@eivrah
Try to verify if the content flashed is the same of the file by dumping and compare.
The board could also be incompatible with certain PCI-E add-on, XHCI are know to cause power issue.
Dunno about SATA type.
@noInk yep I verified it and it was the same. ohh bummer maybe I couldn’t use the PCI-E sata 3 then.
@noInk tried removing and adding ram also tried the jumper in different position no luck it didn’t boot also compared after flashing it was same any thoughts? Can I try your backup bios?
It boots now @noInk what a dumb board it is I flash a lot of files and compared the readback bios non of them is identical. What a weird bios Intel has. Also the PCI-E to Sata 3 is working just need to be in Legacy mode.
@eivrah
Do the SOL bios work now and you sure of the content by byte x byte?
If so, keep it on and try to not use the EV jumper. Also, re-test the SATA addon. I recently acquired a 501 revision and it boot fine ( i7 and XEON ) without the EV modification.
Anyway, I’m having issue with the hardware programmer, the FlashCatUSB 2.1 I got doesn’t seem capable to manage in-circuit SoIC what you are using to read and write in place to the 501?
@earthoo7
Technically I think what you’re asking is possible, but it is very difficult. If you study the structure of a dumped Intel bios and the equivalent .bio file from Intel you can often find the same volumes and structures. A bit of work with a hex editor and FIT you might be able to reconstruct something - provided you had a reference dump to begin with, or you’re happy to keep making educated guesses.
But it’s difficult because sometimes one Intel .bio file is used for different motherboards (eg one .bio for DX79SR, DX79SI, & DX79TO). In that case it is really difficult to see what volume belongs to which motherboard.
I believe the simple answer to your question is: no.
@noInk I’m only using Ch341A with Clip, it’s cheap and does the job. I did compare it, not 100% identical when I checked via winmerge some but most of the lines are the same. My board had a custom Logo I made, so I can confirm that the SOL file is in it now cause it changes the Logo. And yes I remove my soldered pins, my i7-950 and Xeon x5675 works. My Sata add-on also works. Let me know how high you can clock your new board. I can only go 180 multiplier with mine higher is unstable, my RAM might be the problem.
@eivrah
Well, I ordered one CH341A since getting the SPI socket on December from the other side of the world is tuff.
The FlashCatUSB 2.1 just doesn’t work with or without external 3,3VCC on basic things like that, dunno for more advanced. I also shorted out all the 501 quartz crystal just to be sure…
The funny side is I can detect the chip on GROUND and the CHIP SELECTOR without any VCC (no power cable or battery)
kinda pissed by the disorder having stuff all around since I paid like a lot plus the shipping cost to be sure to not get issue like these.
I mean, what an hardware programmer does except reading and writing stuff? taking space? wasting time? MEH
Can you make a diagram or take a photo out the way you are connecting the CH341A on the IC CLIP and also send me a link to the software you used?
Thanks!
@noInk CH341A Program
I made a quick diagram but I only emphasize the red wire should be connected to Pin 1 and on the BIOS chip with the dot, the rest on the other pins:
Here’s mine
@eivrah
Thanks a lot, gonna try it once it arrive. Asking cause I don’t think is coming with any driver or software.
The clip is already attacked. Will report back.
–
The CH134A didn’t work either.
I desoldered the 25x16 form the 501 and the PIN 3 and PIN 4 were GLUED to the PCB PAD underneath the leaded free (tin/zinc) solder.
So, to anyone reading… don’t do that without lab experience or lucky fell, dunno if it was done by manufacturing process or by someone after. Overall the board was in very bad condition.
Also watch out for the nearby PCI-ex x16 connector: the plastic gonna melt away at 270° already (super cheap!) require insulation.
END of the story;
I’m going to use(NOT that I really do) the 202 without overclock-ability since it has already the SPI (SOP)socket and it’s way lot faster to flash\reprogram.
Thanks by the way.
–
I was finally able to get an untouched bios from a working DX58SO by hand, following the full modified programmable file with some squeezed microcode in (it’s pretty much the same as the one I’ve been re-building from since I got the board)
DX58SO 501 UPDATED SOL
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
PCH straps: 1
PROC straps: 1
GbE region offset: 1000h
MAC: 70:71:BC:76:BE:A5
Version: 16.8
BIOS region offset: 3000h
Region access settings:
BIOS: 0Bh 0Ah
ME: 0Dh 0Ch
GbE: 08h 08h
BIOS access table:
Read Write
Desc Yes No
BIOS Yes Yes
ME No No
GbE Yes Yes
PDR No No
Flash chips in VSCC table:
C22015 (Macronix MX25L16)
1F4600 (Atmel AT26DF161)
EF3015 (Winbond W25X16)
898911 (Intel 25F160S33B8)
BF2541 (Microchip SST25VF016B)
CPU microcode
CPU signature: 000106A5h
Revision: 0000001Dh
CPU signature: 000106A4h
Revision: 00000013h
CPU signature: 000206C1h
Revision: 00000006h
Additional CPU microcode
CPU signature: 000206C2h
Revision: 0000001Fh
My NUC6i3SYH stopped working on reboot, with no monitor output, no USB Power. Fan running, power LED showing blue, no blink codes. It is out of warranty, about 3 1/2 years old.
