Is there any info on whether these cards work with Windows 10 in BootCamp with backlight and brightness control? It would be helpful if that info was added to the first post and/or PDF to help clarify.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
iM 4,1-14,4 2011 iMac Graphics Card Upgrade
- Thread starter MichaelDT
- WikiPost WikiPost
- Start date
- Sort by reaction score
- Status
Ah so that's UEFI Windows 10, not BootCamp. First post says k2100m and k1100m support UEFI Windows 10 with backlight. Not clear about BootCamp. I guess if backlight control doesn't work with BootCamp I can just reinstall Windows with UEFI. Should the first post be more clear whether backlight control works with BootCamp or not?
I would assume you get backlight control using the bootcamp drivers. If not try UEFI. We can only document what others checked before and posted here. There is no magic entity overcoming the combinatorial explosion of checking n OS versions, in m iMacs using k different cards and this for each OS update, again and again. This thread solely depends on positive posts of members. Search the thread or check it out yourself! The history is seven years. You cannot expect to find every poster active here to answer your special questions. And post #1 is no support list, there is at best some sort of community support if somebody is sharing your config. If not, you are the explorer. We need explorers to get ahead!
Our main concern was and is the usage of recent MacOS versions.
Hope that some of the (legacy) Windows users can comment on your question!
Greetings to everyone! I'm a newcomer on the forum, however I have been reading this thread for a looong time 
. I got a little problem, recently I bought an GTX770m on aliexpress (it suppose to be a Dell/Alienware) but I discover the brand (and the bios) are from MSI.
Here is my first problem, iMac starts with win10, with black screen but ok I can access through RDP and TeamViewer, but I can't flash VGA with nvflash (I'm stuck on id mismatch and Mismatch in 256 bytes ...), I tried several nvflash versions and options -4 -5 -6 -f ... and always the same result.
Also, tried to flash bios from dell and Alienware not just @Nick [D]vB bios. Could be a problem with nvflash on windows? I'm getting stuck with these... I tried to install WSL but does not run nvflash on 32bits.
Any ideas? Thanks so much!
. I got a little problem, recently I bought an GTX770m on aliexpress (it suppose to be a Dell/Alienware) but I discover the brand (and the bios) are from MSI.Here is my first problem, iMac starts with win10, with black screen but ok I can access through RDP and TeamViewer, but I can't flash VGA with nvflash (I'm stuck on id mismatch and Mismatch in 256 bytes ...), I tried several nvflash versions and options -4 -5 -6 -f ... and always the same result.
Also, tried to flash bios from dell and Alienware not just @Nick [D]vB bios. Could be a problem with nvflash on windows? I'm getting stuck with these... I tried to install WSL but does not run nvflash on 32bits.
Any ideas? Thanks so much!
The other option is to try the Linux USB stick described on post #1, too.Greetings to everyone! I'm a newcomer on the forum, however I have been reading this thread for a looong time
Here is my first problem, iMac starts with win10, with black screen but ok I can access through RDP and TeamViewer, but I can't flash VGA with nvflash (I'm stuck on id mismatch and Mismatch in 256 bytes ...), I tried several nvflash versions and options -4 -5 -6 -f ... and always the same result.
Also, tried to flash bios from dell and Alienware not just @Nick [D]vB bios. Could be a problem with nvflash on windows? I'm getting stuck with these... I tried to install WSL but does not run nvflash on 32bits.
Any ideas? Thanks so much!
You are sure to have a 770M (read the chip type from the GPU)? Unless you can use a clip you need to make one of the nvflash versions work. I do all Nvidia and most AMD flash operations using this tool.
Yes, the model is correct a geniune GTX770m Kepler arch (check with GPU-Z). OK I'm going to try with linux...but a little question, if don't have bootscreen how can I get into USB boot? Maybe bootcamp (from win) would let me select an USB as boot? Thanks!
You're right, all my questions were there... Sorry! I'm just getting mad last 24h dealing with my iMac.Try - at least try - to read the post regarding the Linux USB stick linked on page one! The docs are created for a reason
I understand. In principle no single step is really complicated, but the mass of tasks to manage makes the difference at the end. As I wrote some times before: Make a short plan covering all topics, start with a supported OS like High Sierra (even there we face now the black screen), try all (and in particular the flash) tools in dry runs before and then start the real operation. And nobody can (unfortunately) guarantee that you bought a working card, at all. Having an external display helps a lot, I got tired of reinstalling the internal LCD...
Hello
My imac 27 2011 works like a charm upgrade K2100m and BT4.0/Wifi ac > catalina 10,15,5 + Opencore
I got another one, and two graphic cards another K2100m and a k1100m
I both flash them with the Nick vbios but.....
Chime Ok, Black screen, 2/4 leds on from the left on the "new" iMac, it seems son video cards found
Try to test the cards in the functional iMac same results
Any idea?
Thx
My imac 27 2011 works like a charm upgrade K2100m and BT4.0/Wifi ac > catalina 10,15,5 + Opencore
I got another one, and two graphic cards another K2100m and a k1100m
I both flash them with the Nick vbios but.....
Chime Ok, Black screen, 2/4 leds on from the left on the "new" iMac, it seems son video cards found
Try to test the cards in the functional iMac same results
Any idea?
Thx
Would be nice if you add a signature with the model you are working on - I assume it is a 2011, too?Hello
My imac 27 2011 works like a charm upgrade K2100m and BT4.0/Wifi ac > catalina 10,15,5 + Opencore
I got another one, and two graphic cards another K2100m and a k1100m
I both flash them with the Nick vbios but.....
Chime Ok, Black screen, 2/4 leds on from the left on the "new" iMac, it seems son video cards found
Try to test the cards in the functional iMac same results
Any idea?
Thx
If you have an external display, connect it! Then you fence out problems with the LCD panel and cables and inverter board....
Nvidia:
Using this Kepler cards you can simply try to press just alt/option on boot. You should see a boot selector screen. Card is working then.
AMD:
Use the OpenCore SD card and wait if you can get the OpenCore to show you the "boot screen" on the external display.
Check the card installation on the heat sink, may be that the boot into graphics mode fails - then the card is most likely damaged or less likely badly installed.
First 2 LEDs on mean the card has been recognised and in 99% of my cases I got the card working then. 3rd. LED means communication of card's vBIOS and system is fine (only Kepler and AMD with boot ROM mod and original cards), 4th LED means the internal LCD has been recognised...
Check the vBIOS version, again?
After following Linux Guide... no luck, same error. Maybe MSI build is not compatible...
Last console output (tried with -f without and -4):
root@grml /lib/live/mount/medium/flash # ./nvflash_linux -f -6 ./Kepler_Beta3/770M_EG2.rom
NVIDIA Firmware Update Utility (Version 5.414.0)
Simplified Version For OEM Only
Checking for matches between display adapter(s) and image(s)...
Adapter: GeForce GTX 770M (10DE,11E0,1462,10E8) H:--:NRM S:00,B:01,D:00,F:00
WARNING: Firmware image PCI Subsystem ID (1028.05AA)
does not match adapter PCI Subsystem ID (1462.10E8).
Please press 'y' to confirm override of PCI Subsystem ID's: y
Overriding PCI subsystem ID mismatch
Current - Version:80.06.54.00.11 ID:10DE:11E0:1462:10E8
GK106 Board - 23031e00 (Normal Board)
Replace with - Version
E.AD.00.BE.EF ID:10DE:11E0:1028:05AAGeForce GTX 770M by Nick[D]vB (Normal Board)
Update display adapter firmware?
Press 'y' to confirm (any other key to abort): y
Identifying EEPROM...
EEPROM ID (7F,9D22) : PMC Pm25LD020 2.7-3.6V 2048Kx1S, page
Storing updated firmware image...
EEPROM programming failed.
NOTE: Exception caught.
Nothing changed!
ERROR: Mismatch in 256 byte page-mode programming
This is so frustrating...
Did you follow the guide, i.e. first protectoff and than flash using -6?
Probably you get better help on other more specific threads or even on different sites? Here we are mostly nvflash users, less the trouble shooting experts...I am a user
Signature updatedI already wrote my signature
in my functional imac the 4 leds are on with a k2100m with vbios flashed Nick
Did you follow the guide, i.e. first protectoff and than flash using -6?
Probably you get better help on other more specific threads or even on different sites? Here we are mostly nvflash users, less the trouble shooting experts...I am a user
Yes, indeed. I'm wondering if the problem is the EPROM size, the original firmware it's only 99840 bytes and the mod 230912bytes.
I've been googling since last 24h and several times this error appears on the forum with no solution. I hope to find someone with an MSI working. Maybe @Nick [D]vB would know the answer...
Thanks btw!
I guess, we are coming closer to your issue. The files supplied for Nvidia cards here have the full eeprom chip size of 262.144 bytes. Your file has been corrupted somehow. Please download, again!
All nvflash version read only the non zero bytes of the installed vBIOS, while @Nick [D]vB and others supplied padded files to be used with the clip programmer, too. If you read the installed vBIOS version using nvflash, you get a smaller file. If you want to write this file using the clip you have to make in larger fitting to the chip size. There is a one line unix command to get this done:
Code:
truncate -s 262144 smallbiosfile.romtruncate is not part of macOS, but you can get it installed using brew....
Last edited:
Bonjour,
I am quoting myself because it does not fascinate many people since the subject is Big Sur.
Bug or reality?
Bonne journée / soirée.
I am quoting myself because it does not fascinate many people since the subject is Big Sur.
Bug or reality?
Bonne journée / soirée.
Hi everyone,
Like many others, I didn't read carefully enough and ordered one of the non-functioning HP ZBOOK WX4170 L30654-001 916613-002 cards. This card does not come with a v-bios, and as some have mentioned, the circuit is slightly different around the vbios area.
