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

§kynet;1041821032 said:
You need to rethink this. The end user finances the production of said product, 100% on the consumer.

What it costs AMD to produce is irrelevant to the end user. What matters is the card itself and what it costs. It doesn't matter if it costs AMD $100 to make it or $1000. If someone doesn't think it's worth the price that's all there is too it. If AMD can't afford to price it lower once stock catches up to demand than that's their problem. I'm not going to feel sorry for AMD. They've managed to dig their own grave and I don't pity companies that ruin themselves through horrible mismanagement.
 
The 390x is pointless and confusing with the Fury/Fury Nano. I would still like to see a 390 review. I know it is a rebrand, but anything that can edge a gtx970 on performance/$ is worth a review.
The 390 has been wildly popular in PC building communities. Probably doesn't mean much to AMD's bottom line but it's getting more attention than the 290 or 290X did.

Almost entirely due to this review.
 
AMD is is a CPU/GPU manufacturer not a memory developer.
Actually not true AMD has a history of introducing new types of memory. AMD was also not an API developer but they went down that road and look at the result. Unfortunately for AMD they are not adept at monetizing their ideas, at least not up to this point. HBM and Mantle/DX12 could change this will will have to wait and see.
Personally, I wish they had put all of their HBM efforts into improving their GPU architecture instead... to compete better with Nvidia.
HBM allowed AMD to increase the power budget to the GPU and compete with Nvidia.
Presently, HBM does nothing but reduce PCB size and save about 50W over GDDR5.
This is a tremendous difference.
It also limits them to 4 GB on these first-gen cards, as well as limited supply. AMD is playing the long game on HBM but that does nothing to help them today.
That's why they call it the long game.
 
If anyone is curious how I would re-design the AMD lineup this year, I have my opinions, these are my own, but I think they make some sense.

Firstly, I would not have release re-branded 390 and 390X. I would however have kept the re-branded 285 to 380. IN fact, I would have increased the memory performance even higher than they did.

Fury X - I would price this card at $549. It offers better performance than a GTX 980, but doesn't perform as well as a 980 Ti. Therefore, this price right between the GTX 980 and 980 Ti makes a lot of sense. The price is justified because it does allow a better experience than a GTX 980, so the higher price would make sense. But, it isn't near 980 Ti performance, so the price is lower than 980 Ti by $100. This gives gamers a great alternative between 980 and 980 Ti.

Fury - I would keep the specs the Fury is at, and release the card at $449. At this price it would give the GTX 980 a run for its money being $50 cheaper MSRP. 980 was popular, and AMD really needed to target competition with that card, the Fury would have been perfect to do so with. The Fury would offer the same, if not a little better, performance than 980 for $50 cheaper.

Nano - I would price this card around $329, it would take the place of the 390X/390 position. I would make the Nano the same cut down specs as the Fury, same exact shader count, texture units, rops and then do my magical TDP monitoring to keep it within a target power performance. This would still be a 6" card, and would run even cooler and more power efficient than the current Nano making a real "wow" factor in this price range. This card would sit between the 970 and 980 on performance. It would give the GTX 970 a run for its money.

It would also bring Fiji architecture and latest techonlogy down to more users. You'd now have the latest GCN filling the high-end to midrange, and get rid of the confusion of the fact that 390/390X is a Hawaii GCN 1.1 part. You'd fill this space with your brand new GCN Fiji technology, as it should be today and end the confusion that the "new" 390/X is actually older technology than the 380! In this way, from the 380 up the scale it would all be the latest GCN architecture.

Having a small card, that runs super cool and power efficient sitting where the 390/X is, replacing them, giving the 970 a run for its money would be a real show stopper. People would notice. This card would sell.

Then you have a nice lineup for cards, 380 (with improved memory clocks) to Nano, to Fury to Fury X. Everything below 380 can be the other re-brands we have now.

This would be my lineup.

That lineup would make sense if amd was producing butt loads of Fiji gpus, but their not. Being a limited avaibality part, there in no reason to lower the price. I would have made a nano x and a vanilla nano. Milk it for all it worth. That being said, I would touch a furry gpu. If I was buying today, tomorrow, next week or next month, I'd buy nvidia.
 
That lineup would make sense if amd was producing butt loads of Fiji gpus, but their not. Being a limited avaibality part, there in no reason to lower the price. I would have made a nano x and a vanilla nano. Milk it for all it worth. That being said, I would touch a furry gpu. If I was buying today, tomorrow, next week or next month, I'd buy nvidia.

