It's not really, Yes the 1070 is faster but as Digital foundry showed us with the Fury X v 1070 in Doom, When using Vulcan it has the Fury x well ahead by something like 25% meaning the Fury pro would edge a win too, Now while that's only one game and every other game today has the 1070 ahead having a 1080p freesync monitor and looking ahead the Fury will be great for a few years yet, especially with the adaptive sync advantage he loses by getting an Nvidia card with that monitor. So if he plans to stay on a 1080p freesync screen for a couple of years and wants a graphics card that will do a good job for a couple of years it's a good choice, It'd be better with more ram but it will be okay still. If not an 8gb 480 would also compliment the 1080p freesync panel very well and if I was buying today I would go with the 8gb option. If someone wants to spend there game time staring at an fps counter and going "Ooh this is ten frames faster than if I had the other card" then fine pay the extra for a 1070 but actually playing games without an on screen counter and you'll be hard pressed to tell the difference.
I think there are issues with your argument, and i think you have misinterpreted the results from that particular benchmark. Yes the FURY X (not the card in question) pulled slightly ahead of the 1070 (not an overall performance of 25%, rather a boosted 25% under vulkan which took the Fury X to just above 1070 performance). Additionally, you mentioned that this is only one game.
However, obviously the poster wishes to use AMD for his monitor.

