Yes I did try with just the apple ram installed but made no difference. I bought a SATA cable and HDD left and right Drive brackets/mounts before starting. I removed the blade. installed a SATA drive and fresh OS and saw FW had updated. With the blade out of the iMac I put it in to a MacMini (as a secondary disk and formatted it). Removed SATA drive, reinserted blade to iMac and fresh OS install and correct firmware showing.
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.
iMac 17,1 (Retina 5K, 27-inch, Late 2015) EFI firmware update failures
- Thread starter zorinlynx
- Start date
- Sort by reaction score
Thanks.
I plan on doing the same thing. Further searching leads me to the conclusion that Apple has blocked the upgrade because of instability with the chipset and the blade drive in later OS upgrades.
We have lift off! A big thank you to Philuk20 and many others on this and other sites. I can now confirm that Apple is blocking the EFI update, if you have an Apple blade SSD. Apparently, the chipset in this model is limited so they blocked upgrades with the SSD installed. Once I took the machine apart (I can now do it without instructions), removed the blade SSD, installed a SATA SSD and installed Big Sur from USB stick.... the firmware now reads 429.60.3.0.0!
Apple should have owned up to this a long time ago. They have fixed other design flaws. I don't even want to count the hours I have invested in fixing this problem. So, If you want to upgrade the firmware from 170.0.0.0.0 - you know what needs to be done.
Apple should have owned up to this a long time ago. They have fixed other design flaws. I don't even want to count the hours I have invested in fixing this problem. So, If you want to upgrade the firmware from 170.0.0.0.0 - you know what needs to be done.
Last edited:
Happy that you seem to have found the cause for it
. But what leads you to the conclusion Apple *actively* blocks it because of instability with the chipset? Mind you - if above conclusion *is* true, forcing a firmware update like this could potentially lead to an unstable machine. In my case, I actually didn't even notice the lack of firmware updates for all these years until I ran into the Internet Recovery issue and stumbled upon this thread.
Search more, you can find it. Also, other Mac SSD drive sellers exclude this machine for that reason. I haven't put the 1tb SSD back as I'm waiting for a faster one to arrive. Others have reinstalled the SSD and reported it working.Happy that you seem to have found the cause for it
Mind you - if above conclusion *is* true, forcing a firmware update like this could potentially lead to an unstable machine. In my case, I actually didn't even notice the lack of firmware updates for all these years until I ran into the Internet Recovery issue and stumbled upon this thread.
Followup: I did replace the Apple drive and mine works fine.
Also, Internet Recovery is fixed as well!
Last edited:
We have the exact same problem. PCI SSD as originally supplied from Apple. Running on Mojave. It is stuck at 170.0.0.0.0.
My solution, aside from whining to poor Howard and the people who follow him (like Rob), and spending more than 10 hours on the phone with Apple Support (they were either lied to or trained to lie to me), is two fold.
One, don't upgrade the operating system. Big Sur may be the greatest computer innovation ever, but it's not clear at all that iMac 17.1 models will be fully stable using Big Sur.
Two, think twice about buying another Apple product. I'm not angry that there's a fault. That happens. It's the run around lots of other people and I have gotten. Just imagine what other bad things are going on. This one stinks because I've been using Apple desktops since the Apple IIc days. But, I guess, nothing lasts forever. Windows isn't the answer, but there must be a Linux variation that works.
My solution, aside from whining to poor Howard and the people who follow him (like Rob), and spending more than 10 hours on the phone with Apple Support (they were either lied to or trained to lie to me), is two fold.
One, don't upgrade the operating system. Big Sur may be the greatest computer innovation ever, but it's not clear at all that iMac 17.1 models will be fully stable using Big Sur.
Two, think twice about buying another Apple product. I'm not angry that there's a fault. That happens. It's the run around lots of other people and I have gotten. Just imagine what other bad things are going on. This one stinks because I've been using Apple desktops since the Apple IIc days. But, I guess, nothing lasts forever. Windows isn't the answer, but there must be a Linux variation that works.
Hey BKDad,
I've been running Big Sur since the Beta came out and find it more stable than previous releases. Once the firmware updated I ran it for a while on an external disk and then today reinserted the Apple blade. So far so good.