I think Brent is discussing this from a bigger picture sort of view rather than how the market is constrained right not by availability.
 
Well, my first thought when I heard of this card was: 2 x R9 Nano Crossfire around $800, great deal. Now that I hear that it is $649 each, I just walk away shaking my head. Mind you, I would not buy at 980Ti either because that is beyond what I would want to spend for a card. (My last card was $400 and that was R9 290 Reference in November, 2013 and it is still going strong, even a 4k.)
 
The principal objection to the Nano is the price. And, to a lesser extent, the omission of HDMI 2.0.

Fiji's HDMI support-level is internal; it's intrinsic to the current GCN architecture it's based upon. Some seem to think that HDMI 2.0 is something that is "bolted-on." That it is something AMD can choose to either include or omit, and that they are being stingy and/or foolish to not provide it. The truth is, it's absence is literally baked into the silicon. And while I agree that it would have been wonderful to have had the foresight to include it, I also recognize that Fiji's design was probably finalized a LONG time ago, and HDMI 1.4a was, I guess, deemed sufficient.

As to the price, I think including the price of the Titan, 980Ti, and Fury X does have some relevance to the price of the Nano. (I'll state at this point that I feel the Nano is priced higher than I anticipated. But, at the same time, I admit that I didn't think we would be getting the full Fiji GPU, either.)

Why? It's plainly obvious that there isn't a $350 difference in gaming performance between the Titan and 980Ti. Would anyone dispute that? In a few cases, there is hardly any difference between these two cards at all. As for the Radeon, the Fury X clearly doesn't perform as well as the 980Ti in several games; it mostly meets, or more often slightly under-performs, the 980Ti. So, we have:

Titan at $999
980Ti at $649
Fury X at $649

The obvious outlier here, purely on performance, is the 980Ti. There is no reason I can think of, save one, why we're getting this level of performance from Nvidia at this price. (Although I'm certainly glad that we are.) That single reason would be that with the 980Ti launching roughly a month prior to the Fury X, Nvidia was willing to trade entirely realizable profits on the sale of the 980Ti at, for instance, $749 to $799, for the opportunity to go full-MMA with the Fury X at $649. This is, of course, complete conjecture on my part. In support of it, I'll say that if the retail price of the 980Ti was $749 all along, they would still have sold quite well at launch, and I think sales would still be very good. Why wouldn't they be? Your getting probably more than 92% of the performance of the Titan, on average, for 75% of it's price. This exercise in alternate reality would also have served to take some of the heat off the Fury X being priced at $649.

I don't know if anyone has published what the BOM costs of the Fury X are, but I'll bet that the Fiji's HBM design is quite possibly heart-stoppingly expensive to manufacture. HBM development costs were probably pretty stiff, too. Don't know if yields are good or bad, but I also wonder how much time is needed to produce these as compared to a conventional 28nm GPU. In other words, the manufacturing rate.

Even if one sets aside the expense, and success-rate, to find a "Nano-suitable" candidate through binning, I really have to wonder how much cheaper any of the Fiji-based cards could actually be at this point. Arguing that lower prices would increase demand makes perfect sense only if your supply-line can keep pace, and you have the time to spread out your R&D and manufacturing costs. I don't believe AMD has the luxury of time.

It might just be that a Fiji GPU that can operate consistently at around 900MHz and stay within the 175W envelope that AMD has specified for this tiny 6" card isn't all that common. I guess AMD feels it's price is justified. AMD can look at the bottom-line on what both the Fury X and Nano should sell for in a way no one else can. Since this isn't a company that is sitting atop a mountain of money, I can only assume they're pricing the Nano at what they feel is fair-market value.

I was certainly expecting a lower price, but I was also expecting something like only 3328 SPs, so that not only would AMD be able to hit their TDP target, but find an additional home for any chips that didn't pass muster with all 4096 SPs, or even 3584 SPs, enabled. That probably would have lowered manufacturing expenses across the entire Fiji line, and give AMD a bit more flexibility in pricing. Instead, AMD chose the harder path for them, and unfortunately, the more expensive one for us.

Not sure that AMD ever stated that the Nano was going to be the go-to graphics card for the great unwashed masses, but it obviously is just the opposite. This is clearly a very specialized product for a very particular application. I'm certainly not among it's potential owners, but I wouldn't criticize someone else's choice to buy one, as long as it's an "eyes wide open" decision.

