AMD Radeon R9 Nano Video Card PAPER Launch @ [H]

Vittra said:
There is an SFF subforum on [H]. Few of the people who post there bother with the video card section, and vice versa. There's a reason for that. ;)
Click to expand...

Hopfully i join there sone, the force is strong in this one :) (nano)
 
It would be kinda silly for [H] to give the Nano an award of any kind, if it is lets say 90% of the performance of the fury-x.

If [H] start to think as the few SFF builders, they would loose a lot of readers, its a niche card, for a few buliders, maybe, just maybe will it trigger a new popularity of real SFF builds, but for the avarage pc user and gamer, this littel dimond will suck, both in price, performance.

Edit: And beside [H] does not have and reward what maybe would fit this littel card, they give gold silver for performance and thats it, evryone that read [H] knows that, and well it would be akward if they should give awards for something they never ever given an award for before. U have to take [H] for what it is :)
 
Stryke1983 said:
Because so far the evidence that Nvidia GPUs are going to "suck" at DX12 is based on one game in it's Alpha stage. So anyone with some common sense is going to withhold judgement until we actually see more examples of DX12 games.
Click to expand...

The way Kepler, Maxwell and Maxwell II architecture is designed prohibits it from utilizing all the features of DX12, including ASYNC and multiple drawcalls.
 
Nenu said:
When it came out it was by far the fastest card, allowing an experience nothing else could provide, there was no competition.
Its award was deserved.

When the Fury cards came out, there were already cards performing the same and much faster.
They did not come across with the same impact, in fact, the opposite.
They are priced so high its hard to recommend them let alone give them an award, especially the Nano.
Click to expand...

Award deserved? really thats a $1000 card LOL
 
Snus said:
*Wantapple, im a littel concered abaut the 175w thou, it will be a real challange to get that silent without loosing to much performance.

Any advices what is the best (fastest) 45w cpu out there, becouse well 45w, u can cool passiv with the right gear. But it has to be abel to not bottelneck the 175w nano, i do however recon Nano has to be a bit underclocked to get it silent.

Edit: Probally will have to set up, diffrent modes, one for silent and regular use, and one for gaming.
Click to expand...

i was thinking on Core i5-6500T 35w, hopefully it will fit my Noctua NH-L9i CPU Cooler. that cpu along with R9 Nano shoud be about 210w. ill be using silverstone sff 450w psu. but if i mod my other little case ill be using a 280w "bronze" flex psu.
 
DeadSkull said:
The way Kepler, Maxwell and Maxwell II architecture is designed prohibits it from utilizing all the features of DX12, including ASYNC and multiple drawcalls.
Click to expand...

We're talking about cards with THOUSANDS of little programmable cores. There may be solutions available through drivers that no one has yet seen.

We also don't yet know if full games will use these features. Most games about to hit the shelf have been in development for at least as long as the GPUs on which they will run.
 
wantapple said:
Award deserved? really thats a $1000 card LOL
Click to expand...

That review is 5 moths old. In that time, the Fury X and 980ti were both released, and in TW3 at least, the Titan X still outperforms them. Not to mention if you do work in CAD or graphic design and the extra memory is a benefit. From that POV, anyone who bought a Titan X at launch made a good investment.
 
I could only imagine the language that spewed out of Kyle's mouth after he disconnected that call - probably would've made even the most crude sailor blush.
 
wantapple said:
i was thinking on Core i5-6500T 35w, hopefully it will fit my Noctua NH-L9i CPU Cooler. that cpu along with R9 Nano shoud be about 210w. ill be using silverstone sff 450w psu. but if i mod my other little case ill be using a 280w "bronze" flex psu.[/QUOTE

35w is a beuty, how powerful is it ?
]
Click to expand...
 
wantapple said:
yes because the titan z was $3000
Click to expand...

AND slower. The 295 X2 was an awesome card by AMD, and still kicks ass today. The Titan Z was a 'me too' product that was really no competition. Even at the same price, the AMD card was faster.

