Ah, ok. I'll wait for your result.
I was allready thinking, I was loosing my mind ;)

I only flashed the cleaned image you provided me, when we tought the me was not functioning anymore. We didn't knew that the oem had disabled it in the bios.

After that I only flashed the me once. I changed 2 min and max clock values and flashed that me. ( me only)

I build that one with the latest 8.1 tools we have on the forum.

So we should be looking for a newer 8.1 toolkit?

P.s. do you have a x79 bringup guide as well?
Because the 7 series bringup guide we use, only seem to have documentation for all 7-series chipset except the x79 series.

Is there any newer version of fitc for this me, that you are aware of?

I've got a fixed bios now.
But next week they will modify that one to enable the me again.

However,
I wanted to fool around, to get the bclk unlocked in the intel xtu.
So I hope we can get a newer leaked fitc.

Zitat von plutomaniac im Beitrag #32

---

Yes, it seems like a FITC bug or FITC incompatibility. I don't see any problem with the Z77 SKU and the me region has the same settings as the full dump at FITC. It seems to affect the Patsburg SKUs only like X79.

Intel changed the SKUs at FITC v8 multiple times which is apparent just by running older versions (even 8.1) which show slightly less or different C600 entries at FITC. Every version before the one we have (8.1.40.1456 - was part of 8.1.65.1586 firmware package) shows a missing/unknown SKU error when your PBG ME is loaded. It seems to me that 8.1.40.1456 does not show the error but is bugged and cannot actually understand the SKU which makes it restore everything to default without warning the user. You can even tell it doesn't understand the SKU because the ME region (me.bin) image shows an empty field at FITC SKU whereas the full dump (full.bin) does not. Personally, I would trust the FITC "view" (case scenario) which has yellow entries at ICC and other places. Ironically, that case scenario is the one after loading me.bin and not full.bin. So, I believe that your actual ME settings can be seen when loading the single me region (me.bin) and not full dump (full.bin) even though FITC seems to understand the SKU at the full dump and not single me region.



Either way, FITC v8.1.40 which we currently have, is either bugged or incompatible. Incompatible would mean that there is a newer FITC version which we don't have (for example the one that comes with packages after 8.1.65 and before 8.1.70 which seems to be the last). The image would be incompatible only if it was built with a newer than 8.1.40 FITC. From MEA, I can see that is was built with FITC v8.1.56 (newer) so maybe that's the problem in your case.

Regarding the FITC version shown at MEA: If you used the cleanup guide at some point with FITC 8.1.40 and ME 8.1.56 from the repository, then the FITC value will be equal to the firmware your chose as a clean basis (8.1.56 in such case) and not the actual FITC tool version. It's a small (harmless) visual side-effect of the cleanup guide. In such case, your image was not built with 8.1.56 FITC and it might not be incompatibility but just a FITC bug at the current version we have. However, I don't think that's the case because you shouldn't have used the cleanup guide with firmware 8.1.56 found at the repo since it's dirty (EXTR). Thus, I assume that the OEM actually build the firmware with FITC 8.1.56 and that version shown at MEA is not the side-effect product of the cleanup guide used wrongfully. Let me know if my assumption is wrong and you made such a mistake.

---

Zitat von plutomaniac im Beitrag #32

---

Yes, it seems like a FITC bug or FITC incompatibility. I don't see any problem with the Z77 SKU and the me region has the same settings as the full dump at FITC. It seems to affect the Patsburg SKUs only like X79.

Intel changed the SKUs at FITC v8 multiple times which is apparent just by running older versions (even 8.1) which show slightly less or different C600 entries at FITC. Every version before the one we have (8.1.40.1456 - was part of 8.1.65.1586 firmware package) shows a missing/unknown SKU error when your PBG ME is loaded. It seems to me that 8.1.40.1456 does not show the error but is bugged and cannot actually understand the SKU which makes it restore everything to default without warning the user. You can even tell it doesn't understand the SKU because the ME region (me.bin) image shows an empty field at FITC SKU whereas the full dump (full.bin) does not. Personally, I would trust the FITC "view" (case scenario) which has yellow entries at ICC and other places. Ironically, that case scenario is the one after loading me.bin and not full.bin. So, I believe that your actual ME settings can be seen when loading the single me region (me.bin) and not full dump (full.bin) even though FITC seems to understand the SKU at the full dump and not single me region.



Either way, FITC v8.1.40 which we currently have, is either bugged or incompatible. Incompatible would mean that there is a newer FITC version which we don't have (for example the one that comes with packages after 8.1.65 and before 8.1.70 which seems to be the last). The image would be incompatible only if it was built with a newer than 8.1.40 FITC. From MEA, I can see that is was built with FITC v8.1.56 (newer) so maybe that's the problem in your case.

Regarding the FITC version shown at MEA: If you used the cleanup guide at some point with FITC 8.1.40 and ME 8.1.56 from the repository, then the FITC value will be equal to the firmware your chose as a clean basis (8.1.56 in such case) and not the actual FITC tool version. It's a small (harmless) visual side-effect of the cleanup guide. In such case, your image was not built with 8.1.56 FITC and it might not be incompatibility but just a FITC bug at the current version we have. However, I don't think that's the case because you shouldn't have used the cleanup guide with firmware 8.1.56 found at the repo since it's dirty (EXTR). Thus, I assume that the OEM actually build the firmware with FITC 8.1.56 and that version shown at MEA is not the side-effect product of the cleanup guide used wrongfully. Let me know if my assumption is wrong and you made such a mistake.

---

@ring00 - If they are the exact same board, and ME works on one, you can just flash ME region to the other, either way of dumping/flashing with FPT below will work

To dump >>
FPTw.exe -me -d ME.bin
FPTw.exe -d SPI.bin

To write back >>
FPTw.exe -me -f ME.bin (or SPI.bin, using either file from above dump method will work with this command and flash ME region only)

I can extract ME region with UEFI tools - that MB had gone. Also there is just the same ME firmware in the repository. What is better to use? The question is to clean ME data region with data initialisation - I had such problem several years before, but don't remember the way.

ME FW in repo is NOT configured, use the ME FW from the working motherboard that has working ME FW, as you described above.
DO NOT clean anything, flash the ME FW as-is, how I described, from the board with working ME FW to the board that does not have working ME FW.
If they are the same board, using the same BIOS, then the ME FW will then start working on the other board too

[Guide] Clean Dumped Intel Engine (CS)ME/(CS)TXE Regions with Data Initialization
"If you want to flash a ME region from a dump then yes, you need to clean it first." How can You explain this post from
plutomaniac? I also met this situation and had to clean.