JEMC said:
Captain_Yuri said:
So here's my thoughts overall.
Positives:
In terms of Raster, there's a good chance 7900XTX will get within 10-15% of 4090 while being much more efficient and cheap.
DP2.1 is something that I was ragging on about for the better part of the year so that's great as well
Having two 8 pin connectors or three 8 pins for AIBs is a win considering how terrible this connector is
Negatives:
Ray Tracing is absolute dogshit, make no mistake
Ai cores... Great but where's the Ai tech? Is FSR 3 going to be powered by their Ai cores similar to DLSS 3?
No Reflex? Like seriously???
Mixed:
Pricing. $900 and $1000. It feels like its one of those things that I should be cheering for but we know why it's priced like that. This isn't a situation where it's like oh, AMD is doing everyone a favor. This is a situation where Radeon is once again, a one trick pony. Nvidia prices their product the way they do is because they give you a next generation experience. Radeon once again however is meant for a last gen experience.
Conclusion:
RDNA 3's best hope is Nvidia not discounting the price of the 4080. If Nvidia sticks with the $1200 price tag, then I think AMD will start to steal some of the customers. If Nvidia discounts the 4080 to say $900 to compete against 7900XT, then AMD is in trouble. The 4090 should be the king in every area overall but it does come at a hefty price.
|
Bold: I thought HYPER-RX is their answer to Reflex. It should unify boost and anti-lag techs, and the later is what Reflex does, right?
And who cares what's best for RDNA3. What we should care about is us. Therefore, let's hope Nvidia does lower the price of the 4080 and AMD is forced to respond, to start a price war WE will benefit from. It's about damn time!
|
It's not clear what Hyper Rx is. From what I have seen, it ups the framerate while decreasing latency which isn't what Reflex does. Reflex only decreases input latency due to managing the render queue so there isn't any bottlenecks.
Because Hyper rx increases the framerate in order to decrease input latency, it sounds much more similar to AMD's Boost. What Boost did is dynamically lower to render resolution to meet a certain framerate. But no one used it in practice because it lowers the render resolution, sometimes to insanely low levels which makes it impractical for most situations.
And yea, competition is great.