Today the reverse is true...
 
Well AMD could hang up by not supplying a Nano for review. I hope not.

Since this card is so much of a power SFF niche card, not sure if HardOCP should even bother to review it. Also 175w is actually a lot for a very small case which will also need a sufficient and capable CPU to drive it. Since the card is fan cooled and dumps the heat inside of the case does pose issues.

Will it actually work in a tight configuration and deliver usable performance or can it even beat a 970 in the same configuration (tight as possible)? Which leads into if it can't then the purpose becomes more vague since other bigger options come into play.

To me it looks like a very fun card to play around with, now if practical is a whole new question.
 
AllBlackFan said:
Well that's a lot less rude than what I thought you'd done. My bad then clearly and I apologise. Given what you originally said I think you can understand how I got the wrong end of the stick though.

I'd imagined some poor AMD employee deflated, listening to a dial tone mid-explanation after you'd hung up on them.
Click to expand...

Most GPU briefings are conference calls with other editors. I stayed till the end, nothing really major. It was an overall quiet conference, there were no hardware questions at all at the end till I asked some hard hitting technical questions that no one else in the conference asked. I was the only one talking about the way the clock speed works, no one else was concerned. The vibe I'm getting is that the hardware aspects are going to be downplayed at the launch, with the emphasis on the size and space saving and power in that form factor from most media.
 
Zarathustra[H];1041827073 said:
Which is the appropriate thing to do. A review site that goes publishing recommendations or strong opinions based on speculation is going to have a serious confidence problem with its readership over time.

Until there are a series LAUNCHED titles with DX12 support, any benchmark is nothing but a preview, and not to be taken too seriously. The [H] has built its reputation on this approach. Relying on REAL game performance rather than speculation, or canned benchmarks. Hopefully that will never change, because that is what keeps me coming back.

It very well may be that the GeForce 9xx architecture doesn't handle some aspects of DX12 well. Seeing that it was designed before the API spec was completed, this wouldn't be a huge surprise.

On the other hand DX12 has so many features and leaves so much tweaking in the hands of the game developers, that even if the 9xx has dreadful issues in some titles, it might not in others.

That, and as I mentioned above, I'll play tomorrows games on tomorrows hardware. A GPU doesn't have a very long lifespan in my system anyway. No matter what happens, I won't be too disappointed in my 980ti's. heck, the newest games I'm even playing right now were launched in 2010 (Civ 5) and 2011 (Red Orchestra 2) so I don't think I'll have too much to worry about short term.

Once the DX12 titles become more readily available, I will use benchmarks of those titles with next gen GPU's from Nvidia and AMD as my deciding factor for my next GPU purchase, which by the looks of things is going to be a year away at least.
Click to expand...

I think game developers are still going to have to be concerned with DX11 performance in their games for a very long time to come. The base of DX12 gamers and Win10 users isn't going to be that high for a while. Devs will need to ensure DX11 works good and well for a few years more at least, so I see the DX11 experience still being supported and optimized in coming games.
 
Creig said:
Which would seem to apply to both the Titan X and the Nano. Yet the Nvidia Titan X gets an [H] Editors Gold Choice award while the AMD Nano gets, "I actually hung up on the call when they told us the price".
Click to expand...

The TITAN X and Fury X are worlds apart in performance and gameplay experience. They are no where near each other. AMD doesn't offer a single-GPU video card that can compete with the TITAN X on performance.

Also, the Fury X is allegedly faster than the Nano, but they are the same price. Buying a slower card, at the same price you can get the faster card, from a pure performance perspectivive (SFF aside) doesn't make logical sense.

And again, the Fury X cannot compete with the 980 Ti, let alone the TITAN X.

One offers a high level of gameplay experience no holds bar, the other does not.
 
I am not sure the TitanX is significantly greater in gaming experience over the 980Ti or FuryX. Does it allow one higher resolution with same IQ settings? Or just one or two more minor settings at the same resolution which you wouldn't notice unless looking for it?

