@noInk I’ll chat the seller if he has other store location maybe he will reply a day or two. You mean this free clip PCB?
@eivrah
Nope the CH134A it self, I’m not using the adapter i’ll pin in the dupont wire. Your has different resistor, fuse, TEXT and some trace on the back are on totally different path.
Will certainly do something different it arrive Thursday.! I already order one, hope it’s the same as your like in the photo!
Thanks eivrah I hope that the issue! If it work you don’t know how much time this will save for me! I’m feeling good chance with the 200 TRFC 
@eivrah
I didn’t received yet the CHA341A with the other PCB type, perhaps tomorrow the package will make it trough. UPDATE: Got another one with the “PCB B” (doesn’t work).
Anyway someone should try to compile a list of supported in offline system SPI hardware programmer know to work on intel EOL board.
DX58SO2 SPI
FLASHCAT 2.1 Cannot.
XGecu PRO TL866II Cannot.
CHA341A CLONE of the CLONE Cannot.
Revelprog-IS Cannot.
MCUMALL GQ-4x4 Partially.
CHA341A “ORIGINAL” Can.
DEDIPROG SF 100 Can
–
RK-SYSTEM UPROG-ICP Dunno ( didn’t arrived )
Optional
ULX3S Lattice ECP5 SPISPY (might require for later uses)
(where it ending on arrival)
ACA-SPI-004-K01
Soldering that socket in such nonexistent space it’s way harder then look.
FLASHCAT “PCB 2.1” - know to identify only without the VCC PIN connected, NO read.
XGecu PRO “TL866II Plus” - ZIF to CLIP NO identify, NO read | ICSP port embedded VCC programmer “OverCurrent Protection”.
CHA341A - “PCB B” know to identify only without the VCC PIN connected, NO read.
REVELTRONICS “Revelprog-IS” ZIF to CLIP | ICSP port NO identify, NO read "too much current " consumption error.
MCUMALL “GQ-4x4” - know to identify and partially read, unknown write, wrong verify.
CHA341A - “PCB A” reported to work by @eivrah on identify, read, write and verify.
DEDIPROG SF 100 - know to identify, read, write and verify.
–
RK-SYSTEM UPROG-ICP Dunno ( didn’t arrived )
New year, new chance, I hope.
I’ve got a spi programmer and could read the content of the chip.
Now I don’t know how to get further. Is it somehow possible to “restore” the image without a second NUC to take the flash image from?
nuc_spi_dump_20200107.zip (4.55 MB)
@elektron
Your ME region is corrupted. Did you used any third party tool? The image has still the descriptor and relevant unique data.
Unfortunately if you don’t have the backup it would be too much time consuming for me right now ( I’m dealing with stuff that doesn’t work ) to build your ME region from 0.
I sent you one quick reconstructed ME bios image that might or might not work trough PM.
Flash the image with the hardware programmer and report back.
If it doesn’t work you need to build and configure a compatible ME version and replace the ME region on the binary output of your programmer.
@noInk
Thank you very much for the reconstructed images. I’ve tried both but they don’t seem to work. The blue power LED stays off and fan is spinning at full speed. The same behavior as after the unsuccessful bios update.
As I understand you right, the last chance would be to (binary) paste/replace a working and compatible ME part to the programmer window and flash this “combined” BIOS to the chip?
Sounds crazy, I don’t know if I’m able to do all these steps the correct way… but I will try.
@elektron
Yea, I just extracted your initialized data and merged that portion with a stock uncofigured ME image.
You need to read it all and follow the full procedure it involve many step which I didn’t.
There guide on the forum.
[Guide] Clean Dumped Intel Engine (CS)ME/(CS)TXE Regions with Data Initialization
@noInk
I have very "basic" problems with following the steps:
How can I open my ME-corrupted flash image in any of the mentioned tools, like FIT or MEA?
Your reconstructed image shows SKU Consumer H and Chipset SPT/KBP-LP C for the ME part.
Is this the correct one for the nuc platform?
@elektron
You can’t open your “original” ME because it’s not there.
Probably its the right PCH SKU, I have no idea. Unless you see that PCH\SKU from a twin dump you should trial by error the 11.0.0.1180 ME version (check the @plutomaniac thread Intel Engine: Firmware Repositories Intel CSME 11.0 Firmware Repository).
Once you get one configured to boot and defined the PCH\SKU eventually you can try to merge your older data back.
Try to ask @Lost_N_BIOS maybe he can help you prepare the 11.0.0.1180 ME SKU variant, or know already the PCH\SKU for the NUC series.
@elektron - I can maybe help, if you link me to stock BIOS download page for your system, and or give the exact and FULL model name.
