• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

FRONTPAGE AMD's R9 Nano Details Emerge! Launch Imminent!

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
Definitely an interesting card for ITX builders. The power limit and how it affects actual average core speed will be something to watch.

It sorta seems like the actual will be 900Mhz and "max 1000" is for marketing and only seen in spikes. If 1000 can only be hit in "lighter load" scenarios, it seems fairly useless to even hit 1000...

Nice write-up though, ED. Can't wait to see benchmarks when you get your hands on one :thup:
 
It sorta seems like the actual will be 900Mhz and "max 1000" is for marketing and only seen in spikes. If 1000 can only be hit in "lighter load" scenarios, it seems fairly useless to even hit 1000...

Sounds like modern mobile CPU power management. Instead of putting an upper limit on the speed, let it jump to max speed to get stuff done quicker so it can go back to idle speed. I don't really like that, as I tend to have lots of background tasks running that keep the CPU constantly at full power if I don't force an upper MHz limit.
 
Yea, sounds the same. Except GPU load usually lasts longer, so I don't know how it would work here??
 
I think for builders of super small form factors are gonna have a field day with this. Cost/performance ratio's have lately fallen on deaf ears, else everyone would be running r9 290's for ~220-230$. Sound, heat, and size seem to play in as much as performance nowadays. It seems largely irrelevant that its more than the 970, as its more powerful; but in a super small form factor, and as long as they arent lying... also quiet and relatively power efficient. If someone wants the fastest card in a tiny itx box, the nano will certainly be it.

This really seems for the people with the 4k tv's that want to game from time to time on their little svelte box in the living room. I don't think they are expecting this to be the 'next big thing' so to speak.
 
The problem, for those who care about competition and the resulting lower prices, is that the Nano along with the entire Fiji lineup will not do anything for AMD's sinking market share. All are high cost, low volume, overpriced, niche products. Everything else they came out with (380, 390X) are the usual enhanced re-badges. IMO the best AMD GPUs to buy are still leftover 290s and 290Xs if your system can handle the amps. When I built a mini ITX a while back, I used the Fractal Design Node 304 which had no problem holding the huge tri-fan HD 7970 I was using at the time. It's still small from the front, and the long dimension went back into the TV stand and was not really visible.
 
+1

I'm actually looking at the same case, Dave. You have a build log thread for yours I could reference? Which 7970 did you have? Thanks!
 
No build log, but I posted some photos somewhere. I'm pretty sure the 7970 was a Gigabyte. I used my trusty Cooler Master Hyper 212 to cool the OC'ed i7 4770K. The wiring was a mess to look at but everything stayed nice and cool. If I find some photos, I'll post them here.

I found a list of the parts:

i7 4770K (overclocked to 4.4 GHz)
Cooler Master Hyper 212
MSI Z87I mini ITX motherboard with WiFi
8GB Kingston HyperX DDR3-1600 (2x4GB)
Gigabyte HD 7970
2 7200 RPM drives
1 Kingston HyperX 120GB SATA III SSD Boot Drive
Xclio 550W Modular PSU

Found a photo of the Node 304 with the Gigabyte HD 7970 installed.

1zgrgnd.jpg
Fractal Design Node 304 Mini ITX Case
 
Last edited:
Would love to know how they're claiming 175W when it's supposed to pull Fury levels of performance.
 
Thanks, Dave! :D

Would love to know how they're claiming 175W when it's supposed to pull Fury levels of performance.

Easy.

anandtech said:
R9 Nano is expected to power throttle under normal circumstances. AMD tells us that the typical gaming clock will be around the 900MHz range, with the precise value depending on the power requirements of the workload being run. As to why AMD is shipping the card at 1000MHz even when they don’t expect it to be able to sustain the clockspeed under most games, AMD tells us that the higher boost clock essentially ensures that the R9 Nano is only ever power limited, and isn’t unnecessarily held back in light workloads where it could support higher clockspeeds.

It will hit 1000 when your game first loads, then throttle down to some lesser average when actual load is applied and draw ramps up past 175w. You know, right when you'd want it to be running at its max ;)
 

AMD is going to have a broadcast on twitch today showing what some modders did with the Nano... check it out!
 
For those that can't access twitch at work, can you put together a slide deck?
 
Well.. OCF, like some other major players, were shut out of reviewing the Nano...so we are 0-3 on fiji... :(

If you would like to see an article on why some feel they were shut out, read this from [H].. warning, there is strong language here....
http://m.hardocp.com/article/2015/09/09/amd_roy_taylor_nano_press

I'll admit, I am not happy about it either, but not that unhappy!

I see a lot of feeling there and not much reasoning, but when when somebody's Nano build thread gets deleted just because the admin is unhappy about not getting the same free hardware, I think "feeling" isn't the right word. That's bitching and throwing a tantrum.
I would go so far as to point out that AMD sampled a Nano card to HardForum member (Elmy) so that he could post the build in our own forums! (Which I have since removed. Given that our community does not have a Nano focus.) I was fine with our "Nano-less" review state before I read what Roy Taylor had to say this week. I asked and included Roy in on the email chain with Antal.
 
I feel that it is a petty action to take (deleting the build thread), but I understand their reasoning. AMD apparently left them out because the Nano did not meet their target audience, yet they give one out to one of the members there (part of their audience) to make a build thread for the card. While free hardware is always nice to receive, the amount of hours and work involved to make up the cost of the hardware is so skewed that it generally isn't worth it except for out of the love of doing the reviews.

We've been seeing this more and more where companies are terrified of getting bad reviews for their mediocre products so they only ship to those that they can almost be certain won't give a strong negative opinion. I look at this similar to when game publishers are putting NDAs for games at the same time (or even after (!?)) launch of their game because it is so crappy they don't want to lose the pre-order money and day-one purchases.
 
I have a feeling that limited sampling is also related to limited production. Fury/Fury-X is total fail if you look at amount sold since premiere. Many even bigger stores sold 0-5 cards.
For some reason Nano was announced as cheaper card but we see it for $650 listed in Amazon. In my local stores after adding tax etc, price will be not much lower than full Fury or cheaper versions of GTX980Ti ... really ? who will decide on a card which is throttling under load while can get something faster what cost not much more or something not much slower like GTX970 ITX but much cheaper.
 
Back