Caporegime
On the flipside if they sell them too 'cheap' they'll be no one lining up to buy the 5000 series when software developers haven't, developed. If games in 2yrs time are no more demanding than games now then you could argue consumers will not need a more powerful GPU. Someone has to blink first, and as games currently stand I see very little use case for either the 4080 or 4090.
I think you misread what I said. Nvidia's pricing means that there is a stagnation in price/performance meaning lower end models will be similar performance(or a bit better) in RT performance for similar money.
Hence there is a stagnation,and games developers won't care outside the odd sponsored game. The RTX3060 is probably the most popular Ampere dGPU on Steam now.
It's no point if the RTX4050 costs the same and offers 20% better RT performance? This is why games are not pushing effects anymore - the mainstream/entry level enthusiast dGPUs are getting progressively worse and worse on average compared to the top end. Outside the odd blip,it seems to be very inconsistent in improvements now. In the past dGPUs such as the 6600GT,8800GT,etc offered a decent performance uplift compared to the previous generation. It gave a target for developers to aim for.
The 8800GT was one of the reasons why many people could actually run Crysis - if the pricing and performance per tier had stayed the same as the last generation,far less people would have played Crysis!
It's all about numbers - the vast majority of developers will only target fancy effects,if they know there is a big install base. If not they are not going to hamper their sales. Its why tessellation,which first appeared on ATI dGPUs in 2001,took nearly a decade to be commonly used.
Last edited: