from my own experience, not to dig my feet into the sand and convince everyone what is best. I would rather people make decisions for themselves based on factual information instead of simply "taking my word for it." I'll admit, I've had very few issues with my 6870s but that doesn't mean other's haven't had issues with AMD drivers. Simple fact is, AMD does makes a decent product and both AMD and nVidia has put a ton of effort into making good hardware.

For anyone bashing either the red or green camp, I dare you to design a GPU that does better. I personally think both camps have done a good job (minus minor PR issues that come with any company at some point(s),) and it's important to recognize that.
Posted on Reply
#141
CrAsHnBuRnXp
raghu78u mad at me for calling out their paid FUD and pure drivel.
Do you not know what a question is? See that? And that? Those are questions. Stuff like the above quote is a question. Not a statement. This site is one of the most reputable tech sites around. If you dont like what is here, then you can just exit our forums and we wont miss you.
Posted on Reply
#142
anolesoul
This is "all" very true. I have been a AMD/ATI person for many many years,due to the fact of better performance at a MUCH lesser cost. BUT,I have been disheartened with AMD...for Way too long...and IF they are going to make these recently stated advances that they have been "claiming",they had better DO it---SOONER...than "LATER"..in my book! AND...NOT WAIT TILL 2016!!!
Posted on Reply
#143
Rei86
raghu78you guys did squat till the issue blew up on forums. you and the rest of the press were just trolled by Nvidia. For 4 months you guys knew nothing. The users did all the hardwork analyzing the problems they faced at high res and high settings and in multi-GPU when the last 0.5GB was used. So how about not bragging about something you should be ashamed of. That you got fooled because your testing did not catch the problem. Instead of writing useless editorials like this improve your testing to catch such situations.
You do realize that no one did the test till owners of said card started to push them to the limit right?

Review websites aren't like owners. They have testing parameters where they feel like an avg person would use it for with defined settings. No testing website that I know of does a Skyrim test with a bunch of mods running in the background to see how fast they can fill up the VRAM of a GPU. And you know what, the GTX 970 even after finding out the last .5GB of the card is slow/cut off, didn't change the fact of its original test that was initially done by all websites. NONE of it changed the way it performed unless you're the type that wants to use the card at ridiculously high end resolution with a bunch of texture mods.

As for you bro, y so mad? I don't see the point in getting so angry at an OpED. You don't have to agree with the opinionated article but geez you're fully committed with the tinfoil hat.
Posted on Reply
#144
Tatty_Two
Gone Fishing
james888I remember when people used to think tpu was amd biased. I only joined in 2011 too.
I remember that too, funny isn't it, much of the leading AMD related software and overclocking tools that have been developed have been done so by the owner of this site over the last 10 years or so, if that's a good indicator of NVidia bias then I need to go buy Intel integrated graphics I think....... ohhh wait I have some of that :)
Posted on Reply
#145
Relayer
CrAsHnBuRnXpDo you not know what a question is? See that? And that? Those are questions. Stuff like the above quote is a question. Not a statement. This site is one of the most reputable tech sites around. If you dont like what is here, then you can just exit our forums and we wont miss you.
You really need to think about what you are saying when you post something like that.

Whether someone uses proper grammar or not isn't relevant. It's the content of their post. The free press can and should be called out when bias is noted. If the press is going to have an agenda it has forgotten it's purpose and bertrayed us all. What do you suppose the point of having a comments section attached to the editorial is?

There are many many people, who have given their lives to defend the right and existence of the free press. When someone notices what appears to them to be bias in the press, they should be angry. If it looks like someone is selling out to corporations, someone needs to speak up. Ironically, what you are doing by attacking him and telling him to go away is nothing more than an attempt at censorship. It's not community building and hiding behind that is wrong.

btarunner went on an editorial rant. Not everyone is going to agree with him. Raghu78 posted some valid reasons for his concerns.

I just have one question for TPU.

When did you get so concerned with drivers? I've witnessed W1zzard getting very angry on these forums when he's been called out for not redoing his benchmark suite when new drivers are released.
Posted on Reply
#146
HumanSmoke
RelayerThe free press can and should be called out when bias is noted. If the press is going to have an agenda it has forgotten it's purpose and bertrayed us all.
Very true, and the precise reason that I personally quit frequenting such sites as Fudzilla and SemiAccurate, whose content and personality is a reflection of the unfettered bias of its owner/main contributor. Vote your unwillingness to be part of the complicity by denying page clicks and linking seem like a reasonable stance to take in general.
Having said that, the editorial is the work of a single contributor - not the site in its entirety.
RelayerWhen did you get so concerned with drivers? I've witnessed W1zzard getting very angry on these forums when he's been called out for not redoing his benchmark suite when new drivers are released.
Can't speak for W1zzard, but as someone who has some extensive benchmarking experience meant for public consumption (that is to say a scrupulous attention to a consistent methodology), it can be a bit galling when people harp on relentlessly about retesting thinking it some trivial matter - neither thinking of the backlog it can create, juggling content already slated for publication, or the sheer work involved. W1zzard's latest graphics card encompassed 19 games at 4 resolutions with 28 different cards. Again I can't speak for W1zzard, but assuming all preparation done beforehand ( fresh install drives w/ appropriate drivers for each vendor etc), that would seem like a very solid weeks work - and of course, missing out a few cards (or resolutions) brings its own howls of complaint.

I would say that W1zzard's stance comes from the weight of the testing procedure, where just about everybody else's stance would be more directed at the immediacy of just about anything tech related - but then everybody else doesn't have to go through an endless repetition of benchmark runs.
Posted on Reply
#147
Aquinus
Resident Wat-man
RelayerI've witnessed W1zzard getting very angry on these forums when he's been called out for not redoing his benchmark suite when new drivers are released.
Do you know how much time it takes to do that entire process of benchmark suites? It's very time consuming and there is no easy way to automate it. So yeah, I don't blame W1zz for getting pissed off. I would too.
Posted on Reply
#148
CrAsHnBuRnXp
RelayerYou really need to think about what you are saying when you post something like that.
I'm sorry Mom. Ill do good by you next time. I promise.
Posted on Reply
#149
Relayer
I understand why W1zzard doesn't retest everything with every new driver release. He even stated that typically the difference is negligible to nonexistent (paraphrasing), and I agree. Here we have an editorial though that makes it sound like not only do we need drivers released more often but they have to be WHQL certified or AMD is dropping the ball. Even though, odds are the same sight he is posting on wouldn't even take the time to use the drivers anyway. Unless they were a legit performance driver release.

This might be an editorial for one person, it is TPU's rep though that is giving it weight. Thus people saying that you can't accuse the editorial of bias because TPU has a clear rep for being unbiased.

So, I'm not calling W1zzard out. 0:)
Posted on Reply
#150
W1zzard
RelayerI understand why W1zzard doesn't retest everything with every new driver release. He even stated that typically the difference is negligible to nonexistent (paraphrasing), and I agree.
The issue for me is time. Every full rebench takes around 2 weeks non-stop. Obviously I have other things to do than babysit benchmarks. So I retest every few months only, not every driver release, in which case the differences are negligible. I rebenched ahead earlier this month to be ready for upcoming cards from NV and AMD. This rebench also includes GTA V. Next rebench (which could be on Windows 10 if it's awesome) includes Project Cars and Witcher 3.
Posted on Reply
Add your own comment
May 4th, 2024 03:40 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts