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.
Ok, so I'm starting to experiments to oberseve the behavior of different BIOSes for the GTX880m in macOS and Windows on my 12,2 iMac.

A gave a first try to a stock DELL Bios:
+ has instant login screen on Windows
+ shows Displayport Audio output in macOS
- does not fix sleep in windows and macOS, tried macOS with iGPU enabled & disabled
- of cause, does not support Bootscreen and so on

...i will try more BIOSes if i find some...

Several question coming to my mind:

  1. Regarding the sleep/wake issue in both OSes: I wonder if the CPU upgrade to i7-2600s may cause any problems as the machine shipped with an i5 originally. I will probably switch back the CPU just to rule out any potential source of problems.
  2. What was the initial source BIOS used for the macOS tweaks, could the same tweaks applied to another BIOS bring support for BootScreen without braking displayport sound and causing the long windows login screen delay
Thanks for sharing your thoughts...
J
 
Ok, so I'm starting to experiments to oberseve the behavior of different BIOSes for the GTX880m in macOS and Windows on my 12,2 iMac.

A gave a first try to a stock DELL Bios:
+ has instant login screen on Windows
+ shows Displayport Audio output in macOS
- does not fix sleep in windows and macOS, tried macOS with iGPU enabled & disabled
- of cause, does not support Bootscreen and so on

...i will try more BIOSes if i find some...

Several question coming to my mind:

  1. Regarding the sleep/wake issue in both OSes: I wonder if the CPU upgrade to i7-2600s may cause any problems as the machine shipped with an i5 originally. I will probably switch back the CPU just to rule out any potential source of problems.
  2. What was the initial source BIOS used for the macOS tweaks, could the same tweaks applied to another BIOS bring support for BootScreen without braking displayport sound and causing the long windows login screen delay
Thanks for sharing your thoughts...
J

I don't think the i7 has relation with the sleep/wake issue, as both share the same iGPU. I agree with @Ausdauersportler that this sleep/wake failure is a hardware problem others have faced before, and does not look like it can be fixed by software/drivers.

Do you have brightness control working with DELL bios ?

It's interesting to know that the stock DELL bios has instant login screen on windows, I wonder, if it has UEFI/GOP driver in bios, if bootscreen can be emulated using OCLP, same way as it is done on AMD cards ??
 
Last edited:
Hello folks!

First I like to thank everyone who made this thread, the introductions and of course the guide which helps us all running "old" but good Machines on newer software! Less electronic trash and more sustainability :)

My first steps I have made with dosdude patcher, than linux usb and flashing my K2100m with Nicks Bios.
Now after a fresh clean reset I tried the OCLP method and im so grateful for this program. It worked like charm!

A little problem which appears under Big Sur 11.5.2 is missing Airplay with my Airport Express 2.

iMac 12,2, K2100m, OCLP latest, Big Sur 11.5.2

Is there any solution at the moment to get back Airplay? Since I can stream over my iPad, its not a big thing but would be - in Germany we say "cherry on the cake" or " icing on top of an already excellent cake" :D

Greetings!
 
My first steps I have made with dosdude patcher, than linux usb and flashing my K2100m with Nicks Bios.
Now after a fresh clean reset I tried the OCLP method and im so grateful for this program. It worked like charm!

A little problem which appears under Big Sur 11.5.2 is missing Airplay with my Airport Express 2.

iMac 12,2, K2100m, OCLP latest, Big Sur 11.5.2

Did the airplay was working before? I think that it's a hardware issue with the version of bt these machines have. I have never personally used airplay on my mid2010 iMac.

Also, we have the same GPU. Did you get to have brightness control from the keyboard keys? It' still not working to me (I don't know if even needed something else to do apart from OCLP. I'm on 11.5.1.
 
AirPlay is working under High Sierra and Catalina on other APFS Volumes.

