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.
Good Evening to all,

I am Bryce from Singapore. I have been following this thread for a while. Took a plunge into Xeon processor upgrade as well. Found a E3-1265L. Better performance with less heater emission but i lost the wake feature. it will go into deep sleep as well even though the spec says it have iGPU. Just like to contribute my share of unchartered territory. I also managed to upgrade my wireless card to 943602CS to connect ac network with 4.2 bluetooth to allow this iMac 2011 to continue to stay relevant in 2020. Next step would be getting a GPU upgrade which i have yet to make a final decision on.

I would like to thank everyone in the forum for making this happen. Unleashing the potential for aging model to stay relevant and powerful again.

Cheers.

Have a nice weekend and stay safe.

Screen Shot 2020-08-15 at 8.44.23 PM.png

Screen Shot 2020-08-15 at 8.59.42 PM.png


Screen Shot 2020-08-15 at 9.04.33 PM.png
 
Wait that's a 3.4GHz CPU in a 21.5" iMac? And it's an 80w CPU and it boots, I wonder if another 95W Xeon could book in the 21.5" iMac. Unfortunately I think the HD3000 Kexts needed to be loaded on boot for sleep to work.
I upgraded the CPU to a Intel Xeon 1270 and I've problems with waking from sleep and the higher TDP is a real problem. When I read about the IGPU I thought I would fix it with a different processor with a IGPU and went for Xeon 1245 big mistake the TDP is killing the PSU and my temp raises gradually within 10 mins it reaches over 75c and the system shutsdown. I've replaced the HDD with SSD and removed the ODD and the moved back to the original GPU still no luck the systems heats up (tried adding another fan inside into the ODD space couldn't fit). So my opinion stay away from higher TDP processors. The machine becomes unusable.
 
Last edited:
Good Evening to all,

I am Bryce from Singapore. I have been following this thread for a while. Took a plunge into Xeon processor upgrade as well. Found a E3-1265L. Better performance with less heater emission but i lost the wake feature. it will go into deep sleep as well even though the spec says it have iGPU. Just like to contribute my share of unchartered territory. I also managed to upgrade my wireless card to 943602CS to connect ac network with 4.2 bluetooth to allow this iMac 2011 to continue to stay relevant in 2020. Next step would be getting a GPU upgrade which i have yet to make a final decision on.

I would like to thank everyone in the forum for making this happen. Unleashing the potential for aging model to stay relevant and powerful again.

Cheers.

Have a nice weekend and stay safe.

View attachment 944379
View attachment 944384

View attachment 944385
May I know how did you manage to screw down the wifi card to the 2011 mac. I was checking out the card and the hole dimension to screw down is different for this card.
 
  • Like
Reactions: highvoltage12v
Hi!
Ive been following this thread for a while with great interest as i am about to start a refurb project with my old 2011 21.5 imac.
I already ordered a 770m and a SSD but have not yet seen the value of upgrading the 2.5 I5 as the difference in performance to the 2600s i7 isnt really motivating it enough.
Now, with the potential of the xeons it makes it makes it much more meaningful so it would be great to hear your input on how well it works currently?

Im not sure if I understood the steps you took above, did you flash the eeprom of the CPU? did that fix sleep issues?

Im probably planning to mainly run Win10 on it, do you have sleep issues in win10 as well?

I can sacrifice DVD and other drives if power economics is important, any tips or suggestions are most welcome.

Thanks for an interesting thread! keep it going!
/A
Beware of the Xeon processors TDP. 1275 or 1245 with iGpu, TDP are high and if you decide to upgrade your GPU then definitely the system will heat up and shutdown. As far as I searched Xeon 1265L has the lowest TDP. I tried the 1270 and 1245( this was a real disaster with over heating).
 
Finally made it!
Xeon in my iMac, couldn't be happier.. 😁

Thanks @StephN999 for the info.
Next Graphics upgrade, Quadro, maybe K610m for the cheapest price.

Some evidence here:
DO you have any Imac heating up problem? What does your PSU respond (how high did the temp go?). I upgraded it to 1270 and has some heating problem and the system shutdown (of course the system wake from sleep problem exists). I put in a external fan and it was ok but the fan was too loud.
 
May I know how did you manage to screw down the wifi card to the 2011 mac. I was checking out the card and the hole dimension to screw down is different for this card.
[/QUOTE
May I know how did you manage to screw down the wifi card to the 2011 mac. I was checking out the card and the hole dimension to screw down is different for this card.

Hi Dubukey,

A friend of mine whom design circuitry, built a convertor for this wireless card. I bought one from him with a used wireless card as well after he tested working.
 
It stays interesting.

Today my Xeon E3-1285L V2 (Ivy Bridge) arrived.

