Fury X has joined the game

[Rolf]

That's what W1zz @ Techpowerup says.

[wgmmmx]

I think we will see real spec after compute spec avalible to this GPU.
Strange is that card that AMD claim that got 8,6TFLOP power got same or lower power in game like card that got 6,1TFLOP power(titan x).

[Flomac]

One TFLOP just means "Floating point operations per second" and that's a very rough and simple and very much telling figure of computing capacity. Especially gaming is complex with many factors to consider. Just like a car with 300HP is not necessarily faster than one with 200HP.

[epixoip]

All of the reviews I've seen today prove that this card is exactly what I said it would be. It's a 450W monster that violates the PCI-e spec and makes extensive use of PowerTune to attempt to keep things in check. So there is absolutely no doubt in my mind that it will be just as worthless for Hashcat as the 295X2, possibly even moreso with this next version of PowerTune.

Are you sure this card is GCN v1.2 ??
I see here that they write that is v1.3

It genuinely makes no difference whether they call it "GCN 1.2" or "GCN 1.3" -- it's all the exact same architecture as the HD7000. Nothing has changed. They're just tacking yet more cores onto an architecture that is now 3 generations old (which is what I believe the 1.X nomenclature is all about -- the X indicates how many generations they've been running with the same architecture.)

[c4p0ne]

Unless they do something really dramatic, and very rapidly, AMD is finally dead. Nvidia killed it.

[Flomac]

So there is absolutely no doubt in my mind that it will be just as worthless for Hashcat as the 295X2, possibly even moreso with this next version of PowerTune.

I see it the other way round. Having a cooler capable for 500W but producing only 350W with some peaks on 450W, the card shows no big need to throttle. It needs less power then the 295X and stays cooler, though having the same capabilities. This card has now power problem like the 295X and it does not have a heat problem thanks to a well-sized cooing system.

To me that means it can ride its power all way long with only little throttling. Changing the cooler-fan is obvious. It brings down the temps by 10-12° and though noise goes up it'll be just on the level of a 980Ti or maybe even lower depending on the fan you choose.

As always that card will outperform NVidia in some parts in hashcat and so will NVidia in some other. That's nothing new, every system has its favorabilitis.

It shows to me AMD is on par with Nvidia in performance/price ratio, though of course this card will suck more power so I'd prefer the 980Ti. But did AMD completely fail? Hell no, especially not when you're a gamer and it might be a good solution on hashcat as well, because they got temps and power under control (like they did not with the 290X and 295X).

[Rolf]

PCI-E specifications state that one 8 pin cable can supply up to 150W of power, with the PCI-E slot on the motherboard providing up to 75W of power.

One Fury X has two 8 pin connectors, which means, according to specification, it should draw at most 150+150+75=375W of power.
Instead of putting a third 8 pin power slot (ol' GTX 480 Lightning had 2 x 8pin and 1 x 6pin connectors, for example), they violate the spec, with unpredictable potential damage to the hardware.
In order to minimize the damage from the violation, they throttle the power draw using software/firmware shenanigans.
And they market this BS as an advantage.
Jeremi told me that it wasn't the first time AMD violated the PCI-E standard, so no surprises here.

Additionally, AMD knew darn well what could come out of the violation, so they decided to additionally minimise risks and go the "liquid cooling 4 stock" route.
Liquid cooling solutions I know of can dissipate 300~2700W of power (ERM-3K3UA a notable example for the latter).
But good luck even trying to install 8 Fury X cards without custom racks.
A card which cannot reasonably be cooled by air is an over-engineered card in my book.

Fury X is the product of staff cuts and desperation, that's what it is.
According to my rough interpolations using data for current GCN 1.2 cards, one Fury X could have made 27.5B p/s for NTLM (vs 35B p/s for Titan X), if it could draw as much power as necessary.
But I highly suspect it will be throttled, so that shouldn't be the stable speed.

Last, but not least, the last thing people care about here on hashcat.net is games.
In them, as long as you can get the GPU to meet the Vsync target (or twice as much for 3D), you're cool.
In other words, required performance is inherently capped.
And quick benchmarks do not reflect the real hourly gaming scenarios.

[blazer]

The 450W is total system draw not just the card. The 980ti is listed as 430W in the same chart for the power consumption test.

[inoo]

Do somebody have any data of hashcat BMT for oclHC against Fury X?
It would better posting BMT result and real power consumption data.

[Rolf]

432W power draw specifically for Fury X under hairy doughnut (Furmark).