Feedback

AMD Radeon R9 Nano Preview - Small in Stature, Big on Performance

Author:
Manufacturer: AMD

The Tiniest Fiji

Way back on June 16th, AMD held a live stream event during E3 to announce a host of new products. In that group was the AMD Radeon R9 Fury X, R9 Fury and the R9 Nano. Of the three, the Nano was the most intriguing to most of the online press as it was the one we knew the least about. AMD promised a full Fiji GPU in a package with a 6-in PCB and a 175 watt TDP. Well today, AMD is, uh, re-announcing (??) the AMD Radeon R9 Nano with more details on specifications, performance and availability.

View Full Size

First, let’s get this out of the way: AMD is making this announcement today because they publicly promised the R9 Nano for August. And with the final days of summer creeping up on them, rather than answer questions about another delay, AMD is instead going the route of a paper launch, but one with a known end date. We will apparently get our samples of the hardware in early September with reviews and the on-sale date following shortly thereafter. (Update: AMD claims the R9 Nano will be on store shelves on September 10th and should have "critical mass" of availability.)

Now let’s get to the details that you are really here for. And rather than start with the marketing spin on the specifications that AMD presented to the media, let’s dive into the gory details right now.

  R9 Nano R9 Fury R9 Fury X GTX 980 Ti TITAN X GTX 980 R9 290X
GPU Fiji XT Fiji Pro Fiji XT GM200 GM200 GM204 Hawaii XT
GPU Cores 4096 3584 4096 2816 3072 2048 2816
Rated Clock 1000 MHz 1000 MHz 1050 MHz 1000 MHz 1000 MHz 1126 MHz 1000 MHz
Texture Units 256 224 256 176 192 128 176
ROP Units 64 64 64 96 96 64 64
Memory 4GB 4GB 4GB 6GB 12GB 4GB 4GB
Memory Clock 500 MHz 500 MHz 500 MHz 7000 MHz 7000 MHz 7000 MHz 5000 MHz
Memory Interface 4096-bit (HBM) 4096-bit (HBM) 4096-bit (HBM) 384-bit 384-bit 256-bit 512-bit
Memory Bandwidth 512 GB/s 512 GB/s 512 GB/s 336 GB/s 336 GB/s 224 GB/s 320 GB/s
TDP 175 watts 275 watts 275 watts 250 watts 250 watts 165 watts 290 watts
Peak Compute 8.19 TFLOPS 7.20 TFLOPS 8.60 TFLOPS 5.63 TFLOPS 6.14 TFLOPS 4.61 TFLOPS 5.63 TFLOPS
Transistor Count 8.9B 8.9B 8.9B 8.0B 8.0B 5.2B 6.2B
Process Tech 28nm 28nm 28nm 28nm 28nm 28nm 28nm
MSRP (current) $649 $549 $649 $649 $999 $499 $329

AMD wasn’t fooling around, the Radeon R9 Nano graphics card does indeed include a full implementation of the Fiji GPU and HBM, including 4096 stream processors, 256 texture units and 64 ROPs. The GPU core clock is rated “up to” 1.0 GHz, nearly the same as the Fury X (1050 MHz), and the only difference that I can see in the specifications on paper is that the Nano is rated at 8.19 TFLOPS of theoretical compute performance while the Fury X is rated at 8.60 TFLOPS.

Continue reading our preview of the AMD Radeon R9 Nano graphics card!!

The memory system is also identical between the R9 Nano and the Fury X: 4096-bit wide high bandwidth memory bus, 4GB of capacity, 500 MHz memory clock rate and up to 512 GB/s of available memory bandwidth. Again, very impressive!

View Full Size

There is one very big difference that we have to point out though: The R9 Nano is rated with a 175 watt TDP, the Fury X at 275 watts. That is a difference in power consumption that we just haven’t seen in any other card to card variance when based on the same GPU. It almost seems impossible and until we have a card in our hands to test, I’m going to continue to have my doubts.

As I see it there are only two ways that AMD can make this magic happen. The first is that they have been binning the Fiji GPUs for the release of the R9 Nano since day one, saving any GPU that shows incredible power efficiency for this launch; knowing that meeting that 100 watt difference goal was going to be a top priority. Finding a GPU that could run at 1000 MHz at a lower voltage would save AMD a lot in the thermal dissipation department since we know that voltage and frequency do not scale linearly.

The second option is that AMD has built the R9 Nano with an “up to” clock speed of 1.0 GHz but that it doesn’t hit that as often as you might expect. If you test a Fury X or even an AMD Fury today (or even the R9 390/390X cards) you will see that even though the cards all list “up to” clock speeds, you almost never see frequencies drop below that level. That is a very different experience than using NVIDIA’s GeForce cards with GPU Boost technology – NVIDIA specifies a base clock speed and a “typical” boost frequency know that the majority of the time its hardware will be above even that number. That keeps everyone honest and makes sure there aren’t undue expectations placed on the clock rate over time in real-world gaming scenarios. When we get our hands on the hardware this will obviously be one of the key things we look for – does the Fiji GPU on the R9 Nano throttle in an open-air or small form factor chassis?

