Post Reply
AMD/ATI Pixel Clock Patcher
09-21-2013, 08:01 PM
Post: #161
RE: AMD/ATI Pixel Clock Patcher
(09-20-2013 07:49 PM)imzo Wrote:  When installing the patch i get an error " Debug: CryptUIWizDigitalSign failed: 0x80090016" followed by " File successfully patched, but signing failed".
That error code means the key container couldn't be found or accessed, but that doesn't make sense because the patcher is providing the key. Either there's an issue accessing the folder under %APPDATA%\Microsoft\Crypto\RSA, or something else is screwed up. Try deleting that folder, and if that doesn't work, try creating a new administrator user and running the patcher from that account.

You can also try the older version, which requires test mode: https://www.monitortests.com/atikmdag-patcher-1.1.2.zip

If the older version doesn't work either, then something is screwed up on your system. Nobody else has ever reported that particular error code.
Find all posts by this user
Quote this message in a reply
09-21-2013, 09:01 PM
Post: #162
RE: AMD/ATI Pixel Clock Patcher
(09-21-2013 08:01 PM)ToastyX Wrote:  
(09-20-2013 07:49 PM)imzo Wrote:  When installing the patch i get an error " Debug: CryptUIWizDigitalSign failed: 0x80090016" followed by " File successfully patched, but signing failed".
That error code means the key container couldn't be found or accessed, but that doesn't make sense because the patcher is providing the key. Either there's an issue accessing the folder under %APPDATA%\Microsoft\Crypto\RSA, or something else is screwed up. Try deleting that folder, and if that doesn't work, try creating a new administrator user and running the patcher from that account.

You can also try the older version, which requires test mode: https://www.monitortests.com/atikmdag-patcher-1.1.2.zip

If the older version doesn't work either, then something is screwed up on your system. Nobody else has ever reported that particular error code.

Thanks ToastyX, i deleted the RSA folder and tried to patch again, and this time is was successful Smile

Been trying various refresh rates using CRU, does 96Hz no problem, but at 120Hz i see lines of red flickering over the screen, is there anyway to get my Qnix stable at 120Hz or is my particular monitor just not capable of running stable at 120?

Thanks again for helping me with the patcher issue - much appreciated Wink
Find all posts by this user
Quote this message in a reply
09-22-2013, 12:39 AM
Post: #163
RE: AMD/ATI Pixel Clock Patcher
Try the "LCD reduced" timing option in CRU. If that's still problematic, try reducing the horizontal total to 2664. If that still doesn't help, then you might need a better quality cable, or the monitor can't handle it.
Find all posts by this user
Quote this message in a reply
09-22-2013, 08:44 AM (Last edited: 09-22-2013, 08:52 AM by imzo)
Post: #164
RE: AMD/ATI Pixel Clock Patcher
Thanks again for the help ToastyX !

Tried the LCD reduced option, and it helped, but not much - so i then tried your second suggestion of setting the horizontal total to 2664, this helped a LOT. I still see the occasional red line when in game, but rarely, i think getting a better cable would eliminate the problem completely.

Thanks again ToastyX - Much appreciated Wink


Ps. Is there an optimal setting in CRU - eg. is it better to have the timings higher or lower? And what about pixel clock - would increasing the pixel clock be of any benefit or should that not be touched?

EDIT: Forgot to mention - it seems that your patch has issues when playing video content in an application such as MPC or in a web browser (flash), i get instant display corruption that is only fixed by restarting Windows.
Find all posts by this user
Quote this message in a reply
09-22-2013, 09:10 AM
Post: #165
RE: AMD/ATI Pixel Clock Patcher
Did you right click the video and disable hardware acceleration? It's a known issue
Find all posts by this user
Quote this message in a reply
09-22-2013, 09:32 AM
Post: #166
RE: AMD/ATI Pixel Clock Patcher
(09-22-2013 09:10 AM)Black Octagon Wrote:  Did you right click the video and disable hardware acceleration? It's a known issue

Thanks for the reply Black Octagon.

Ah, so its a known issue... Hopefully a fix can be found.

I had thought to do disable HW Acceleration, but the problem is that as soon as the video starts to play, the entire screen becomes corrupted and i cant see anything anymore. And it isnt just limited to flash, playing any movie in say Media Player classic causes the same corruption when using DXVA. I would rather not disable HW acceleration - especially in MPC as this disables the various filters.