Then maybe I can find someone else’s dump to use as base to redo your ME FW, otherwise nothing can be done as you can’t simply use a stock unconfigured ME FW, it needs to be configured for your exact system
What did you do that corrupted the ME FW like that? Nothing from your dumped ME FW can be used to correct any other ME FW to put back in there, it’s all messed up so don’t even waste your time trying anything like that.
I don’t know where else to post this tbh. @LOST_N_BIOS
I have been running the same rig, i7 920 on an older intel DX58SO motherboard with 8gb DDR3 for like 8 years now, and I want to swap it out for a xeon x5660 that I bought on ebay to squeeze a couple more years out of my PC. When I put the cpu in, the PC turned on with no display, turned off after a few seconds and turned on again with still no display. I think the issue is I need to update my motherboard’s BIOS, but intel has removed any downloads for the DX58SO motherboard from their website as of a few months ago I think. Is there any workaround for this?
@sirrelll
latest offical version for the DX58SO is SOX5810J.86A.5 ̶5̶9̶9̶600
–
Got the composite socket in place running on the DX58SO2 and updated the bios for microcode, RAID and NVME support.
Better board already on stock overall.
Unfortunately Mine is a v1, I can’t find a place on the board with the socket like yours does?
[quote="sirrelll, post:375, topic:30311"] Unfortunately Mine is a v1, I can\'t find a place on the board with the socket like yours does? [/quote]
Unfortunately Mine is a v1, I can't find a place on the board with the socket like yours does? and I still can't find anywhere to download a working .exe file for the BIOS.
@sirrelll
Yea, intel removed all their download for legacy product.
Try to search for "DX58SO\SOX5810J.86A.5600.BI.ZIP" or something along.
This is last update to mine DX58SO status.
Since it couldn’t be overclocked at all and the replacement board (501) was in pretty bad condition (broken PCI-ex plastic holder) I’ve boxed the DX58SO for an DX58SO2.
Due some inability to use any mid-low cost SPI programmer and the time constrain I’ve opted to add the required hardware adjustment by installing a composite socket for bios programming and implemented all the software changes\update on the new board.
So, far nice! valid.x86.fr/1m4i2s I like it more then the bloodrage. Too bad I didn’t bought it back then, way less power drawn by the wall and by a lot.
Better bios interface with the hardware, sort of EFI and up-to-date solid component and line assembly, it’s also overclockable…
and stable unlike the bloodrage which is more a miss then an hit unless crazy voltages and clock are supplied.
Found a good all-time voltage spot @ 4GHz (REAL v-drop 0.896/1.224 | Bios V-CPU 1.054 = dynamic 1.0500 & high v-drop and low power idle) as shown in the valid.x86.fr link above for the x5675.
Added 16GB of more ram and will think about a more powerful or updated GFX card someday or the next year, probably by closing the loop and upgrading mine main.
Will check if I can enable AMT by adding the OLD emulated ME (or whatever is bootstrapped to-by the MCH) with an 32mbit chip and see the impact on term of stability and performance loss.
–
After reading few post on the forum I’ve investigated a little on the bloodrage an X58 board…
Turned out somehow AWDFLASH was corrupting always the bios on write operation causing the CPU PPM and other stuff to malfunction.
Except for a way higher temperature and harder low overclock along a greater amount of consumed energy it’s not really that noticeable, took me like 12 year to bare check LOL
Found out by reading-back one of the chip content with an in-socket programmer after flashing on-board newer microcode.
The DOS version of the Intel Flash Programming Tool (from version 5) can also eventually be used to write\read on X58 without error if the system is bootable and the descriptor is unlocked.
The ““ISA”” BIOS on the bloodrage is horrific anyway… I really need to retire that board for good.
@Lost_N_BIOS
The original stock bios for my NUC is this one.
The system was on a very old version (0024) before.
Exact (mainboard) model name is NUC6i3SYB.
The system wouldn’t boot some day. There was no monitor signal at all, only fan running, power LED on. I managed to find a combination of pulling CMOS battery, lifting the yellow jumper on the board (for BIOS recovery mode) and setting it back to get the Power Button Menu.
From there I tested a lot, but everytime I tried to access (write) to the BIOS a sudden reboot without monitor output was the result.
One time I managed to get into the Power Button Menu with the yellow jumper open. Then I tried to start a BIOS recovery / update with 0072.
After that the board seemed completely unusable. No chance to get a screen output anymore, only fan running.
@elektron
I’ve sent you few customized bios without ME to try out.
Probably none will work but if your NUC is bricked and have nothing better to do give it a try while you look for an full backup.