AMD Radeon R9 Nano Review by TPU...Not

[ZoneDymo]

Do we really need a million websites to say the same thing and show the same tests with the same outcome about one card?

Guru3D has one, Im sure you all agree they are far from biased so the truth about whether this card is good or not will come out with their review.
Its pretty much just facts, numbers, very little opinion involved.
Perhaps some different testing rigs and settings but other then that it should all provide the same numbers.
A few more then 1 review is wanted to be sure but I dont really see it as such a crying shame that TPU is not getting this card.

 

[RCoon]

I dont really see it as such a crying shame that TPU is not getting this card.

No card means no review, no review means less traffic, less traffic means less salary for the paid staff whose lives depend on said salary. This isn't just a tech website for fun, it actually gives a couple of people jobs, and the servers cost to run. Sure there are news articles and other reviews, but I imagine ANY GPU review of any card brings in a fair amount of traffic.

Disclaimer: I'm not paid TPU staff

 

[jigar2speed]

No card means no review, no review means less traffic, less traffic means less salary for the paid staff whose lives depend on said salary. This isn't just a tech website for fun, it actually gives a couple of people jobs, and the servers cost to run. Sure there are news articles and other reviews, but I imagine ANY GPU review of any card brings in a fair amount of traffic.

Disclaimer: I'm not paid TPU staff

You have generated more traffic than an actually review would have with this news - NO

 

[RCoon]

You have generated more traffic than an actually review would have with this news - NO

That's a good point. 19 pages.... jesus christ.

 

[john_]

That really isn't the point.
The vast majority* of people actually spending $650 on a card will do their homework before purchase.
The issue is the way AMD is massaging the launch and message. You could argue that AMD has a long history of doing this ( remember the Trinity launch for example), but it doesn't make it any more palatable.

Why isn't it the point? You say that people who spend $650 will do their research, so why is it important to have reviews at day one?

How AMD promotes it's hardware is their problem, as Nvidia's promotional lies about their hardware specs, is also a problem that they should deal with it and not keep doing it in the future. But they will keep doing it in the future.

You want AMD to be flawless, but how about a 6 months lie about 970's specs? How about a fake support about async shaders? How about Fermi still not being ready for DX12? You seem to worry about AMD's image because it will take 15-30-60 days to see a review of a card that CAN NOT be manufactured in large quantities, but at the same time what is your opinion about a company that gives (deliberately?) incorrect specs about it's products. They probably thought that no one will test the VRAM speed on GTX 970. They probably thought that there wouldn't be a program using async shaders, before they manage to create something that looks like async shaders in their drivers. They probably thought that they could fake DX12 support on Fermi cards, before Windows 10 comes out. They probably thought that no one will notice. Still you worry about AMD? Really?

* The launch of the card isn't just about the card, it is about the message AMD want to get out - not just about their own product, but in comparison with the competition, and a massaged company profile. Limiting reviews to companies prepared to spout the company line brings into question the objectivity of those reviews.

The reviewers who will make the reviews in two days, will have to worry for the reviews that will come out 2-3 weeks latter from the sites that didn't got that card. If those reviews contradict their findings, they will have to give plenty of explanations. So, don't expect to see reviews in two days showing Nano crushing Titan X.

If TPU had a card and every other side didn't, what would have been your conclusion. That TPU sold out to AMD? And would that conclusion being valid today before seeing the review and before being able to compare it with other reviews, or would you wait for a month to see the other reviews first? But it is so easy for you to bash AMD.

Most people don't do anything more than gloss over the facts, look at the pretty pictures, and come away with an impression shaped by the summary. If this wasn't the case, why are so many AMD fanboys up in arms about having Project CARS being part of a benchmark suite? and why AMD themselves are vociferous in their shoutcasting and guerrilla marketing against GameWorks ? You seem unhappy at Nvidia's Gameworks program for its lack of direct access, but seem OK with AMD restricting access to its own product. No one is holding a gun to the game developers heads, or the prospective buyers of the game for that matter.

It seems you know pretty much about "most people". The way you describe them, I would say double digit IQ, best case scenario. Probably that's why you are on a mission to save them from the claws of bad AMD and offer them to the angelic company named Nvidia(as a sacrifice). Seriously now, I never had problem with Project Cars being included in the database. As it was already being said, it is a game that people play, so it is logical to be there.
Gameworks. You are rebelling because AMD wouldn't give a card at TPU and you will find the slimy truth about Nano 15-30 days latter. But then you cannot understand why a closed library that no one can have access in it, coming from a GPU manufacturer is considered bad? It's like watching a basketball game where the heads of referees are covered by hoods, so you can't tell if those referees are in fact the owners of one of the two teams playing.