Update: Just before press time I got some additional information from AMD on this topic. Depending the workload (dependent on the game) AMD claims that the R9 Nano GPU clock will "settle" on an average somewhere between 800 MHz and 1000 MHz, always picking the most power efficient operating point. That means that the GPU will be downclocked in order to keep the card at the 175 watt TDP level it is advertised. AMD claims that the GPU will not be thermally constrained until it hits 85C and clocks wouldn't be throttled until that point.

View Full Size

It's important to note that even though AMD says the GPU is not thermally constrained, all aspects of the ASIC performance come into play during use. Efficiency is defined as the ability to run at certain performance level and clock speed within a set TDP (which AMD has already defined at 175 watts). So if a particular game plays with a heavy GPU workload and attempts to draw more power than the 175 watts allowed, AMD's Fiji implementation will downclock until it arrives a voltage that hits ~175 watts draw. That will be different for all different kinds of software so its something we will pay particular attention to in our review.

For users that want to tweak things you will be able to adjust the power limit in the AMD Catalyst Control Center, thus decreasing efficiency, but pushing you closer to that 1.0 GHz frequency regardless of the GPU workload. This obviously takes AMD's new R9 Nano outside the range that it wanted to be able to claim for this product, and what it could guarantee in tight quarters that might be more thermally constrained, but gamers will have that flexibility if they wish, gaining as much as 10% in graphics performance based on AMD's quotes. Once you hit that 1000 MHz mark, however, you revert to the more standard overclocking models that we have already seen aren't spectacular with Fiji GPUs.

End Update

The R9 Nano isn't going to be cheap: AMD will be setting the MSRP of the card at the same level as the Fury X: $649. That's a lot to pay for a card that is performing slightly lower than the $549 R9 Fury but it has some characteristics that you just can't find in another high end GPU product. I'll await final decision on the value of the R9 Nano after I put it through its paces in a week or so.

AMD is adamantly painting the R9 Nano into a unique position in the market, and deservedly so. Spoiler alert: you can't get more performance in a smaller package anywhere else. But in order to set that up we have to have some kind of initial comparison point. For that purpose, AMD is comparing the R9 Nano to the R9 290X, the original Hawaii GPU released in October of 2013 and the basis for the R9 390X released this past June.

Here are the primary claims AMD makes about the Radeon R9 Nano:

  • 40% shorter than the R9 290X: Sure, that makes sense. The R9 Nano has a 6-in PCB compared to the reference R9 290X that measured 11-in. Even the Radeon Fury X PCB measured 7.5-in.
     
  • Up to 30% faster than the R9 290X: Now this is more interesting. If this is accurate it puts the performance of the R9 Nano very close to the performance of the Fury, non-X model.
     
  • 20C cooler than the R9 290X: Compared to the reference design for the R9 290X, which was not a good showing for the product in many ways, this is easy to do. With a target operating temp of 75C on the R9 Nano and 95C on the R9 290X, the math checks out.
     
  • 30% less power than the R9 290X: With a TDP of 175 watts for the Nano and 250 watts for the R9 290X, again the math checks out.
     
  • 16 dbA quieter than the R9 290X: Heh, well then. Comparing the R9 Nano to the reference R9 290X cooler is an easy win as that was one of the louder flagship launches in memory. The Nano is rated at 42 dbA while the 290X is measured at 58 dbA! Keep in mind that the Fury X is rated at <32 dbA so the Nano isn’t going to quite as silent as that was (when you found one without coil whine).

All of that adds up to a piece of GPU hardware that has up to 2x the performance per watt and 2x the performance density (in terms of card length) compared to the R9 290X from AMD’s previous portfolio. What that means for gamers will really be determined by the hard numbers and how close in performance the Nano is to the likes of the Fury. And of course pricing; that usually matters too.

Now, let’s talk about the potential benefits to the R9 Nano form factor that AMD is touting for today’s re-announcement.

View Full Size

The Nano is small, damned small, even when compared to the likes of a Mini ITX system build as demonstrated in the image above. The Mini ITX form factor designates a 6.7-in square motherboard platform and with only a 6-in board length, the AMD R9 Nano fits within that easily, even considering standard PCIe plug lengths.

View Full Size

There are quite a few cases on the market, with more coming all the time, that have very cramped quarters for the graphics card when it comes to length. Take these three designs shown above, with card length maximums ranging from 8.66-in down to as little as 7.09-in! Clearly there several thousand times as many cases that can handle the full run of available graphics cards on the market, but this trend of smaller cases and smaller PCs is growing as more and more PC gamers get older and want to continue gaming in a domesticated environment.

View Full Size

Though this slide is exaggerated slightly with the case selections, the truth is that as recently as 1-2 years ago if you were going to game on a 4K monitor you would need a dual-GPU configuration and thus a standard ATX capable chassis. The Corsair 900D shown on the left is one the largest cases you can buy today, so…sure. The middle chassis could also easily hold a second R9 290X card for 4K gaming but instead it shows the installation of a single AMD Fury X card. (Also, not sure that June/July releases count as “early” 2015.) Finally, on the right, is one of the smaller Lian Li mini ITX cases with the R9 Nano resting comfortably inside.