I totally agree on the run around. Apple knew about this a long time ago and played dumb until pressed on it. The last promised call back didn't happen. With that, I was left with one choice to fix the machine so I took it and it worked.
I'm not recommending everyone take apart their machine and do what I did. Although, I've done it now three times and I can do it in under an hour. (not auditioning for the job)
We have likely seen the last Apple machines with replaceable parts. The new iMac, will be an M2(?) and all hardware failures will require a new motherboard. It will be fast though... real fast!
I get you.
My feeling is that Apple optimizes new operating systems for new hardware these days. Regression testing takes time. It can affect delivery times, which, from what I've read, is secondary at Apple to getting it truly right. Fixing bugs is not one of the more coveted jobs there. (The problem with that is bugs migrate forward.)
So, maybe it's best not to plan on updating the operating on an existing computer. The problem then is when you get a release like Catalina, which nobody exactly raves about. Then, you're stuck.
Catalina might be the poster child of not wanting to work on old stuff. I bet Big Sur got tons more resources than Catalina did. Catalina may have been released just so Apple could say that they released a new operating system last year.
Remember that both releases had some major under the hood stuff. They split out the system vol and then sealed it. Also APFS has evolved a bit. Time Machine runs way better now.
These are my observations.
I have really appreciated your reports, rgmenke, and those of Philuk20. I have the same non-updating EFI problem (iMac 17,1 with the 512 GB Apple SSD). While I will probably just live with the old firmware, it’s good to know that there’s evidence toward a solution, should it be needed.
Many thanks to both of you brave souls!
I wish I had an easier solution but I tried them all (many times) with no luck. BTW, I updated a previous post stating that the firmware update fixes the Internet Recovery issue.
I'd like to summarize this in my own words to make sure I got it right.
rgmenke and Philuk20 both own iMac 17.1 models with PCI "blade" SSDs. (Same here.)
The firmware upgrade process stopped upgrading the firmware at revision 170.0.0.0.0. (Same here.) Reformatting the SSD, reloading the operating system and all sorts of other clever ideas were tried in an attempt to get past that. Nothing worked.
Finally, replacing the blade SSD with one attached to the SATA interface allowed the firmware upgrade to install without a problem.
It also appears that re-installing the blade SSD in place of the SATA drive works OK with the new firmware.
This experience leads you to believe, reasonably I think, that Apple put an actual block on upgrading the firmware on the 17.1 model with blade SSDs. This is further supported by the fact that most aftermarket SSD companies don't offer blade replacements for the 17.1 model, and explicitly say so, which indicates some kind of limitation with this logic board and PCI based SSDs.
This is not some crazy conspiracy concept. There's actual evidence.
That right?
I presume that Apple did this for a reason. Perhaps the PCI SSD controller chip won't do something that the firmware beyond 170.0.0.0.0 expects to do. Who knows? Adding a block certainly had to have taken extra time and effort, so I'd presume they didn't just do this to play with some iMac owners. They did mess it all up by not fixing the EFIcheck process so that the blade SSD 17.1 iMacs don't show an error code for having the "wrong" firmware version.
So, here are my questions then:
Will the firmware upgrade process - assuming there are further firmware upgrades - continue to work from this point onward with the blade SSD back in place?
What kinds of problems did Apple try to avoid by withholding upgrades beyond 170.0.0.0.0? Will this lead to a problem in actual use?
Clearly, the drive vendors like OWC knew that there's an issue with 17.1 iMacs and PCI based SSD's. They knew some time ago. So, Apple must've known as well.
The way Apple has handled this is really not appropriate and could probably make for a far better lawsuit than most of the complaints brought against them. They have a defect which they chose to workaround. That part is fair. Then they continually chose to lie about it. That part is not fair. Litigation is not worth the time and effort, unless you have a hidden desire to put some attorneys' kids though college.
Apple has gone down in my estimation because of all this.
rgmenke and Philuk20 both own iMac 17.1 models with PCI "blade" SSDs. (Same here.)
The firmware upgrade process stopped upgrading the firmware at revision 170.0.0.0.0. (Same here.) Reformatting the SSD, reloading the operating system and all sorts of other clever ideas were tried in an attempt to get past that. Nothing worked.
