M70q Gen 1 - NM-E101 board, weird config?

[acotam2]

Hi, I bought a used M70q on Vinted and recieved an E101 board, so here are some high res pictures (feel free to use them):

(Those are the pictures i took after some tinkering with the PCIe slot and configs, sorry for the mess)
The main difference I see is the lack of the WWAN slot and some other unused footprints around the front I/O (and I think some change in the audio circuit - two additional coils next to the Realtek ALC chip)

The next thing I did was to check the resistor config to check if it's not the amazon revision, and i saw that:

And the config doesn't fit any known setup listed on the schematics? I currently cannot check the BIOS behavior while changing the config, because I don't have an LGA1200 CPU with iGPU, and trying to solder a mining riser to the port doesn't seem to work (I might have simply missed something, idk). Gonna update you in couple of days when other CPU will arrive.

Here are the labels - it seems to be a base config with G5905T, 4/128GB:

[a-little-wifi]

Very interesting, thanks for posting! I checked Lenovo support's images of NM-E101 boards and they have the same config straps as your image, interesting that it's not listed in the schematic,. Taking a look at NM-C621 board pics the resistor config is 110100 (same order as in your picture), which matches the M70q LI entry as expected. NM-E101 seems to almost match M70q LC except for ID3. It also looks like the IDs in the table are sequential, with this one coming 2 positions after the last one in the table.
I would guess it's a version they added later after the table was already embedded in the schematic.

Does make me wonder what the relevant difference between this one and NM-C621-based M70q's is, from the bios perspective. Only thing I can think of is disabling the non-routed WWAN PCIe lane, but that slot isn't populated on NM-C621 models anyway...
Is the audio IC the same model as on NM-C621? I can't read the numbers on the chip in your photo.

About the riser slot footprint's PCIe not working, I think that's because it's connected to lanes 8-15 of the CPU on these boards. The routing looks identical to NM-C621/NM-C901 at least, where this is the case. Using it normally would require the CPU to be set to x8x8 or x8x4x4 bifurcation, but the H470 chipset in the M70q doesn't allow this. I'm not sure if anyone's tried setting the bifurcation strap anyway despite the spec saying it's not supported, would be interesting to know how it behaves.
Alternatively, I think if you set the PCIe lane reversal strap you might be able to use it, but with the lanes reversed of course. You'd have to solder the mining riser to the last CPU lane of the slot.
It also looks like H470 does support the lanes for the riser slot's PCH link, so you could also use that.

Thanks for pointing out the removed WWAN footprint btw, not sure how I hadn't noticed that before. I'll have to see how I work the difference into the board info table lol.

I'm looking forward to hearing about anything more you find out about this board! Thanks again for posting, and thanks for the pictures, I'll add them to the boardpics directory soon.

[a-little-wifi]

I just checked M80q gen 1 board pics as well for the config straps and they are as follows:

M80q NM-C621
110010 -> M80q LI

M80q NM-E101
111100 -> between Tiny Amazon and M70q NM-E101

So it looks like the NM-E101 board configs were both added on to the end of that table, with the M80q coming before the M70q (the regular M80q also comes before the regular M70q in the table, so that makes sense)