Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Status
The first post of this thread is a WikiPost and can be edited by anyone with the appropiate permissions. Your edits will be public.
Hello, I hear something about a vbios for s7100x made for mac that could output edp?
sorry, I'm not smart enough to know what any of that means. (What is the S7100x? EDP?)

Though I've done a lot of 'flashing' and programming of many things, I've never tried the process(es) outlined here, on the WX4130, in order to install on my old iMac 11,1.

Really just looking for some confidence-building points from a couple of the experts

OK - Have ordered the card now, so am committed! From notes here, looks like 11,1 and WX4130 should work nicely together. Can anyone share any specific mileage?
 
Last edited:
sorry, I'm not smart enough to know what any of that means. (What is the S7100x? EDP?)

Though I've done a lot of 'flashing' and programming of many things, I've never tried the process(es) outlined here, on the WX4130, in order to install on my old iMac 11,1.

Really just looking for some confidence-building points from a couple of the experts

OK - Have ordered the card now, so am committed! From notes here, looks like 11,1 and WX4130 should work nicely together. Can anyone share any specific mileage?
AMD FirePro S7100X is mxm card gpu and edp is embedded displayport (like a displayport) is video output (for what i know) for laptops. I want to ask anyone here which vbios is the one that have video output/edp support. Because I saw a 2 year old post in reddit of about the gpu and someone posted this year, 11 months ago saying a modded vbios for mac have that edp enable. I own that gpu and see 3 files in this website. s7100x_imac, gop-lvtma, and s7100x_6939_imac. I have laptop I'm experimenting with some 2 gpus but currently don't have a screen with edp but lvds screen installed, my main laptop have screen edp installed but I prefer avoid disassemble it if possible to see which is it.
 
AMD FirePro S7100X is mxm card gpu and edp is embedded displayport (like a displayport) is video output (for what i know) for laptops. I want to ask anyone here which vbios is the one that have video output/edp support. Because I saw a 2 year old post in reddit of about the gpu and someone posted this year, 11 months ago saying a modded vbios for mac have that edp enable. I own that gpu and see 3 files in this website. s7100x_imac, gop-lvtma, and s7100x_6939_imac. I have laptop I'm experimenting with some 2 gpus but currently don't have a screen with edp but lvds screen installed, my main laptop have screen edp installed but I prefer avoid disassemble it if possible to see which is it.

Assuming that you don't know how to navigate from post#1..., below is the direct link to S7100x for iMac description.


More in depth information from the expert
 
Hi,

I have 2 iMacs, I want to make some upgrades to the GPU but I have some questions, maybe you can guide me or recommend what to do:


iMac (27-inch, Late 2009)
iMac11,1
NVIDIA Quadro K3100M
High Sierra

Running good, black screen when it goes to sleep and no wake up.
Browsers ask me to update OSX.

Questions:

- Can I update OSX with this GPU? or do I need to upgrade the GPU?
- Which GPU will you recommend to upgrade?



iMac (27-inch, Mid 2011)
iMac12,2
Dead GPU

- Which GPU will you recommend to upgrade so I can used beyond High Sierra?

Thanks in advance...
 
iMac11,3 W5170M have removed (desoldered) the jumper wires in the picture below, still the same old problem of random black screen not displaying, both Windows and MacOS systems are purely in this kind of problem? How do I fix this?
1.png


 
Hi JAY,

If you have heatsink B, you need copper plate on sink as stated on

your card may be overheating or nearly dead…….
your config.plist is patched like mine with correct pciroot for graphic?
 

Attachments

  • Capture d’écran 2025-09-14 à 18.01.56.png
    Capture d’écran 2025-09-14 à 18.01.56.png
    58.4 KB · Views: 20
Hello all, a big problem occurred: Apple released Tahoe. I know this is not the quite the forum for this but it really affects all of us who upgraded the old iMacs. I sold quite a few of those and now some of the buyers upgraded to Tahoe with the result that the can't access the Mac anymore. I don't exactly know what the status of OCLP is regarding TAHOE but apparently it does not work properly. I will re-install Sequoia if possible but most of the buyers are not in the position to bring the iMac back to me.
Do you have any other work-around for this problem?
 