Yes, I patched the Bios with Nicks Version and after a fresh Setup of supported High Sierra I created EFI with the OCLP and made a Big Sur APFS Volume on another SSD. I stick through the entire OCLP installation guide and it worked. After checking keyboard controls I copied back the EFI like written in the sticky.
 

A mod for HP WX4150 cards​

I'd like to report another successful POST of a modded HP WX4150 card in an iMac12,2 (27' 2011 iMac). I now have two modded HP WX4150 cards happily running in two 27' 2011 machines, reported temperatures are sensible (ie. they are no different from those reported by a machine with a Dell WX4150 built in). I also ran some GPU benchmarks - I did not notice any anomalies after 30 minutes of benchmarking.

On the other hand, while the card did POST without a problem in an iMac11,2 (21.5' 2010 iMac), some sensors reported abnormal values, CPU and ODD fans ran at full speed, and I believe that CPU was throttled. (That's still better than iMac11,2 with an unmodded HP WX4150 card - sometimes it POSTs, sometimes it doesn't, definitely does not POST with a HDD connected.) I replaced it with a Dell WX4150 (from my 27' 2011 machine), which works perfectly. I'm not sure (when) I'll investigate this (iMac11,2 + modded HP WX4150) further.

If your soldering skills are above basic and you have the tools required, please do try this. Here's how I did it:
  • a (at least) 50W soldering iron with precision tip and temperature control (80W+ is better, 50W should be enough for leaded solder)
  • flux (pen)
  • solder wick (to remove excess solder)
  • tweezers
  • magnifying glass (I did not use it, but I should.)
  • a (non-POSTing) HP WX4150 card
  • two 0R 1206 resistors (Buy more. They are really cheap and you will probably destroy one...)
  • 0.5mm (or thinner) solder wire
Two jumpers (the 0R 1206 resistors) are needed between the pads in the orange rectangles under the red arrows (see attached pictures). Those pads had some excess solder on my cards. I removed it with a solder wick to facilitate soldering later. Note that those pads are a part of a (relatively) large copper area, which means that a fair amount of heat will be required to melt the solder. Be very careful not to remove tiny capacitors and other components which are in close proximity. Cover cleared pads with a generous amount of flux. Place the resistor and align it with tweezers. Take your time. Use tweezers (in your left hand) to press down on the middle of the resistor (if it moves, align it again), then pick a small amount of solder with the soldering iron (in your right hand) and solder one pad of the resistor. Remember the 'large copper area'? A certain amount of heat will be required again. (This is really important to avoid cold joints. Believe me...) Again: Be careful not to short the nearby components. If you do, remove the shorts with the solder wick. Repeat for the second pad and second resistor. (Add flux as needed, you can remove excess flux when finished with a toothbrush and isopropyl alcohol.)
I attached some photos of my card before and after the mod.
Made the mod on an HP WX4150 which wasn't working in the 2011 before - it's working perfectly now!
I did it using a "normal" 50W soldering iron, without precision tip, and used two pieces of wire - totally doable that way although not looking that prolific.
Thanks for your investigation and sharing it here!
 
