Fury Nano: First results in!

[RussianSensation]

No for several reasons.

Its a pointless product in SFF because its directly unfit.

Is that why mini GTX970 are unfit for SFF? I guess all those people who bought them are stupid you are saying? Lian Li is a stupid company for showcasing their miniITX cases with a mini GTX970, right?

I guess it didn't occur to you that some gamers who think outside the box could considering adding an AIO CLC on the Nano in a SFF case?

Silverstone is a stupid company too right for putting a GTX980 with an open air cooler into their MiniITX Raven case?

Maybe you should start a blog on teaching people to build miniITX systems the right away since you seem to know it better than AMD/SilverStone/Lian Li, etc.

Same reason why I wouldnt touch a 970 mini or anything like it. If you owned a SFF setup you would understand.
I am not emotional about it as you.

Show facts and evidence that a card like the Asus mini GTX970 or Fury with 175W TDP is unsuitable for a SFF PC. I can't wait until there are SFF PCs with Nanos from users who will all prove you wrong. If your SFF PC can't exhaust 175W of power, the problem is your chosen case/airflow management not the card.

Cards with the same performance and power usage already existed for a year.

Oh really? I wasn't aware you could buy a 6" card that has the flexibility of being nearly as fast as a Fury X once overclocked to 1.05Ghz.

Just as small for about 300$ too.
I don't think people will say "hey look a new card 12% faster for 479$" and dump their gtx970 mini's just yet.

HD7950 = operating clocks 800mhz
Real world operating clocks in the hands of enthusiasts = 1150-1275mhz.

Also, last time I checked, the Fury was almost 40% faster than a GTX970 at 1440P. Even an 800-900mhz fully unlocked Nano should have no problem beating the GTX970 by 25%+ if those marketing slides are not BS.

Since GTX970 is cut-down, it's not possible to overclock it to reach GTX980Ti speeds. If Nano is a full blown Fiji in mini-ITX form factor, slap a $60 AIO CLC and it may hit 1050mhz and thus match the Fury X. Did you ever think of that being a possibility? Or I guess overclocking doesn't count when it comes to AMD cards?

Further, miniITX GTX970 was never regularly on sale for $300. On various occasions the 970 Gigabyte card was $280-300, but it has a lot of issues with stability, temperatures, noise levels from reviews with its 3/5 rating.

And AMD did seppuku by pricing it at $649...
What a dissaster from AMD and a clear advise:

Source for the $649 MSRP on the Nano?

Intel has the golden chance to take AMD and Nvidia down since HBM will be hella expensive

You repeating and wishing for Intel iGPU to destroy NV/AMD's discrete GPUs in the CPU section might be cool cuz no one cares to waste time proving you wrong in the CPU section. Right now Intel may have caught up to crappy cards like GT630/730 but once NV/AMD have HBM2 ready on lower end cards, it's game over as far as performance goes.

I know some people here love throwing the idea that discrete GPUs are going the way of sound cards any minute now but even though the discrete GPU market has shrunk, there are still 9-10 million dGPU products sold each quarter.

Only casuals are OK with Intel's drivers and its control panel.

 

[n0x1ous]

If Nano is a full blown mini-ITX card, slap a $60 AIO CLC and it may hit 1050mhz and thus match the Fury X. Did you ever think of that? Or I guess overclocking doesn't count when it comes to AMD cards?

why not just buy a Fury X at that point then? overclocking barely counts because Fiji overclocks so poorly unfortunately.

Maybe someday we'll be able to overvolt them. can't get more than 50mhz out of mine.

if those marketing slides are not BS.

the fury x marketing slides vs reality don't inspire a lot of confidence here.......

 

[happy medium]

that has the flexibility of being nearly as fast as a Fury X once overclocked to 1.05Ghz.

says who? what review are you reading?

if those marketing slides are not BS.

really marketing slides?

 

[AnandThenMan]

How out of touch are you with the market that you consider AMD to be first to market this gen because it was the first to have HBM?

How out of touch do you have to be to realize that HBM allow a product like Nano to exist, and allowed AMD to significantly lower power consumption as well. And it allows AMD to gain experience with HBM.

HBM isn't some feature I can utilize as a consumer.

Really? That's odd I guess you don't care about form factor or power consumption. I guess you'll be skipping all future graphics products that utilize HBM which will be pretty much all of them.

