Paul MaudDib posted:

I think that's just Ryzen users though?



Maybe don't buy the budget processor and expect to play on launch day, is all I'm saying.

Paul MaudDib posted:

I mean, there wasn't a price premium for RTX cards this generation, AMD got greedy and increased prices to follow NVIDIA, it was pretty much just a question of whether you wanted the 2060S, the 5700XT, or the 2070S. Frames per dollar were within about 10% any way you went. Except the AMD cards came with poo poo-tier drivers that crashed constantly.

Equal pricing probably can't happen this generation though given that NVIDIA may have a huge generalized performance advantage in any engine where nvidia integrates DLSS one time and gets it "into the app store" so to speak. This is not game-specific code, it's going to be as widespread as TAA currently is, and NVIDIA will be knocking doors to get this in titles. If AMD doesn't make their cards quite a bit cheaper then they're going to be a hard sell given that NVIDIA gets to double their framerate in any DLSS title.

This will probably be widely celebrated as "AMD dropping prices and NVIDIA trying to gouge you for DLSS!11!!". But, same thing as usual, AMD reacting to a market disadvantage and dropping prices as a result.

shrike82 posted:

MLID is an odd channel - 70K subscribers and his videos get <100K views. Can't imagine it's worth his time. He should pivot to InfoWars-style conspiracy theories; that seems the way to get eyeballs these days.

Taima posted:

Even if relative stock for the 3080Ti is better than the 2080Ti launch, I’m assuming that quarantine will cause feverish, insane demand at almost any price point they care to set.

My plan is to try to grab an FE the nanosecond that’s possible, and if that fails, keep my ear to the ground on a few forums to track partner boards as they arrive. I fully expect this to be a fuckin’ battle.

I sat out the 2080Ti, what were the winning strategies then?

I got a 1070 right when they released by ocd trolling amazon listings / nowinstock

OhFunny posted:

I don't see nvidia stagnating both in hardware development or corporate culture (à la no price cuts) as Intel did with Jensen at the helm.

Nvidia immediately countered the 5000 series with their Super line and forced AMD to cut prices before launch to remain competitive.

K8.0 posted:

DLSS is a factor too. CP2077 has a good chance to be THE game everyone cares about benching for several years the way TW3 was, and Nvidia has a huge lead baked in. Unless they have some incredibly unlikely tricks up their sleeve, AMD is in a worse position in the GPU space than they've ever been.

Taima posted:

The craziest thing to me (besides the fact that DLSS2 is better than native, which still amazes me)

Mr.PayDay posted:

In the 2018 forum threads I was
But I was told I was wrong and Turing is an overpriced shitshow with features no one needs.
“DLSS is rear end, Raytracing costs 50+ % frames, this is such a fail, lmao Nvidia”.
That was the consensus and I was the dumb tech nerd who fell for Nvidia to buy an overpriced 2080Ti in their eyes.

Rinkles posted:

Maybe the AMD customer is more discerning about sharing data.

Whiskey A Go Go! posted:

Is "Moore's Law is dead" pandering to the same crowd? His videos always clickbaiting that Big Navi is going to crush X.

Zedsdeadbaby posted:

That would be hugely optimistic, let's put it that way. The new cards are competing with Turing, as much as loyal AMD users hate to face it. I don't expect anything much better than 2080ti-tier speeds on their fastest cards.

AMD simply aren't the ones to go to for the extreme high-end, they haven't been for years. They're all about the segments underneath it (and they do well there)

Cygni posted:

in what, tflops? if tflops were all that mattered, the Vega 64 would be faster than a 2080 Ti. it is not.

DrDork posted:

The XBoSxX claims 12 TFLOPs vs the 2080Ti's 14+ TFLOPs, so even on paper it's trailing by 15% off the start. The released clock rates of ~1.8Ghz is about where the base clocks of current Navi cards are, with boost clocks around 2Ghz. So if they just ported it over directly like that with ~2Ghz clocks, yeah, it could probably be somewhere close to or a bit below 2080Ti performance under optimal conditions (never forget that TFLOPs don't tell the whole story thanks to drivers and game optimizations--and RDNA1 tended not to scale nearly as well with resolution vs Turing in some games).