When AMD began talking about the size benefits of integrating HBM in this generation of products, this is exactly the kind of form factor implications it was pointing towards.  From NVIDIA though, the options are a bit more tame.

View Full Size

The highest performing Mini ITX friendly graphics card on the market with a GeForce GPU is the ASUS GTX 970 Mini ITX (that we have in for testing already). As you can guess based on the generic metrics listed above for the R9 Nano, it seems very likely that this ASUS card will be drastically outclassed by the new smaller Fiji-based option. Pricing will still have a lot to do with this battle in the market but it would seem that for enthusiasts that care about sizing and space, the R9 Nano is going to be king of that hill for some time.

View Full Size

Here is a little preview of performance, as tested by AMD of course, to see what I mean. My “measure with my fingers on my laptop screen math” shows in Shadows of Mordor the R9 Nano gets around 50 FPS while the GTX 970 Mini ITX card gets something around ~35 FPS. That would be a difference of about 40% if it holds up.

Finally, AMD surprised me by telling me that it would be opening up the R9 Nano design to partners and we will likely see custom variants of the Nano later in the year.

 

So there you have it, you now know as much as about the Radeon R9 Nano as I do. Cards should be incoming to reviewers in the near future and I am looking forward to pitting this card against the claims that AMD made here with today’s re-announcement of the product. I’d be shocked if NVIDIA and its partners didn’t already have some kind of response in development but without the benefit of HBM (High Bandwidth Memory) I don’t know if it’s possible for Maxwell to get to the same form factor and performance / power levels that Fiji is showing us today.

As we await precise performance results, AMD’s Radeon R9 Nano looks like it might be the most interesting graphics card release so far this year.

Video News


August 27, 2015 | 09:13 AM - Posted by Senpai (not verified)

I am very curious to see how well the card does with the limited 175w power "limit" (I did read the Updated section).

I have a gut feeling that this will be similar to how Hawaii GPU's throttled speed based on thermal performance, but this time it will be based on Wattage.

PCIe 1 and 2 can provide up to 75w of power, the 8 pin power connector can do 150w, so 225w in total. PCIe 3 on the other hand can do 300w by itself. Add the 8 pin and you have 450w.

Welp, I guess we have to wait 2 more weeks or so until the NDA lifts and we can see how the Nano works out.

August 27, 2015 | 09:27 AM - Posted by Sebastian Peak

PCI Express 3.0 still only outputs 75W from the slot itself. The PCI-SIG 300W output spec is based on the use of 8-pin and 6-pin connectors in conjunction with the slot output.

August 27, 2015 | 09:21 AM - Posted by Master Chen (not verified)

Well, crap. This is what I was most afraid of happening.
Wake me up when it goes at ~400$ for non-reference. For now - useless.

August 27, 2015 | 09:28 AM - Posted by Anonymous (not verified)

You want a card that can beat a 980 for less than $400?

August 27, 2015 | 10:13 AM - Posted by Master Chen (not verified)

I want an HBM ITX card that doesn't cost the same as full-sized HBM card.

August 27, 2015 | 11:42 AM - Posted by allons (not verified)

It is a full fiji HBM card though.

August 27, 2015 | 07:41 PM - Posted by Master Chen (not verified)

It's an ITX card, though. Downclocked, on air. Same money? Pass.

August 27, 2015 | 01:15 PM - Posted by Anonymous (not verified)

Well keep dreaming idiot. No one sells a more compact product for LESS money.

August 27, 2015 | 07:43 PM - Posted by Master Chen (not verified)

Your ass should be moderated some more, it seems.

August 27, 2015 | 02:45 PM - Posted by StephanS

It does not. AMD showed at hot chip that the nano to be about 10% faster then a 290x.

The big thing with the nano is not performance, its power usage.

Is in the same league as the GTX 980 @ 165w, the nano being rated at 175w.

But look at the 1440p benchmarks... A card that is 10% faster then a 290x is going to be crushed compared to a GTX 980.
(GTX 980 are now about $480)

Yes, the nano is faster in 4K gaming, compared to the GTX970, when the game pref are tweaked to give the advantage to the nano. But that advantage disappear at 1440p

nano got better HW then a GTX 970 mini, but in Dx11 games, at 1440p or below that advantage goes away... And the GTX 970 is only 145w.

August 27, 2015 | 04:31 PM - Posted by arbiter

That is pretty much what i had to say in a few posts. GTX970 isn't ment for playing game at 4k so AMD comparing their card to it is stupid compare. 970 is only ment for many 1440p at most. Plus the whole issue of tweaking game settings which they likely did to give the most advantage to their card is very possible since they did it with the fury X when comparing it to 980ti

August 27, 2015 | 04:16 PM - Posted by Anonymous (not verified)

Everyone knows that AMD should beat NVidia with at least 50%, at half the price to be even considered as an option.
(I would like to say that was sarcasm, but reality begs to differ.)

August 28, 2015 | 01:00 AM - Posted by renz (not verified)

yes. that's what AMD always did? at least that's what most people expect from AMD. cheaper alternative to the much premium priced nvidia.