Searching here and there I stumbled across the thread on updating a Mac (Pro) microcode.
Apparently Petri Krohn and Larsvonhier gained important insights and doing rapid progress recently regarding patching MacPro 1,1 and 2,1 for Xeon 5100s and 5400s.

Different processors though, but important work maybe for us too...

BTW, tryed out UEFITool in conjunction with dosdude1s RomTool. Is somewhat more verbose than the dosdude1's Apple Microcode Tool and might be useful due to automatic CRC correction according to the cited sentinelone blog.

My imac 11,2 (27") Bootrom has Microcode for CPUID's 206A3, 206A5, 206A6 and 206A7.

1598190445317.png


As far as I understood cpu-world.com and wikipedia, 206A7 is the SandyBridge CPUID, on the other hand the number behind the A might indicate different CPU steppings according a thread on german MSI forums

My Xeon E3-1290v1 is stepping 7 accourding to Intel's MacCPUID.

By the way, the same Information as MacCPUID can bei gained via terminal and sysctl:
Code:
sysctl machdep.cpu.

I would thus assume that replacing 206A6 might be safer than replacing 206A7...Has anyone insights?

Regards,

roscho
 
Last edited:
Bonjour,

@roscho, my idea is to remove all Sandy Bridge microcodes and put only Ivy Bridge microcode, if you want to give me your firmware privately and I'll modify it for you.

Capture d’écran 2020-08-23 à 16.06.40.png


Edit: Otherwise I had already said that the dosdude1 tool wasn't compatible with the EFI rom of the iMac 2011 (and probably others) because it doesn't modify the duplicate microcodes on our firmwares, but it allows to see if the modification is well done. :)

By the way, since I don't have an Ivy Bridge at the moment, I would test a pre-spectre microcodes to see if we can regain performance ...or not. o_O
 
Last edited:
  • Like
Reactions: Rendell001
Salut Stéphane,

Petri Krohn suggests doing a two step approach: first midfying the Bootrom and checking if the SandyBridge CPU still works (=the process of BootRom modding not being harmful), and then in a second step swapping the CPU (=POST would mean the mod has been useful).

From a scientific point of view I find that logical...

Has anyone already compared the microcode part of a 13,2 (Late 2012 IvyBridge Imac) Bootrom with a 12,2 Bootrom?
 
Salut Stéphane,

Petri Krohn suggests doing a two step approach: first midfying the Bootrom and checking if the SandyBridge CPU still works (=the process of BootRom modding not being harmful), and then in a second step swapping the CPU (=POST would mean the mod has been useful).

From a scientific point of view I find that logical...

Has anyone already compared the microcode part of a 12,1 (Late 2012 IvyBridge Imac) Bootrom with a 11,2 Bootrom?
12,1 is the 21.5" 2011 SandyBridge?
 
Salut Stéphane,

Petri Krohn suggests doing a two step approach: first midfying the Bootrom and checking if the SandyBridge CPU still works (=the process of BootRom modding not being harmful), and then in a second step swapping the CPU (=POST would mean the mod has been useful).

From a scientific point of view I find that logical...

Has anyone already compared the microcode part of a 13,2 (Late 2012 IvyBridge Imac) Bootrom with a 12,2 Bootrom?
Re-Bonjour,

The problem and that's why I asked you to pass me your EFI rom in private, is that it's unique per machine, so find someone who will pass you his complete rom... maybe just ask for the boot code of the machine otherwise.

That said, if you want to test my modification anyway, at least we'll be fixed on that.
 
The problem and that's why I asked you to pass me your EFI rom in private, is that it's unique per machine, so find someone who will pass you his complete rom... maybe just ask for the boot code of the machine otherwise.

As far as I understood you, your imac 2011 did never POST with your modded roms - did it?

I hesitate a little to modifiy my ROM yet before doing more Research.
This iMac is my "productive" machine and not just for testing...

The aspect of duplicate microcodes is just one thing to consider.

The other thing might be correct CRC and corrected ROM length (Padding FF etc) and the ominous last four bytes of ZeroVector, but this UEFITool issue might long been fixed as the sentinelone blog dates from 2015 ...

Another question might be, if it is sufficient to only swap microcodes.

Maybe other Bootrom regions might also be important where CPUID is checked during boot process (as it seems to be the case for Mac Pro 1,1 and 2,1 roms -> Petri Krohns work in progress).

As it has be written on netkas (maybe not correct in every detail but giving a general idea):
"Anecdotal evidence has indicated that Mac systems also contain a “boot ROM”, which is executed before the EFI firmware and verifies the integrity of the firmware image including its cryptographic signature at the end of the firmware volume. If the firmware image is not deemed to be valid, the system generates the “S.O.S.” beep sound (literally “S O S” in Morse code) and refuses to boot."
 
Last edited:
As far as I understood you, your imac 2011 did never POST with your modded roms - did it?

I hesitate a little to modifiy my ROM yet before doing more Research.
This iMac is my "productive" machine and not just for testing...