It's extremely funny comparing the situation with Nano and Gameworks. NO ACCESS is totally different with what you like to call as limited access. Also Nano doesn't affect in any supernatural way the performance of your 780 LSI. Don't worry and when you will run at 10th September your favorite benchmarks, scores wouldn't go down because Nano was released. In fact you are the first person that should question the GameWorks performance on Kepler cards compared with the GameWorks performance on Maxwell cards.

And as I have told you in the past, when it comes for you to defend your lovely company you are full of smoke. And do you really argue with what I wrote in that post(the one you linked)? I think not.

Personally, I couldn't give a shit about Nano - I prefer my graphics not to have training wheels attached (power and OC locked down), but it is galling to see some of the sites that I enjoy visiting being denied access simply because they won't kowtow to a companies propaganda. At least when W1zz tore Nvidia a new one on the GTX 590 review, the company didn't pitch a hissy fit in retaliation.

You are one of the biggest Nvidia fanboys and defenders in here. Why give a shit about Nano? As for propaganda, AMD is just an amateur compared to your lovely company.

JMO

 

[Tsukiyomi91]

Take a chill pill... please...

 

[john_]

Take a chill pill... please...

I took five before posting. Nice lemon flavor, but probably had the opposite effect.

 

[arbiter]

How AMD promotes it's hardware is their problem, as Nvidia's promotional lies about their hardware specs, is also a problem that they should deal with it and not keep doing it in the future. But they will keep doing it in the future.

So you gonna compare 1 of the few things in last 4 years to the repeated Lie after Lie AMD has spouted month after month about how good their hardware is in benchmarks that designed to only focus on strongest part of their hardware. Also on top of that trying control reviews of their products as quoted below to limited what sites show only those strength's

The issue is the way AMD is massaging the launch and message. You could argue that AMD has a long history of doing this ( remember the Trinity launch for example), but it doesn't make it any more palatable.

Its sad to see you focus on 1 thing nvidia did wrong and say that equal to least dozen lies and mis truths AMD spouted in last few years.

You want AMD to be flawless, but how about a 6 months lie about 970's specs? How about a fake support about async shaders? How about Fermi still not being ready for DX12?

How could nvidia support an AMD tech they likely didn't even have access to or wasn't likely even added to DX12 til after thei maxwell gpu design was complete. Truthfully kinda sick of pointing out these facts to an AMD fanboy that act's like AMD hasn't done a single thing wrong or hasn't told a single Lie in the last 4 years.

With this post i am done with talking to someone that is a complete and massive fool. (clicks ignore button)

As a side note to TR and that linked story on trinity. I Applaud TR for not accepting those terms. They want keep their integrity as reviews in tact cause when you get cault being a shill like that for a company you lose your credibility as a review site.

 

[HumanSmoke]

Why isn't it the point? You say that people who spend $650 will do their research, so why is it important to have reviews at day one?

Because I want to read a review conducted by W1zzard. I'd prefer to read a TPU review rather than a review that probably is less exhaustive in its breadth than his.

You want AMD to be flawless

I think you have a basic disconnect between what is written and what you think is written.
I, and most people don't expect AMD to be flawless - why would we? They've never demonstrated that capacity in the past.
What I do expect is some kind of consistency. There is no real reason why a well regarded authoritative tech site should be denied access to the card at the expense of some superficial glib social network channel - at least not in my book.

How AMD promotes it's hardware is their problem

Yes it is - and I'm commenting upon it. Are you now placing restrictions on that too?

as Nvidia's promotional lies about their hardware specs, is also a problem that they should deal with it and not keep doing it in the future. But they will keep doing it in the future.

Immaterial. I've already commented on the issue in the relevant forum threads.

You seem to worry about AMD's image because it will take 15-30-60 days to see a review of a card that CAN NOT be manufactured in large quantities

Well done! You've missed the context once again. I'm not worrying about AMD's image - what I am interested in is seeing a review by my preferred reviewer.
The fact that you can't even parse a simple concept such as this without imprinting your own false assumptions all over it doesn't fill me with confidence that you'll be able to parse this post either, but hey, I'll give it one more shot before putting you on ignore.

but at the same time what is your opinion about a company that gives (deliberately?) incorrect specs about it's products.

