Post Reply
AMD/ATI Pixel Clock Patcher
08-29-2013, 03:44 AM
Post: #151
RE: AMD/ATI Pixel Clock Patcher
(08-27-2013 10:46 AM)castrwilliam Wrote:  Would it be possible to patch the VGA dot-clock max limit out?
It might be possible. The main problem is I don't have a monitor that can go that high, so I can't test it myself. I'll see what I can find...
Find all posts by this user
Quote this message in a reply
09-02-2013, 10:37 AM
Post: #152
RE: AMD/ATI Pixel Clock Patcher
I followed the instructions to the T but I'm getting horrible glitches.

I have a 7970 and a QNIX monitor. Now, when I built the computer the first time, I had amazing results at 120hz other than the red lines for a while. After I brought it down to 110hz which got rid of the red lines, I began to run into problems where the screen would flicker and eventually black out.

So I began troubleshooting. Reset both CRU and ATI Patch. Reinstalled 13.8beta-> applied current ATI patch-> reboot computer-> Open CRU and copy/paste stock 59hz settings-> tweak 2nd profile refresh rate to 110Hz->bring it to the top-> reboot computer -> BIOS splash good! Windows 7 64 bit Loading logo GOOD! -> the MOMENT it enters the desktop, screen flickers and then blacks out immediately. I have to go to safe boot to reset CRU to get back into desktop normally.

What am I missing? It worked the first time and I followed the Youtube videos, what has changed? I tried it with 13.4 CCC as well. Same results.
Find all posts by this user
Quote this message in a reply
09-09-2013, 08:09 PM
Post: #153
RE: AMD/ATI Pixel Clock Patcher
I currently run crossfire and am running the two bridges per your recommendation. I however would like to go trifire and I understand that there is a pixel clock limitation using one bridge however I may have come up with a workaround if you would bear with me and validate my logic. I know of an individual that does extensive prototyping and has resources to get custom PCB's made. Would it be possible to create a PCB like those found for Nvidia that populates both crossfire slots on each card? Would that not effectively solve the pixel clock limitation of running more than 2 cards as they will all have full bandwidth? I may be overlooking something however as far as I understand this should work.

Thanks for your insight....
Find all posts by this user
Quote this message in a reply
09-10-2013, 12:15 AM (Last edited: 09-10-2013, 01:05 AM by Hobart)
Post: #154
Thanks ToastyX! 4K (3840x2160) resolution for Radeon HD 66xx (Turks) in Windows
I've got these items, connected by a single HDMI cable:
  • Radeon HD 6670 "Turks" w/ Catalyst 13.4 (GPUZ details)
  • Seiki Digital SE39UY04 39" 4K UHDTV (MONINFO technical details from EDID)
    • Running under Linux (Ubuntu 13.04 64, proprietary Catalyst 13.4), everything automatically comes up perfectly in 3840x2160. (XOrg log)

      Under Windows 7 Ultimate 64, the card would only output 1080p (1920x1080).

      I patched my drivers using the AMD/ATI Pixel Clock Patcher, followed by using the Custom Resolution Utility.

      Windows is now looking great in 3840x2160. (screenshot)
Visit this user's website Find all posts by this user
Quote this message in a reply
09-10-2013, 08:00 AM
Post: #155
RE: AMD/ATI Pixel Clock Patcher
(09-09-2013 08:09 PM)RequiemFiasco Wrote:  I currently run crossfire and am running the two bridges per your recommendation. I however would like to go trifire and I understand that there is a pixel clock limitation using one bridge however I may have come up with a workaround if you would bear with me and validate my logic. I know of an individual that does extensive prototyping and has resources to get custom PCB's made. Would it be possible to create a PCB like those found for Nvidia that populates both crossfire slots on each card? Would that not effectively solve the pixel clock limitation of running more than 2 cards as they will all have full bandwidth?
I don't know if that will work. I've never tested CrossFire. It might be possible to get around the limit with three cards by using three bridge cables. Someone will have to check if that helps. The limit might also depend on the rendering mode.
Find all posts by this user
Quote this message in a reply
09-10-2013, 12:26 PM
Post: #156
RE: AMD/ATI Pixel Clock Patcher
(09-10-2013 08:00 AM)ToastyX Wrote:  
(09-09-2013 08:09 PM)RequiemFiasco Wrote:  I currently run crossfire and am running the two bridges per your recommendation. I however would like to go trifire and I understand that there is a pixel clock limitation using one bridge however I may have come up with a workaround if you would bear with me and validate my logic. I know of an individual that does extensive prototyping and has resources to get custom PCB's made. Would it be possible to create a PCB like those found for Nvidia that populates both crossfire slots on each card? Would that not effectively solve the pixel clock limitation of running more than 2 cards as they will all have full bandwidth?
I don't know if that will work. I've never tested CrossFire. It might be possible to get around the limit with three cards by using three bridge cables. Someone will have to check if that helps. The limit might also depend on the rendering mode.