Getting back to Nano:
  • What I would be looking for is does this card actually work in the smallest case possible with the best airflow? Maybe some examples or tests here. AMD should in this review give out a whole computer to show an ideal configuration to show what advantages it may have
  • There are many options now for small cases using virtually full size cards, what application or use would the Nano really work well in? If any.
  • What size case will it work compared to one that can support larger cards? For example if the next size up case is only a few inches deeper but will support even a TitanX what is this card for?
  • A Low power, very small size CFX configuration with performance would also be very interesting but then SLIing two 970s/980s would always be an option as well with virtually the same power requirements. Now for cases and mobo that would support two cards pcie 3 8x/8x) or better. What I've found for the mobo requires a case that seems like you could just use larger cards.
 
Last edited:
The only way I see Nano being a true win is if it is as fast as the FuryX. If it runs an avg 10 frames slower, then it is in line with the GTX 970 and the R9 290 and not worth the money.

The Fury X would of been a nice SFF card if they would of just made it with a slimmer radiator. I was not able to fit it in a Lian Li PC-TU100B because it hits the motherboard. The one great quality was suppose to be its size. So unless you can fit the cards into a SFF case, why not just buy the R9-290(x) or R9-390(x) for half the price.
 
Brent_Justice said:
Naturally, else it'd be a Fury X :p

The only thing separating the two is the artificial cap on TDP.
Click to expand...

And of course, they have to maintain the artificial differentiation between the two because the Nano sells for so much less and they don't want to poach Fury X sales with Nanos that perform just as well...um, wait, no... it's because...the...huge SFF market...no, wait...<FATAL ERROR: PERFORMING MEMORY DUMP>
 
Brent_Justice said:
Most GPU briefings are conference calls with other editors. I stayed till the end, nothing really major. It was an overall quiet conference, there were no hardware questions at all at the end till I asked some hard hitting technical questions that no one else in the conference asked. I was the only one talking about the way the clock speed works, no one else was concerned. The vibe I'm getting is that the hardware aspects are going to be downplayed at the launch, with the emphasis on the size and space saving and power in that form factor from most media.
Click to expand...


If only you'd asked "Will this thing overclock like no tomorrow"
 
Brent_Justice said:
I think game developers are still going to have to be concerned with DX11 performance in their games for a very long time to come. The base of DX12 gamers and Win10 users isn't going to be that high for a while. Devs will need to ensure DX11 works good and well for a few years more at least, so I see the DX11 experience still being supported and optimized in coming games.
Click to expand...

Exactly. This is what every kid screaming "DX12" in every post doesn't get. it's going to be years before developers are actually targeting DX12 and building games around it, rather than bolting on a couple extraneous DX12 features onto a DX11 game and calling it a "DX12" title. It's just not the magic saviour of gaming that some people assume.

Hell, the guy that helped AMD finalize Mantle isn't even shipping the first DX12 EA titles based on Frostbite until holiday 2016, and he's planning to co-ship them with Vulkan last he mentioned.

DX11 will continue to be targeted for years, in some cases even DX9. No matter how obnoxious Microsoft gets with trying to hit people over the head with near-forced Windows 10 installations, a massive pool of Win7 PCs will remain.
 
If Nvidia taunted DX12 support from Fermi and up and AMD Hawaii and up plus major consoles are DX12 hardware level, I think DX12 will pick up much faster then DX11 ever did. You already have DX12 hardware out there for years. If Nvidia is just that much slower in things then that is their problem.

Now on the PC side with everything before Win 10, DX11 maybe here for awhile. So AMD users will be able to play games in DX12 mode if on Win 10 while many Nvidia users will just use DX11. Probably not an issue.
 
DPI said:
Exactly. This is what every kid screaming "DX12" in every post doesn't get. it's going to be years before developers are actually targeting DX12 and building games around it, rather than bolting on a couple extraneous DX12 features onto a DX11 game and calling it a "DX12" title. It's just not the magic saviour of gaming that some people assume.