August 27, 2015 | 09:44 AM - Posted by Heavy (not verified)

well it's a full fiji, i wish the price was lower too.i would of been happen with the cut down verzion of the fiji just for the lower price

August 27, 2015 | 09:36 AM - Posted by Anonymous (not verified)

Very interesting card. If you are putting an htpc for a top end gpu, I guess you don't spare expense.

August 27, 2015 | 10:06 AM - Posted by Anonymous (not verified)

Testing with 0xAA and that is more important by far, without anisotropic in 2015.... OMG.

AMD is the way to return back to the past. GOING TO THE MIDS OF THE NINETY's years of the past century/millennium!!!

WTF is the sense of testing 4K without AA and aniso??

Great AMD, you are making so good!!

August 27, 2015 | 10:33 AM - Posted by killurconsole

"WTF is the sense of testing 4K without AA and aniso??"

I'm quite curious ,where do they mention that ,cuz that would be really ridiculous

September 5, 2015 | 11:46 PM - Posted by arbiter

http://techreport.com/review/28912/tiny-radeon-r9-nano-to-pack-a-wallop-...

They have about 60% way down the page a table using footnotes from AMD what settings they used to test. Its pretty much same Settings they used to compare Fury X vs 980ti. They turned off any option that didn't shaders on the card. In that test they said fury x was 20% faster when it ended up being 5-10% slower when you use settings more gamers are likely to use. Probably see same thing on the nano.

August 27, 2015 | 11:01 AM - Posted by Master Chen (not verified)

You really don't need AA above x1 or AF when you're using full-blown 4K with 4K textures. And it won't be needed anymore right until the point when 8K moves in and 4K content on 8K would have obvious jaggies again due to mismatch of resolutions. It's an ever-spinning wheel of the circle of life.

August 27, 2015 | 12:35 PM - Posted by Ryan Shrout

I hate to just be that guy, but this statement is just incorrect. AA and AF are still very valuable with matched-resolution 4K rendering.

August 27, 2015 | 07:50 PM - Posted by Master Chen (not verified)

No, that's what you LIKE/PREFER to BELIEVE, to justify purchase of a typical 800+$ monstrosity of a card. In reality, 90% of those modern games which both support 4K resolution AND have actual 4K textures to match that - simply run ideally on a 4K monitor without the need of an AA above x1 or massive AF addition. In games like "Shadow of Mordor" additional AA above x1 doesn't make jack shit of a difference pure visual fidelity-wise, while hurts performance quite noticeably.

August 28, 2015 | 01:58 PM - Posted by LeoDS (not verified)

Your guy's comments are interesting. The only 24" 4K Monitor I could find has 6ms Response time.

Considering the ammount of Money you would be investing to go 24" 4k, the gain to be had, and the compriomise you have to make along the way it's really easy to undertstand why Ryan would think 24" 4K is wrong to begin with. I mean for gaming obviously, that is what we are talking about.

But then again, the most sad Thing, is that we are taling about thousands of Dollars to Play damned dumbed down casual games like shadows of mordor, when you're better off just getting a 100 bucks PC off some second Hand and playing some asteroids.

These companies really achieved a ridiculous end user perception of programmed obsolescence and false perception of what People actually Need. Good for them, because without that they'd be hurting even more, as People who actually want to Play Videogames don't Need to spend much Money at all in Hardware at this Point.

August 28, 2015 | 06:13 PM - Posted by Master Chen (not verified)

Anything that's 8ms or below, is absolutely fine for gaming. Not for every gaming, but in gaming in general. As for response times for high-paced fast gaming like that of competitive tourney FPS or fighting games, it's been long proven that 8ms AND anything below that is more than enough for those. 1ms and 2ms "GAYMEN" monitors is nothing more than just a marketing fad, because 1 frame of animation is actually 16ms exactly, so in all actuality both 8ms and 6ms are VERY playable, even if bad latency kicks in. Now, 10ms and higher are NOT good for gaming indeed, but still tolerable. Where it REALLY starts getting ugly, is those monitors which EXCEED the "16ms" wall. One last time - 8ms, 6 or 5ms, 4ms, 3ms, 2ms, 1ms - ALL of them are absolutely great for games, but "1 and 2ms" is just a marketing fad aimed at inexperienced gray masses, because in reality there's almost never (EVEN in the most fastest games) an actual noticeable difference even between 1ms and 8ms, 8ms don't make the response slower, it doesn't make response aquarium-like (10ms does, though), AND it really DOES NOT affect playing performance. It's all just a fad and placebo effect for inexperienced "modern age geymurr" kiddies who really don't understand/refuse to admit how these things actually work. And I'm saying this not just for a debacle purpose, but because I am myself a highly sophisticated and very experienced tourney player at fighting games, mainly at extremely fast-paced Japanese 2D titles like KoF and Guilty Gear, so I know first-hand out of my own personal experience which monitor timings are absolutely great, which are "just acceptable", and which are completely unsuited for fast competitive gaming. I've tried them all (old tube TVs, CRT monitors, Plasma, LCD and it's different sub-types, and even OLED) during those 19 years of me being a competitive tourney player at fighting games (I've started with Samurai Spirits II and KoF '96, back in 1996), so I know what I'm talking about. ~8ms is more than enough. Yes, 1ms, 2ms, 3ms, all of this is obviously fast, no one doubts it, but they're NOT oh-just-so-much better than 4ms, 5ms, 6ms, or 8ms. They are simply NOT. There is almost never a difference, and it sure as hell won't affect your playing performance as long as you don't give in into this stupid fad's placebo effect.

