ATI HD5800 / Catalyst Improvements, Bugs, Fixes

I've stopped trying to reason whether it is developers not reaching out to AMD, or AMD not doing enough for developers. I like AMD products and now own a 5770 but this has always been a sore spot with me and of course you can never get a straight answer from anyone. It's always the dev stating AMD didn't help them at all, then AMD stating the dev never contacted them blah blah. And no dev or publisher will ever state publically that they won't (or can't) work with ATI due to Nv TWIMTBP program.

In the end Nvidia obviously and very publically works close with developers, AMD don't seem to at all from a consumers point of view and that fact is rather depressing.

 

Yes, that is not the problem.

 

Update on Ultra Large Cursor issue I've had.

I broke down and tried the 9.11 beta cats and it had no effect. No clue whether the fix made it into those cats or not. However...

Broken Hope clued me into reports that a new beta bios for Gigabyte boards that addresses a problem with GOM player, have been able to fix this issue.

So far, so good. So looks like there's some conflict between the old Gigabyte bios and the 58xx series of cards (didn't happen with 4890 on the same MB). It's only been a couple days but so far no large cursors, and it used to pop up within 5 min - 2 hours of starting up windows.

Regards,
SB

 

Uhm, links or something to back up what you're saying please?

 

Digi powerplay is bugged with multiple monitors. This is a pretty well documented issue. I can't even run multiple monitors.

 

It's not "bugged". As Dave has posted previously, you can't fiddle with memory timings during vsync because there are two vsyncs to worry about.

 

If it's not a bug it's a regression as this worked perfectly fine on my R700, but now it doesn't on my R870.

(edit) And I'm not entirely sure we are both talking about the same issue.

 

R700 wasn't actually using PowerPlay - or at least the hardware PowerPlay - it was just using software to detect "3D exclusive mode" and switching from 2D states to 3D states. Of course, if an app doesn't declare 3D exclusive mode it will be stuck in the 2D states.

Memory clock switching does not occur in mutli-monitor scenarios for the with hardware PowerPlay systems for the reason OpenGL guy commented on (which, as he pointed out, I've also commented on previously). With R700, the flash in multimon mode wasn't so much of an issue because it basically only does it when you boot a game, and in many instances it may also be changing resolution as it loads up, so that wasn't deemed an issue; with hardware PowerPlay, because we can get much lower lows and because it is hardware activity based PowerPlay states will move around frequently - both at the desktop and even in games - and so in multi-panel mode we have restricted the memory clock switching so that the screen flashing as it moves states on some of them do not occur.

Even though, in multi-mon modes, we have to restrict the engine clock a little more than we can in single panel modes, we still achieve better power savings than we could with 2D/3D switching.

 

I understand that, but at the end of the day, I still can't use multiple monitors...

 

I don't get your problem Willard, can you explain it in small words for me?

 

If hardware PowerPlay is based on hardware activity what is the reason that the Flash 10.1 beta will bug out PowerPlay and prevent it returning to idle clocks after viewing a HD flash video?

 

PowerPlay works with a number of software/system detected states (i.e. "standard single monitor", Multi-Monitor, UVD, Overdrive, etc., etc.) each of which can have high, medium and low values. The UVD state tends to have the high, medium and low states set at the same values.

 

But shouldn't the UVD state only be working when UVD is accelerating decoding? The Flash 10.1 bug keeps the GPU at UVD clock speeds when idle until you reboot the system even when I'm not viewing a video/HD flash video.

 

Yes, that indicates that something that has activated UVD is still lingering in memory,

 

Not sure what it is then since even when closing down Firefox which should unload the flash plugin the clocks still don't return to idle clocks.

 

Are you certain nothing is still resident? Check the task manager to be sure.

 

I'm certain, I closed everything down, the only processes running under my name were stuff that's there at boot up. At a fresh start it's fine, it will run at idle clocks, once a HD flash video has been run it's stuck at UVD clocks, this only occurs with the Flash 10.1 beta.

 

Hello Dave, I can't see this listed anywhere so I thought I would ask.
It just seems like ATI doesn't know about the 120hz problems with their cards. The problem is that most ATI users can't use the new 120hz lcd screens in 2D mode and some also have problems in 3D mode, since that causes distortion. It's just been over a year since the screens came out, and it still doesn't work unless you use a nvidia card

If you google ATI 120hz you can find lots of threads about it, and on the ATI forum there are many threads too. There is a thread on Anandtech as well with some videos of it here.
http://forums.anandtech.com/showthread.php?t=327884
It just seems like a really really easy thing to fix. All that has to be done is to increase the 2D clock a bit if it runs on 120hz..so it seems weird it takes so long. And as it's pointed out on the last page on the Anandtech thread, it's not even recognized as a know problem in the driver changelog

Thank you.

 

Read this thread. I was updated a few drivers back. You may still see an issue if you're overclocking, but remove the OC and use the latest drivers and give it a go.