so which version am i looking for which has a good chance to flash?Ver 1.0 => High chance that card would not POST.
so which version am i looking for which has a good chance to flash?Ver 1.0 => High chance that card would not POST.
Read post 1 again. (It has been updated recently...)so which version am i looking for which has a good chance to flash?
Have you tried running any benchmarks? It is very likely your iMac will simply reboot from time to time - you have mounted a power hungry GPU into a 21.5' iMac, if your signature is correct. NVIDIA table on post 1 clearly states that this is an incompatible combination.Finally I was able to successfully swap my GPU to a pre-patched GTX880M from AliExpress. So far so good. I went straight to upgrade to upgrading to Big Sur using the OC legacy patcher and the system runs smoothly.
However, I'm having a sleep issue. The system goes to sleep but does not come back and I have to reboot it. Not sure but I think this more related to OC than the graphics card? Is this a known problem?
Thanks
Sorry, I've corrected my signature. I'm running it in a 27" model with 3 heat pipes installed...Have you tried running any benchmarks? It is very likely your iMac will simply reboot from time to time - you have mounted a power hungry GPU into a 21.5' iMac, if your signature is correct. NVIDIA table on post 1 clearly states that this is an incompatible combination.
May be your panel type is a non standard. Find it out using this and post your type.There is no Backlightinjecor - is it supposed to be loaded permanently or is it just used during the boot?
What else could I check?
Did you try to flash using amdvbflash?Looking for help moving the circuit, before looking at the forum someone shared a Dell Red PCB HD6770M replacement 128k capacity EEPROM to write a modified Dell HP VBiOS support iMac. i myself also have a Dell Red PCB HD6770M can work fine on iMac 2011 21.5 2011 27.
I recently acquired 10 HP M5950 cheaply, the pcb color has two kinds of blue and green compared to the Dell red PCB front missing EEPROM to make up for themselves, the back circuit has a difference part of the empty solder.
I'm pretty sure there is no problem soldering EEPROM, soldered 6 sheets are the same result can not read from the EEPROM to identify the correct VBiOS information written VBiOS written a different Dell HD6770M forum modified Dell HP iMac M5950 6770M are the same. It is blank in any case.
EEPROM is to purchase dozens of brand new Huabang 256kb capacity actual written VBiOS information is 128kb.
These M5950M 6770M are HP 8560W 8760W tested to determine that the graphics card is normal workmanship without any problems. In the PC PCIE adapter MXM development debug card test graphics card can also work properly run scores also no problem. gpu-z view VBiOS information is blank did not correctly read the various tests written by CH341A VBiOS information.
Translated with www.DeepL.com/Translator (free version)
May be your panel type is a non standard. Find it out using this and post your type.
Yes, there is no BacklightInjector any longer, @khronokernel found another way to achieve the same mod using a DeviceProperties entry with OCLP.
I am not quite sure if this new solution will work in case you come up with a new formerly unknown display type. In the old times we had to add this new display type to the BacklightInjector plist.
EDIT: But this does only effect the full range of backlight, not the backlight control itself.
For what it's worth on my iMac 12,2 with a K3100M I get a null answer to the panel type. (I too have an issue with sleep and figured it would be worth looking)May be your panel type is a non standard. Find it out using this and post your type.
$ ioreg -lw0 | grep IODisplayEDID | sed "/[^<]*</s///" |xxd -p -r | strings -6
$
Check each of @nikey22 vBIOS upload posts - he described another method.Well... the output does not look good...
View attachment 1818161
Hi @internetzel Using amdvbflash graphics tool with CMD command to refresh Dell original M5950 HD6770M shows programming success, test GPU-Z tool shows VBiOS information is still blank no correct reading of VBiOS information written by programming. I guess it's still a hardware problem.Did you try to flash using amdvbflash?
amdvbflash shows valid version and product name for the EEPROM contents - so I'd say that flashing has indeed been successful. It doesn't show anything for the new VBIOS you are flashing, but if I remember correctly that happens when old and new VBIOS have identical version and product name.Hi @internetzel Using amdvbflash graphics tool with CMD command to refresh Dell original M5950 HD6770M shows programming success, test GPU-Z tool shows VBiOS information is still blank no correct reading of VBiOS information written by programming. I guess it's still a hardware problem.