Here my latest GOP (this means you'll need OpenCore in order to have a boot picker) VBIOSes for all AMD Polaris cards, except the WX7100 for now.
Those VBIOSes have been tested for a lot of time now, with the exception of the VRAM configurations, which appeared only recently and were collected by @Ausdauersportler and @edwardgeo. (That means that effectively only the WX4130 and RX480 VBIOSes have been tested in their final state!)
They should allow for proper boosting and throttling, before and after sleep, as well as in Windows. However, boosting and throttling may behave differently after the first time waking your iMac from sleep.
My first impressions are, well: Impressive! I am now testing/using those VBIOSes on 4 machines with WX4150/WX4170 cards. I'll just list my observations (in no specific order):
  • iMac11,2 with a Dell WX4150 (with Samsung K4G80325FB-HC28 RAM): Both WX4150_GOP.rom and WX4150_GOP_ALT_VRAM.rom work, offer the same performance before and after sleep - several sleep cycles actually. (54 MRays / Second average with MetalBench)
  • iMac11,1 with a modded HP WX4150 (with Samsung K4G80325FB-HC28 RAM): Same as with iMac11,2 above.
  • iMac12,2 with a modded HP WX4170 (originally EEPROM-less, with Hynix H5GQ8H24MJR-R4C RAM): The WX4170_GOP.rom does not work. (It doesn't fire up the display to show OpenCore boot picker, nor does it load into macOS.) WX4170_GOP_ALT_VRAM.rom however has no issues! It offers the same performance before and after sleep. (74 MRays / Second average with MetalBench)
  • iMac12,2 with a WX4170 (originally with an EEPROM, with Samsung K4G80325FB-HC28 RAM): Neither of WX4170 ROMs work - they both give the OC boot picker, both load into macOS, but the machine just freezes a few seconds into benchmarking - I can still see and move the mouse pointer, but GUI is unresponsive to either mouse clicks or keyboard presses. However, it works perfectly with either of WX4150 ROMs - no sleep issues, but slightly lower performance. (65 MRays / Second average with MetalBench)
This is the most stable performance I was able to achieve with that last WX4170 so far. It appears that that card does not tolerate higher Core clocks well. (Although it offered stellar performance with @edwardgeo 's VBIOS before sleep.) I don't think there's anything wrong with the card, maybe those very early (2016) editions just don't support those higher Core clocks. (Another observation: That card invariably reports higher Power consumption when compared to the modded WX4170: it was 115W vs. 77W with @edwardgeo 's VBIOS, with WX4170_GOP_ALT_VRAM.rom it is 58W vs. 40W before it freezes. If that means anything...)
I'll soon let those machines sleep and I'll test the performance again tomorrow. I'm pretty sure there will be no issues. Also, tomorrow I'll replace that 'problematic' WX4170 with another modded WX4170 (with Samsung K4G80325FB-HC25 RAM) for a test. I'm positive it will perform as well as the one with Hynix RAM.
 
Last edited:
Hello folks!

First I like to thank everyone who made this thread, the introductions and of course the guide which helps us all running "old" but good Machines on newer software! Less electronic trash and more sustainability :)

My first steps I have made with dosdude patcher, than linux usb and flashing my K2100m with Nicks Bios.
Now after a fresh clean reset I tried the OCLP method and im so grateful for this program. It worked like charm!

A little problem which appears under Big Sur 11.5.2 is missing Airplay with my Airport Express 2.

iMac 12,2, K2100m, OCLP latest, Big Sur 11.5.2

Is there any solution at the moment to get back Airplay? Since I can stream over my iPad, its not a big thing but would be - in Germany we say "cherry on the cake" or " icing on top of an already excellent cake" :D

Greetings!
Airplay on the iMac12.x depends on the SandyBridge iGPU (called HD3000) and the AppleGVA.framework from High Sierra copied back to Mohave and Catalina.

This approach did not work on Big Sur. So we disabled using OpenCore the HD3000 entirely to avoid other technical problems.
 
BTW, one nasty side effect of these hard shutdowns on my iMac / 780M when stressing the card is that sometimes it won't POST again until I remove the card and make 3 PRAM resets without any card installed.

That's one of the issues I'm having with my 12,2 with a K3100m. if I do a "restart" from the apple menu the machine starts up nice and good, but if I "shut down" or the power goes off POSTing is kind of hit-and miss. I've found that (1) unplugging it and then plugging it in, (2) leaving the power plugged in for ~10 minutes or so and (3) pushing the power button a couple of dozen times will eventually get it to POST and then boot. I've done PRAM resets (with the card in, I'm not into taking this thing apart just to reset the PRAM) to no avail.

I don't allow the system to go to sleep as that also puts it in a state where it has to POST. I just put the display to sleep and live with the power being on all the time.

