News for hackers. Stuff that matters.
Re:Adrenalin driver breaks Command & Conquer games by bwang29 on Tuesday January 02, 2018 @03:43AM (#16050713)
Response doesn't make sense at all

"This title is from 2007, so we are unlikely to devote any valuable engineering resources to this issue, which is most likely caused by outdated API modules"

Doesn't make sense #1: Reason to not fix because it is from 2007, what about counter strike?

Doesn't make sense #2: Fixing the issue implies devoting "valuable" engineering resource, does it mean they don't have enough engineers or such issue is extremely difficult?

Doesn't make sense #3: "outdated API modules" was that likely from an educated guess and what does it mean even.

It sounds really foolish for a reply of this calibre to a community of supporters. I love AMD for what they're doing technically but I hope someone can at least promise to fix it soon before this hits the larger community.

Re:Adrenalin driver breaks Command & Conquer games by hvidgaard on Tuesday January 02, 2018 @04:02AM (#16050786)
This is about politics and nothing else. I hope, and believe, AMD will fix this.

The main issue is probably due to a bug in the old driver. After a while, bugs become a feature, and it's not uncommon to implement bugs to retain backwards compatibility. This is where the politics enter the scene. Some engineers believe that all bugs must be squashed, while others have a more pragmatic approach. IMO, any behavior that is a problematic deviation from the old drivers, should be changed. Otherwise there is no backwards compatibility, only a best effort approach, which is unacceptable for something as crucial as a graphics card driver.

Re:Adrenalin driver breaks Command & Conquer games by digi_owl on Tuesday January 02, 2018 @04:36AM (#16050900)
Sadly AMD seems to be in a rush to drop stuff from their Windows driver.

I had the unfortunate experience of watching them drop support for the GPU part of my APU while it was still one of their headline APUs.

This because they had based it on a previous gen discrete GPU, and they didn't want to support that GPU range any more with future drivers.

It is experiences like these that makes me understand were RMS came from, and it pains me when so many even within the FOSS world poo poo his message and seem all too willing to break things left and right just because it is "old and crufty" code.

Re:Adrenalin driver breaks Command & Conquer games by ricardobeat on Tuesday January 02, 2018 @04:11AM (#16050817)
When using a gl API, doesn’t code declare a version number that they could use to selectively patch known bugs, if that’s the case?
Re:Adrenalin driver breaks Command & Conquer games by kevingadd on Tuesday January 02, 2018 @08:22AM (#16051679)
There are specific version profiles like "opengl 3.3 core profile" but a lot of older apps aren't versioned in any fashion, they just request specific extensions. Behavior for specific extensions has varied (old drivers had major bugs, and apps had workarounds for them). If you're on mobile you can assume with near-100% reliability that your GLES driver has a bunch of bugs and you'll need to ship workarounds for them. Chrome (Blink) and Firefox (Gecko) both have huge blacklists for specific GPUs/drivers along with sets of workarounds for issues, many of them specific to individual mobile GPU chipsets or drivers. I shipped a game targeting GL 3.3 core a couple years ago and we hit multiple driver-specific bugs, despite 3.3 being pretty old.

Now imagine you're a driver developer: All this software out there assumes your driver is broken. What happens when you fix it? Well, hopefully the app's workarounds don't actually cause it to break or produce incorrect results. What if the workarounds make it really slow? You want it to run fast on your hardware so people won't buy your competitor's hardware. So now you have workarounds for their workarounds...

OpenGL shaders have (iirc) always been versioned, so that's good at least. The introduction of profile versions also helps but at present even those tend to have major bugs depending on OS and driver, so workarounds still end up in games and applications.

Direct3D was historically much more explicit about versioning and feature sets but there's a lot of variability through things like caps flags (hardware can advertise different levels of functionality) and the app explicitly requesting specific features like a given texture format. Shaders in Direct3D are versioned, but it's quite possible that a driver is only built to handle shaders generated by a given shader compiler, so a new compiler targeting the same shader version will still break the driver. Some shaders probably were generated by old broken compilers, and the driver probably has workarounds for those.

Also, modern graphics drivers typically have a bunch of hacks they quietly apply to big-name game titles and applications to make them run faster or suppress crashes. This goes as far as driver developers actually rewriting the shaders that a game uses, in order to improve performance on their hardware. An infamous old example of this is 'quack3'. https://techreport.com/review/3089/how-ati-drivers-optimize-...

On the bright side, a lot of this madness is gone in Vulkan and Direct3D 12. But very little software is going to end up using those APIs because they're much, much harder to use correctly.

Re:Adrenalin driver breaks Command & Conquer games by wastewaste on Tuesday January 02, 2018 @08:22AM (#16051676)
> for something as crucial as [computer games]

fixed

Re:Adrenalin driver breaks Command & Conquer games by giancarlostoro on Tuesday January 02, 2018 @07:16AM (#16051441)
> Doesn't make sense #3: "outdated API modules" was that likely from an educated guess and what does it mean even.

I hope it doesn't mean they lost the source code. Just as bad: did they properly use VCS back in the day? You'd be surprised, I still hear of companies that don't use VCS to this day.

Re:Adrenalin driver breaks Command & Conquer games by viraptor on Tuesday January 02, 2018 @07:29AM (#16051481)
RA does not have as many active players as CS. They may mean that it's the business calculation of (no_developers * time * salary) > potential loss.

Either way - hard to figure out what's the move and agenda is about from a single paragraph from a forum post.

Re:Adrenalin driver breaks Command & Conquer games by solarkraft on Tuesday January 02, 2018 @11:24AM (#16052814)
If they did it like that they sure messed up by forgetting PR. They can't afford not being the"good guys" in the processing space.
Re:Adrenalin driver breaks Command & Conquer games by rasz on Tuesday January 02, 2018 @06:27PM (#16056901)
Yes, AMD doesnt have engineers. They rather throw money into garbage fire like Seamicro acquisition resulting in total loss of ~$300mil than actually invest in developers.

- This is why AMD drivers require 15-25% more single thread CPU power to reach fps parity with Nvidia, basically if you want AMD GPU you need to go Intel CPU).

-This is why their educational outreach is a joke, all GPU computing uni courses are teaching Cuda with Nvidias provided support (materials and hardware).

-This is why there still is no viable answer to Cuda. Forget about AMD and Tensorflow.