The Pascal generation GPUs have a low TDP, which enabled Nvidia to offer very similar configurations for desktop and laptop versions of their cards for the first time in history. I hope to be wrong, but sadly it seems we are back to the old days, with the desktop Turing cards vastly different than the laptop variants (my aging laptop's GTX 770M is way slower than the desktop 770, for example).

If that's the case, at least it would be good to see a naming convention that makes clear the performance difference, though the names referenced by Eurocom are a bit odd.

Interesting naming, I wonder if we can read anything into the specs regarding these names, I'll give it a go:

GTX 1170 & GTX 1180:  these might be the existing Pascal architecture just on 12nm rather than 14nm, with increased clocks and cores - i.e. not having ray tracing nor tensor cores.

RTX 1070 & RTX 1080:  these could be a cut down version of the new desktop Turing architecture.  Turing architecture uses more power than Pascal, so that's why they would need to cut it down for mobile.  RTX obviously referencing the ray tracing & tensor core aspects of the new Turing architecture.  So these would have less cores and lower clocks than the desktop RTX 2070 & 2080.

This could end up in the curious situation of the GTX 1170/1180 cards having more brute gaming power than the RTX 1070/1080, due to the fact that the RT cores & Tensor Cores have a high power usage budget; also, we're in the somewhat confusing numbering scheme that 1170 is a bigger number than 1070, so the GTX version has a bigger number associated with it, which maybe supports that it could have more brute gaming power.  This would seem like a strange trade off though, as it seems counter-intuitive that the new RTX cards would have less gaming power than the more basic & older tech GTX cards.

All just some thoughts!