The machine otherwise works fine, benchmarks are what you would expect them to be. I just got my hands on a cheap 11,3 and I'm going to try the card in that one just to see if it's the card or the box, when I get a free minute that is.
 
Hi, everyone! I have an iMac 11,1. I put installed a gtx780m. At overview graphics I had: GeForce GTX 780M by Nick(D)vb 4096 MB. I had no bootscreen neither brigthness control but everything else worked good. I tried to install a patched catalina, i dissabled SIP and when it restarted it never booted up properly. only chime and black screen. Is there anything I can do? Thanks in advance! :(
 
(Snipped…)
-----Just some quick questions about this:
  • Are those thin- or thick-film resistors, if that matters? I went looking around for them and saw at least the 0R 0603 SMD resistor provided as both types.
  • Might you have any particular sources you'd recommend for parts?
  • My WX 4170 has a sheet of black (plastic?) shielding covering its entire back side (presumably for heat distribution?) How would one get it off to work behind it, then put it (or a replacement for it) back on afterwards, at least hypothetically?
(This is for the HP ZBook 17 G6 I've been trying to turn into a Hackintosh for more than a little while now. I may well want to find someone else to do the actual soldering, though; this level of job may be beyond me…)
 
  • Like
Reactions: edwardgeo
My first impressions are, well: Impressive! I am now testing/using those VBIOSes on 4 machines with WX4150/WX4170 cards. I'll just list my observations (in no specific order):
  • iMac11,2 with a Dell WX4150 (with Samsung K4G80325FB-HC28 RAM): Both WX4150_GOP.rom and WX4150_GOP_ALT_VRAM.rom work, offer the same performance before and after sleep - several sleep cycles actually. (54 MRays / Second average with MetalBench)
  • iMac11,1 with a modded HP WX4150 (with Samsung K4G80325FB-HC28 RAM): Same as with iMac11,2 above.
  • iMac12,2 with a modded HP WX4170 (originally EEPROM-less, with Hynix H5GQ8H24MJR-R4C RAM): The WX4170_GOP.rom does not work. (It doesn't fire up the display to show OpenCore boot picker, nor does it load into macOS.) WX4170_GOP_ALT_VRAM.rom however has no issues! It offers the same performance before and after sleep. (74 MRays / Second average with MetalBench)
  • iMac12,2 with a WX4170 (originally with an EEPROM, with Samsung K4G80325FB-HC28 RAM): Neither of WX4170 ROMs work - they both give the OC boot picker, both load into macOS, but the machine just freezes a few seconds into benchmarking - I can still see and move the mouse pointer, but GUI is unresponsive to either mouse clicks or keyboard presses. However, it works perfectly with either of WX4150 ROMs - no sleep issues, but slightly lower performance. (65 MRays / Second average with MetalBench)
This is the most stable performance I was able to achieve with that last WX4170 so far. It appears that that card does not tolerate higher Core clocks well. (Although it offered stellar performance with @edwardgeo 's VBIOS before sleep.) I don't think there's anything wrong with the card, maybe those very early (2016) editions just don't support those higher Core clocks. (Another observation: That card invariably reports higher Power consumption when compared to the modded WX4170: it was 115W vs. 77W with @edwardgeo 's VBIOS, with WX4170_GOP_ALT_VRAM.rom it is 58W vs. 40W before it freezes. If that means anything...)
I'll soon let those machines sleep and I'll test the performance again tomorrow. I'm pretty sure there will be no issues. Also, tomorrow I'll replace that 'problematic' WX4170 with another modded WX4170 (with Samsung K4G80325FB-HC25 RAM) for a test. I'm positive it will perform as well as the one with Hynix RAM.
Great! What is a good source for this card?
 
So I did further testing and comparison between the stock DELL Bios and the Nickey Tuned Bios.

I realized that the DELL is way slower but runs cooler which may be nice for non gamers.