1. It isn't really relevant to the discussion at hand
2. How indolent do you have to be to ask the question rather than run a quick search? Since you seem incapable...

To a degree, it doesn't matter whether the user runs into problems or not [regarding the GTX 970 memory allocation issue]. The product still needs to tally with the advertising and specs. Most people might never run into the issue, but it still doesn't negate the problem for those who do.

If TPU had a card and every other side didn't, what would have been your conclusion.

1. Lucky TPU - an exclusive!
2. One card for review? That's some real crappy yield and launch right there.

Still you worry about AMD? Really?

Sure.
When I see a company attempting to commit ritual suicide and it affects my future hardware choices. I sure do. One glance around the net has already shown how well received this cherry-picked reviewer strategy is, and it hasn't even launched yet. If you think this strategy is a positive move for AMD - great. I personally don't think it is. AMD were always going to sell whatever limited amount of cards regardless - the Fury X is perpetually OoS, so presumably they sell what goes into the channel. If they sell regardless, why accompany the launch with sourness from denying mainstream enthusiast sites at the expense of less authoritative reviewers?

I never had problem with Project Cars being included in the database. As it was already being said, it is a game that people play, so it is logical to be there.

Again....I didn't say you did. Try not to make it personal when that obviously what wasn't inferred - it just makes you sound like a whiner. What I was referring to was a wider sentiment, as should have been apparent from the way I worded the post. Once again, you're interpreting rather than reading.

As for propaganda, AMD is just an amateur compared to your lovely company.

Propaganda is propaganda. Don't excuse one company because they aren't particularly adept at implementing it....oh, and by the way, I've probably owned more AMD/ATI hardware than you have. I think I can come up with a fair amount of verified proof if you'd like to take a wager on it. Loser donates money to the charity of the winners choice via PayPal OK for you?

 

[john_]

Damn, all Nvidia fanboys firing at the same time.

arbiter
So a quick reply. It's not one tiny lie. Have a little dignity when quoting someone and try not to change his post in your answer. They can't control reviews.

I though Nvidia fanboys where saying that Nvidia was first at DX12 with Microsoft and AMD just rush to make it like it was their idea a low level API. Now it is a tech that was added after Maxwell was finished? And why does Nvidia says it supports that technology. You don't mind contradicting yourself in what you and others where writing in the past. You just have to write something I guess.

HumanSmoke

I also want a review of GTX 980Ti from my neighbor. Until Nvidia gives a card to my neighbor that is the only person I trust, I will not trust them. Not to mention that no one forbids W1zzard to test a Nano. TPU was NOT denied access. For some reason it was not given a free card. But that can't stop TPU from writing a review. The review is just going to be done latter and TPU will own NOTHING to AMD because they didn't got a free card.

So this argument and the biggest part of your post, is in fact, smoke

The fact that you can't even parse a simple concept

with a few cocnlusions about me,

Try not to make it personal

but I shouldn't take it personally.
Nice work .

 

[Tatty_Two]

The topic of this news piece is about TPU and some other sites not being selected by AMD to run a review on Nano, therefore it's not about the GTX970 and .5GB of memory, It's not about AMD and it's poor (relatively) DX11 performance, in fact it is not about NVidia at all so just to make things perfectly clear, any more Tit for Tat(ty) discussions that continue to de-rail this thread will be deleted, up until the point that I get tired of deleting them and then it may be just members that get deleted thereafter..... thank you.

 

[the54thvoid]

Can anybody get a list of known Nano receivers? That way assumptions could be made on testing parameters. I'm still sure that Guru tends to be dismissive of noise created (for Nvidia) in that some 980ti cards W1zzard tested that were 'a bit loud', Guru said were quiet.
But, it could all be coincidence, although bloggers getting these 'rare' cards makes that seem unlikely.

 

[john_]

PCPerspective got one also. I bet Tom's Hardware will also have one. I don't know it, but they did got a Fury X in June. PCWorld also is in this list and TweakTown too.

On the other hand, eTeknix published a few hours ago that it didn't got one, and I bet KitGuru will also have to buy AMD cards, not just in Nano's case, but in the near future also.

 

[rtwjunkie]

Just Googled
"" Nano Review ""
https://www.google.co.uk/search?q=Nano+review&ie=utf-8&oe=utf-8&gws_rd=cr&ei=RVzrVe7vN4q7atDVsfgG

Result 3 from top

Now that deserves a card for review 3rd most popular result and the rest are about that indian car the Tata nano

You'll be interested to know that TPU's "non-Nano Review" has dropped to 4th on that link, but is still the most returned result for a GPU named Nano.