The rest of your post is just

 

[happy medium]

allowed AMD to significantly lower power consumption as well

~30 watts is significant? on a 250 watt card? (Fury)

 

[tential]

How out of touch do you have to be to realize that HBM allow a product like Nano to exist, and allowed AMD to significantly lower power consumption as well. And it allows AMD to gain experience with HBM.

Really? That's odd I guess you don't care about form factor or power consumption. I guess you'll be skipping all future graphics products that utilize HBM which will be pretty much all of them.

The rest of your post is just

GTX 970 has already existed in that size bracket. So products of this size were already possible.

HBM memory did not allow AMD to beat Maxwell in power consumption. So EVEN WITH HBM, AMD still can't get it's architecture to use less power than Maxwell. So imagine what Nvidia will do with HBM.

People only care about something if it's a BENEFIT to them. HBM, on AMD's CURRENT cards offers me nothing that I couldn't already get from Nvidia.

But hey, AMD is clearly doing everything right, nothing they could do better, it's all the consumer's fault for not realizing that you should buy an R9 Fury, Fury X, or Nano because of HBM specifically is a great feature!
Lol......

 

[svenge]

I'm getting that "Surely you can't underdeliver three times in a row" vibe....

AMD has underdelivered in x86 innumerably many times in a row, why can't their graphics division get in on the act?

 

[Azix]

should be noted that the card is not completely dumping air into the case. Some of it is headed out of the case due to fin orientation.

Good job on the active vrm cooling. This card is the best choice for SFF PCs now. If ITX boards could do crossfire it would be epic for high end SFF.

 

[tential]

lavaheadache, has a R9 290x in a SFF case....
So really, this whole "You need xyz TDP" is just ridiculous in the first place.
http://forums.anandtech.com/showthread.php?t=2414541

(Not lava's build, but this is the same ml07 case).

So ya... unless the ml07 is no longer a SFF case, I don't need Nano there either.

But ya, I guess we now buy GPUs because of things like HBM, and ignore the performance of the card. Just another reason to buy AMD. You can get a slower card, but at least you have HBM!

 

[AnandThenMan]

HBM memory did not allow AMD to beat Maxwell in power consumption. So EVEN WITH HBM, AMD still can't get it's architecture to use less power than Maxwell.

So in your world AMD should have not bothered with HBM, I mean it's not a feature you are anyone can even use. Why would AMD want to lower power consumption, really now.

So imagine what Nvidia will do with HBM.

All we can do is imagine. And if we want to imagine, consider what AMD will be able to do with HBM2 and a new process node which they are traditionally better at moving to versus Nvidia.

People only care about something if it's a BENEFIT to them. HBM, on AMD's CURRENT cards offers me nothing that I couldn't already get from Nvidia.

So then you are simply not interested in AMD or HBM/HBM2 at all. Good for you.

But hey, AMD is clearly doing everything right, nothing they could do better, it's all the consumer's fault for not realizing that you should buy an R9 Fury, Fury X, or Nano because of HBM specifically is a great feature!
Lol......

This is nonsensical ranting, just stop.

 

[tential]

AMD has underdelivered in x86 innumerably many times in a row, why can't their graphics division get in on the act?

Ouch, I don't even know how to respond... I feel like I should have definitely saw that coming lol. You have to have one profitable/worthwhile division of your company then!

 

[happy medium]

consider what AMD will be able to do with HBM2 and a new process node which they are traditionally better at moving to versus Nvidia

show me the money, when?

 

[tential]

Give a reason that HBM specifically allows an AMD product to become a better choice than Nvidia's alternative.

I am not denying HBM is a good technology.

I'm saying that asking anyone to buy the R9 Fury/x or Nano specifically because of the fact that it has HBM is ridiculous.
HBM doesn't magically make the Fury X a better choice than GTX 980 Ti.
HBM doesn't magically make the Fury a better card than the GTX 980. It at best makes it equal in price/performance ratio, but the card STILL loses out in power consumption. Wow, AMD with a new technology that is supposed to significantly lower power consumption (your words), still can't lower it below the GTX 980?
HBM didn't magically create Mini ITX cards. The GTX 970 existed already, and the GTX 980 could be made into a mini ITX card too.
HBM didn't magically create low power cards. It lowers power consumption yes, but low power cards are possible before HBM. Look at Maxwell....