Speaking of criticism, I think this "PAPER launch" criticism is a bit undeserved. OK, it's deserved, but a bit over-emphasized. They missed their target date by two weeks. In an industry where product launches are hardly rare. At least AMD provided specifications. With those specs, I'm sure that there are Fury X cards being underclocked to 850-900MHz as I write this for obvious reasons.
 
I think Brent's idea makes sense using GDDR 5. Fury X at 8 GB all else the same is probably doable at $550 or so. They could have gotten higher volume and the overall net profit would likely be higher. HBM could have stayed for the Nano and X2 due to thermal considerations and frankly that differentiation could probably have supported strong pricing as they would be lower volume niche products.

Perhaps AMD's biggest issue is that GCN doesn't seem to mesh with a DX11 world, and so it was probably introduced about two years too early. GCN should really shine under DX12. But with a limited budget and resources you make do with what you have.

Brent's pricing strategy would be awesome for consumers but unfortunately would probably be bad for AMD's bottom line.
 
If anyone is curious how I would re-design the AMD lineup this year, I have my opinions, these are my own, but I think they make some sense.

Firstly, I would not have release re-branded 390 and 390X. I would however have kept the re-branded 285 to 380. IN fact, I would have increased the memory performance even higher than they did.

Fury X - I would price this card at $549. It offers better performance than a GTX 980, but doesn't perform as well as a 980 Ti. Therefore, this price right between the GTX 980 and 980 Ti makes a lot of sense. The price is justified because it does allow a better experience than a GTX 980, so the higher price would make sense. But, it isn't near 980 Ti performance, so the price is lower than 980 Ti by $100. This gives gamers a great alternative between 980 and 980 Ti.

Fury - I would keep the specs the Fury is at, and release the card at $449. At this price it would give the GTX 980 a run for its money being $50 cheaper MSRP. 980 was popular, and AMD really needed to target competition with that card, the Fury would have been perfect to do so with. The Fury would offer the same, if not a little better, performance than 980 for $50 cheaper.

Nano - I would price this card around $329, it would take the place of the 390X/390 position. I would make the Nano the same cut down specs as the Fury, same exact shader count, texture units, rops and then do my magical TDP monitoring to keep it within a target power performance. This would still be a 6" card, and would run even cooler and more power efficient than the current Nano making a real "wow" factor in this price range. This card would sit between the 970 and 980 on performance. It would give the GTX 970 a run for its money.

It would also bring Fiji architecture and latest techonlogy down to more users. You'd now have the latest GCN filling the high-end to midrange, and get rid of the confusion of the fact that 390/390X is a Hawaii GCN 1.1 part. You'd fill this space with your brand new GCN Fiji technology, as it should be today and end the confusion that the "new" 390/X is actually older technology than the 380! In this way, from the 380 up the scale it would all be the latest GCN architecture.

Having a small card, that runs super cool and power efficient sitting where the 390/X is, replacing them, giving the 970 a run for its money would be a real show stopper. People would notice. This card would sell.

Then you have a nice lineup for cards, 380 (with improved memory clocks) to Nano, to Fury to Fury X. Everything below 380 can be the other re-brands we have now.

This would be my lineup.
How is the Nano $130 less in your scenario? Is the overclocking headroom of the Fury really worth that much more than the Nano with it's "and then do my magical TDP monitoring to keep it within a target power performance."?

I would say shave $50 off the nano from the Fury making it $400. I can't imagine the performance gap between the two to be more than $50. Otherwise I would think the Nano would cannibalize the Fury sales at $329 vs $450. Perhaps I am missing something there.
 
Thats what I don't understand how they can still suggest people get it but bash anything from AMD that they deem not price/perf.
I don't either, at all. Seems like AMD asking a premium price has ruffled more than a few feathers, almost like people don't want AMD to be able to sell products at a premium regardless of the performance. Which brings us back to Titan X, terrible value but it keeps getting praised.
I think Brent's idea makes sense using GDDR 5. Fury X at 8 GB all else the same is probably doable at $550 or so.

This would be around a 320 watt part if not more and you can guess the flak AMD would get for that.
 
