Macbook pro 16" A2141 not updated

[PITERPENY]

Hi. I've encountered this problem for the first time. I need help. Has anyone encountered it before and how did you solve it? Only MacOS Catalina can be installed on a MacBook via online recovery, and the Mac works fine (although it's also running slowly). But when I try to update the MacBook, it constantly crashes, and nothing else works except Catalina. Attempts to install Venture or Monterey Jack from an external drive are also impossible. I tried DFU recovery, but I can still only install Catalina and can't update, even though the system offers to update to Tahoe. It's impossible to run any installer from the system itself, like Venture or Monterey Jack, except Tahoe, which crashes at the end of the installation. The MacBook worked on MacOS Tahoe before the wipe. Diagnostics show everything is fine, of course.

[reformatt]

My main suspect would be RAM. I had an old A1502 circa 2014 with similar issues. It'd either crash loading the installer or half way through. Memtest showed memory errors (there are 16 chips on this board so I just swapped out the board with a spare one I had lying around). If you have external USB booting enabled and signing turned off, see if you can run a memtest boot drive.

[PITERPENY]

The MacBook starts up and runs with a pre-installed Tahoe on an external drive.

[Boardview Bios and Schematic Bot]

The MacBook starts up and runs with a pre-installed Tahoe on an external drive.

This is a classic internal drive issue. The drive has likely reached the end of its life and needs to be replaced. Why do I say this? Let me explain.

You mentioned you can install the OS via online recovery, but not with a USB installer. Catalina is the factory OS for the A2141. The official online installation might just be a bit more stable—I don't know exactly how to explain the underlying mechanism. But why is it running so slowly? Because the drive on the motherboard is practically dead. The system is frantically retrying read and write operations in the background, which causes extreme lagging.

The drive is already failing. When you use a USB to install a newer OS, the installation package is massive, and it performs extremely high-intensity, continuous writes to the storage chip. This is a bit different from how the online install runs. There are probably too many bad blocks inside the drive, which directly causes the writing to fail and crashes the installer.

If everything before this was just a guess, the fact that you successfully ran the OS from an external drive completely confirms it. It proves beyond a doubt that your original drive is faulty.

You can buy a new storage chip or desolder a good used one and install it. If you can then install the OS via USB (or online) without any lagging, that fully proves the drive was the problem.

[reformatt]

My main suspect would be RAM. I had an old A1502 circa 2014 with similar issues. It'd either crash loading the installer or half way through. Memtest showed memory errors (there are 16 chips on this board so I just swapped out the board with a spare one I had lying around). If you have external USB booting enabled and signing turned off, see if you can run a memtest boot drive.

[PITERPENY]

This is a classic internal drive issue. The drive has likely reached the end of its life and needs to be replaced. Why do I say this? Let me explain.

You mentioned you can install the OS via online recovery, but not with a USB installer. Catalina is the factory OS for the A2141. The official online installation might just be a bit more stable—I don't know exactly how to explain the underlying mechanism. But why is it running so slowly? Because the drive on the motherboard is practically dead. The system is frantically retrying read and write operations in the background, which causes extreme lagging.

The drive is already failing. When you use a USB to install a newer OS, the installation package is massive, and it performs extremely high-intensity, continuous writes to the storage chip. This is a bit different from how the online install runs. There are probably too many bad blocks inside the drive, which directly causes the writing to fail and crashes the installer.

If everything before this was just a guess, the fact that you successfully ran the OS from an external drive completely confirms it. It proves beyond a doubt that your original drive is faulty.

You can buy a new storage chip or desolder a good used one and install it. If you can then install the OS via USB (or online) without any lagging, that fully proves the drive was the problem.

Thanks for the comprehensive answer. I think I'll take your advice and get to work on the SSD right now. First, I'll try to rule out SSD failure and reball these SSDs, then I'll program the new SSDs using JCID 13. Why didn't I think about SSD problems? The SSD checker showed 100% health.

[PITERPENY]

My main suspect would be RAM. I had an old A1502 circa 2014 with similar issues. It'd either crash loading the installer or half way through. Memtest showed memory errors (there are 16 chips on this board so I just swapped out the board with a spare one I had lying around). If you have external USB booting enabled and signing turned off, see if you can run a memtest boot drive.

Thank you for sharing your experience and advice. I managed to erase the system and am currently trying to install a new system on the reinstalled old NAND drives. If all the SSD-related hassles fail, I'll try reinstalling the memory chips as well. Fortunately, they aren't coated in compound and don't require the same numbering as with SSDs. It's definitely not worth discarding this option without checking.

[Boardview Bios and Schematic Bot]

Thanks for the comprehensive answer. I think I'll take your advice and get to work on the SSD right now. First, I'll try to rule out SSD failure and reball these SSDs, then I'll program the new SSDs using JCID 13. Why didn't I think about SSD problems? The SSD checker showed 100% health.

Thanks for adopting my suggestion. I won't explain the reason for the software issue for now. If it is confirmed to be a drive problem, we'll talk about it then; if it's not a drive problem, explaining it is useless.

Then, I saw a user suggest it might be a memory problem. If a memory failure occurs, it usually causes a kernel panic. In that case, it is absolutely impossible for the system to still run normally when booting from an external drive. So, it most likely shouldn't be a memory problem.

Now I feel the issue is a bit intriguing. I am indeed very curious and want to see what the true root cause of the fault actually is.