Hello all, a big problem occurred: Apple released Tahoe. I know this is not the quite the forum for this but it really affects all of us who upgraded the old iMacs. I sold quite a few of those and now some of the buyers upgraded to Tahoe with the result that the can't access the Mac anymore. I don't exactly know what the status of OCLP is regarding TAHOE but apparently it does not work properly. I will re-install Sequoia if possible but most of the buyers are not in the position to bring the iMac back to me.
Do you have any other work-around for this problem?
Should be in this topic I think: https://forums.macrumors.com/threads/macos-tahoe-26-on-unsupported-macs-discussion.2458481/
 
  • Like
Reactions: dorninko
OK, trying to follow @Ausdauersportler 's GRML-FLASH instructions to the letter...

[I]If you have a USB device with more capacity format it using FAT/GUID (not ExFAT!),[/I]
[I]I do! just open the image in Finder, and copy all folders over to the new and bigger drive.[/I]

Done! And the newly created USB stick boots into GRML nicely. No prob!

This seems to suggest I might be able to directly copy ROM files into some directory which would then be visible after booting this USB? For example, where would I access the flash/video directory after creating it, in order to place the ROM file? Seems to me the first step in process is flashing the WX4120.

IE, am I able to copy files directly onto the (mac-)mounted GRML filesystem? This would be great, but how/where would these files be visible on the booted GRML disk?

GUID allows you to add OpenCore via OCLP to the EFI partition on the USB to get an emulated boot screen on Macs with PC graphics cards.

OK; have been able to install OCLP to the EFI partition on GRML. But how to get ROM file(s) onto the USB?

Can see it would be doable over the network, but cannot stat WiFi on GRML...

An old Unix guy, got no problem with any of the command-line stuff. But I'm clearly not getting it here... Can someone clarify?
 
Last edited:
OK, trying to follow @Ausdauersportler 's GRML-FLASH instructions to the letter...

[I]If you have a USB device with more capacity format it using FAT/GUID (not ExFAT!),[/I]
[I]I do! just open the image in Finder, and copy all folders over to the new and bigger drive.[/I]

Done! And the newly created USB stick boots into GRML nicely. No prob!

This seems to suggest I might be able to directly copy ROM files into some directory which would then be visible after booting this USB? For example, where would I access the flash/video directory after creating it, in order to place the ROM file? Seems to me the first step in process is flashing the WX4120.

IE, am I able to copy files directly onto the (mac-)mounted GRML filesystem? This would be great, but how/where would these files be visible on the booted GRML disk?

GUID allows you to add OpenCore via OCLP to the EFI partition on the USB to get an emulated boot screen on Macs with PC graphics cards.

OK; have been able to install OCLP to the EFI partition on GRML. But how to get ROM file(s) onto the USB?

Can see it would be doable over the network, but cannot stat WiFi on GRML...

An old Unix guy, got no problem with any of the command-line stuff. But I'm clearly not getting it here... Can someone clarify?

Whenever I use the GRML USB to boot, I have to open a website to get guidance about Linux command.
Not very complicated ones, just simple ones to list the directories, change directory etc.
The GRML USB image file posted inside this threat contain all the necessary files to flash vBIOS, including both flashing program for AMD and nVIDIA GPUs.
I would suggest you surfing through the directories on the USB once you have made it, to know the locations of the files.
I myself often create a new folder, copy necessary files for the GPU I'm about to flash, change their names to a much shorter version so it will take me less time to type in the command.

For an old Unix guy like you, I think you can try to do the auto-run batch files, which auto run after Linux boot.
This is a blind flashing method described in TechPowerUp to reflash GPUs bricked from previous wrong flashing attempts (Overclocked vBIOS etc)
Using the auto-run (autoexec.bat) help you to bypass the remote login procedures, but it is advised to know for sure each and every command in the autoexcec.bat files, as well as the GPU.
 