Speaking of criticism, I think this "PAPER launch" criticism is a bit undeserved. OK, it's deserved, but a bit over-emphasized. They missed their target date by two weeks. In an industry where product launches are hardly rare. At least AMD provided specifications. With those specs, I'm sure that there are Fury X cards being underclocked to 850-900MHz as I write this for obvious reasons.

Here is why I decided to highlight this. I hate paper launches and I always have. I see these "coming back" again so to speak. I intend to draw attention to this trend in our coverage from here out. When it is a paper launch, I am going to call it as such. AMD wanted us to call it a "soft launch," or that is how AMD referred to it. Had we had hardware and drivers to test, but no availability on the street, I would have been good with that. This paper launch puts us in a position to simply regurgitate the company's slide deck, and quite frankly, I do not like that position. It is not good for the enthusiast community.
 
Others have been warned, and bannings have already started. If you decide to get off topic, you will get a week off. Should you like to discuss the Witcher story we did, I highly suggest you do it in that thread, not this one.
 
§kynet;1041821138 said:
So if AMD said absolutely nothing for another 2 weeks you'd be 100% fine when it does launch? So what are the rules here, how much can a company say about an upcoming product and when? BTW if you want to call it a paper launch I respect that, your call. I guess I just have a higher tolerance for better or worse.

I would imagine he has no issue with what a company says or when they say it. However, a paper launch is still a paper launch and calling it that is not a bad thing. Parts are not available at all so, what are they going to test?
 
§kynet;1041821138 said:
So if AMD said absolutely nothing for another 2 weeks you'd be 100% fine when it does launch? So what are the rules here, how much can a company say about an upcoming product and when? BTW if you want to call it a paper launch I respect that, your call. I guess I just have a higher tolerance for better or worse.

This is usually how it goes. Hardware and drivers come to us that are under NDA. Specifications and presentations come to us that are under NDA. Product launches with stock on the shelves on that day and if not, in a day or two. Embargoes are lifted and reviews are published. I have done this for nearly two decades and I have helped push this industry away from paper launches. If you recall that used to be the norm.

AMD can do what it wants. This was a paper launch and that is exactly what I am going to call it.
 
The principal objection to the Nano is the price. And, to a lesser extent, the omission of HDMI 2.0.

Fiji's HDMI support-level is internal; it's intrinsic to the current GCN architecture it's based upon. Some seem to think that HDMI 2.0 is something that is "bolted-on." That it is something AMD can choose to either include or omit, and that they are being stingy and/or foolish to not provide it. The truth is, it's absence is literally baked into the silicon. And while I agree that it would have been wonderful to have had the foresight to include it, I also recognize that Fiji's design was probably finalized a LONG time ago, and HDMI 1.4a was, I guess, deemed sufficient.

As to the price, I think including the price of the Titan, 980Ti, and Fury X does have some relevance to the price of the Nano. (I'll state at this point that I feel the Nano is priced higher than I anticipated. But, at the same time, I admit that I didn't think we would be getting the full Fiji GPU, either.)

Why? It's plainly obvious that there isn't a $350 difference in gaming performance between the Titan and 980Ti. Would anyone dispute that? In a few cases, there is hardly any difference between these two cards at all. As for the Radeon, the Fury X clearly doesn't perform as well as the 980Ti in several games; it mostly meets, or more often slightly under-performs, the 980Ti. So, we have:

Titan at $999
980Ti at $649
Fury X at $649

The obvious outlier here, purely on performance, is the 980Ti. There is no reason I can think of, save one, why we're getting this level of performance from Nvidia at this price. (Although I'm certainly glad that we are.) That single reason would be that with the 980Ti launching roughly a month prior to the Fury X, Nvidia was willing to trade entirely realizable profits on the sale of the 980Ti at, for instance, $749 to $799, for the opportunity to go full-MMA with the Fury X at $649. This is, of course, complete conjecture on my part. In support of it, I'll say that if the retail price of the 980Ti was $749 all along, they would still have sold quite well at launch, and I think sales would still be very good. Why wouldn't they be? Your getting probably more than 92% of the performance of the Titan, on average, for 75% of it's price. This exercise in alternate reality would also have served to take some of the heat off the Fury X being priced at $649.

I don't know if anyone has published what the BOM costs of the Fury X are, but I'll bet that the Fiji's HBM design is quite possibly heart-stoppingly expensive to manufacture. HBM development costs were probably pretty stiff, too. Don't know if yields are good or bad, but I also wonder how much time is needed to produce these as compared to a conventional 28nm GPU. In other words, the manufacturing rate.