That's alot of site traffic and name recognition for TPU!

 

[RCoon]

Can anybody get a list of known Nano receivers? That way assumptions could be made on testing parameters. I'm still sure that Guru tends to be dismissive of noise created (for Nvidia) in that some 980ti cards W1zzard tested that were 'a bit loud', Guru said were quiet.
But, it could all be coincidence, although bloggers getting these 'rare' cards makes that seem unlikely.

I heard through the grapevine AMD are sending out cards to Vloggers and Bloggers, mainly the hipster types this time around. That means people saying "ooh maybe AMD doesn't have enough cards" aren't really accurate in their assumption. AMD has the cards, they're just trying to advertise to a different market - the wrong market, but a different one nonetheless.

I wouldn't class PCWorld as a PC Enthusiast site. That place is more for the middletons.

 

[64K]

I heard through the grapevine AMD are sending out cards to Vloggers and Bloggers, mainly the hipster types this time around. That means people saying "ooh maybe AMD doesn't have enough cards" aren't really accurate in their assumption. AMD has the cards, they're just trying to advertise to a different market - the wrong market, but a different one nonetheless.

That's my impression as well. Full Fiji has been in short supply all along. That didn't stop AMD from getting a Fury X to this site. I think there are going to be some issues with the Nano's performance at the same price as the Fury X.

It seems that one of the selling points for the Nano is "I want one because it's so cute". That's part of the appeal of the Nano. A thorough test of the Nano, like it would get on this site, isn't part of AMD's game plan for this card.

 

[remixedcat]

Most of those mainstream sites suck specially for networking hardware reviews... they don't prove thier graphs like I do and SnB does, They test for like 10 minutes and then stop, they don't show screenshots that are worth a damn, they scatter the review over 10 pages, and they can't just TLDR them effectively. Anyone can shit out a graph made by tableau or crappy MS office 2003 and just shit out numbers and don't prove anything other then they can use graphing software. My 2 layer approach using PRTG and a traffic generator makes my results provable and repeatable. Companies actually love my results and I quite often get selected for betas very easily and even have hardware companies request interoperabilitiy tests with other network hardware vendors. Yah think with the effort TPU goes thu that's way more in depth, provable, and so tightly controlled MFRs specially like AMD would realize this. Soccer moms that even go out and buy routers or stuff actually even hate cnet, recode, pcworld, etc becuase thier sites are clunky, take too long blabbing about the company's history that makes said hardware and not really getting to biz on how the reliability is.... consumer reports has even sunk to fanboyism by ranking macbook pros and apple's phones ans such higher then PCs or windows tablets or android phones becuase of some dumbassery like "user experience" and 50 million buzzwords.

Oh and one thing if you have to purchase hardware for a review instead of having it given to you you will go thru more lengnths to justify that purchase, thus rate higher, that's one reason review sites depend on donor hardware so they don't have the purchase bias and they don't have any emotional attatchement to the hardware. It's not theirs so they will be more likely to beat it up for benches and more likely to call out BS. If you bought something with your hard earned money you are more likely to succumb to finding some way you didn't "fuck up" by getting it.

 

[dorsetknob]

If you bought something with your hard earned money you are more likely to succumb to finding some way you didn't "fuck up" by getting it.

Thats SO TRUE

Before CD and DVD Roms were Cheap enough to buy for everyday use i went out and bought a LS120 drive ( for moving big files )

Drive still works but times have passed on and the drive / format never caught on
because of the pricedrop / availability of CD's
Even today i can still find a use for it Excuse coming
its ide and 1.44 backwards compatable so can be used on not so old systems that do not have a 3 1/2" floppy drive for flashing from 1.44 media or loading retro programs from "Floppys" excuse ended and thats the justification i'm sticking to

 

[EarthDog]

Oh and one thing if you have to purchase hardware for a review instead of having it given to you you will go thru more lengnths to justify that purchase, thus rate higher, that's one reason review sites depend on donor hardware so they don't have the purchase bias and they don't have any emotional attatchement to the hardware. It's not theirs so they will be more likely to beat it up for benches and more likely to call out BS. If you bought something with your hard earned money you are more likely to succumb to finding some way you didn't "fuck up" by getting it.