I want to attempt to get this working, and in doing so - contribute my bit to the zeitgeist of this fantastic thread.
I have studied the circuit around the vBios pads on my card, and have cross-referenced it with photos of the working WX4170 cards. It appears that the cards both have two large smd caps next to the vbios, and then three smaller smd caps in the area too. With the non-working cards the caps are differently positioned, but I am hoping they are connected to the same places.
The real difference comes from a missing R1342 resistor on the non-working cards and the addition of Q6, a 6679 mosfet. One side of R1342 connects to pin 8 of the vbios, which is the VCC (power supply) pin. I suspect that this simply needs to be jumpered or better yet - have a 0 ohm resistor soldered in place. What i'm less sure about is this mosfet, what it is doing and whether it should be here at all in the presence of a vbios. I notice from photos of the working cards that there is a small black component where this mosfet would be, it's extremely hard to tell from the photos but I think this component is called U2.
Furthermore, there are some strange shorts on the pcb of the non-working card which surprise me. Pins 8,7 and 3 of the vBios pads are shorted together by the PCB. These pins are the VCC pin, the HOLD# (IO3) pin and the WP# (IO2) pin respectively. Studying the schematic of the GD25Q40CTIGR vBIOS chip here - I cannot fathom why these pins would be shorted together.
So... I need someone with a working WX4170 to do some checks for me, which I will use to hopefully get the non-working models to function in the iMac - after which I will put together a guide and post it here.
Specifically, I need someone with a working WX4170 to give me the following.
1. ID of the U2 chip (if i am even right that it is called U2)
2. Resistance measurements of U10 - are pins 8, 7 and 3 shorted together on a good board?
3. Does R1342 exist on the good boards? If so - what resistance does it measure?
If anyone can wap out a multimeter and make those checks with their working WX4170 card - I will be eternally grateful. I promise I will do everything I can with the information to try to solve this problem!
Best,
Dave.
Like many others, I didn't read carefully enough and ordered one of the non-functioning HP ZBOOK WX4170 L30654-001 916613-002 cards. This card does not come with a v-bios, and as some have mentioned, the circuit is slightly different around the vbios area.
I want to attempt to get this working, and in doing so - contribute my bit to the zeitgeist of this fantastic thread.
I have studied the circuit around the vBios pads on my card, and have cross-referenced it with photos of the working WX4170 cards. It appears that the cards both have two large smd caps next to the vbios, and then three smaller smd caps in the area too. With the non-working cards the caps are differently positioned, but I am hoping they are connected to the same places.
The real difference comes from a missing R1342 resistor on the non-working cards and the addition of Q6, a 6679 mosfet. One side of R1342 connects to pin 8 of the vbios, which is the VCC (power supply) pin. I suspect that this simply needs to be jumpered or better yet - have a 0 ohm resistor soldered in place. What i'm less sure about is this mosfet, what it is doing and whether it should be here at all in the presence of a vbios. I notice from photos of the working cards that there is a small black component where this mosfet would be, it's extremely hard to tell from the photos but I think this component is called U2.
Furthermore, there are some strange shorts on the pcb of the non-working card which surprise me. Pins 8,7 and 3 of the vBios pads are shorted together by the PCB. These pins are the VCC pin, the HOLD# (IO3) pin and the WP# (IO2) pin respectively. Studying the schematic of the GD25Q40CTIGR vBIOS chip here - I cannot fathom why these pins would be shorted together.
So... I need someone with a working WX4170 to do some checks for me, which I will use to hopefully get the non-working models to function in the iMac - after which I will put together a guide and post it here.
Specifically, I need someone with a working WX4170 to give me the following.
1. ID of the U2 chip (if i am even right that it is called U2)
2. Resistance measurements of U10 - are pins 8, 7 and 3 shorted together on a good board?
3. Does R1342 exist on the good boards? If so - what resistance does it measure?
If anyone can wap out a multimeter and make those checks with their working WX4170 card - I will be eternally grateful. I promise I will do everything I can with the information to try to solve this problem!
Best,
Dave.
Here right now:
K2100M, Catalina 10.15.5, 2692 Geekbench Metal - something went wrong with your run
K2100M, Catalina 10.15.5, 2692 Geekbench Metal - something went wrong with your run
Sorry, my fault. Read your post on the iPhone - it is hiding the signature.I already wrote my signature
Hi,Here right now:
K2100M, Catalina 10.15.5, 2692 Geekbench Metal - something went wrong with your run
I have a very clean installation (well almost since I had my TimeMachine backup) I do not understand what I missed, if you have an idea? 😅
Thank you! 😉
It is alive...
man i feel like Frankenstein. So the HP Firmware from Powertechs did the trick and i can confirm the K3000M is working. Valley test will follow.
Valley: It only sees 256mb.
man i feel like Frankenstein. So the HP Firmware from Powertechs did the trick and i can confirm the K3000M is working. Valley test will follow.
Valley: It only sees 256mb.
Last edited: