AMD/ATI Pixel Clock Patcher

[Vennto]

(12-11-2016 12:20 AM)ToastyX Wrote:  

(12-10-2016 10:30 AM)Vennto Wrote:  Since I'd like to create a program to disable bios check automatically for the next drivers, I'd like (if possible) to know the process (or simply a small copy of the code) that this tool use to avoid conflict with windows driver signature enforcement.

Do you have a kernel-mode code signing certificate?

Nope, but I was thinking about sign them with a test cert (https://msdn.microsoft.com/en-us/windows...er-package) and import it as trusted... yeah yeah there are a few security warnings but these things should not be used by common user.


... or... I could make the program that patch the bios check and simply continue to use your tool to sign the drivers

[ToastyX]

(12-11-2016 12:56 AM)Vennto Wrote:  Nope, but I was thinking about sign them with a test cert (https://msdn.microsoft.com/en-us/windows...er-package) and import it as trusted...

That will not work for kernel-mode drivers. Test-signed drivers will only load in test mode or with driver signature enforcement disabled.

[PanzerIV]

MAJOR BUG:
I've done a clean uninstall and reinstall of the latest driver for AMD on Windows 10 64bit for my (R9 390 8Gb) because I kept having visual corruption as if I had overclocked way too much my card, for no damn reason since I updated to that last driver... the cleaning fixed all of it, however here's my issue with the latest version of the (Pixel Clock Patcher 1.4.3)

If I apply it then reboot, my monitor (BenQ XL2730Z) that's 2560x1440 through Display Port at usualy 144Hz, well it won't have signal anymore and it remains black if I attempt using 144Hz... it will only work at 120Hz. If I remove the patch and reboot, it works again at 144Hz.

Actualy is there any advantage of using the patcher for simply being at (2560x1440 144Hz) with a DP cable, or I gain nothing at all from doing it?

[ToastyX]

(12-14-2016 06:08 AM)PanzerIV Wrote:  MAJOR BUG:
I've done a clean uninstall and reinstall of the latest driver for AMD on Windows 10 64bit for my (R9 390 8Gb) because I kept having visual corruption as if I had overclocked way too much my card, for no damn reason since I updated to that last driver... the cleaning fixed all of it, however here's my issue with the latest version of the (Pixel Clock Patcher 1.4.3)

If I apply it then reboot, my monitor (BenQ XL2730Z) that's 2560x1440 through Display Port at usualy 144Hz, well it won't have signal anymore and it remains black if I attempt using 144Hz... it will only work at 120Hz. If I remove the patch and reboot, it works again at 144Hz.

That's not a bug in the patch. The XL2730Z defines 2560x1440 @ 144 Hz twice for some reason, and the new HBlank patch is unlocking the first one. I don't know why they included that if the monitor can't handle it. Use CRU to edit the extension block and delete the first 2560x1440 @ 144 Hz (538.76 MHz). It should be using the 143.855 Hz (586.00 MHz) one.

(12-14-2016 06:08 AM)PanzerIV Wrote:  Actualy is there any advantage of using the patcher for simply being at (2560x1440 144Hz) with a DP cable, or I gain nothing at all from doing it?

What are you using the patch for? What limits were you encountering without the patch?

[PanzerIV]

(12-14-2016 07:28 AM)ToastyX Wrote:  What are you using the patch for? What limits were you encountering without the patch?

Thanks a lot for the very quick reply. It's a really odd and specific bug on that particular monitor, but it's great that you were already aware of it and able to tell me about it.

To be honnest like I've said, I have absolutely no idea if the patch give 0 color, sharpness or image quality improvement from having extra "bandwidth overhead" or if you have no issue then, it won't change anything. I know it used to be a total must-have for many years ago when I was with my previous (XL2420Z) with DVI-D and that I needed to remove the limit when I was creating a custom resolution with different vertical values as recommended at (BlurBusters.com) when I was using the (BenQ Motion Blur Reduction) feature but that was only possible with the 1920x1080 resolution, so now that vertical trick is useless by being at 2560x1440.

So if you say there is absolutely no reason at all to use the patcher in my case, then I'll follow your recommendation ^^

[ToastyX]

(12-15-2016 09:31 AM)PanzerIV Wrote:  To be honnest like I've said, I have absolutely no idea if the patch give 0 color, sharpness or image quality improvement from having extra "bandwidth overhead" or if you have no issue then, it won't change anything.

The patch doesn't provide any performance benefits. It only removes certain restrictions to allow resolutions that are normally blocked by the driver.

[Vennto]

Do this patch break video acceleration and hdcp as soon as I apply it or only when I actually go above the older unpatched limits?

[ToastyX]

(12-17-2016 09:59 AM)Vennto Wrote:  Do this patch break video acceleration and hdcp as soon as I apply it or only when I actually go above the older unpatched limits?

As soon as it's applied. Any change to the driver code causes this to happen. I'm not sure to what extent video acceleration is broken. Since 14.4, there are cases where it appears to work, but I don't have a way of knowing for sure. Some of the workarounds might not be necessary anymore. I don't have anything that requires HDCP either. Supposedly the Netflix app does, but I'm able to use it with the patch.

[Black Octagon]

Video acceleration no longer seems broken on my 7970. 2 years ago or so it was an issue, though I can't remember what version of Catalyst drivers they were


Sent from my iPhone using Tapatalk

[ToastyX]

(12-18-2016 04:51 PM)Black Octagon Wrote:  Video acceleration no longer seems broken on my 7970. 2 years ago or so it was an issue, though I can't remember what version of Catalyst drivers they were

When video acceleration appeared to start working, I still had playback issues like the beginning of videos not displaying correctly. I just left acceleration disabled for a while to avoid these problems, so I don't know when things changed. Even recently, I had to disable acceleration because of driver bugs unrelated to the patch causing Firefox to crash with RX-series cards.