|
To be fair, they achieved those ~30% gains by making the dies ~50% larger, so it was far from a free gain. Ofc that increased die size also went heavily to non-shader units this time around, the features cost NVidia money to implement so in a way it makes sense they cost consumers too, but if most consumers can't use the features yet then they're obviously not going to assign as much value to them as they cost in materials, so it's still a bad business move for NVidia as most consumers probably would have preferred "low fat Turing" upto higher pricing tiers for better value on existing titles, but conversely I'd say it's probably a relatively good one for the industry as developers will have had earlier motivation to commit resources to DXR as a result of the much wider spread of hardware accelerated DXR cards on the market and have had a much wider audience to test & receive feedback from especially in lower resource situations.
Essentially I think the annoyance was primarily with the fact some consumers felt they had little viable alternative at Turings launch, so they didn't really have a choice in whether they assigned value to the genuinely fairly expensive features, but of course everyone has the choice of just, not upgrading(If they didn't fancy what AMD had), which I'm sure NVidia weighed up when they took the risk of full fat Turing for mainstream consumer cards.
|
Linear Mode