My friend has a 12.2 imac 27. The k3100m video card works perfectly in High sierra. Sleep-works perfectly. It works perfectly in Big Sur! Do you have OC installed correctly?For what it's worth on my iMac 12,2 with a K3100M I get a null answer to the panel type. (I too have an issue with sleep and figured it would be worth looking)
Code:$ ioreg -lw0 | grep IODisplayEDID | sed "/[^<]*</s///" |xxd -p -r | strings -6 $
I may be thinking of something slightly different but I know that a mods team in Malaysia HAVE made a special connector that fits in place of the imacs internal dvd drive and allows the connection of an external GPU via the dvd drive PCIE connector.I've just contacted with the seller and he confirmed that the internal LCD will not display anything.
Project dropped.... sadly.
I don't have OC installed because I am not trying to run an unsupported OS... on the list of things to do but busy with other stuff. (I have a 2010 11,3 imac that I'm going to try to put a GTX880 into and upgrade the hard disk to an SSD that is in front of more work with this imac 12,2. I just don't allow it to sleep and it's good unless I have a power failure and then rebooting takes a number of button pushes to get it to come on.My friend has a 12.2 imac 27. The k3100m video card works perfectly in High sierra. Sleep-works perfectly. It works perfectly in Big Sur! Do you have OC installed correctly?
I also have a green PCB 1GB version that had no bios chip, I soldered one from a dead 4670 and put it in the correct vbios but as you experienced already it is not read like the card still has no bios chip as before, I suspect it has some missing stuff, resistor or whatever on the bios line, if you find a fix please let me know.Hi @internetzel Using amdvbflash graphics tool with CMD command to refresh Dell original M5950 HD6770M shows programming success, test GPU-Z tool shows VBiOS information is still blank no correct reading of VBiOS information written by programming. I guess it's still a hardware problem.
I also have a green PCB 1GB version that had no bios chip, I soldered one from a dead 4670 and put it in the correct vbios but as you experienced already it is not read like the card still has no bios chip as before, I suspect it has some missing stuff, resistor or whatever on the bios line, if you find a fix please let me know.
On the other hand I have 2 K2100m that have the so called black screen bug. they work just fine with windows but not with any version of OSX/macOS unless I press alt on boot to fire up boot manager. I think there is a "bug" in the VBIOS.
Before anybody ask, no, I do not use OC or plan to use it, so I am looking for a proper fix, thanks.
I understand but this is a very wrong approach, this shouldn't be done like this, for example DSDT can be patched directly in the BIOS, I don't need a bootloader to do that for me and add possible other zilion bugs...As OCLP has solved this issue, I don't think any other "proper fix" will be researched in the future.
I understand but this is a very wrong approach, this shouldn't be done like this, for example DSDT can be patched directly in the BIOS, I don't need a bootloader to do that for me and add possible other zilion bugs...
I will look on OCLP code to see how is implemented, rather I will write a simple efi driver(if is not possible to do it in the ACPI) and put it into the BIOS to solve this then use a bootloader.
The correct approach would be reading the first post, again!I understand but this is a very wrong approach, this shouldn't be done like this, for example DSDT can be patched directly in the BIOS, I don't need a bootloader to do that for me and add possible other zilion bugs...
I will look on OCLP code to see how is implemented, rather I will write a simple efi driver(if is not possible to do it in the ACPI) and put it into the BIOS to solve this then use a bootloader.
You really have a fetish with the first post, that is not a proper fix, a proper fix would be at VBIOS level or BIOS level, but as usual instead having a chat with REAL information we lose time pointing to stuff that I saw already.The correct approach would be reading the first post, again!
A so called "proper fix" has been mentioned there half a dozen times, or more. A simple plist modification.