Even if one sets aside the expense, and success-rate, to find a "Nano-suitable" candidate through binning, I really have to wonder how much cheaper any of the Fiji-based cards could actually be at this point. Arguing that lower prices would increase demand makes perfect sense only if your supply-line can keep pace, and you have the time to spread out your R&D and manufacturing costs. I don't believe AMD has the luxury of time.

It might just be that a Fiji GPU that can operate consistently at around 900MHz and stay within the 175W envelope that AMD has specified for this tiny 6" card isn't all that common. I guess AMD feels it's price is justified. AMD can look at the bottom-line on what both the Fury X and Nano should sell for in a way no one else can. Since this isn't a company that is sitting atop a mountain of money, I can only assume they're pricing the Nano at what they feel is fair-market value.

I was certainly expecting a lower price, but I was also expecting something like only 3328 SPs, so that not only would AMD be able to hit their TDP target, but find an additional home for any chips that didn't pass muster with all 4096 SPs, or even 3584 SPs, enabled. That probably would have lowered manufacturing expenses across the entire Fiji line, and give AMD a bit more flexibility in pricing. Instead, AMD chose the harder path for them, and unfortunately, the more expensive one for us.

Not sure that AMD ever stated that the Nano was going to be the go-to graphics card for the great unwashed masses, but it obviously is just the opposite. This is clearly a very specialized product for a very particular application. I'm certainly not among it's potential owners, but I wouldn't criticize someone else's choice to buy one, as long as it's an "eyes wide open" decision.

Speaking of criticism, I think this "PAPER launch" criticism is a bit undeserved. OK, it's deserved, but a bit over-emphasized. They missed their target date by two weeks. In an industry where product launches are hardly rare. At least AMD provided specifications. With those specs, I'm sure that there are Fury X cards being underclocked to 850-900MHz as I write this for obvious reasons.

Yeah these are probably the best binned chips. Perhaps only a tiny fraction of Fury GPUs can operate within that power envelope and have all of those shaders enabled. So the price is justified. It's a niche product that takes advantage of HBM's strengths, for a tiny minority of customers. It's going to cost more because it's new tech and there is no competitor with the same level of performance within that form factor and power consumption range. AMD also needs to be careful not to cannibalize sales of normal fury and R9 390.
 
Well, I guess that's it for AMD's future. Their pricing folks have clearly completely lost it.
 
It's clear that AMD's marketing department has lost its mind. If they had any sanity left, they would have marketed this is "X% smaller than the Titan X - finally, THE small enthusiast card!" and sold it for $999. All the AMD fanboys would still buy it, and their sales for everyone else would still stay at what it already will be, zero. Free money. What kind of dysfunctional company turns down free money?
 
If I was going to make a mini htpc I would still go with the fury x because it will probably be quieter and keep things cooler.

actually, the temperature of the Fury-X GPU is irrelevant because you have no headroom for overclocking, even at 60'C

and because the Fury-X uses noticeably more power than the 980 TI, the temperature that the Fury-X is generating via the radiator is actually putting more heat into your room than the 980TI does even though the 980TI is running at 75-80'C

the watercooler on the Fury-X has made people ignorant to this shit
 
For what it is, i dont think the price is that bad thou, had expected it somwhere between fury and fury x, but for thoose of you who build regular pcs, this card is not ment for you, its ment for SFF, a small niche in the pc marked, and its contender is the SFF NV 970, and well i would have wished it a tad lower, but can live with that price.

Why would you that build in regular cases even need the Nano, for heavens sake you have tons of diffrent brands to choose from on both nv and amd side. Fury, 980 TI, 980, Titan, Fury X, and when there finally comes a SFF card that can offer great performace, you even manged to missunderstand its place. Figure that.

Silent, powerful, small, thats the holy grail for an SFF builder, always been.

Edit_ is that, hey its a cute littel card that is powerful, let us put it our big ugly pcs, maybe thouse also get cute ?????? (teasing)
 
Last edited:
Thats why I hope they review it against that card and do judge it accordingly

Are you going to compare it to only small form factor boards like the 970 Mini-ITX? http://www.newegg.com/Product/Product.aspx?Item=N82E16814125706

If so, if its only a few FPS faster will you also say something like "If you want the best experience from a small form factor, the Fury Nano is your best bet" or will you denounce it because its $350 more?

Thats why I asked about the Titan X vs 980 TI. You've said that for performance you are willing to spend $350 for 2fps, will that stay the same with Nano form factor?
 
Form factor might be great but how many ITX cases exist that don't use half-height gpu and can't fit bigger card while they have space for Nano ?

Aren't gaming oriented ones designed to git 10,5 " cards anyway ?
 
Thats why I hope they review it against that card and do judge it accordingly

Dont even excpect that, there at best will be a few HW sites for regular users that even grasps the concept, its nothing new, and it want be only [H], SFF has been compared to those big monstrues ugly cards up true the years, they cant understand the goal of a SFF powerful build, because its easy to throw in what ever you want into a big ugly pc case and get the performace, even quite easy to make it silent, with watercooling, all what you thought that you knew from building in thouse big ugly cases doesnt count, when you build , for small , silent and powerful.

So why isnt building SFF so popular, well u have quite powerful laptops, and they are bye design SFF, and they even comes with a screen , keybord and stuff.

So yes SFF relly is a small niche marked, but its darn fun to build, when you have to figure out the max size of evry component, find the lowes tdp cpu to drive your gpu, find thouse fans that is small and silent (thats not easy), SFF can only get popular again, if we mangade to build SFF that can beat thouse gamer laptops, and be quite close to those big ugly monster pcs.
 
Form factor might be great but how many ITX cases exist that don't use half-height gpu and can't fit bigger card while they have space for Nano ?

Aren't gaming oriented ones designed to git 10,5 " cards anyway ?

Yes , ask urself why.... the answear is as easy as it gets, there was no powrful SFF gpus, nvida has done a great thing with makeing the SFF 970, and im happy with that AMD follows up with the Nano, there are comming more parts SFF psu is out there , they come at a price premium, with 2 powerful SFF cards SFF970/Nano, maybe its a chance there will come more SFF cases, even thou, you have to excpect to do modding, dremling and so on, as in the good old days before powerful laptops beated the H*** out of what we could build in SFF.
 
Last edited:
Cases like this is what i want for nv970/nano, http://www.quietpc.com/st-fc5-alpha, can it be done, well as long as a rizer card can turn the gpu card then it can be done, but, yous see, thats not even all, you have to fit a powersupply in there as well, becouse the 150w that case comes with, dosnt need to tell why.

Well thats my taste in SFF, but other will have other prefrances.

Edit , how can i even see a fit for that nano card in that case http://www.quietpc.com/st-fc5-evo , well look there is room for a singel slot grapics card, turned with rizer card, then max you can use is a mini itx board, that should give hight enough for a dual slot grapic card, you alse see you have to drill more hols in the bottom so the gpu can get air for its fan, well the cpu is cooled bye the chassis, so yout have to find the best 45w cpu to drive ur gpu, and the most difficult par will be fitting and finding a psu that can handel what the componets you have choosen need, so why am i telling you this, becouse it is what building SFF is abaut, you cant just slap componets in a big ugly chassis, even thou i see i maybe can be done in that chassis i linkted to, i am not sure, but god im already having fun figuring that one out. I guess thouse who read my posts can see that im really excited abaut the nano, yes i am, but i bet only a few can understand that. :)
 
Last edited:
As some people said here, the TitanX does not offer a 350$ performance premium over the 980Ti and it still costs 999$. The TitanX offer premium for those who can afford it. Nvidia is making money like this, AMD is following up.
 
Well, it's kinda obvious what's happening:
Yield is low. Guessing on the fact that this is a fully enabled chip, the yield on the actual Fiji chip is pretty good, but the yields of the whole die + interposer + ram stacks is low. Kind of sad but it's the reality right now.

In any case, due o the basic laws of supply and demand, AMD knows that they will sell every single one of these pretty much at any reasonable price. I bet they could go higher and still sell them all.

However, there is definitely a negative public image created by the high prices.

Oh and AMD, STOP MARKETING THESE AS 4K ! That is doing more harm than good, seriously. The Nano, especially, could do well enough based on the fact that it would be the fastest in that size, and maybe even at that tdp? At the right price, targeted like that, this could have been a really really great product.
 
Well i think its a great product for what it is ment to be, wich i have tried to explain.