The aspect of duplicate microcodes is just one thing to consider.

The other thing might be correct CRC and corrected ROM length (Padding FF etc) and the ominous last four bytes of ZeroVector, but this UEFITool issue might long been fixed as the sentinelone blog dates from 2015 ...

Another question might be, if it is sufficient to only swap microcodes.

Maybe other Bootrom regions might also be important where CPUID is checked during boot process (as it seems to be the case for Mac Pro 1,1 and 2,1 roms -> Petri Krohns work in progress).

As it has be written on netkas (maybe not correct in every detail but giving a general idea):
"Anecdotal evidence has indicated that Mac systems also contain a “boot ROM”, which is executed before the EFI firmware and verifies the integrity of the firmware image including its cryptographic signature at the end of the firmware volume. If the firmware image is not deemed to be valid, the system generates the “S.O.S.” beep sound (literally “S O S” in Morse code) and refuses to boot."
To give you more confidence: Boot rom mods using UEFI tools works! Just check this link...

This was used to provide former versions of AMD VBIOS with a boot picker. You may watch the video how it works. A simple and fast process. I have a modded BIOS here and have just to disassemble the iMac to flash it back using a CH341A clip programmer. I got the ROM using @dosdude1's RomTool.
 
  • Like
Reactions: Rendell001
To give you more confidence: Boot rom mods using UEFI tools works! Just check this link...

This was used to provide former versions of AMD VBIOS with a boot picker. You may watch the video how it works. A simple and fast process. I have a modded BIOS here and have just to disassemble the iMac to flash it back using a CH341A clip programmer. I got the ROM using @dosdude1's RomTool.

Yeah,

I read Nick‘s eg2 rom post this afternoon. Fortunately the CRC checks and ROM size issues seem to be handled automagically by recent UEFITool versions (Apples EFI implementation had not necessarily to be identical to standard UEFI implementation).Also the creation / identification of the latter is undoubtedly a huge achievement.

We have to see if changing microcode (SEC/PEI Firmware phase) is as easy as / comparable to padding ffs files (DXE EFI Driver phase).

Nice background document by MITRE on Blackhat.
 
Last edited:
  • Like
Reactions: Ausdauersportler
Hi Dubukey,

A friend of mine whom design circuitry, built a convertor for this wireless card. I bought one from him with a used wireless card as well after he tested working.
Thanks, Do you have a picture of the converter I'm curious how does it look. Also if it is available for sale?
 
Bonsoir roscho,

I said to myself, because I don't have a firmware of an iMac 13.1, considering all that's in its update file, maybe there's a way to find the bootcode you're interested in.

By the way, if we can know more about it, thanks.
 

Attachments

  • IM131_0115_B00.scap.zip
    4 MB · Views: 119
  • Like
Reactions: Rendell001
Bonsoir StephN999,

good idea to check the 287.0.0.0.0 FW update, two concerns may be that it contains only the EFi part and secondly structure might be different according to the following link, but I will have an eye on it...

MacProFirmwareToolUpdate
 
Bonjour,

I breacked my installation (Not find a term to say it with understanding "j'ai foutu en l'air ou bousillé") yesterday by making some modifications to eventually not have to use it in the future OpenCore (Thanks Time Machine), I realized that I really need to find a way to modify the EFI Rom, I noticed that at startup (alt) my new Wifi card was unusable, there are a lot of things to change in my opinion, but this is out of my reach. :confused:

20200901_094512.jpg
 
Last edited:
Yes, changing the WiFi/BT card breaks the Boot Rom internal TFTP client - which is the piece of software allowing WAN boot aka remote recovery. One could try to identify this piece of microcode from the 2012/13 models having the BCM94360 cards installed and patch it back, too.

You always may connect the Ethernet cable back to your router in such cases...my solution as of now.
 
  • Like
Reactions: StephN999
DO you have any Imac heating up problem?
Bonjour,

Excuse me I saw your message quickly during the vacations and forgot to answer it, I have no problem with my Xeon, I've talked about it on this topic, to my great surprise I'm well below the PDT max in full load, I have a theory on that by the way.

Are you still having problems today?

Have a nice day.

Edit: Added tests captures of Intel Power Gadget.


Idle.png1 core Max.pngAll Max.png
 
Last edited:
What is the recommended CPU for the 2011 iMac 27"? I want all features (sleep/wake, etc) working, and as much performance as possible.
 
Bonjour,

@gagarin04, apart from telling you the most powerful officially recommended by Apple (1) I don't see!?
and of course don't forget enough ram (2) and a SSD.

1. Intel® Core™ i7-2600

2. I recommend 32, it allows to play quietly with a virtual machine, 24 seems to be the minimum to have the maximum in the benchmarks according to my research.

Edit: Grilled by roscho. 😂
 
Last edited:
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.