Hell, the guy that helped AMD finalize Mantle isn't even shipping the first DX12 EA titles based on Frostbite until holiday 2016, and he's planning to co-ship them with Vulkan last he mentioned.

DX11 will continue to be targeted for years, in some cases even DX9. No matter how obnoxious Microsoft gets with trying to hit people over the head with near-forced Windows 10 installations, a massive pool of Win7 PCs will remain.
Click to expand...

Developers for games on PS4 and Xbox1 are targeting DX12 hardware level already and past couple of years. Those games should have DX12 paths, at least the newer games coming out next year. How big a difference it will make has to be seen. Will AMD just cream Nvidia in performance - I doubt it (maybe in a few titles but that is about it).
 
DPI said:
Exactly. This is what every kid screaming "DX12" in every post doesn't get. it's going to be years before developers are actually targeting DX12 and building games around it, rather than bolting on a couple extraneous DX12 features onto a DX11 game and calling it a "DX12" title. It's just not the magic saviour of gaming that some people assume.

Hell, the guy that helped AMD finalize Mantle isn't even shipping the first DX12 EA titles based on Frostbite until holiday 2016, and he's planning to co-ship them with Vulkan last he mentioned.

DX11 will continue to be targeted for years, in some cases even DX9. No matter how obnoxious Microsoft gets with trying to hit people over the head with near-forced Windows 10 installations, a massive pool of Win7 PCs will remain.
Click to expand...

Except that DX12 and D3D12 are backwards compatible with previous versions. So there's no reason for developers not to take advantage of DX12 improvements.
 
noko said:
Developers for games on PS4 and Xbox1 are targeting DX12 hardware level already and past couple of years. Those games should have DX12 paths, at least the newer games coming out next year. How big a difference it will make has to be seen. Will AMD just cream Nvidia in performance - I doubt it (maybe in a few titles but that is about it).
Click to expand...

And by then, Pascal. If Maxwell chokes on DX12, NV can do like Microsoft with Windows 10--shrug their shoulders and say, "Sorry, you'll just have to upgrade.Hand us your money, please."

Side note--what ever happened to (Newton?) and (Ohm?)? Then again, they skipped a lot more of the alphabet between Fermi and Kepler, so who knows...
 
Well, this is going to be fun... Tech Report is saying they've been denied a sample Nano for testing.

May have missed it, but has [H] gotten theirs yet?
 
FYI: AMD has refused our request for a Nano to review.
 
Hehe, I didnt expect they would send you one.
Truth hurts, at least it will when you get a sample to review.
 
How many other sites hit the chopping block?
I did see rumors that AMD was being stingy with review samples this time.
 
I wouldn't send them out to you for review either, Kyle.

Or any other site.

Or anyone who might, you know... Test it.
 
I wish I could say this stopped me from buying one, but I never would have bought one.

I suspect they only have 3 or so samples. Because this card is probably only a marketing gimmick and they'll only ship 100 of them in total.
 
The fact that AMD has denied any up front sampling shows they KNOW this card will be a dud. I have doubts that it will even get within 10% of the real Fury X, especially once thermal throttling kicks in.
 
Maybe this is why AMD keeps failing, they focus on idiotic things. I mean stop trying all this nonsense and just make a good product that we want at a good price. Stop being that other guy that offers good price/performance ratios and try kicking some ass for once.
 
AMD knows they have a crap product and dont want egg on there face faster than whats coming. You know when your graphics card launch sales pitch is based on the size of the card and irrelevant 4k gaming, you just failed.
 
Evil_SPanKY said:
The fact that AMD has denied any up front sampling shows they KNOW this card will be a dud. I have doubts that it will even get within 10% of the real Fury X, especially once thermal throttling kicks in.
Click to expand...

Well my expectations of it is that it will be slightly faster than a 390x.
 
You must log in or register to reply here.
Back
Top