So yes, AMD is first to HBM. Yes, HBM has benefits. AMD was unable to utilize any of those benefits to do anything that is not already out on the market right now. How can I make this any more clear?

 

[happy medium]

So yes, AMD is first to HBM. Yes, HBM has benefits. AMD was unable to utilize any of those benefits to do anything that is not already out on the market right now. How can I make this any more clear?

That's about as clear as it gets and spot on.

 

[bystander36]

So in your world AMD should have not bothered with HBM, I mean it's not a feature you are anyone can even use. Why would AMD want to lower power consumption, really now.

All we can do is imagine. And if we want to imagine, consider what AMD will be able to do with HBM2 and a new process node which they are traditionally better at moving to versus Nvidia.

So then you are simply not interested in AMD or HBM/HBM2 at all. Good for you.

This is nonsensical ranting, just stop.

I think the point is that we don't really care about HBM/HBM2 specifically, but the product that it may or may not be in. I'm sure it is part of our future, but just because AMD used HBM on Fury doesn't mean we are going to go out and buy it. It is just one piece of a product. The whole product is what matters.

It does help shrink video cards, which were starting to get big, but this only matters if you are going to put together a small PC. I have a large case, and have no use for a small video card if that is its only advantage.

 

[AnandThenMan]

Yes, HBM has benefits.

People only care about something if it's a BENEFIT to them.

??????

 

[desprado]

AMD have make sure that nano will non selling that for sure.

 

[happy medium]

??????

translation...HBM has benefits but not for us consumers yet.

 

[SimianR]

Honestly it seems like there's a lot of debate going on without seeing any real numbers yet. Until we have reviews we have very limited info. If the Fury Nano ends up being around the same as a 980 - then it has delivered performance in a form factor we haven't really seen yet. I personally think $650 is insane - but all we have is a WCCF table showing that price. At the right price point it could be an attractive option for certain systems, I think $500-550 would be the sweet spot. AMD is in a tough spot because of how they have priced the Fury/X.

 

[happy medium]

I think $500-550 would be the sweet spot.

I think that would be a disaster.

Did anyone hear if the Nano will have HDMI 2.0?
I thought I read somewhere it will.

 

[Abwx]

HBM memory did not allow AMD to beat Maxwell in power consumption. So EVEN WITH HBM, AMD still can't get it's architecture to use less power than Maxwell. So imagine what Nvidia will do with HBM.

Lol, they payed a premium to have a custom process at TSMC, but here we have people that think that it is due to clever enginering..

Ask any EE, once he see the frequencies and voltages he can instantly compute the transistors caracteristics differences..

As to imagine what Nvidia would have done with HBM this is just laughable, truth is that they would had been incapable to develloppe it in time, that s why they are waiting that AMD clean the road before jumping on the ship..

Here the timeline of HBM with the tests chips manufactured during those 8 years :

http://www.computerbase.de/2015-08/amd-fiji-vom-ersten-schritt-zum-fertigen-produkt-in-8.5-jahren/

 

[Azix]

I think that would be a disaster.

Did anyone hear if the Nano will have HDMI 2.0?
I thought I read somewhere it will.

if it has the performance of a current $500 card, no.

dp to hdmi 2.0 incoming http://www.paradetech.com/products/displayport-format-converters/ps176/

 

[tential]

if it has the performance of a current $500 card, no.

dp to hdmi 2.0 incoming http://www.paradetech.com/products/displayport-format-converters/ps176/

AMD - "Just wait, it's coming."

Just like how people had to wait for the 290x to beat the 780ti, or how they had to wait for dx 12 to fully utilize their cards.

The "It's incoming" is NOT a good sign for AMD, yet something they repeatedly have to say.

 

[Abwx]

AMD - "Just wait, it's coming."

Trolling and derailing the thread, that s generaly the only thing left to the obviously argumentless...

 

[Azix]

AMD - "Just wait, it's coming."

Just like how people had to wait for the 290x to beat the 780ti, or how they had to wait for dx 12 to fully utilize their cards.

The "It's incoming" is NOT a good sign for AMD, yet something they repeatedly have to say.

not AMD that makes the adapter, but .... w/e ...

end of the day the hdmi 2.0 excuse has a short shelf-life.