That they'll have something at least a bit higher than that seems natural. I'd expect that they didn't wring every bit of performance out of something they had to slap into a SoC and share a power budget with a CPU on. But how much higher they bother going? Total mystery right now, and prices are also an open question. If I were AMD and knew I had a competitive product, though, I'd start messaging that real goddamned soon, or NVidia is gonna release Ampere into an environment where people are salivating for new cards.

On the other hand, AMD has traditionally been quite ok with tapping out around the xx60/70 level, on the argument that that's about where volume profits start to drop off. Sure, the xx80Ti and Titan make $$ profits, but they don't sell anywhere near as many of them as they do lower-tier cards, where the "real" money is made via volume sales. I honestly could see them hanging around in the mid tier for one more generation, then using the hot cash injection they've gotten from Zen to try to turn it up to 11 for whatever their next gen will be.

BOOTY-ADE posted:

Same goes for video cards, people are sort of jumping the gun at the moment saying they'd gladly pay $1500-2000 for a card...forgetting exactly how many people have been affected by the pandemic (job/wage losses, no 2nd stimulus yet, no additional unemployment, etc). Unless people get some government help within like the next couple weeks tops (yeah loving right), I don't see NV selling many cards at those ridiculous prices.

shrike82 posted:

Because there might actually be games out that need the power in a year’s time?

Arguing that not rushing to buy Ampere at launch is equivalent to never wanting to upgrade is a galactic brain take.

repiv posted:

Flight Simulator doesn't even use DX12 currently, their first priority rendering-wise will probably be that to help alleviate the massive CPU bottlenecks.

Space Gopher posted:

Funny, I thought everyone who was buying hardware for day one sick graphics in Cyberpunk would have already picked up a 2080Ti for that April release. Maybe even one of those co-branded promotional cards.

Preordering games is silly, preordering singleplayer games is especially silly, and dropping a bunch of bills on hardware for your preorder is just ludicrous.

shrike82 posted:

if games don’t support DLSS at launch, that’s going to be a problem

imagine the lols if CP2077 doesn’t support DLSS or RTX at launch, that’s another game that’s been in the oven for a while. and there hasn’t been much preview-wise highlighting Nvidia tech outside of a couple screenshots

Kraftwerk posted:

So is the 3000 series cancelled now? I thought it was supposed to be awesome but it seems the mood here has soured.

Mr.PayDay posted:

So how many more weeks or months will AMD be stalling leaks and news?

Will they openly allow losing the 3070 and 3080 target group? Because they will if the rumors are true they won’t launch new AMD GPUs before November.

That’s like 2 months of Nvidia selling next gen GPUs without any competition.

DrDork posted:

I disagree. While I wouldn't be dropping price points or specific performance notes yet, they could certainly have started priming the hype train by now and at least keeping themselves in people's minds so they don't automatically just buy NVidia for a lack of any other options. They did this before, albeit poorly, with the "Poor volta" stuff. Instead there's been...nothing.

It's gonna be real hard for them to do anything useful once Ampere launches unless they're ready to commit to both price and performance targets like the day after Ampere is revealed, and even then they'd have to have cards near ready to go, which by all accounts they do not.

DrDork posted:

But to have pretty much not even mentioned that they're working on new cards for the last 6 months or so does not suggest much confidence in their product. Maybe that's the take, then: they know even Big Navi is gonna struggle against anything above the 3070 (and maybe even there), so why waste marketing dollars on trying to get hype and mindshare and attention when they're just gonna have to play the value card again?

DrDork posted:

Thing is, none of that's actually been shown. We have TFLOPs out of the XBox, which is...uh...vaguely useful in a general sense, but as we've seen before, AMD cannot always translate raw FLOPs into meaningful gaming performance.

And, sure, RDNA2 "exists." And that's...about it. No indication of how big they're even making it--the 80 CU part is a rumor, that's it. The only part they've actually put much into is saying that RDNA 2 should offer up to 50% better perf/watt than RDNA 1, which is cool, but doesn't say much for final performance.

AMD has said pretty much diddly about any actual PC cards using RDNA2, and instead has let Sony/Microsoft chatter about FLOPs and leave everything else as a wild unknown.


Eh. If an 80-CU part does exist, and if we can assume a flat 50% increase in performance (lol), it could very well beat a 3070. Of course, an 80 CU part would also be hilariously expensive, so it might be a case of a $1000+ AMD card trying to trade punches with a $600-$800 NVidia card.