After disassembling the mainboard and many testing hours with the yellow jumper disconnected and set back, disconnecting/reconnecting CMOS battery and
pressing power button, I found a sequence, to (sometimes) boot to the restore menu / UEFI.
But changing settings in UEFI or flashing a new BIOS file is not possible, because the NUC won’t start up on next reboot. But obviously it is not possible to flash a new .bio file without rebooting the machine.
It seems, that part of the UEFI has been corrupted in a way, that makes changes and regular boot impossible.
The current installed version is 0024 (very old).
It has a W25Q64FV bios chip I could try to replace (if this makes sense at all), but I don’t have a spare NUC hardware to take a working chip from.
What could have been the problem I also saw someone posted that his BIOS flasher didn’t work on DX58SO. Have you tried replacing the clip wires to a shorter one? Maybe your clip has a longer wire attached same with others.
@eivrah
No clue, maybe the color used to mark the bios chip also indicate how it’s soldered\glued.
That 501 board had an red mark but most of the online photo are blue marked on PIN 1.
I’ve used 3.3volt 150mA provided by the flashcatUSB and also 3.3volt from an external power module with 700mA but in both case nothing was readable with the VCC connected.
The CH341A didn’t even recognized the 25x16 in all scenario.
If I find another board for a reasonable price and without broken stuff I’ll check again.
Anyway, SOP socket for bios chip should be mandatory for consumer board, not having one is like asking for issue when the manufacturer drop the support few month later or the warranty end.
It also fall in the GPDR stuff since user data/telemetry and other things can be written on some partition\volume.
@elektron
See these pages and the linked resource BIOS Recovery by Power Button Menu or BIOS Recovery by Security Jumper
I’ve already done what the intel instructions say.
But as I tried to explain in my first post, this is no option, because the system is not surviving a reboot, after trying to reboot it always hangs in a no screen, no USB power, only fans running state.
Then I have to put jumper and CMOS battery away and back and power cycle many times, only to get back to the Power Button Menu.
Regular BIOS recovery/update is not possible without reboot.
@elektron
With a corrupted bios the display will never turn on with or without any procedure provided by intel.
If something is shown on display somehow;
Detach the NUC from the AC/POWER
Remove the configuration jumper
Insert a FAT32 USB with only the *.BIO file for the NUC
Attach the AC\POWER and turn it on (do not interrupt)
This procedure does not work for me.
It simply does nothing.
More likely it’s a hardware issue then. I found an interesting discussion about a hardware change for the skylake NUCs:
https://forums.intel.com/s/question/0D50…?language=en_US
Perhaps the aging capacitors may be my problem, preventing the system to run… ?
@elektron
If it worked before with that memory module unless the board get overclocked by default on newer release it shouldn’t be an issue.
The capacitors change will just increase the voltage capacitance to allow better clock. Mine 202 board suffered a similar fate ( it stuck at 135MHZ ).
Is the UEFI shell available? what the working option? can you take a photo? If you are able to display something either find a way to flash the bios or make a backup with an hardware programmer.
I did one more attempt to boot and flash and tried to document the necessary steps.
The system only shows the Power Button Menu after the CMOS battery and the yellow jumper was removed and set back. This needs several attemps until it works (until LED goes orange, then booting to Power Button Menu). This only works once a time.
After one reboot the screen is off, the whole procedure has to be done from the beginning (jumper, CMOS Bat, …).
So, I managed to get into Power Button Menu and from there with USB-Stick and Iflash2 started the Flash progress. The system seems to read the data from USB and then goes to restart. After that the system stays off (no fan, no USB power) with a orange blinking LED (2 sec. interval).
@elektron
If the flash end successfully with a message and it doesn’t boot after there might be an hardware or software issue.
In both case if it’s either hardware or software you must check integrity. What you do next depend by how much time and resource you gonna pour into getting it back.
Mine board was a pre-production board likely made to run the first batch of a new architecture, so, tossing time for finally sacrifice another sort of working cheap board to see the difference was due.
I’m not an expert and I’ve got no free inventory to check on possible behavior, also, without the actual dump, it would not possible to check on anything.
To restore your bios from zero in case of software failure you will need the .BIO and a working content backup from a donor twin board.
There is no message or anything else that indicates, that flash process was successful. It reads from attached USB stick for a short moment ( 1-2sec.), screen goes blank after that and system stays off in "orange blink mode".
When BIOS is in updating process it should give a 0.5sec blue blinking pattern, I think.
My next attempt would be to get a cheap hardware flash programmer, but if there is no hope without a second board to take a flash image from, or if it’s probably a (bigger) hardware issue, I think I would give up.
Too bad for this little NUC, that’s really not old enough to go to trash.