The only option i see is to remove the patch, then disable HW Acceleration in flash, then reapply the patch. Not sure what can be done with none flash content other than disabling DXVA.

Thanks again for your reply.
Find all posts by this user
Quote this message in a reply
09-22-2013, 01:38 PM
Post: #167
RE: AMD/ATI Pixel Clock Patcher
Lower is better if the monitor can handle it because it uses less bandwidth. The pixel clock is tied to the refresh rate and the totals, so reducing the totals is the only way to reduce the pixel clock without reducing the refresh rate. I think 2652 is the lowest horizontal total I've heard that works with the QNIX, but you might need a better cable for that to work. The vertical total can also be reduced to 1445.

The screen corruption is a separate issue unrelated to the patch. That happens if you're overclocking the memory on your video card when the vertical total is reduced. AMD cards are programmed to change the memory clock back to stock when video acceleration is active, but the memory clock is not supposed to change when the vertical total is reduced because there isn't enough time for link retraining without screen corruption. I don't know why AMD hasn't fixed that. It's been a problem since the 5000-series at least. The problem also affects 144 Hz monitors and multi-monitor configurations.

When you undid the patch, the problem stopped happening because it switched back to the default 60 Hz with the standard vertical total. "LCD standard" keeps the blanking high enough to avoid the problem, but most 2560x1440 monitors can't handle that at 120 Hz. The easiest way around the problem is to leave the memory clock at stock. It's possible to overclock the memory by messing with the PowerPlay tables, but that's a huge pain and not worth the trouble.

That won't fix the video acceleration issue though. There seems to be some sort of integrity check that prevents video acceleration from working if any part of the driver is modified. I don't know if a fix will ever be found.
Find all posts by this user
Quote this message in a reply
09-22-2013, 02:20 PM
Post: #168
Thumbs Up RE: AMD/ATI Pixel Clock Patcher
(09-22-2013 01:38 PM)ToastyX Wrote:  Lower is better if the monitor can handle it because it uses less bandwidth. The pixel clock is tied to the refresh rate and the totals, so reducing the totals is the only way to reduce the pixel clock without reducing the refresh rate. I think 2652 is the lowest horizontal total I've heard that works with the QNIX, but you might need a better cable for that to work. The vertical total can also be reduced to 1445.

The screen corruption is a separate issue unrelated to the patch. That happens if you're overclocking the memory on your video card when the vertical total is reduced. AMD cards are programmed to change the memory clock back to stock when video acceleration is active, but the memory clock is not supposed to change when the vertical total is reduced because there isn't enough time for link retraining without screen corruption. I don't know why AMD hasn't fixed that. It's been a problem since the 5000-series at least. The problem also affects 144 Hz monitors and multi-monitor configurations.

When you undid the patch, the problem stopped happening because it switched back to the default 60 Hz with the standard vertical total. "LCD standard" keeps the blanking high enough to avoid the problem, but most 2560x1440 monitors can't handle that at 120 Hz. The easiest way around the problem is to leave the memory clock at stock. It's possible to overclock the memory by messing with the PowerPlay tables, but that's a huge pain and not worth the trouble.

That won't fix the video acceleration issue though. There seems to be some sort of integrity check that prevents video acceleration from working if any part of the driver is modified. I don't know if a fix will ever be found.


Thanks for the info ToastyX.

I did actually have my card OC'd, didnt think that that could have been causing the display corruption though - but i set everything back to stock settings in CCC overdrive and tried playing a video again with DXVA enabled - sure enough, it now plays perfectly, so thanks for your help with that.

I will have a play around with trying to lower the totals some more and see how low i can get it, but as you said i may need a better cable...

Thanks again for all your help and insight, much appreciated Wink
Find all posts by this user
Quote this message in a reply
10-01-2013, 12:19 AM
Post: #169
RE: AMD/ATI Pixel Clock Patcher
Hi Toastyx,

I am currently using the qnix270led on a 670gtx but will be upgrading to the new amd r290x when it's available.

Will the monitor have issues with the new cards? or will i be okay?

Thanks Wink
Find all posts by this user
Quote this message in a reply
10-01-2013, 12:42 AM
Post: #170
RE: AMD/ATI Pixel Clock Patcher
There's no way for me to know. So far, everything starting with the 5000-series shares the same limits.
Find all posts by this user
Quote this message in a reply
 Post Reply


Forum Jump:


User(s) browsing this thread: 9 Guest(s)