Gabriel S. posted:

Is anyone dead set on pre-ordering?

Taima posted:

They are going completely and utterly out of their way to avoid saying absolutely anything about the 3090 and its weird as gently caress

They even seem to have taken it off that loving chart because it goes to 1400, the before-rumoured price, like they just decided to bump that poo poo to 1500 and remove it from the chart last minute.

shrike82 posted:

no new slate of DLSS games announced during the presentation which is disappointing - not sure why they aren't pushing partnerships with devs more aggressively

Mr.PayDay posted:

Herkelman (CVP & GM AMD Radeon) just answered with a 🤔

https://twitter.com/sherkelman/status/1300842481886662658
Is this a smirk “too soon” ? Is it defeat?
No counter ? No leaks ? Allowing AMD to lose the first wave to 3070 and 3080?

The AMD Reddit already had nice meltdowns and deleted hysterical posts and threads.

Or in other words: On a scale from 0 to 3090, how hosed is AMD now?

Subjunctive posted:

How do you think that AMD will manage to exceed NVIDIA on efficiency? What evidence has there been that they'll even be able to hit the 3080 mark ignoring DLSS capability? Many second-place market contenders have learned that just wanting really badly to catch the leader doesn't get you there, and AMD's record on this stuff--especially in terms of running cool--doesn't make me at all confident that they'll catch the 3080 in 2020 at least.

movax posted:

I was going to ask — one of the historical differentiators all the way back to the ATI R100/R200 days was that driver quality differed greatly. I've never first-hand owned or run an ATI/AMD GPU — is this historical FUD or do AMD drivers still leave a lot to be desired vs NV?

VorpalFish posted:

RDNA was basically at parity vs turing efficiency wise, and that was tsmc7 vs 12.

I don't think the situation improves vs ampere on Samsung 8. They may still be on a better process, but that gap is shrinking.

VorpalFish posted:

Given the amount of reliable information available right now, any claims made by anyone is just "blowing smoke." But I'd feel pretty confident betting that performance will not scale linearly with power consumption, and I'd also be pretty confident that rdna2 won't beat ampere in efficiency.

edit: most of that SOC power will be consumed by the GPU as well, probably putting it close to 200W, and on top of that, we don't actually really know what the gaming performance even is. Extrapolating that RDNA2 is somehow "fixed" based on the consoles doesn't track.

Some Goon posted:

Multiple mainstream reviewed reported crashes, it just wasnt apparent from the handful of reviews and limited testing how widespread and persistent it was.

repiv posted:

except moores law is dead, who should be thrown in the trash and never listened to again

he made so many extremely specific claims and none of them panned out

BurritoJustice posted:

What an unnecessary low shot at DF. Not the first time HU have slung poo poo at better reviewers without solid basis. He could've framed this in a way that is less accusatory and made a more compelling point

redreader posted:

Yeah I'm not ruling out the idea of getting an AMD card. My new lg monitor supports g-sync and freesync. Either way I should be able to get a massive upgrade over my 980.

gradenko_2000 posted:

Whatever AMD's ray-tracing solution is going to be, it won't use tensor cores, since obviously AMD's cards won't have them. The worst-case scenario would probably be something like an Ampere card being less performant than an RX 6000 if the Ampere card is trying to do ray-tracing without tensors, but I'd consider that really unlikely, considering Nvidia is ahead of them in this sector, and knows that it has to deploy its development arm to make sure that its cards are well-supported. And especially if the ray-tracing implementation is going to be done through a more generic API/library (sic?) in the form of DX12 Ultimate. And especially since ray-tracing isn't all about tensor cores anyway.

Rolling Scissors posted:

When people say this do they mean the future potential of DLSS is hard to beat? Because right now, the amount of games that support it, is frankly pathetic. And looking at the list of games that are coming out, it's not impressed either.

v1ld posted:

shrike82 posted:

FS2020 on a 2080 Ti runs at 35FPS at 4K ultra so that's going to be the game to watch

VorpalFish posted:

Who's actually saying that though? AMD being forced to offer better price/performance does not at all mean they're finished.

Malcolm XML posted:

If RDNA2 cannot at least be a decent competitor to the 3080/3090 then AMD is finished in the discrete GPU segment since there's very little reason to buy AMD if the development work is going nVidia, where you get value-adds like RTX voice and NVENC that simply won't work on AMD