Last edited:
Hi JAY,

If you have heatsink B, you need copper plate on sink as stated on

your card may be overheating or nearly dead…….
your config.plist is patched like mine with correct pciroot for graphic?
Yes replaced two W5170M a Dell red PCB M5100 also padded with 1MM copper The problem is still a random black screen, can't find the cause.
 
OK, so I have this problem with an m5100 on iMac 11,2 (21.5-inch, Mid 2010) I mentioned in another thread - basically, what I'd get is yellow screen and pointer after Monterey fresh install, after it reboots and starts post-fresh-install stuff (Language/Timezone/Account creation/etc.). Despite kind advice by Ausdauersportler and other people, nothing would work for me until I removed the m5100, put back my old Radeon HD, did the post-fresh-install stuff, rebooted into Monterey post-install, removed the Radeon HD, put back in the m5100 and then it would miraculously work like nothing happened.

So, I am sitting there thinking - is there anything at all I can do to address this? I mean every time I think of a potential re-install, I cringe. Also, I was thinking of trying Ventura (should I even on this HW?), but if it would do the same yellow screen stuff or if it didn't work for me and I wanted to roll back to Monterey, I'd have to do all the removing-repasting-putting-back circus again, which is kind of annoying.

What's more, is I have this OCLP bug (?) which I suspect might be related - no matter what I do, it spits out an error at me when I try to build it in Host Mode (please, see screenshot) it does build only in target mode.

Can it be a wrong VBIOS, for example (I bought the card off Ali with it pre-flashed for iMac)? Should I try to re-flash? Any other thoughts? Thanks!

Screenshot 2025-09-26 at 20.09.01.png
 
Yes replaced two W5170M a Dell red PCB M5100 also padded with 1MM copper The problem is still a random black screen, can't find the cause.
Just my two cents. Do I get it right that 3 different GPUs show same symptoms? Does it happen only in Win/macOS or, say if you were to boot Linux or WinPE from USB? Does the system freeze entirely or is it just the image? If it is everywhere - can it be your power supply, or LCD?
 
Just my two cents. Do I get it right that 3 different GPUs show same symptoms? Does it happen only in Win/macOS or, say if you were to boot Linux or WinPE from USB? Does the system freeze entirely or is it just the image? If it is everywhere - can it be your power supply, or LCD?
Two w5170m a m5100 are randomly black screen, black screen when the key can still hear the system feedback sound, if it is in the playback of the black screen, the system is running normally can also hear the video sound.

This only happens in Windows and macos.



Switching back to the previous k2100m everything is fine.
 
Two w5170m a m5100 are randomly black screen, black screen when the key can still hear the system feedback sound, if it is in the playback of the black screen, the system is running normally can also hear the video sound.

This only happens in Windows and macos.



Switching back to the previous k2100m everything is fine.
Scrap the power supply (internets say m5100 has slightly lower power draw than k1200m) and LCD ideas then.
 
OK, so I have this problem with an m5100 on iMac 11,2 (21.5-inch, Mid 2010) I mentioned in another thread - basically, what I'd get is yellow screen and pointer after Monterey fresh install, after it reboots and starts post-fresh-install stuff (Language/Timezone/Account creation/etc.). Despite kind advice by Ausdauersportler and other people, nothing would work for me until I removed the m5100, put back my old Radeon HD, did the post-fresh-install stuff, rebooted into Monterey post-install, removed the Radeon HD, put back in the m5100 and then it would miraculously work like nothing happened.

So, I am sitting there thinking - is there anything at all I can do to address this? I mean every time I think of a potential re-install, I cringe. Also, I was thinking of trying Ventura (should I even on this HW?), but if it would do the same yellow screen stuff or if it didn't work for me and I wanted to roll back to Monterey, I'd have to do all the removing-repasting-putting-back circus again, which is kind of annoying.