Also, UP2414Q (pretty much the only worthwhile 24" 4K monitor on the market, and there's really not that many of those out there to begin with) in reality has overall response timings lower than the ones listed in it's official specifications. It's always like this with Dell: unless TFT Central, or Prad.de, or some other professional monitor and panel reviewing source out there makes clear that the reviewed monitor's timings are matching officially listed specs, it's not really always the case. The GODLIKE U2414H, for example, has officially listed timings of 8ms. It has been proven by TFT Central that in actuality it has overall response timings of 4ms exactly, way faster than what Dell lists it at, so it's better that you don't always pick your choice by reading officially provided specs only (always check for quality professional reviews first).

August 27, 2015 | 01:41 PM - Posted by Patrick Proctor (not verified)

Because AA at 4K is damn pointless unless you have a massive 34" screen you sit less than 6" away from.

August 27, 2015 | 07:57 PM - Posted by Master Chen (not verified)

Exactly. The main problem with that comment of Ryan's, in my opinion, lies in that he automatically assumes that everyone out there already sits on those 40 inch Seiki monstrosities, so he basically disregards a possibility that someone out there might use a 24" or 27" 4K panel these days. He thinks (and looks like truly believe in that) that if you're using a smaller 4K panel to get higher PPI (and thus less need in applying any kind of additional filtration) - you're doing it wrong. I find this quite cynical as a PC hardware enthusiast.

August 27, 2015 | 10:27 AM - Posted by SRowe (not verified)

Your BULLET POINTS are broken...

Up to faster than the R9 290X: Now this is more interesting. If this is accurate it puts the performance of the R9 Nano very close to the performance of the Fury, non-X model.

UP TO ....HOW MUCH FASTER?
Just an FYI

August 27, 2015 | 10:32 AM - Posted by Randal_46

Great card, shame about that price point.

August 27, 2015 | 11:06 AM - Posted by Anonymous (not verified)

$649?

DEAD ON ARRIVAL

August 27, 2015 | 11:57 AM - Posted by mAxius

its a full Fury X in a sff card i do not see your logic

August 27, 2015 | 12:15 PM - Posted by StephanS

much slower then the fury x because of slower sustain clock.
AMD officially put the nano at 10% above the 290x hawaii card.
(check the hot chip slides from AMD from 2 days ago)

Also, the driver situation will still make this card slower at 1080p then the tx 970.. and probably just be a match at 1440p.

so $330 vs $650 for nvidia fan. non will switch
And $650 for a card thats just a little faster then a 290x at <$300?
what AMD fan that refuse to get a gtx 970 mini will spend over $300 just for the AMD brand?

no, the price is just ridiculous... its that way because AMD only plan to have very, very, very few units made.

August 27, 2015 | 01:19 PM - Posted by Anonymous (not verified)

They put it at 30% above, drivers are completely fine, even better in dx12 and certainly not killing cards or frying laptop displays like the nv drivers are doing. You'e a complete biased retard.

August 27, 2015 | 02:16 PM - Posted by StephanS

hawaii vs Fiji nano:

31FPS vs 33FPS

http://www.pcper.com/image/view/60784?return=node%2F63647

And no driver problems ? AMD only uses 4K benchmark because it reduce the driver problem they are having. case in point? ALL hawaii / fiji reviews.

Biggest Red flag ? Ashes of the Singularity in Dx11 mode
r9-390x : 28.3FPS
GTX-980 : 49.9FPS

Is a GTX 980 really almost 2x faster then a 390x ?
Of course not... AMD Dx11 drivers are not fully leveraging the GPU.

Here is the core of the issue:
http://www.bit-tech.net/hardware/graphics/2015/06/24/amd-radeon-r9-fury-...

The Fury x is 5% slower then the GTX980 in GTA 5 at 1080p
but its then 22% faster with just a resolution bump to 3840x2160

The Fury X is easily 30% faster then a GTX 980, but still is 5% slower... 100% driver related.

At 1080p the GTX970 will EAT the fury nano alive. because the nano is SLOWER then the Fury X by about 20+%

Those are cold hard facts. The AMD dx11 driver stack to holding back its HW by as much as 30%.

And how many games are out to made Dx11 irrelevant in a game card purchase ? none. and you probably just a few title in the next 12 to 24 months.

The nano at $650 is a joke, mainly because of the poor dx11 driver architecture & performance.

BTW, I own a 290x . And as so the nano is very un-impressive.
If I needed a mini card I would go with the GTX970 without hesitation. Now, 2 years from now, when the title I play are all Dx12, I might grab a nano off ebay for $250...

