You know, I have had this nano thingy in my rig for years. Check my system spec. It's a mini-itx rig with a mini 760.

TarkheinThat's because US prices are pre-tax. Tack on our Goods and Services Tax (10%) and you'll find very similar pricing.

I think there's some retailer price gouging going on with these Furies because they are in short supply. Here in the USA Amazon is selling the Fury X for $1,000. That would be almost $1,400 AUS. If the Nano is priced high too because of Fiji scarcity and it comes in at a little over the 290x in performance then I don't think they will sell well unless priced at around $300 right now since you can still buy a 290x for less than $300 until the supply dries up.

I imagine TPU will have review out for us to disect on Thursday. Hopefully the vapour chamber (is it?) cooler, low clocks (binning not required) and less PCB translate to reasonable price. Anything over £350/$350 is too high IMO.

the54thvoidI imagine TPU will have review out for us to disect on Thursday. Hopefully the vapour chamber (is it?) cooler, low clocks (binning not required) and less PCB translate to reasonable price. Anything over £350/$350 is too high IMO.

Agreed, I was thinking of getting a R9 290 for $249 but the heat and power consumption scares me. If the Nano falls in the sub $400 bracket, think I'm gonna get one to replace my aging HD 6870.

I doubt it will go below $400 if it's indeed a full fury X with only the core clocks underclocked. The fact alone with its small footprint and given the projected peformance I think that demands a premium price right there.

That thing is going to be so loud.

the54thvoidI imagine TPU will have review out for us to disect on Thursday. Hopefully the vapour chamber (is it?) cooler, low clocks (binning not required) and less PCB translate to reasonable price. Anything over £350/$350 is too high IMO.

I would consider its performance point before you suggest something like that. ;)

I also feel we are going to be pretty surprised at its performance point as well. Only time will tell. :)

I keep posting these not productive in any way, unlike me troll things, but here it goes:

#AMD + #HYPE = #DISAPPOINTMENT

The universal formula. Why people expect so much from this card is beyond me.

I wonder if waterblocks for the Fury Nano will be cheap due to the compact size. I'm really interested in these things, but I'm tired of looking at even 390x prices, let alone anything Fury.
For the sake of competition, I'm really hoping Team Red will have a nice little hit on their hands here.

Sihastru#AMD + #HYPE = #DISAPPOINTMENT

The universal formula. Why people expect so much from this card is beyond me.

Because this card features a new promising technology HBM (resulting in new innovative graphics cards dimensions without sacrificing performance) which has the potential to change the game and offer us universally better graphics cards.

That is why.

the54thvoidI imagine TPU will have review out for us to disect on Thursday. Hopefully the vapour chamber (is it?) cooler, low clocks (binning not required) and less PCB translate to reasonable price. Anything over £350/$350 is too high IMO.

I think it is a vapour chamber - it probably has to be for the available surface area. My question would be: A effective is a vapour chamber that compact with a centrally mounted fan that places the fan hub directly over the hottest portion of the chamber base. Should be OK at stock (you'd hope), but I'm guessing that the range of overclocking will be hardly worth mentioning. I would have thought having the fan mounted as far forward as possible to maximize airflow across the chamber might have been a better design (even if it makes the aesthetic asymmetric). Most vapour chamber designs generally favour the fan-at-one-end implementation favoured by usual reference designs.

What motherboard is that on which the Nano is installed?

krimetalWhat motherboard is that on which the Nano is installed?

Looks like an ASRock X99E-ITX/ac. They used the bundled cooler too.

Xzibit provided some good pictures yesterday, from what I see it appears to be a Hybrid Vapor Chamber with Heat Pipes. In the first picture I see what looks like flattened heat pipes poking out in several spots.
www.techpowerup.com/forums/threads/amd-radeon-r9-nano-nears-launch-50-higher-performance-per-watt-over-fury-x.215486/page-3#post-3334915

I just find it so marvelously delightful how compact and fully utilized of a package it is. It’s just so different to imagine in the next shrink we’ll have High-end Enthusiast level packing 16Gb in a package this size. We are watch the graphic card changing before our eyes, from what we’ve come to know for most of their modern existence. Those 10-11 inch, 3-fan behemoths’ with power sections that appropriated huge sections of PCB real-estate, while not probably completely disappearing reflect on the revolution seen in this more than any 28 nm Fiji/HBM iteration.

xviLooks like an ASRock X99E-ITX/ac. They used the bundled cooler too.

Powerfull beast, 2 IC on a X99 platform. Pair it with a 5930X and a Nano GPU :)

This thing is probably going to sell for $400 or more and it's just not worth it at that price point. Disappointing because it has a lot of potential and fills a nice niche, but the price/performance just won't be worth it for most people.

xenocideThis thing is probably going to sell for $400 or more and it's just not worth it at that price point. Disappointing because it has a lot of potential and fills a nice niche, but the price/performance just won't be worth it for most people.

Two days until we find out. :P

That's a full fury x if that slide is legit...

Xzibit

Edit: No Mantle support on the Nano? I find that hard to believe.
For comparison:

Texture first rate and pixel fill rate are different...and the core clock.

EarthDogTexture first rate and pixel fill rate are different...and the core clock.

The first two are a product of the third.
Pixel fill is clock speed x ROP (1000 x 64 = 64000MPix/sec)
Texture fill is clock speed x TMU (1000 x 256 = 256,000 MTex/sec)

Will be interesting under what conditions (and for how long) the Fury Nano maintains a 1000MHz core clock.

HumanSmokeThe first two are a product of the third.
Pixel fill is clock speed x ROP (1000 x 64 = 64000MPix/sec)
Texture fill is clock speed x TMU (1000 x 256 = 256,000 MTex/sec)

Will be interesting under what conditions (and for how long) the Fury Nano maintains a 1000MHz core clock.

So the Fury Nano is basically a Fury X with a reduced clock?

Meaning all this energy savings that magically materialized are the result of a meer downclock?

Meaning the poor performance of the Fury X can be explained by them merely taking the Fury Nano and clocking the snot out of it, energy efficiency be damned?

The question I have then is WHY DID THEY NOT OPEN WITH THE NANO INSTEAD WHEN IT MIGHT HAVE BEEN COMPETITIVE?

AMD seems as clueless as ever... sad.

R-T-BMeaning all this energy savings that magically materialized are the result of a meer downclock?

A mere 50MHz downclock. :p

Given its compact size (and likely reduced cooling capability), I (and many others, it seems) suspect it won't spend nearly as much time up in the boosted clock zone as the Fury (X) does.

xviA mere 50MHz downclock. :p

Given its compact size (and likely reduced cooling capability), I (and many others, it seems) suspect it won't spend nearly as much time up in the boosted clock zone as the Fury (X) does.

Gah... that just boggles my mind. Why pair it with such a piss poor cooler? Why not simply let OEM board vendors make a Fury X chip into something truly awesome?

Don't get it, honestly.