Finally, replacing the blade SSD with one attached to the SATA interface allowed the firmware upgrade to install without a problem.
It also appears that re-installing the blade SSD in place of the SATA drive works OK with the new firmware.
This experience leads you to believe, reasonably I think, that Apple put an actual block on upgrading the firmware on the 17.1 model with blade SSDs. This is further supported by the fact that most aftermarket SSD companies don't offer blade replacements for the 17.1 model, and explicitly say so, which indicates some kind of limitation with this logic board and PCI based SSDs.
This is not some crazy conspiracy concept. There's actual evidence.
That right?
I presume that Apple did this for a reason. Perhaps the PCI SSD controller chip won't do something that the firmware beyond 170.0.0.0.0 expects to do. Who knows? Adding a block certainly had to have taken extra time and effort, so I'd presume they didn't just do this to play with some iMac owners. They did mess it all up by not fixing the EFIcheck process so that the blade SSD 17.1 iMacs don't show an error code for having the "wrong" firmware version.
So, here are my questions then:
Will the firmware upgrade process - assuming there are further firmware upgrades - continue to work from this point onward with the blade SSD back in place?
What kinds of problems did Apple try to avoid by withholding upgrades beyond 170.0.0.0.0? Will this lead to a problem in actual use?
Clearly, the drive vendors like OWC knew that there's an issue with 17.1 iMacs and PCI based SSD's. They knew some time ago. So, Apple must've known as well.
The way Apple has handled this is really not appropriate and could probably make for a far better lawsuit than most of the complaints brought against them. They have a defect which they chose to workaround. That part is fair. Then they continually chose to lie about it. That part is not fair. Litigation is not worth the time and effort, unless you have a hidden desire to put some attorneys' kids though college.
Apple has gone down in my estimation because of all this.
I would say you summed it up well. I did have one of those blade replacement companies tell me that they were experiencing erratic behavior with this model. I'll go a step further and propose that it may be the 17,1 iMac and certain OS releases. They (vendor) said that their engineers were re-evaluating (my words) possible support for this machine.
Apple used a proprietary PCIE connector for the drive so standard blades won't connect without an adapter. Even then folks had to uses work arounds for sleep/wake issues. With the current OS and the Apple blade I haven't experienced any issues - time will tell.
What is clear is that Apple has adopted the Ostrich method for handling this problem. Support calls start with folks eager to help and end with a more senior person trying to convince you there is no issue. (6 times)
As for future upgrades... who knows? Others, who have achieved the initial upgrade, have reported moving from 428 to 429 so there is hope.
I guess time will tell.
My plan is to not upgrade beyond Mojave. At some point, a new iMac will come along with much better performance and we'll buy one of those, complete with the operating system the computer was originally designed for. Probably. This model really does everything we want as fast as we would want, so it may be a matter of just avoiding deteriorating hardware.
But, we also have a white iMac 5.1 that I maxed the memory in and replaced the HD with an SSD. It runs Snow Leopard. It's obviously not as fast as the 17.1 and doesn't have a retina display, but it really is entirely usable for 95% of what we do here. (That version of Safari is lame now, but the latest Firefox is fine.) It works great on the workbench with test instruments and will run Windows 7 through Boot Camp just fine. So, maybe obsolescence is in the mind of the user to a large degree.
Somewhat OT - I saw an advertisement for Subarus last night. One of their selling points is that more than 90% of the Subarus produced in the past ten years are still on the road. That counts for something amongst people. Maybe not in the computer world.
The Ostrich scheme of managing customers is hardly appealing. At some point, maybe, Apple will get lumped in with other tech companies like Twitter and Facebook and will get to discuss things with the government, whether here or elsewhere. Some of this behavior will undoubtedly surface.
I got the same run around as you did, but only persisted through it twice before I decided it was hopeless trying. My hat's off to you for persevering.
I'll just add another data point here with what I've been going through with my 17,1 with the 512G SSD.