DELL
+ instant driver load on windows
+ DP audio
- way slower, around 989 OpenGL
- no Bootscreen/Brightness Control

880M_BR.rom
+ way faster, around 1800 OpenGL
- driver load takes forever on Windows (around 30 seconds+)
- no DP audio

About the very long time until driver initializes in windows: can this be fixed in the 880M_BR BIOS file? What could be the cause? Does it only affect some card or it is a general problem?
 
Does anyone ever experience a chime & black screen and fans going up to „11“ with the 12,2? I have another machine showing this behavior and have not opened it up yet because i‘m in a dispute with the seller who claimed it was working before shipping.
I remember experiencing this once on my working machine but it was gone after some reassembling…
 
Ha ha, my question to "stupid" you then would be - why do you want to do so ? :p😜🤪
firstly, a BIOS chip is not expensive. if you want to change the physical chip, why recycle an old one?
Secondly, if you are aiming for its vBIOS content, it is specific to the GPU. You cannot graft one to another of different type.
Thirdly, if it is meant for two same type GPUs, why not just flash the BIOS chip in the old one instead of pursuing the difficult task of chip desoldering and soldering?
1. i have the old card and with that the old bios chip
2. i dont have to wait for a new chip to come in from china
3. i dont have to research what eeprom will work with my programmer and the new video card
4. i have all the tools and skills and knowledge to do this, only question is, would it work?
 
-----Just some quick questions about this:
  • Are those thin- or thick-film resistors, if that matters? I went looking around for them and saw at least the 0R 0603 SMD resistor provided as both types.
  • Might you have any particular sources you'd recommend for parts?
  • My WX 4170 has a sheet of black (plastic?) shielding covering its entire back side (presumably for heat distribution?) How would one get it off to work behind it, then put it (or a replacement for it) back on afterwards, at least hypothetically?
(This is for the HP ZBook 17 G6 I've been trying to turn into a Hackintosh for more than a little while now. I may well want to find someone else to do the actual soldering, though; this level of job may be beyond me…)
  • The kind of those two 0R resistors is irrelevant. Even bare solid wire would do, but I still somewhat refuse to recommend it. Moreover, those two resistors are needed only to make the card POST in an iMac12,2. If you will be using the card in a HP laptop, you can (probably) omit them.
  • I buy parts at a local electronic parts store. (You can DM me about the EEPROM if you're from EU - my usual sources did not have any.)
  • You can remove that black plastic sheet - it's for insulation. The card will function just as fine without it. (Of course, if this insulation is needed in your particular 'mounting scenario', just reattach it with some kapton tape when finished.)
I'm not sure if adding an EEPROM + RP1 is enough to force that HP laptop to read VBIOS from the card instead of the laptop BIOS. (If I understand correctly, this is how HP laptops load VBIOS information.) I think @edwardgeo knows the answer to this.
 
  • Like
Reactions: RandomDSdevel
Just for your info:
The original (unmodded) HP WX4150 was recognized and functioned well in mini computer HP 800G1 USDT.
Even after flashing modded GOP vBIOS for iMac, it still functions well in windows environment.
Which gop was used? Yesterday I build g4 and g5 zbook gops and they are working fine, but same procedure with iMac gop making it unbootable.
 
  • Like
Reactions: Nguyen Duc Hieu
Which gop was used? Yesterday I build g4 and g5 zbook gops and they are working fine, but same procedure with iMac gop making it unbootable.

I used the vBIOS rom from post#1.
Both AMD_GOP (by Nickdvb) and Dell RX560 (internetzen) roms are working fine in my HP 800G1 USDT.
I haven't tried the stock vBIOS rom of HP WX4150 on iMac. I flashed it immediately after purchasing.


 
  • Like
Reactions: edwardgeo
@nikey22 , @internetzel, @Ausdauersportler, I have a few questions:
- The iMac have CoreFG2 aka co-gop program driver. How this one is different to MacPro driver?
- the iMac 2017 contains Elsemer (can’t remember) and BaffinM output drivers . Creating gops from them prevents iMac from POST, even if they are compressed as pci 2.3... the other problem that they are using supported device list and patching have no results. MacBook Pro 13.3 have supported gops for rx4xx series.
- yesterday I build two different gops for my zbook and they are working, so something in iMac prevent them to execute..

