No-one has considered the other alternative? NVidia is out for blood and looking to annihilate AMD so they are pushing wattage higher?
"Poor Volta"
AMD's hype is saying 50% better performance per watt them RDNA. That would put a 300 watt card roughly on par with those 3DMark scores you are talking about which most people I saw thought was the 3080, not Ti/3090.
The problem is that AMD has been hysterically dishonest with their hype in the GPU segment in the past, they threw down poor Volta years ago and have since introduced three new tiers none of which has matched the hype they claimed.
This time they claimed "nVidia killer", perhaps they just ticked JHH off enough that he's out to humiliate them this generation and reinforce the widely accepted view that AMD just can't compete.
Do I think the above is going to happen? No, but it's far more likely than what you people are setting yourselves up for.
Go back and check the Twitter claims again, the same guy saying it's 100% SS8 was saying no SS8 all SS7 previously. Then he swapped to SS8 and now he comes out with this 600 watt power plug? Sounds like someone is mole hunting.
Even if we assume this is a SS 8nm, SS 10nm was comparable to TSMC's 10nm, and supposedly this is on 8nm EUV which is a very large step up over "12nm". That would still be a full node drop and simply looking at history a full node drop for nVidia has resulted in *massive* gains every time. Meanwhile AMD is claiming a 50% performance gain per watt on a slightly refined node- something they have never come close to demonstrating they could do.
Despite all of this, you guys are truly trying to convince yourselves that your scenario has a serious chance to end up right?
Upside to all of this, everyone has pages worth of posts so when these products hit we can all see exactly how good everyone is at speculating.
On mobile and quoting sections is a bit of a pain, so I'm gonna be a bit lazy. I'll edit this into proper forums formatting in the morning, it's late and I'm half asleep.
> No one has considered the alternative...
Yeah, of course we have. Actually, Kopite isn't the source of the retaping rumours, those existed before he said anything, and the alternative where Nvidia actively tries to prevent AMD from taking any crowns is what I thought was happening. I thought NV were switching from 7LPP to N7.
> AMD is saying 50% power efficiency gen on gen... on par with what most people thought is the 3080
What most people think is irrelevent as it's spawned from baseless expectations of next gen.
It's also wrong.
> Prior bad marketing from RTG
You know the whole 50% perf/W statement? Let me run through what it's based off:
One RDNA and one RDNA2 chip. Both are at the same clocks they will release at. Both using the memory configs they will release with. Both of which are compared at the same CU count, the larger die will have CUs disabled for the test if needed.
In terms of power pulled by the GPU core alone, the 5700 and Series X should both be roughly on par at each other within the 120-130W area. The 5700 has a max boost that's 100mhz lower than the Series X and a "game clock" that is 200mhz lower than the Series X... and 16 less CUs on top.
That clock difference doesn't seem like much until you realise it's still 70mhz above the much more power hungry 5700XT.
Why am I bringing this up? Because at the 1825mhz of the Series X, its clear that there's actually a greater than 50% perf/W efficiency gain. I'm a bit lazy to try and do actual maths, but rough guesstimation says it's more like a 60-65% perf/W gain.
You can bring up the past, but it's generally not advisable to live in it. I'd instead look at the evidence we have today.
> Claimed "Nvidia Killer"
Man if Leather Jacket man got seriously annoyed at what might potentially be an internal codename at AMD and not one they ever actually publically made then maybe he should stop slinging out complete rubbish like next generation consoles being weaker than a 2080MQ.
> Twitter guy said no 8SS only 7SS but now is saying only 8SS.
Firstly the names are 8LPP and 7LPP ffs.
Secondly we've discussed what happened already. Something something 7LPP is a flop, something something retape. You get the idea.
> 8EUV
Doesn't exist.
> Full node shrink yadda yadda
Yeah, and you'd have seen major perf/W improvements if clocks weren't ramped up to extreme degrees, but alas.
Also, 8LPP isn't a full node shrink either way. It's extremely close, but quite not a full node shrink.