But AMD at the moment is loosing money on its gaming cards market and they might have to let go of this division. So in 2 years who knows, Intel might have something better :)

September 5, 2015 | 11:52 PM - Posted by arbiter

You have to remember that 30% claim, is using 4k and AMD's custom settings. Those custom settings had fury x beating a 980ti by 20%. Independent reviewers seem 980ti winning by ~5-10% in most tests.
Here is a link to techreports podcast, they talk about AMD's bench markers guide they put out and settings they used in their claims.
https://www.youtube.com/watch?t=2177&v=O0sLUWlvU18

AMD used settings that took advantage of their card's shader count and everything else was turned off.

August 27, 2015 | 01:23 PM - Posted by mAxius

your logic is undeniable... its a fury x meant for people using sff cases it is equal to a freaking 980 in performance and only uses 175w. In 2 weeks it is the king of small form factor cards if you arnt using a sff system please stop trolling.

August 27, 2015 | 02:27 PM - Posted by StephanS

The core compute of the nano is in line with a GTX 980, yes.
but PC games dont talk directly to the HW.

DX11 is a huge factor. AMD can drop about 30% of the GPU performance in 1080p.
This can make the nano slower at gaming then a GTX 970 at 1080p.
But it should be a match at 1440p. (4K, its just to slow)

note the GTX 970 mini are in the 145w range.
The GTX 980 is 165w, less then the nano

So what stop nvidia from selling a GTX980 for the SFF market ?
The GTX 980 is now in the $480 range.

AMD gaming market is going to tank *massively*...

August 27, 2015 | 04:34 PM - Posted by arbiter

yesturday you could get a mini gtx970 on newegg for 290$.

August 27, 2015 | 11:09 AM - Posted by fkr

this should be good. i hope they do charge a premium for this product. it is one of a kind. go make some money AMD.

August 27, 2015 | 11:56 AM - Posted by akaRonin (not verified)

Greetings Ryan,

Have you had the opportunity to ask AMD why Joe Macri lied on June 16th during the E3 announcement ?

August 27, 2015 | 01:20 PM - Posted by Anonymous (not verified)

about what?

August 27, 2015 | 12:03 PM - Posted by Anonymous (not verified)

Don´t care about the price. I buy at least 1 maybe 2 AMD R9 Nano.

I Just want the BEST GFX on this planet, that fits my needs. And of couse made by :

-- AMD, Leaders of the pack --

It looks like it will perform about the same, as and R9 290+/- under DX11, which is "now and past", but under DX12 "which is about now, and future" it turns into a true Beast, that actually WORKS - dig´it.)

August 27, 2015 | 12:43 PM - Posted by StephanS

I was thinking $350 was a stretch considering it would fall below a GTX 970 at 1080p because of the bad dx11 drivers.

But $650 is so out of touch... it tells me AMD expect to only make very few fiji cards.

At hot chip they show the 290x, a ~$280 card only being edge by 10% by the nano.

Who would pay almost $400 more for 10% ? both card at 4GB....

I would have been slamming AMD for pricing the nano at $350 because they would be crushed by the GTX 970 at retail... at $650, its INSANITY.

August 27, 2015 | 01:22 PM - Posted by Anonymous (not verified)

God your comments are so toxic. No once cares, go buy another 970 then.

August 27, 2015 | 02:36 PM - Posted by StephanS

I own a r9-290x, and will probably grab a nano off ebay in 2 years when dx12 become relevant.

Having said that, its AMD situation and pricing that is toxic.

This out of this world pricing and Dx11 driver situation will continue to have people go nvidia for their next card. (AMD market share is plummeting, even after Fiji release. the nano will accelerate AMD decline in the GPU market)

August 27, 2015 | 01:25 PM - Posted by mAxius

again its a flagship gpu meant for sff users nv has nothing that can match it. keep on trolling

August 28, 2015 | 01:29 AM - Posted by renz (not verified)

you know the best part is? it doesn't affect nvidia at all. at this rate i'm not going to be surprise if nvidia will continue gaining market share.

August 27, 2015 | 12:41 PM - Posted by collie

I just wana hear it. Way more performance that I need, but I wana know if it's a lawnmover or not.

August 27, 2015 | 01:01 PM - Posted by Keven Harvey (not verified)

Shouldn't be worse than the mITX version of the 285/380, that's 190w tdp and while louder than normal versions, it's not bad.

August 27, 2015 | 05:16 PM - Posted by Anonymous (not verified)

It's going to swallow all matching Nvidia cards, and cut them into small pieces.. but wait, there's no one.)
So it must be quiet then.)

August 28, 2015 | 02:42 AM - Posted by collie

Naw, AMD fanboy over here, I just wana know how loud it is. There are people in my circle with the budgets for top end game rigs, and either I build them for them or advise on parts/systems. But I honestly wana know, how is the volume on this guy. It (the fan) is big, it could be very effecient at low power and when it's maxed it's drowned out by the game/8k video, or maybe it's loud as fuck. Or something else. I just wana be close to one to hear it at desktop/game and hear it.

August 27, 2015 | 01:24 PM - Posted by icebug

I wonder if we are going to see a still small form factor sized version of this card with external fan exhaust. I would think that would help sustain the higher clock speeds for a little longer since it wont be contributing to the case temperature as much. Not sure how that would work though from an engineering perspective.