Any ideas?

Reverse engineering of iMac 6770m decompressed gop and BaffinM gops shows that they are loading framebuffer too and all additional info like efirom
 
I used the vBIOS rom from post#1.
Both AMD_GOP (by Nickdvb) and Dell RX560 (internetzen) roms are working fine in my HP 800G1 USDT.
I haven't tried the stock vBIOS rom of HP WX4150 on iMac. I flashed it immediately after purchasing.


If you’re interested in coop, welcome to my thread for HP,

 
@nikey22 , @internetzel, @Ausdauersportler, I have a few questions:
- The iMac have CoreFG2 aka co-gop program driver. How this one is different to MacPro driver?
- the iMac 2017 contains Elsemer (can’t remember) and BaffinM output drivers . Creating gops from them prevents iMac from POST, even if they are compressed as pci 2.3... the other problem that they are using supported device list and patching have no results. MacBook Pro 13.3 have supported gops for rx4xx series.
- yesterday I build two different gops for my zbook and they are working, so something in iMac prevent them to execute..

Any ideas?

Reverse engineering of iMac 6770m decompressed gop and BaffinM gops shows that they are loading framebuffer too and all additional info like efirom
Searching the forums here for CoreEG2 will show you the discoverings of @Nick [D]vB, the following one containing the key information:
Thanks, that looks very useful, will add to my list of reasons to install W10 !
To be clear I didn’t get standard PC GOPs working on the 2011 iMac either,
just Apple’s “mutant” GOPs pulled from later iMac / Macbook bootroms.

I did some more firmware tests with a desktop RX480 recently,
and found at least 5 reasons why the test we did before failed:

- I stupidly missed the sub-system ID check in the IM18,3 Polaris GOP EFI
Is this something you and @nikey22 can shed some light onto?
- You can’t load multiple EFIs from AMDs PCI rev2 vbios (but you can with Nvidia!)
I found out how to achieve that, with the caveat of having to strip off the legacy VGA firmware so the whole VBIOS fits within 128 kB.
- CoreEG2 requires an external EDID parser (AGP has that built-in?)
- The MP6,1 CoreEG2 was too old to properly load the IM18,3 GOP EFI
- IM18,3 CoreEG2 was too new for MP5,1 (as you found out the hard way, sorry!)

But… it turns out the CoreEG2 from the iMac 17,1 is “just right”
99% sure it would work on MP3/4/5,1 (it doesn’t brick a 2007 iMac !)
It loads & links the 18,3 Polaris EFI ok but sadly the un-patched iMac
frame-buffers are hopelessly mismatched to any desktop cards vbios.

That’s the killer, which sadly I don’t have the time, or IDA pro skills, to fix.
It’s a shame Fl0r!an didn't work this out a few years ago,
Netkas & MVC seem to have it covered now anyway.
@nikey22 is working on this now.
The M380 EFI might be slightly easier to patch for the R9 290X and
I’m not sure how the Vega iMacs “skeleton” vbios handles display outputs
but if it’s EFI builds them dynamically somehow that might be an option…
 
Last edited:
Well, that's a bummer. It seems that a (modded) Rev.F WX4170 card does not fit onto iMac11,1 (2-pipe) heatsink. Two surface mount tantalum electrolytics are in the way. (And that's with a 1.5mm copper shim...) Apparently I'll have to file a part of the heatsink for a proper fit. (This also means I'll have to postpone testing @internetzel 's latest VBIOS with that card.)

(I had no problems fitting that card onto the large (3-pipe) iMac12,2 heatsink. Those two problematic electrolytics do touch, but are not pressed against the heatsink.)
 