[PITERPENY]

Thanks for adopting my suggestion. I won't explain the reason for the software issue for now. If it is confirmed to be a drive problem, we'll talk about it then; if it's not a drive problem, explaining it is useless.

Then, I saw a user suggest it might be a memory problem. If a memory failure occurs, it usually causes a kernel panic. In that case, it is absolutely impossible for the system to still run normally when booting from an external drive. So, it most likely shouldn't be a memory problem.

Now I feel the issue is a bit intriguing. I am indeed very curious and want to see what the true root cause of the fault actually is.

Yes, I'm also curious, since this is the first time in my practice that I've encountered a system that doesn't improve beyond Catalina, while still operating stably on Catalina. What's the bottom line? I reballed the NAND to rule out the stock NAND, and unfortunately, that didn't work. I installed new ones after programming them, and that didn't help either. Only Catalina works. Transferring from a working board also didn't work. In the end, I can assume that the problem is definitely not in the NAND itself (since there aren't many solutions, the next suspect is the RAM). Now, on Reformat's advice, I'm running a RAM test from a MemTest86 bootable flash drive, and it seems to me that all the errors are on one RAM chip.

[PITERPENY]

My main suspect would be RAM. I had an old A1502 circa 2014 with similar issues. It'd either crash loading the installer or half way through. Memtest showed memory errors (there are 16 chips on this board so I just swapped out the board with a spare one I had lying around). If you have external USB booting enabled and signing turned off, see if you can run a memtest boot drive.

Thanks for this tip, it really solved the problem. I removed all 16 RAM chips from the Scanix, reballed them, and reinstalled them, but that didn't help. Then I took the same donor board with 16 Micron RAM chips and transferred them to my board. To my surprise, the board booted after recovery, but there was no image. So I realized that you can't just replace the Scanix with a Micron, so I had to look for a board with the same revision of Scanix. As soon as I transferred them to the board, the board immediately worked correctly and displayed an image even without DFU. I still did a DFU recovery and the MacBook installed the Tahoe OS on the first try. (I've never reballed so many chips at once, but the result was worth it). I hope my experience will help someone else too!

[reformatt]

No problems. That was the most likely thing for me. That's an incredible amount of work and for it to work after that amount of rework is a great effort.

If you are changing manufacturer or size, you need to change the RAMCFG resistors on the board. See the following from the 820-01700 schematic. If you are taking off a donor, then just match the resistors from the donor board.

[Boardview Bios and Schematic Bot]

Thanks for this tip, it really solved the problem. I removed all 16 RAM chips from the Scanix, reballed them, and reinstalled them, but that didn't help. Then I took the same donor board with 16 Micron RAM chips and transferred them to my board. To my surprise, the board booted after recovery, but there was no image. So I realized that you can't just replace the Scanix with a Micron, so I had to look for a board with the same revision of Scanix. As soon as I transferred them to the board, the board immediately worked correctly and displayed an image even without DFU. I still did a DFU recovery and the MacBook installed the Tahoe OS on the first try. (I've never reballed so many chips at once, but the result was worth it). I hope my experience will help someone else too!

I just finished reading your post, and the problem is indeed resolved now. Just as reformatt pointed out, the crux of the issue really was the RAM. That guy's experience is truly next-level—he got straight to the core of it!
Troubleshooting such an extreme "edge case" is insanely difficult. If only this motherboard came with a physical expansion slot, we could have just disabled the onboard memory and easily ruled it out as the culprit.
However, what I admire most is your execution. Your problem-solving ability and that relentless, never-give-up drive of yours are just hardcore!! I'm so glad to have personally witnessed this entire troubleshooting process. Desoldering and inspecting those 16 individual RAM chips one by one... my god, I haven't even done this kind of troubleshooting work myself.
For anyone out there troubleshooting memory issues on a MacBook Pro 16-inch model, this thread is simply a priceless treasure trove.

[PITERPENY]

No problems. That was the most likely thing for me. That's an incredible amount of work and for it to work after that amount of rework is a great effort.

If you are changing manufacturer or size, you need to change the RAMCFG resistors on the board. See the following from the 820-01700 schematic. If you are taking off a donor, then just match the resistors from the donor board.

Thank you for helping me understand my mistake and for helping me figure this out. I studied the schematic for this board and this is exactly it, and all these resistors around the U1200 chip need to be changed. Why is this important?! Because out of the 15 motherboards I have, 14 were Micron and only one was HiNix. Thank you.

[PITERPENY]

I just finished reading your post, and the problem is indeed resolved now. Just as reformatt pointed out, the crux of the issue really was the RAM. That guy's experience is truly next-level—he got straight to the core of it!
Troubleshooting such an extreme "edge case" is insanely difficult. If only this motherboard came with a physical expansion slot, we could have just disabled the onboard memory and easily ruled it out as the culprit.
However, what I admire most is your execution. Your problem-solving ability and that relentless, never-give-up drive of yours are just hardcore!! I'm so glad to have personally witnessed this entire troubleshooting process. Desoldering and inspecting those 16 individual RAM chips one by one... my god, I haven't even done this kind of troubleshooting work myself.
For anyone out there troubleshooting memory issues on a MacBook Pro 16-inch model, this thread is simply a priceless treasure trove.

Thanks for the words of support. This is the first time I've encountered this kind of problem, and thanks to Reformatt for setting me on the right path. I hope this difficult journey will show someone the way to the right solutions for such problems. Thank you.