August 27, 2015 | 02:03 PM - Posted by Anonymous31276 (not verified)

Especially since this is for a SFF case, a proper exhaust type fan should have been used. But I think they wanted to make it look as small as possible.

August 27, 2015 | 02:53 PM - Posted by Lithium

Unbelievable

This is insult for normally IQ-ed human

Reality is that there is some fully functional Fury chips, but they cant touch 1000MHz in normal gaming.
More like 800MHz, maybe 50MHz up or down.
So, lets invent hot water, lets talk SFF, bla bla bla,
what a SCUM.

CUDA cores count:

GTX 970 = 1664
GTX 980 = 2048
--- HERE IS PLACE FOR ONE MORE CUT-DOWN VERSION OF GM200
GTX 980 Ti = 2816
GTX Titan X = 3072

So buy buy AMD aka SFF - IFX Universe Masters

August 27, 2015 | 03:21 PM - Posted by Anonymous31276 (not verified)

SFF gaming has never been about bang for the buck, it's about building something powerful and small.

August 27, 2015 | 03:49 PM - Posted by Sebastian Peak

Excellent point. Look at the pricing of SFX power supplies, the premium paid for Mini-ITX motherboards and existing ITX versions of GPUs. ITX is in no way an affordable form-factor. For a budget build Micro-ATX makes a lot more sense and the total cost would be a couple hundred less.

August 28, 2015 | 02:03 AM - Posted by Snus (not verified)

And add silent to, to that, then you have what its all abaut, and that dosnt come cheap :), pity its so few of us that love that challange when building, maybe things would be more aqffordabel then.

August 27, 2015 | 04:16 PM - Posted by Anonymous (not verified)

THE PRICE IS VERY RIGHT IN MY EYES.

If you want the best SUPER mITX "TITAN-KILLER" CARD EVER MADE, then you have to pay for it.!
- Just like a Titan card, but MUCH cheaper, "and under DX11, it is even better then the nvidia 980, but in DX12(which is now and in the future) it will be a mini FURY BEAST, thats will crush its opponent with ease!
Ps. and it actually WORKS. -- can you dig`it.)

I WILL BUY ATLEAST 2 of these mini BEAST.

KEEP UP THE SUPER GOOD WORK AMD, AND PCPER.)

Nice day to all.)

August 27, 2015 | 07:39 PM - Posted by Anonymous (not verified)

$650 for that? And people wonder why AMD keeps loosing more market share.

August 27, 2015 | 07:41 PM - Posted by Anonymous (not verified)

LOL *losing

August 27, 2015 | 08:09 PM - Posted by Snus (not verified)

BBest write up of the hw sites so far, finally some who understands what a SFF build is , and why this card is exciting, evrything with SFF comes with a premium, you have to balance the powerdraw, on cpu /gpu, get the most performance possibel with the least amount of noice and size, that is what true SFF builds are abaut, and ofcours custom cases, been googling and found a lot of companies, that can print ur case design, this can become so much fun- And if the performace that amd claims is accurate, then Nano is the king of SFF, for us that care abaut SFF.

August 27, 2015 | 09:05 PM - Posted by StephanS

The GTX 970 is 145w, so plenty of room to overclock.
And maxwell overclock extremely well. (unlike fiji)

The mini GTX 970 now sell for ~$300
And if you overclock it (145w to 175w+) you might see that at 1440p the nano will have a hard time keeping up in most games.

If you game at 1080p, GTX 970 mini is the clear winner for SFF build. faster then then the nano and $350 cheaper. (and on paper less heat/power)

Lets hope AMD got brand new driver next month, if not , I dont think the nano will be the fastest option of SFF beside 4K gaming ?

I know nothing is announced, but the GTX 980 is 165w.
So if a SFF gaming market even exist for that class of cards, nvidia can release a $450 GTX 980 to fully counter the nano.

But then again, it seem that the mini GTX 970 is already here for $350 less. nvidia would only compete with itself.

August 27, 2015 | 09:45 PM - Posted by Snus (not verified)

Yes, the king of SFF for now is nv 970, but value is not a thing that you can concern urself with, when u build SFF, and well for 1080 i guess the nv 970 will du, but for a powerhouse u need 2 nano in CF, u want to game at 4k, am just surching after the smallest aliminium case i can find that will support cf of nano, later on maybe a furyx2 will be a better option, that depends on the size and the powerdraw, small size require small psu, you see building a SFF GHTPC, near silent and well if the 42 db noise that amd claim is not true, the to H*** with the card, for one i really would want an externel nano psu, but sadly i only find it as 150w, and well i need at least 300w, 175 for the card 65w for gpu and a littel for the rest. I am for sure excited for the first time in years. SFF i a niche,, and not even a big one, hey considiering laptops are well SFF, so a impressive SFF build has to beat the best a gaming laptop can offer, afterall u have tons of room to build compared to a laptop :), this i just exciting, and no i dont really care for hybrid builds, But well the size and the noice of the fury x2 will tell if its good for true SFF.

As a side not, i have not found a good i want yet. But i have time.

August 28, 2015 | 03:14 AM - Posted by Anonymous (not verified)

Your right, nivida is competing with itself, because they DONT have a match for AMDs line-up of ITX-cards. And wait, the GTX 980 is not a mITX, get it?