Edit: there is a if there , is it silent enough to reach to the few SFF builders left :)
 
And well since i am surching for what i would need to make http://www.quietpc.com/st-fc5-evo, ready for the nano, the chassis is 450 x 345 x 72mm (W x D x H), this silverstone 450w psu http://www.silverstonetek.com/product.php?pid=342&area=en , is 125mm(W) x 63.5mm(H) x 100mm(D) SFX Form , that meens all the parts is there to make that case a good home for SFF cards that is dual slot high.

Most of you probally want care, becase SFF is nothing to you, but well there are still some like me, who nv 970 and nano is cards that gets us excited.
 
Well i think its a great product for what it is ment to be, wich i have tried to explain.

Edit: there is a if there , is it silent enough to reach to the few SFF builders left :)

I've heard an awesome saying that goes a bit like "There is no such thing as a bad product, just a bad price". And that's the issue with this card: 980Ti price is just too damn high. I can understand this card as being a 'luxury' SFF card, but $649 is beyond luxury for this product. Even $559 would have been a much better price-point. It still commands a luxury tag, it outpaces other cards near the ~450-500 price but is half their size...

but 649!? That's a joke... and nobody is laughing.
 
Nvidia cards only do 8 bit on HDMI 2.0 at 4k. There is simply not enough bandwidth with HDCP overhead. Nvidia is also is not 10 bit desktop capable at any resolution (or on Dp), only in DX. Read it on their website yourself. This means no support for ultra wide gamut media in future.

That's a big reason why AMD does not include HDMI 2.0 when marketing a 4k card, that does 10 bit on desktop and in games. Now you can get 40"+ DP 4k 10 bit screens.
By not having HDMI2.0, this has forced the market to build such screens, I am glad to see this. Of course I'd be not so happy if I'd rushed out and bought a 4k 8 bit HDMI2.0 screen + gfx card.

But AMD does not push this marketing re:8 and 10 bit support.

But lets just hate everything AMD does and be angry fanboys, to troll my post which is 100% correct regarding Nvidia/HDMI capabilities. I've done my homework, as I'm in the market for a dual 4K card setup on a 40" plus screen, as [H] says there is no single card 4k solution currently viable for max settings, with good fps, in most games..

I do agree a slight price drop would've been perfect for the Nano though, the tiny ITX builds that are possible do excite me, even though I'm a full size builder.
 
I've heard an awesome saying that goes a bit like "There is no such thing as a bad product, just a bad price". And that's the issue with this card: 980Ti price is just too damn high. I can understand this card as being a 'luxury' SFF card, but $649 is beyond luxury for this product. Even $559 would have been a much better price-point. It still commands a luxury tag, it outpaces other cards near the ~450-500 price but is half their size...

but 649!? That's a joke... and nobody is laughing.

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.
 
Last edited:
And well since i am surching for what i would need to make http://www.quietpc.com/st-fc5-evo, ready for the nano, the chassis is 450 x 345 x 72mm (W x D x H), this silverstone 450w psu http://www.silverstonetek.com/product.php?pid=342&area=en , is 125mm(W) x 63.5mm(H) x 100mm(D) SFX Form , that meens all the parts is there to make that case a good home for SFF cards that is dual slot high.


Excellent PSU, we reviewed it 2 years ago. http://www.hardocp.com/article/2013/03/14/silverstone_sfx_st45sfg_450w_power_supply_review

Still very expensive though, so it will fit right in with the Nano. ;)

https://pchound.com/442647/silverstone-450w-st45sf-g/?x=0
 
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.

You should put this system in the general hardware thread. There's things that can be better.

Personally if I was doing a low profile PC I'd buy the chassis for "show" (power and optical disc) then use long length wires to a normal PC somewhere else (behind the TV stand or basement, attic, whatever). 15'+ cables are very cheap. Even with buying the cables it saves $ and you don't have to sacrifice power. Also makes it completely silent.
 
Last edited:
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.

Have you put into any consideration on what impact of placing a relatively large source of heat (whether the Nano or 970 doesn't matter, both are still 150w+ at load) into such an enclosure (essentially no active airflow direcly out of the case)? None of these mitx solutions are fully exhausting.
 
Again with the half-assed marketing benchmarks, ie 0xAF.

Can't wait to see the Arctic Island card marketing slides: "Beats Nvidia when we disable hardware triangle setup and transform and lighting!"
 
kyle, brent

when are you expecting a nano on the test bench?

is should have a little oc head room right? its a 175w card, but it gets 150w from the 8pin and 75w from the pcie slot.
 
Back
Top