AMD Radeon RX 9070 benchmarked in Call of Duty, could be a match for Nvidia's RTX 4080 Super

AMD's handling of release of information, good grief. Personally, I feel it would have done them better to have released Benchmarks vs their own 7000 series cards (leave nvidia cards all together out of the picture, no need to run free ads for nvidia after all). So that being said, if we take pre-information info and compare 9070 XT to 7000 series it's almost a modern next gen 7900 XTX and say that as it looks really close in raster (we may see with real drivers it will beat it) however better in all other aspects from rays and fake frames etc... If AMD ran with a $479 2nd Gen 7900 XTX paints an interesting offering vs 4080 and 4070 comparisons and then nvidia saying how much better 5070 is (even if it isn't) paints AMD 9070 XT as a low-class product -- AMD, drop that marketing and move to the idea its $479 (2nd gen 7900 XTX) GPU.
 
Then, nine months later when it actually DOES drop to $449, everybody says "great, now that it's even cheaper, hopefully it will drive down the cost of the 5070 so I can buy one" while simultaneously claiming that the reason AMD cards sell less than nVidia is because they have inferior hardware and don't have aggressive enough prices.
And then when Nvidia street prices actually go up not down, still buy the 5070 and still complain AMD should have done more to bring the prices down.
 
"The RX 9070 averaged 99fps in native 4K in Black Ops 6's in-game benchmark tool using alpha drivers. For comparison, the RTX 4080 Super can reach roughly 129fps in 4K in DLSS quality mode, which upscales the game from 2560 x 1440 resolution." ...

You can not compare a GPU that cheats on resolution then call it a 4k resolution by using upscalers and say, OH this gpu has 129 FPS compared to competition, DUDE, call it a spade if it's a spade. I could do the same, use FSR to upscale 4k from whatever resolution I need and get a bazillion FPS, what is the point of the F* TEST THEN ?

Let's go back to RASTER REAL WORLD PERFORMANCE tests that dont use inflated fake resolutions like upscalers do.
 
Back