August 27, 2015 | 09:47 PM - Posted by Snus (not verified)

*with just one nano, i probbaly can find a 45w cpu that can feed it.

August 27, 2015 | 09:51 PM - Posted by Snus (not verified)

Edit, sorry for typos, am reading true tips and tricks for silent gaming micro atx, and trying to see how small i can get, so my mind is elsewhere.

:)

ps, i recon the case will be very expensive, u see i want one of this, http://www.quietpc.com/htpccases. so you its not easy, but easy is not fun.

August 27, 2015 | 10:13 PM - Posted by Snus (not verified)

This is the case i really want, let the chassie cool the cpu, and well the case has to modifed to fit a micro or mini atx powersupply, and it might work there should be room if i only go for singel nano. u see u have to drill and mod at the back, this things take time,-

http://www.quietpc.com/images/products/st-fc9s-04-large.jpg

http://www.quietpc.com/st-fc9

or maybe i can mod it so it fit cf of nano, you see my friend these kinds of builds are not the kind of builds u do for tun. its a passion, silent and powerfull is the game, and well 42 db, is not excatly quiet, so its a challange you see, well u want understand if its not ur poisen :)

August 28, 2015 | 02:50 AM - Posted by Justin150 (not verified)

A few thoughts

The price point is all wrong at $500 it is interesting, at $650 it is not.

I watercool my kit, the nano with a full board waterblock, top of range skylake (watercooled) and a single 180x180 radiator makes for a very powerful and very quiet SFF. Although personally I would go SLI and a micro-atx format for a serious OTT gaming rig.

I really want AMD to succeed, bringing in HBM is a great move although I will probably wait a year for version 2 so that they can iron out the kinks and stick more memory on the cards.

The biggest problem after the price point is the drivers, with AMD the drivers will be worse, upgrades to the drivers slower than NVidia.

August 28, 2015 | 03:28 AM - Posted by Snus (not verified)

Why would thouse of u that build in regular big ugly pc cases even need and think og putting a cute littel card in thouse gigaqntic ugly cases, u have enough to choose from before at the high end, 980, 980ti, titan, fury and fury X.

Nvida did a great thing realeasing the SFF 970, and im happy that amd follow up with the nano.

This is ment for SFF as the nv SFF 970, is that so hard to figure out ???????

August 28, 2015 | 02:32 PM - Posted by StephanS

The issue is price, even if this is only for die hard stupidly wealthy SFF gamer.

Second, why can AMD make a non nano version for us with "big ugly cases' ? (The Fury is castrated, the X is a water block monstrosity, and is unnecessary)

Most people system can easely deal with 250w + video cards but dont want a closed loop water cooler. (specially when the card cant really overclock)

The nano is simply a marketing mistake. Its a limited die harvested product that is overpriced. While leaving 2 market wide open.

The $350 SFF GPU gamer market.
The $300-450 "Ugly case" gamer market.

This is where the big margins are at.

Even AMD does have superior HW with fiji, the driver and price make it a non option to gamers (AMD reputation in the gaming circle is down the drain).
Look at recent market share releases, AMD is plummeting in the gaming GPU sector. Look at the top seller on Amazon... AMD pricing / and power configuration is a disaster.

AMD gaming GPU is a money losing division. AMD cant continue to fund money losing R&D.

And the Nano pricing exemplify everything that is wrong with AMD.

August 28, 2015 | 05:40 PM - Posted by Anonymous (not verified)

Try hard trolling alert

August 29, 2015 | 09:04 PM - Posted by Snus (not verified)

Well, for what i am planning for the most powerfull SFF card awalibel:

FC5OD ALPHA Black Fanless HTPC Aluminium Chassis+ $306
SilverStone SFX Series ST45SF-G 450W $109
GA-H97N-WIFI $126
Xeon E3-1265L V3 $290
Corsair Vengeance Pro Series 16GB $379
Intel SSD DC S3500 Series 600GB 2.5 $570
Silverstone SST-SOB02 $145
Amd Nano $649

so far $2574

And more will come to, i have ofcourse to do more resurch, but u see, when wanting to build powerful SFF, you just have to forget abaut price, it is losy bang for buck- But spend round 25% of what you need totally for building a powerful SFF on a grapic card, is not so bad, if the performace is there.

its a lot to consider, ram for inctase, does fastram make an impact on gaming with hbm 4gb and so on, can that 45w cpu feed the gpu. But my point , at system with this cost 50$ more or less does not realli matter, ofcourse i can go nv 970 and save 300..., well i ocourse need win 10, but this is a process as evry SFF buid that aim for gaming is.

August 28, 2015 | 06:36 PM - Posted by Anonymous (not verified)

The BEST ITX CARD in the WOLRD. - AMD R9 NANO.!

I will grab one, when it hits the stores

August 29, 2015 | 02:07 AM - Posted by Anonymous (not verified)

$650... WTF?
Epicfail

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Lines and paragraphs break automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd> <blockquote><p><br>
  • Web page addresses and e-mail addresses turn into links automatically.

More information about formatting options

CAPTCHA
This question is for testing whether you are a human visitor and to prevent automated spam submissions.