Monday, April 25th 2022
AMD Radeon RX 6750 XT Surfaces on GFXBench, 2% - 12% Faster than RX 6700 XT
The upcoming AMD Radeon RX 6750 XT has recently appeared in a benchmark listing on the GFXBench website where it performed 2% faster than the AMD Radeon RX 6700 XT. The Radeon RX 6750 XT is an upcoming refresh of the RX 6700 XT featuring faster GDDR6 memory running at 18 Gbps and increased clock speeds. The RX 6750 XT achieved a score of 366.5 FPS in the single GFXBench 5.0 Aztec Ruins High Tier off-screen entry which is 2% faster than the average RX 6700 XT score at 362.2 FPS for the same test. This initial test is not likely to be a representative sample of the card's performance but we would expect further tests to leak in the coming days. The AMD Radeon RX 6750XT is set to be officially unveiled on May 10th and will gradually replace existing RX 6700 XT offerings.
Update: The card has also been tested in various other GFXBench 5.0 tests including Aztec Ruins Normal Tier, Manhattan, and T-Rex where it achieves scores 6% - 12% above the RX 6700 XT.
Sources:
GFXBench 1, GFXBench 2, VideoCardz
Update: The card has also been tested in various other GFXBench 5.0 tests including Aztec Ruins Normal Tier, Manhattan, and T-Rex where it achieves scores 6% - 12% above the RX 6700 XT.
34 Comments on AMD Radeon RX 6750 XT Surfaces on GFXBench, 2% - 12% Faster than RX 6700 XT
They are just there to get reviewed and occupy the space and get some press attention while we wait for next gen. They are just there because yields are better now and they can improve the clock a little bit.
As for the price, they will be priced according to the market or they won't sold any. Availability is getting there and you can now buy a GPU easily so the scalper are having a hard time these days.
www.newegg.ca/p/pl?d=rx+6600&Order=1
only $80 above MSRP now lol I will keep waiting
the RX 480 - 580 - 590 were roughly in the in 5-10% increase neighbourhood
www.techpowerup.com/review/sapphire-radeon-rx-590-nitro-plus/32.html
This is 55% scalping.
The RX 590 was on a tweaked, smaller process node that was actually different (moved from 14LPP to 12LP+) but I didn't ever mention that - you've just brought it into the discussion for the first time, so how is it relevant here or to my point?