I have to disagree with that. In fact, I feel quite the opposite. It would be ANNOYING to me to have buy hardware to review and the hardware would have to, for lack of better words, climb its way out of that hole. IF the website I write for would buy it, that is not money out of my personal pocket so I really wouldn't care either way honestly. But surely, because it was purchased, I would not hold it up on a pedestal like you are suggesting... no way.

 

[GhostRyder]

You know, my big question on the Nano is why they didn't use cut down Fiji chips for it to save not only on power but cost overall by using chips that won't meet the standard of the full Fiji XT. Its obvious they are binning these chips out for Nano, but that only makes it harder to produce and even shorter in quantity. Though to be fair I doubt the problem with the quantity is to blame on the Fiji chips them selves and more likely HBM itself slowing things down.

Seems like it would have been logical to use the cut down Fiji pro chip, bin those out a bit to keep power down, then put them out there for a reasonable price (Like maybe 550 ish). Heck, they could have just scrapped the Fury Pro and just did the Nano as the reference design of Fury (While reducing the clocks maybe even a bit more to accommodate the power reduction needed) while allowing non-reference soon after and kept the size but allow for more coolers. Then we could have had the extreme OC edition cards along with micro sized cards available at the same time.

That is all just IMHO about the Fury Nano...

 

[Tatty_Two]

That's my impression as well. Full Fiji has been in short supply all along. That didn't stop AMD from getting a Fury X to this site. I think there are going to be some issues with the Nano's performance at the same price as the Fury X.

It seems that one of the selling points for the Nano is "I want one because it's so cute". That's part of the appeal of the Nano. A thorough test of the Nano, like it would get on this site, isn't part of AMD's game plan for this card.

In some respects if it were found that the card cannot compete with the opposition (I don't know, I am only using this as an example) then arguably AMD need a different approach, they know that in recent times they cannot get the market share based on outright performance (so little point in spending a fortune to market that approach), they adopted a "Bang for Buck" approach which is understandable but has cost them a fortune, some would therefore argue that they need to try a new approach and perhaps this "niche" thing is the way to go, I cannot help but associate the word "niche" to small quantities though, good, bad or indifferent it's hard to see how this card may help turn their fortunes around, I hope it does but as I am no marketing expert (clearly!) then what do I know.

 

[Tsukiyomi91]

I took five before posting. Nice lemon flavor, but probably had the opposite effect.

Ok...

 

[arbiter]

In some respects if it were found that the card cannot compete with the opposition (I don't know, I am only using this as an example) then arguably AMD need a different approach,

If you use history as benchmark, aka Fury x claims vs 980ti. Apply how they claimed it was 20% faster then a 980ti but end reviews mostly had it as 10% slower, you could have a pretty good idea where a nano could fall vs card they want to compare it to. They compare it vs a gtx970 which they said the nano is 30% faster so using those metrics could make a pretty safe bet that nano will be almost same performance to maybe 10% faster in most games, that could very since lets be truthful that gtx970 isn't really geared for 4k video like they marketed their card for. If those numbers end up being right i would say that would be impossible to compete vs a card that costs half the price.

I heard through the grapevine AMD are sending out cards to Vloggers and Bloggers, mainly the hipster types this time around.

Lets not sugar coat it, they are sending to RTP members mostly (red team plus). That way they can hype it with their biased views.

 

[Tsukiyomi91]

The R9 Nano won't be the "best bang for your buck" card with that $650 price tag... comparing it with the GTX980Ti, clearly the 980Ti is... but comparing with the GTX970, it's a little not fair as they're comparing a sub-$350 card, yet alone an ITX version of the card.

 

[the54thvoid]

The R9 Nano won't be the "best bang for your buck" card with that $650 price tag... comparing it with the GTX980Ti, clearly the 980Ti is... but comparing with the GTX970, it's a little not fair as they're comparing a sub-$350 card, yet alone an ITX version of the card.

I actually hadn't even thought about that. The price versus performance level versus the 970 (albeit limited ITX) cards. I'd hope a card that costs several hundred dollars more is substantially better.
In fact, 30% faster by AMD metrics but 90% more expensive..... Hmm.. Now we can argue the 970 only has 3.5gb memory but most people (including Guru) have said its performance is still stellar for its price and power consumption.
So, the Nano, if priced at the presumed level with 970 beating performance is almost 'Titanesque' price/perf metric.
And I did not know the RTP stood for Red Team Plus. That's just embarrassing.