- Had an Airport card failure (WiFi wouldn't turn on at all, even in recovery)
- Got the repair approved to be covered by my credit card extended warranty
- Had an AASP repair it; they replaced the Airport card and logic board
- The AASP told me they had to order 2 logic boards; the first was defective
- Brought it home, booted into recovery, wiped the drive, reinstalled Catalina, and restored from Time Machine
- Immediately started to experience kernel panics
- And now I couldn't boot into recovery. Now only Internet Recovery with El Capitan (10.11).
- At that point, the machine was experiencing kernel panics while in Internet Recovery
- Took it back to the AASP
- They ruled out bad RAM and decided to order another logic board
- It took almost 3 weeks to get the new logic board -- which was unusual since the first repair took under a week
- They replaced the logic board, got Big Sur installed on it, released it back to me
- I booted into Big Sur Recovery, did a clean install of the OS, restored from Time Machine
And then I stumbled across this thread. I had noted down the firmware after the first repair and prior to the second repair and it was 170.0.0.0.0. I'm looking at my firmware with the 4th replacement logic board and it's 429.60.3.0.0.
I don't know if it all just coincidence and I'm just lucky that my new logic board has the updated firmware, but it seems unusual that it took three replacements (one of which had 170.0.0.0.0). Being unable to boot into Recovery was also odd, as I know I was able to do so with Catalina installed prior to the first AASP visit and immediately after the first repair.
- Had an Airport card failure (WiFi wouldn't turn on at all, even in recovery)
- Got the repair approved to be covered by my credit card extended warranty
- Had an AASP repair it; they replaced the Airport card and logic board
- The AASP told me they had to order 2 logic boards; the first was defective
- Brought it home, booted into recovery, wiped the drive, reinstalled Catalina, and restored from Time Machine
- Immediately started to experience kernel panics
- And now I couldn't boot into recovery. Now only Internet Recovery with El Capitan (10.11).
- At that point, the machine was experiencing kernel panics while in Internet Recovery
- Took it back to the AASP
- They ruled out bad RAM and decided to order another logic board
- It took almost 3 weeks to get the new logic board -- which was unusual since the first repair took under a week
- They replaced the logic board, got Big Sur installed on it, released it back to me
- I booted into Big Sur Recovery, did a clean install of the OS, restored from Time Machine
And then I stumbled across this thread. I had noted down the firmware after the first repair and prior to the second repair and it was 170.0.0.0.0. I'm looking at my firmware with the 4th replacement logic board and it's 429.60.3.0.0.
I don't know if it all just coincidence and I'm just lucky that my new logic board has the updated firmware, but it seems unusual that it took three replacements (one of which had 170.0.0.0.0). Being unable to boot into Recovery was also odd, as I know I was able to do so with Catalina installed prior to the first AASP visit and immediately after the first repair.
Ahh! So there *is* a fix. Does About This Mac still think that your computer is a 17.1? I wonder if a newer logic board is the answer.
Thanks very much for the update.
Apple made changes to Recovery mode recently that the old firmware doesn't seem to follow. This is exactly a reason why the firmware needs to be updated!
This is just a guess but going through a AASP is probably more effective than the direct approach with Apple.
Yes, it still shows as a 17,1.
I don’t know whether the AASP worked out the resolution to be getting the updated firmware (and had to seek out a replacement logic board for that purpose), or if it’s a just a coincidence that the 4th replacement was on the latest firmware. But it seems odd (or very unlucky) that they went through two other replacements that failed before this one.
In any case, the only reason I did the repair in the first place was because of the extended warranty coverage (which has since expired), so it cost me nothing out of pocket. My plan was to replace it with a new iMac or mini M1 and I figured I should get it repaired so that I could trade it in or sell it.
Edited to add: thinking about it further, is it possible that my first logic board replacement had the newer firmware and when Catalina was reinstalled, it attempted to revert it to 170.x? When I received it back from the AASP the second time, it was on Big Sur and I booted into Big Sur Recovery, erased the drive, and reinstalled Big Sur without issue.
Last edited:
OK - thanks.
It sounds to me like our AASP is pretty good. You should tell the world who they are so that their good work gets rewarded.
I'd say stick with what you've got if it is working. I've got an M1 Mini too and it's not a patch on these iMacs
I did the upgrade operation in November and (touch wood) it's been fine since. I'm on 10.15 for now and the FW has since gone from 428 to 429 with updates.