-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
Plenty of other leaks around all seem to corroborate that in raster 7900GRE≤9070 XT<7900 XT.
Last edited:
Associate
- Joined
- 7 Nov 2017
- Posts
- 1,929
so no actual info then.
Leakers, because as we know they're infallible with their info, despite getting specs wrong, release date wrong, name wrong etc etc etc.
That you are, sir.
Err, I am a trusted leaker according to a person on the internet..
and it's not just my old age.
If I'm honest, in the last maybe 5-8 years, leaks have been pretty accurate and correct for virtually everything.
The biggest difference between leaks no and say 20 years ago, we are a more connected world and info coming from some part of the world is much easier these days.
It is important to note that everyone favorite youtubers that test these things have basically holed themselves in testing the same way, just because someone tests something differently doesn't mean its incorrect.
Even the games being tested are meh a lot of the time, half of the testers benchmark CSGO but not other PC exclusives, so theres that too.
We can pretend AMD is doing well but the sad truth is the last Radeon gpu to make a top 20 spot in the steam survey is the RX580.
I want AMD to do well as we badly need some competition in the GPU market but they can’t keep messing up the pricing every generation.
If amd wants a top spot again, price the 9070XT at $299
Last edited:
Anyone can be "accurate and correct" if they change their info dozens of times over the months leading upto a release. And more often than not that's exactly what they do, come launch day or review embargo day they point to the latest info that was close or maybe correct, and the rest of the stuff they were nowhere near gets brushed under the carpet.
If you were to list the stuff they were totally wrong about vs the stuff that they got right, or close to right, the stuff they got wrong would be a lengthy list indeed.
Last edited:
If performance sits between a 7900GRE and 7900XT but with better RT and upscaling then the 4070S has already been offering that option for around £500-550 and sometimes under £500 for the past year so I think anything over £400 will be DoA.
Soldato
Theory is AMD botched their dual chip high performance design and instead put them chips to the 9070xt and then spin some bs about focusing on the mass market. No market disruption, the status quo it is and more market share loss.
Sounds possible. But who really knows.
Were they not meant to have two GPU's on the same package though? Rather than two separate ones like the 295x2?
I have wanted them to get that and have the advantage over nvidia for a couple of generations now. But honestly I just lost all hope with AMD GPUs. If they ever do well it will be a surprise. I won't hold my breath however. Something just not right about the team over there. Certainly the marketing team anyway
Last edited:
Steve from hardware unboxed spoke about this two months ago while he was on the mlid podcast. Steve visited a trade show and there he says he spoke to some AMD guys, as well as other journalists and the mood was depressing and word going around was that AMD's experiments with chiplet gaming GPUs was a failure and didn't work out the way they expected, as a result RDNA4 was going to flop and they needed to pivot back to a monolithic GPu to salvage something from it
If I recall correctly, even rdna3 was a bit of flop, in absolute terms - simply because the 7900xtx that launched was nowhere near the initial predicted performance and depending on which rumours you assume are more accurate, the 7900xtx was meant to be 50% to 100% faster than what it launched with (and this was because the 7900XTX was meant to be a dual GPU chiplet card)
Based on statements from MLID, AMD would work towards correcting the issues and RDNA4 would be RDNA3+Fixes to deliver the original performance that was intended for the 7900XTX. As we've gotten closer to launch it became apparent that these "fixes" did not materialise and whatever was wrong could not be fixed.
Last edited:
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 30,999
Clickbait **** for the softbrains.
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 30,999
Think it was AiB's he spoke to during various dinners, and that RDNA4 had turning out quite disappointing and not to expect too much.
Also to note, took AMD about 14/15 months of driver updates to get RDNA3 anywhere near the slide numbers that they touted at launch so at least they got there in the end. AMD driver guy got a bonus that month
Last edited:
100% this. David Wang said a few years ago that this was impossible due to the ridiculous bandwidth requirement needed to do chipsets for GPUs.
Soldato
Fine wine, launch disappointing and get it where it needs one day far in the future.Think it was AiB's he spoke to during various dinners, and that RDNA4 had turning out quite disappointing and not to expect too much.
Also to note, took AMD about 14/15 months of driver updates to get RDNA3 anywhere near the slide numbers that they touted at launch so at least they got there in the end. AMD driver guy got a bonus that month