Thanks for the reply. I think I will look into it further. I cant help but wonder why AMD never made a pcb style connector. It may be a limitation on the cards end or maybe they diddnt want to copy Nvidia. Ill see what I can find out.
Find all posts by this user
Quote this message in a reply
09-18-2013, 10:45 PM
Post: #157
RE: AMD/ATI Pixel Clock Patcher
(09-07-2012 09:07 PM)ToastyX Wrote:  [*]CrossFire requires two bridges if the pixel clock is greater than 320 MHz. This is only possible with cards that have two connectors. It will not work properly with more than two cards.

I just tested this out without the second connector and I am having no issues so far. I have benched it in unigen and bioshock both ways as well as a few hours of BF3 multiplayer and everything seems the same. My monitor is clocked at 108htz/2560 with a 450ish pixel clock. Is the issue with crossfire on specific titles or is it possible that the pcie lanes and crossfire bridge is offering enough bandwidth at this resolution?

This gives me hope that 3 way crossfire is doable with single connectors. What is different about the way the crossfire bridge and pcie lanes are saturated from a 120htz eyefinity group vs a 2560 monitor running at 120htz? Dont the cards share the frame buffer and workload and since eyefinity is essentially one large display wouldn't they be at a higher combined pixel clock than a 2560 monitor?

I will continue to test this in some other titles and report back. What information were you using when you suggested to use 2 bridges?

Thanks a bunch.
Find all posts by this user
Quote this message in a reply
09-19-2013, 01:31 AM
Post: #158
RE: AMD/ATI Pixel Clock Patcher
Some people reported problems with some games crashing on startup and severe stuttering with vsync enabled. BF3 is one of the titles reported to have problems with vsync enabled, and Dirt 3 is reported to have problems starting up. I don't know if that only affects certain configurations, but multiple people confirmed the problems went away after adding a second bridge.

I don't know what's different about Eyefinity, but something seems to be different because the new frame pacing improvements in 13.8 don't work with Eyefinity yet.

Have you tried the different rendering modes to see if that has any effect?
Find all posts by this user
Quote this message in a reply
09-19-2013, 03:25 PM
Post: #159
RE: AMD/ATI Pixel Clock Patcher
I have not messed with rendering modes lately as it seems the profiles are ideal most of the time anyways, I will try this evening after class and see if anything changes. Its no wonder I have not seen issues, I am fundamentally against vsync and never run it (mouse lag) so that explains why I have not seen that problem. Ill test out vsync later as well.

I wonder if pcie gen 3 has anything to do with it...
Find all posts by this user
Quote this message in a reply
09-20-2013, 07:49 PM (Last edited: 09-20-2013, 07:50 PM by imzo)
Post: #160
RE: AMD/ATI Pixel Clock Patcher
Hi guys!

Just registered here.

Wondering if maybe someone can help me overclock my new Qnix. I have followed all instructions and installed all files and done everything else, but no matter what i do i cannot get the patched file to work. When installing the patch i get an error " Debug: CryptUIWizDigitalSign failed: 0x80090016" followed by " File successfully patched, but signing failed".

I used DSEO to enter test mode, although no test mode text appears on the desktop after restart. I have tried disabling driver enforcement in CMD using bcdedit.exe -set TestSigning On. But even after this i am getting the same error when using the patch. After patching and restarting, i get another error message on screen - this time from CCC telling me that no hardware drivers are installed or are not working correctly.

I have been trying everything i can over the past couple of days, but nothing has worked so far. I have followed guides both here and in youtube vids. I am at a loss as to what to do.

I have tried manualy signing the patched driver, but this did not work either.

I am using a HD7970 GPU with 13.10beta driver package installed on Windows 7 Ultimate 64bit.

ANY help is much appreciated.

Thanks in advance.
Find all posts by this user
Quote this message in a reply
 Post Reply


Forum Jump:


User(s) browsing this thread: 16 Guest(s)