What's more, is I have this OCLP bug (?) which I suspect might be related - no matter what I do, it spits out an error at me when I try to build it in Host Mode (please, see screenshot) it does build only in target mode.

Can it be a wrong VBIOS, for example (I bought the card off Ali with it pre-flashed for iMac)? Should I try to re-flash? Any other thoughts? Thanks!

View attachment 2559345
hi,
you can create an opencore usb stick with your now working config.plist on it and startup on usb with ALT key ,after you can mount your hard drive the way you prefer and copy file to new installation.
clover configurator 2 can mount drive without using terminal .
.i had yellow screen and entered password on that screen ,and started OS...
 

Attachments

  • Capture d’écran 2025-09-27 à 19.47.59.png
    Capture d’écran 2025-09-27 à 19.47.59.png
    344.5 KB · Views: 7
Last edited:
  • Like
Reactions: CthulhuSaur
hi,
you can create an opencore usb stick with your now working config.plist on it and startup on usb with ALT key ,after you can mount your hard drive the way you prefer and copy file to new installation.
clover configurator 2 can mount drive without using terminal .
.i had yellow screen and entered password on that screen ,and started OS...
Thank you, this is a good idea, I will try that, of course (nota bene, using Terminal is not a problem since you can run it after booting from install USB while in pre-install environment). The problem is - I am not absolutely sure my current config.plist is 'working'. 1. As mentioned in this other thread, making at least basic changes to the same config.plist (adding radpg=15 to the boot-args) changed nothing with the yellow screen and, at the same time, the system started fine (including login window) after post-installation stuff without any changes to config.plist. 2. On my currently flawlessly working machine under Monterey, OCLP cannot build itself properly in Host Mode, giving an error, which makes me suspect something must be 'wrong' about my 'host' configuration (how I read it is it can build generic 11,2 config in target mode, but gets confused about mine being 'fully 11,2' or something, although it absolutely is, save for the new m5100).
 
Whenever I use the GRML USB to boot, I have to open a website to get guidance about Linux command.
Not very complicated ones, just simple ones to list the directories, change directory etc.
The GRML USB image file posted inside this threat contain all the necessary files to flash vBIOS, including both flashing program for AMD and nVIDIA GPUs.
I would suggest you surfing through the directories on the USB once you have made it, to know the locations of the files.
I myself often create a new folder, copy necessary files for the GPU I'm about to flash, change their names to a much shorter version so it will take me less time to type in the command.

For an old Unix guy like you, I think you can try to do the auto-run batch files, which auto run after Linux boot.
This is a blind flashing method described in TechPowerUp to reflash GPUs bricked from previous wrong flashing attempts (Overclocked vBIOS etc)
Using the auto-run (autoexec.bat) help you to bypass the remote login procedures, but it is advised to know for sure each and every command in the autoexcec.bat files, as well as the GPU.
OK, thanks to various folks here for their comments... Think I've managed to figure out what's going on here!
Having no experience with GRML - and what a great tool this is! - I didn't really understand its model... (Yes, extensive UNIX and Linux experienced helped a great deal.) However, notes and suggestions all over the map here make this hard to follow for the newcomer.

At some point, I followed a link to get the unmodified GRML from its own website, so I was ultimately not using @Ausdauersportler's pre-built image. I was relying on suggestions everything could be simply copied onto the filesystem directly - well, this doesn't quite work...

Once I figured out how to make my own custom bootable image of GRML on USB - including the persistence flag, it became clear how to place files into the persistent filesystem - I used SCP from another machine... The nonetwork option could also be useful, but my setup works fine with a DHCP-assigned address on a 'hard' TCP/IP line. I've also been able to copy current versions of all the tools required. Still not entirely sure which GPU ROM I should install, but I'll do more research. Open to all strong opinions!

This is all sorta outlined in @Ausdauersportler's notes, but it was not terribly clear to me.

