So if you ignore the stuff that's been in the channel for 2-3 years, the 2070 and 5700 XT start to look okay? Not really. There's still the 1660Ti, which is probably NV's most-compelling buy right now. Do you really think a 2070 is worth ~$170 more than a 1660Ti right now, in terms of raw performance? Where's the "bang-for-buck"? If all you care about is that, you'd be getting the 1660Ti. Nothing on the AMD side looks good in comparison (unless you're going low-end, in which case some of those $160 RX 580s look okay-ish). Only people slumming around for $300 Vega56s that they can overclock can really challenge it.
I have no idea what point you're trying to make. It looks as if you're trying to say something that is opposite of what I suggested, but I don't see it.
I never said the 2070 or 5700xt are ok value.
I'm addressing what I think a lot of buyers are
not doing, which is what you seem to be doing; looking at a company's progress of a product type, getting underwhelmed, and then using that as an argument for not choosing those products.
IF a comparison of price/performance shows a 1660Ti is the best value then a lot of buyers will choose that - regardless of nVidia's history.
IF a comparison of price/performance shows a 5700xt is the best value then a lot of buyers will choose that - regardless of AMD's history.
('all else being equal')
That was my point.
And of course, not everyone will simply choose the best price/performance product, a fair amount of users will have a budget and will try to find the best value within that budget and its priorities. If the budget says they can spend $200 that's one thing. If the budget says 'infinity dollars' for the GPU it's another. And if the budget says $450 then they'll get the best performance possible for that amount of money.
IF that happens to be the 5700xt then that's what users should buy. It's not going to matter that it's an "underwhelming" progress from the Vega 64 for example, because at the time of purchase it's the best product within that budget (not the "if" before).
Really the question should be posed conversely:
- Why should a user NOT buy the 5700xt
IF it performs the best for $450 and that's what the user has and wants to spend on a GPU?
I mean, imagine what an odd conversation it would be if you recommended someone to skip it if the premise above was true;
"Hey I think I'm going to buy the 5700xt."
"No, don't buy it."
"Why not? I've got $450 to spend, and it's the best performing card for the money."
"Because Vega 64".
"Huh?"
"It's a poor upgrade over the Vega 64."
"But I don't have a Vega 64... What's a better card for $450?"
"The 5700xt is the best value for $450"
"Ok, so get the 5700xt?"
"No, because 'underwhelming'".
"???...."
The logic in the argument isn't really there in my opinion. That's all I was trying to say (and now I said it again with a lot more text... sorry)