AMD Radeon R9 Nano Coming Sooner Than You Think?

[btarunr]

AMD's upcoming disruptive performance-segment graphics card, the Radeon R9 Nano, could be arriving sooner than its late-Summer expected launch. One of AMD's promotional heads Anthony "Elmy" Lackey posted two pictures of the card on his Flickr page, which reiterates just how compact the thing is. AMD earlier announced that the R9 Nano will be faster than the Radeon R9 290X, with typical board power well under 190W, making it an exciting product to look forward to. The R9 Nano will be based on the same "Fiji" silicon, which powers the R9 Fury X and R9 Fury. AMD could make a major announcement related to this product very soon, given how Elmy promised to release a few details next week.



View at TechPowerUp Main Site

 

[OolongGoat]

That is a fairly compact card..... geeze.....

 

[64K]

This is a card that will make AMD some cash hopefully. Dependent on price and performance. The Fury X and Fury are fun to read about but most people aren't going to spend the money to get one. It's the midrange cards like the Nano and entry level cards that most will be buying.

 

[ZoneDymo]

so sexy, I must have it!!!

 

[Disparia]

Nope! Because I wasn't thinking of a date before.

For those of us looking for smallish cards it'll be hard to beat. I held off getting a GTX 960 when AMD first announced it; glad to hear it's coming 'soon'.

 

[FordGT90Concept]

So will it beat a 390X?

 

[moproblems99]

A day late and a dollar short. I probably would have bought one had they released it in July with Fury.

 

[Aquinus]

Why am I feeling like this is going to fall somewhere between a 380 and 390? I've seen nothing but just the look of the card isn't giving me the full performance vibe. Maybe it's nothing. I dig the size of the card though. I still think AMD has some work to do before HBM gets utilized properly so I don't regret not waiting.

 

[arbiter]

So will it beat a 390X?

AMD claimed "significantly faster then a 290x" in their announcement conference but don't see how that card will keep full performance without throttling.

 

[FordGT90Concept]

That's an unfortunate possibility. I really hope one fan is enough to keep it running at full throttle.

 

[EarthDog]

So will it beat a 390X?

It said it would beat a 290x... they are the same exact card sans more memory...

Yes.

 

[HumanSmoke]

This is a card that will make AMD some cash hopefully. Dependent on price and performance.

The 390X uses a 438mm² GPU and some relatively inexpensive 6Gbps GDDR5, the Fury Nano will use a 596mm² GPU and a pricey HBM and interposer implementation. I doubt AMD will make much (if any) cash in relation to their conventional GDDR5-based cards.

Technically, you could make a revenue killing by selling $10 notes for $5 each, but your bottom line tells a different story.

 

[Random Murderer]

Holy cow, it looks smaller than the Asus "ITX" cards.
Kind of want, just for the size and power envelope.

 

[BiggieShady]

Even with TDP well under 190W, that single fan will be spinning very fast ... cooling on this little fellow might just be barely good enough, and this is me hoping I'm wrong.

 

[NC37]

Why am I feeling like this is going to fall somewhere between a 380 and 390? I've seen nothing but just the look of the card isn't giving me the full performance vibe. Maybe it's nothing. I dig the size of the card though. I still think AMD has some work to do before HBM gets utilized properly so I don't regret not waiting.

Because it likely will be. This card is very likely the reason why there is no 380X. Simply because the performance is likely on par with what the 380X was going to deliver.

 

[Ralfies]

I have that pen at work. It sucks. Uni-ball vision elite is where it's at.

The card is adorable.

 

[Patriot]

Because it likely will be. This card is very likely the reason why there is no 380X. Simply because the performance is likely on par with what the 380X was going to deliver.

-.- Them some good drugs...

 

[Joss]

Why am I not excited?
oh... it must have to do with all those previous disappointments.

Let me go and watch some proper prOn

 

[Casecutter]

Regrettably, I'd think AMD can't price it much less than $450. While performance might be par/better than the 390, perhaps coming in close to the 390X/980 at times. The problem is it a niche product to take some "ITX Crown", although little solace in such a win. Against a Gigabyte 970 ITX, it might have 7-10% lead @1440p, but with the Gigabyte 970 ITX right now is $290 and the Metal Gear Solid V... It's not going to show great with like 60% higher price.

 

[Patriot]

Regrettably, I'd think AMD can't price it much less than $450. While performance might be par/better than the 390, perhaps coming in close to the 390X/980 at times. The problem is it a niche product to take some "ITX Crown", although little solace in such a win. Against a Gigabyte 970 ITX, it might have 7-10% lead @1440p, but with the Gigabyte 970 ITX right now is $290 and the Metal Gear Solid V... It's not going to show great with like 60% higher price.

290x and 390x are on par performance wise... to be significantly faster than a 290x puts it above the 390x...

at $450 faster and less power than a 390x wouldn't be bad.

 

[HumanSmoke]

-.- Them some good drugs...

It might not be that far from reality. The difference between the Fury and the 390X isn't that pronounced, and I suspect the 15-20% clock reduction will impact Fiji somewhat more than the 12.5% reduction in shader count. It might be more noticeable with the lower tier cards being of reference design/speed, but AIB custom cards with higher clocks at, or near, reference cards (if they can be found) pricing makes that comparison more academic in nature.

 

[Patriot]

It might not be that far from reality. The difference between the Fury and the 390X isn't that pronounced, and I suspect the 15-20% clock reduction will impact Fiji somewhat more than the 12.5% reduction in shader count. It might be more noticeable with the lower tier cards being of reference design/speed, but AIB custom cards with higher clocks at, or near, reference cards (if they can be found) pricing makes that comparison more academic in nature.

The ROPS can't feed the high SP count at low res on fury. That gap becomes significant at 4k, tesselation, and eyefinity.

 

[FrustratedGarrett]

The 390X uses a 438mm² GPU and some relatively inexpensive 6Gbps GDDR5, the Fury Nano will use a 596mm² GPU and a pricey HBM and interposer implementation. I doubt AMD will make much (if any) cash in relation to their conventional GDDR5-based cards.

Technically, you could make a revenue killing by selling $10 notes for $5 each, but your bottom line tells a different story.

Except that the Fury Nano is based on harvested Fiji chips, which is why they're releasing a midrange card based on a high end chip.

 

[AsRock]

Except that the Fury Nano is based on harvested Fiji chips, which is why they're releasing a midrange card based on a high end chip.

And why the hell not .

 

[FrustratedGarrett]

It might not be that far from reality. The difference between the Fury and the 390X isn't that pronounced, and I suspect the 15-20% clock reduction will impact Fiji somewhat more than the 12.5% reduction in shader count. It might be more noticeable with the lower tier cards being of reference design/speed, but AIB custom cards with higher clocks at, or near, reference cards (if they can be found) pricing makes that comparison more academic in nature.

Depends on the game: AMD's DX11 drivers aren't as efficient batch creation and validation wise compared to Nvidia's, especially when the DX runtime tries to manage multiple concurrent threads. If you look at Bioshock Infinite for example, the Fury is around 20% faster than the 390x, which is in line with the difference in FPU count between the two chips.