I haven't yet flashed my WX4130; can't do all this this week. Plus, I have to source copper shim and thermal paste. Hopefully soon!
 
Last edited:
Thank you, this is a good idea, I will try that, of course (nota bene, using Terminal is not a problem since you can run it after booting from install USB while in pre-install environment). The problem is - I am not absolutely sure my current config.plist is 'working'. 1. As mentioned in this other thread, making at least basic changes to the same config.plist (adding radpg=15 to the boot-args) changed nothing with the yellow screen and, at the same time, the system started fine (including login window) after post-installation stuff without any changes to config.plist. 2. On my currently flawlessly working machine under Monterey, OCLP cannot build itself properly in Host Mode, giving an error, which makes me suspect something must be 'wrong' about my 'host' configuration (how I read it is it can build generic 11,2 config in target mode, but gets confused about mine being 'fully 11,2' or something, although it absolutely is, save for the new m5100).
hi ,
upload your config.plist to see how it is ,
your pciroot is also PciRoot(0x0)/Pci(0x1,0x0)/Pci(0x0,0x0)

that s why you have an error in opencore...buildup
plisteditpro is nice editor also free trial
 
Last edited:
hi ,
upload your config.plist to see how it is ,
your pciroot is also PciRoot(0x0)/Pci(0x1,0x0)/Pci(0x0,0x0)

that s why you have an error in opencore...buildup
plisteditpro is nice editor also free trial
Not really!

The iMac11,2 can have two different PCI path. The OCLP is perfectly designed to handle this really special case unless a user decides to mess with settings. Look up the documentation, learn how to unset all user changes and let the software run on the target machine to get a valid and working configuration. There is absolutely no need to make manual changes to the configuration.

On the target system the OLCP software will auto detect both the correct path and the installed graphics card type.

When running on a different Mac the only settings to touch and alter are the target (iMac11,2) and the MXM graphics card type (GCN). It will create a special configuration with entries for both PCI path variants of the GPU.
 
Last edited:
Well, either it was some kind of a Monterey-specific bug all along, or I did something right this time, but I took a leap of faith with a safety net, OCLP-built myself a Ventura installer, installed on a separate APFS volume as a dual-boot, it booted into post-fresh-install stuff (Language/Timezone/Account creation/etc.), did not give me any yellow screens or any other funny business (although pre-root-patching acceleration was definitely off - you could tell because it was one of a hell of a 'turn-based strategy'). I completed the install, did post-root-patching, got my graphics acceleration and (now properly working and offering password prompt) Wi-Fi. Then I deleted the Monterey-containing APFS Volume and am back at single-boot Ventura. I like the way the system has been performing, so I am going to leave it at that. Thank you, guys, again!

NB: Just for the record: iMac mid-2010 (21,5 inch, 11,2, A1311), with Intel Core i5 mod (i5-680), red Firepro M5100 mod, Samsung SSD 860 PRO 256GB mod, 12 GB of RAM - Ventura 13.7.8 (22H730) via OCLP 2.4.1 - Can recommend (subjectively) works almost as good as High Sierra.

P.S. Because I initially freaked out at the yellow screen, thought that I fried my GPU, and ordered another one, I now have two perfectly working m5100's., one laying around unused. If anyone in the EU needs one and would like to take it off me - send me a PM please, we'll work it out.
 
Last edited:
  • Like
Reactions: Nguyen Duc Hieu
Is there any traction currently on upgrading past Monterey on a Mid 2011 27" (12,2) with the AMD RX5500XT cards? Currently they show unsupported for the Navi 10 cards on OCLP with no known date for when they will be added, but, this is the best card available for these cards in terms of pure power and capability (im sure the W7100 is good, but good luck ever finding one for sale now a days..).

Any news on this front?
 
As far as this thread and other projects have stated that card will not work on a Mid 2011 Imac 12,2 though, So. Best card we can get is the 5500/6600 and the most robust for our model is 5500.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.