Attachments

  • RevF_WX4170.jpeg
    RevF_WX4170.jpeg
    798.7 KB · Views: 498
  • WX4170_11.1_1.JPG
    WX4170_11.1_1.JPG
    146.1 KB · Views: 293
  • WX4170_11.1_2.JPG
    WX4170_11.1_2.JPG
    172.4 KB · Views: 288
  • Like
Reactions: Ausdauersportler
Searching the forums here for CoreEG2 will show you the discoverings of @Nick [D]vB, the following one containing the key information:

Is this something you and @nikey22 can shed some light onto?

I found out how to achieve that, with the caveat of having to strip off the legacy VGA firmware so the whole VBIOS fits within 128 kB.

@nikey22 is working on this now.
Okay, but hd6770m contains gop that enables bootscreen. Without that gop or with pc gop bootscreen won’t loaded. That’s why some users used gop from hd6770m (Apple) at HP GPUs and they get bootscreen work on iMac 2011. So Apple gop play common sense in to make it bootscreen working.

Gop rom contains efi compressed gop driver, you can decompress it if you create a new file with efi image only.

My method: extract efi image from 55AA gop rom, then use Parralel desktop and boot from usb with efishell bootloader, mount fs0 and decompress image, then compress new image to see final results of compressed image, it will help to recognize entire image in gop rom. The efi rom doesn’t contain checksum so it can be modified easy.

When Apple end mxm era, the GOP driver was included in platform bios. But reverse procedure make non booting iMac. So EfiRom compression (from new times) not working.

So idea is to use 6770m Apple rom header that contains some info about framebuffer and rom revision... and patch entry data to meet data in BaffinM PE32 efi. Also BaffinM driver is limited to devices 67EF, 67FD or DF... so they should be also patched and checksum of PEI32 correct


The problem here is I made 5-10 test builds per day to make 4150 work on new oses, about 200+ builds and about 30+ gpu unbricking... here is testing not so fast, so I shared all ideas with community, hope that would be useful

To solve bootscreen, you need to back to 6770m or compatible gpu and patch them with custom gop rom, that will help to solve problems with newer gpu!


Thanks for all testers and developers, hope that bootscreen will be solved soon.

I will back to work of 4150 powermanagment issues and output glitches




EDIT1
Added project folder. It could be hard to understand the structure, but could be useful
 

Attachments

  • IMAC 2011 - iMac12,2 - A1312 - 2429 and iMac12,1 - A1311 - 2428.zip
    14.1 MB · Views: 163
Last edited:
What's with that last WX4170?"]This is the most stable performance I was able to achieve with that last WX4170 so far. It appears that that card does not tolerate higher Core clocks well. (Although it offered stellar performance with @edwardgeo 's VBIOS before sleep.) I don't think there's anything wrong with the card, maybe those very early (2016) editions just don't support those higher Core clocks. (Another observation: That card invariably reports higher Power consumption when compared to the modded WX4170: it was 115W vs. 77W with @edwardgeo 's VBIOS, with WX4170_GOP_ALT_VRAM.rom it is 58W vs. 40W before it freezes. If that means anything...)
Unfortunately there seem to exist two grades of Polaris 11/21 (Baffin) chips. There are original VBIOSes that clock the GPU at 1053 MHz (Apple even underclocked those in their MacBooks) and others that clock the GPU at approx. 1200 MHz (Polaris 11) or even 1275 MHz (Polaris 12).
The VRAM clock is commonly 1500 MHz on all models but there also exist variants that clock it at 1750 MHz.
There's also a Polaris 31 chip, which can be clocked even higher, but that doesn't seem to have been put on MXM boards.

The same is true for the Polaris 10/20 (Ellesmere) chips. Likewise there's also a Polaris 30 chip which hasn't been put on MXM boards.

For us I think we should have public VBIOSes that have clock settings that work on all card models, the benefit of the later/better chip types being lower temperatures and hence increased lifetime.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.