AMD R9 Nano official specs (and later, reviews)

[digitalwanderer]

Whether it makes sense for a consumer is something else...

That is the big question right now.

Think it'd fit in a really, really decent sized laptop?

 

[silent_guy]

What? Where did you see that?

Cute.
First: my sentence starts with 'if'. It's a speculation.
Second: you know very well my opinion on Fiji products having much lower margins than gm200 due to their usage of HBM.

The pricing of the Nano doesn't prove but strengthens that argument.

 

[kalelovil]

The idea seems to be that they're offering something unique, in that the Nano performs at a level that is unmatched by any card of this size, which has value for people who want compact systems.

That may be true, but in practice I'm skeptical that it will convince a significant number of buyers.

But unique for how long? If it does gain a market, there are presumably no major barriers to the production of a Geforce GTX 980 "mITX" which can significantly undercut it on price while coming very close in performance.

 

[Silent_Buddha]

But unique for how long? If it does gain a market, there are presumably no major barriers to the production of a Geforce GTX 980 "mITX" which can significantly undercut it on price while coming very close in performance.

But quite likely nowhere even remotely close to its size.

Regards,
SB

 

[Deleted member 13524]

Cute.
First: my sentence starts with 'if'. It's a speculation.
Second: you know very well my opinion on Fiji products having much lower margins than gm200 due to their usage of HBM.

Yes, you've been spending all you forum-strength to convince everyone that the Fury X is much more expensive to make than the GM200 cards, because HBM.

But you had never reached the point of ludicrousness to suggest the $650 Fury X is aggressively priced.
What's next? AMD loses money with each Fury X sold, because HBM?

 

[kalelovil]

But quite likely nowhere even remotely close to its size.

Regards,
SB

Why? The Asus & Gigabyte GTX 970 "mITX" are only an inch longer (or taller depending on how you count the PCI-E connector), and their decent overclocking results suggest the power circuitry could handle a fully unlocked GM204 with reduced overclocking potential.

 

[Deleted member 13524]

Why? The Asus & Gigabyte GTX 970 "mITX" are only an inch longer, and their decent overclocking results suggest the power circuitry could handle a fully unlocked GM200 with reduced overclocking potential.

I think you meant fully unlocked GM204.
But yes, a fully unlocked GM204 could probably fit into that form factor and the performance difference to the Nano would be 15% instead of 30%. Price those at $450 and the Nano's pricing becomes even more ridiculous.

 

[lanek]

Looking how the core clock should be extremely constrained on the Nano, just push the power limit slider on the CCC and you should easely goes to +15% performance.

 

[Deleted member 13524]

Looking how the core clock should be extremely constrained on the Nano, just push the power limit slider on the CCC and you should easely goes to +15% performance.

I doubt the VRMs are capable of handling a 15% overclock, not to mention the cooling system and air exhaustion capabilities of the tiny mITX case models that require the Nano.

 

[hoom]

So its really not a salvage part then.
I was quite wrong.

 

[CarstenS]

I doubt the VRMs are capable of handling a 15% overclock, not to mention the cooling system and air exhaustion capabilities of the tiny mITX case models that require the Nano.

AMD specifically said in the briefing, that given good cooling and a well ventilated case, you could push the Nano all the way up to Fury X power levels.

 

[Deleted member 13524]

AMD specifically said in the briefing, that given good cooling and a well ventilated case, you could push the Nano all the way up to Fury X power levels.

Erm.. then I guess you might as well just buy the Fury X, because the price is the same, the adequate cooling system is already there and you have some overclocking headroom on top of Fury X's levels.

 

[lanek]

I doubt the VRMs are capable of handling a 15% overclock, not to mention the cooling system and air exhaustion capabilities of the tiny mITX case models that require the Nano.

Well the boost is set at 1000mhz... at 175W - 75°C, the card certainly run on an average at 850mhz, but the capacity for get it at 1000mhz is there... the fan have a low db rate, im pretty sure that, it will keep up with no problem at 500-700 rpm more.

As for the price, i heard different sound, we should now when the card will be on retail.

 

[digitalwanderer]

Erm.. then I guess you might as well just buy the Fury X, because the price is the same, the adequate cooling system is already there and you have some overclocking headroom on top of Fury X's levels.

But it's no where near as cute and tiny! :O

 

[silent_guy]

Yes, you've been spending all you forum-strength to convince everyone that the Fury X is much more expensive to make than the GM200 cards, because HBM.

Forum strength is for others to judge. I'm basing it more on 25 years in the semiconductor business, and the experience that complex things with high added value with only a single supplier never come cheap, especially not in low volume. And you're absolutely right: it's all because of HBM. How could it not be? The die sizes are identical.

But you had never reached the point of ludicrousness to suggest the $650 Fury X is aggressively priced.

Compared to its competitor, it is aggressively priced (on the low side of course) in terms of gross margin percentage.

What's next? AMD loses money with each Fury X sold, because HBM?

Not at all. I have stated a couple of times, directly or indirectly, that the gross margins of Fiji must still be very positive.

 

[digitalwanderer]

Forum strength is for others to judge.

C'mon guys, let's not fight over "forum strength"! Let's remember it's all about the post count, now WORSHIP ME!!!

/s

 

[pharma]

With 30 percent more performance and 30 percent lower power than the previous generation Radeon R9 290X AIB, the 175W Radeon R9 Nano becomes the world’s most power efficient Mini ITX enthusiast graphics card available—period. This sets a new bar on size and performance. Its price is going to challenge a few users, but in graphics you really do get what you pay for, so we think this is a good deal if you can afford it.

Cooling is not the only part of the equation; the new Nano gets the best performance per watt and price—check the Pmark scores. (The Pmark is JPR’s comparison of price, wattage, and performance to calculate a products comparative value.)

The AIB’s GPU has 256 texture mapping units and 64 raster operation units. Compared to the Radeon R9 Fury X with 8.6 TFlops and a 1050 MHz GPU clock, the Radeon R9 Nano hits 8.19 TFlops with a 1000 MHz engine clock. It accomplishes that partially by employing greater memory bandwidth over GDDR5 via a 4096-bit memory interface. Normally a wide bus width will cost you in power consumption, but AMD came up with some clever low power switches which they are not talking about.

http://www.eetimes.com/author.asp?section_id=36&doc_id=1327544&

 

[swaaye]

Bring on 16nm and HBM2. And some new games too.

I wonder if there will be a 970-like market shake-up card made with that new hardware.

 

[Alexko]

You're probably right, but it might still be a year away, perhaps even further out. It could also happen sooner, of course, but as far as I'm aware, no information has been made public, and nothing particularly credible has leaked.

 

[swaaye]

You're probably right, but it might still be a year away, perhaps even further out. It could also happen sooner, of course, but as far as I'm aware, no information has been made public, and nothing particularly credible has leaked.

I've seen Q1 2016 mentioned for GP100 but who knows if that's realistic.

I just can't see how anyone can be thrilled with 28nm anymore and be willing to drop big money on these cards. Especially fizzled-out 4GB Fiji.