Thanks Fernando,
Here is the information in the policies settings of the two disks:



The difference seems to be the "Enable write caching field"
I had a look at the Intel forum, and they say the cache writing cannot be enabled by the user due to the structure of the Intel 750 itself (something about data protection).
I also tried to detach the 950, leaving the Intel only, and the speed is the same, so I guess it is not a matter of pcie lanes.

I'm having an issue with my ASROCK IMB-181-L that it wont allow me to insert even compressed as it says file exceeds the volume size.
It identifies the CSMCORE on 02:01-02

Edit:
I removed SATADriver and now it's giving me LIP about mebx ctrl + to enter but I can't enter the bios...
crap.

I checked both options for the Samsung as you suggested Fernando. As for what concerns the Intel, yes I used the latest Intel Nvme drivers (1.5.0.1002) but, having another look at what Intel advertises for the 750 400 gb on the box (2200 read, 900 write), my results does not seem so far. However, the Samsung is definitely faster in my x79 system.

Also, what would you suggest if I wanted to install Windows 7 on one of these drives in the future? (as a boot drive I mean). Logically it would have to be installed in UEFI mode but I read around certain cases of people able to install it in legacy mode by injecting the drivers. Which one between the two drives would you use in my case as boot drive and which one for applications or scratch disk? I mainly use the pc for work, doing 3D reconstruction for archaeological purpose, as well as video editing and photo editing of large full frame (Sony A7R2) and medium format (Phase One / Sinarback) files.

I inserted NvmeExpressDxEmodule GUID 9A4713C2, I prepared the bios as stated in the de-balling AsRock Guide, and flashed the de-balled firmware (from UEFITool) and removed SataDriver and replaced it with the compressed module. I will try to reflash the de-balled firmware over SPI with a BBB to try and get me access back to the machine.

I tried CTRL+P and F2 and Delete all of which do nothing from the machine it just sits at the POST screen without proceeding.

Yes the issue still persists.

edit: I am currently dumping the contents of the SPI EEPROM and will follow similar steps to that of my x200 flashing using flashrom to dump the image before flashing

Edit2: I can't get flashrom to detect the SPI flash chip :/ "No EEPROM/flash device found."

I followed the pinout here https://www.pjrc.com/store/w25q64fv.pdf but I may have connected it backwards the first time, since it's VDD and VCC are flipped from normal D:<

Edit3: PROGRESS!!!!


root@beaglebone:~# ./flashrom -p linux_spi:dev=/dev/spidev1.0,spispeed=512
flashrom v0.9.9-r1955 on Linux 4.1.25-ti-r62 (armv7l)
flashrom is free software, get the source code at https://flashrom.org

Calibrating delay loop... OK.
Found Winbond flash chip "W25Q64.V" (8192 kB, SPI) on linux_spi.

@Fernando I have gotten the system to boot again with the de-capsuled firmware, what would you suggest as the course of action at this point? The ME maybe messed up as the CTRL+P message is still there.

I'm guessing this will need to be run from windows on the host machine? I cannot boot from the nvme drive so I am unable to load any OS...

Okay but, as far as replacing modules how do I go about doing this so that the machine can actually boot? I followed the guide other than deleting the satadriver module and it would not boot.