AMD Vega spotted - performance between the 1080 and 1080Ti

WOOOOAHHHHHHHHHHHHHH

>>59334737
If AMD aims the price to be similar as a the GTX 1070 (maybe between the 1070 and 1080) consider me sold... Considering both 1080 and 1070 just got a price cut.

I don't want to leave my Freesync monitor

>>59334737
Finished and bankrupt?

>>59334737
>The T-Rex is the most relevant because it's a 3D bench
How is it relevant when Vega is a completely new architecture that has 50% higher fp8 performance which that bench isn't optimized for?

File: 1317464430091.jpg (12KB, 472x472px) Image search: [Google]

12KB, 472x472px

>>59334737
>15% higher results than the 1080
>80% larger die size
>HBM2

There is no fucking way AMD is going to be able to compete with Nvidia with this.

With a die size that big it's going to be very expensive to make, definately with HBM2.

>>59334790
>80% larger die size

Is that a good thing or a bad thing?

>>59334802
Bad for production costs.

The larger the die, the less you can fit on a wafer, which means more expensive to produce

>>59334806
At what Price would you say its a good deal. Assuming if OP's benchmark is valid?

>>59334830
15-25% higher than the 1080.

But it might be even more expensive to produce than a 1080ti

Performance obviously doesn't scale linear with price.

>>59334839
And even then, I'm not even taking into account stuff like power draw and heat output.

Which will most likely be significantly higher than the 1080 with a die size that big

>>59334737
>>59334790
>>59334839
1000mhz and 1200mhz boost?
This is clearly the low end model. Perhaps a "Nano" model that's under 175w TDP yet still outperforms the 1080.

The high end model is at least 1436mhz.

Probably something for mITX builds that will sell like hotcakes thanks to that.

Those clockspeeds make no sense, that's not a 12.5tflop chip with 1200mhz.

>>59334790
we'll have some room at reduced margins but you're most likelt correct. 50+/- aimed at the 1080ti price point is what I'd bet. Pleasant surprise if slightly bellow 600 and i'll the praise the lard if it's hovering around the 500's.

>>59334978
that boost clock will make many a fanboy unpleased. I myself was putting some faith into Adored's comments on speculative clocks.

>>59334737
wait raja told us that the base clock will be 1500mhz and it shows 1200mhz..

oh this is going to be good lol

>OpenCL benchmark

>>59335076
As I already said: because what is seen there is likely a R9 Nano successor.

They are likely launching at least 2 different RX Vega cards.

>>59335013
i'll clean Raja's loo if it's 499,99 - delivering on his wildly speculative, informal off-camera, don't take it home, en passant, whisper of a high end desktop at 1000 bucks. Ryzen 1800x + Vega flagship could technically fulfill that in the hearts of fanboys.

Vega Mi25 instinct is clocked around 1520mhz, this is 1200mhz.
So either someone was testing IPC or this is an ES without final clocks

File: screenshot-boards.4chan.org-2017-03-10-12-01-00.png (45KB, 1285x212px) Image search: [Google]

45KB, 1285x212px

kek

File: images.jpg (8KB, 260x194px) Image search: [Google]

8KB, 260x194px

>>59335118
just to let a hint regarding my promise of heavy duty cleaning.

>>59335116
this is a 4096 monster it cant be a small one

>>59335146
I have a nano, and it has the full 4096 core count of thr fury x. It's just down clocked for consumption

File: screenshot-2017-03-10-12-04-18.png (68KB, 549x1177px) Image search: [Google]

68KB, 549x1177px

>>59335115
https://compubench.com/compare.jsp?benchmark=compu15d&did1=26470338&os1=Windows&api1=cl&hwtype1=dGPU&hwname1=AMD+Radeon+R9+Fury+X&D2=NVIDIA+GeForce+GTX+980+Ti

your point?

>>59335134
kek at your sad life of SS'ing anon comments. I could understand the satisfaction of making someone eat their words if there were handles, but to save a copy from a random anon, sometime, somewhere, somehow said something silly is pathetic.

if vega comes out around or under $400 its a victory if its in the 600 range its DOA

and for all you morons out there spending $700 on a 1080ti enjoy playing AAA games at 4k for about a year then you will have to lower settings to low just to stay above 60 fps

4k wait for volta
1080/1440p 144hz vega will be good for years to come

File: gamerfood.jpg (384KB, 1000x589px) Image search: [Google]

384KB, 1000x589px

> 1700X - $349
> Vega Nano - $399
Total: $748 for top notch, 8-core/16-thread hardware

This is the best timeline ever, meanwhile at Intel/Nvidia:
> i7 7700K - $349
> 1080ti - $599
Total: $948 for a 4-core/8-thread rig;

Thanks AMD.

>>59335134
>>59333189

Yeah, this Amdrone faggot was so funny.

>>59335194
you are overrating how fast gfx cards losses its performance, i can still play most games at medium at 1080/60fps with my 670gtx after 5years. Yes this card could max any game at 1080/fps at release.
Yes 1080ti wont be playing 4k/60 for 5years. But it will support highest settings for atleast 2-3years at 4k. Why? Because consoles.

>>59335197
*it's 1048 anon. add another 720 bucks if you're a real good goy buying the greatest from ROG.

>>59335194
I'll lick Raja's loo clean if it's 399,99 for the top end model.
For realsies this time.

>>59335349
500-600 is a more realistic price...

>>59334978
The low end model was rumored to compete around 1070 level. If you look around for the device ID mentioned in the article you'll find that it's the big one too.

>>59335197
Or a 1400X + Vega Nano mITX PC.
Under 250watt total power consumption at load, under 30watts at idle.

I'm going to have to build 2 PCs with all this $AMD stock and all the good options AMD is giving.

>>59335349
Maybe $450 for the Nano and $550-$650 for the top end card if lucky.

The top end card is likely going to roughly match the 1080Ti in un-optimized games and shit-stomp it in optimized ones like Prey.

>>59335364
Shit rumors from currytech. Disregard.

>>59335194
>if vega comes out around or under $400 its a victory if its in the 600 range its DOA

simply because you are moron the world doesn't need to accomodate your retard obtuse ways. If it launches at 800 dollars while destroying the 1080Ti it'll be a success by any metric, including perf/dollar.

I hope it comes out at better pricing scheme, but i don't think it would be very smart from AMD to do charity for you, personally (you).

>>59335364
People assume it competes at a certain level due to the TFlops or memory bandwidth. You have to factor in the wildly different architecture.

>>59335383
> 1400X
no, get the 1600X

>>59335166
Nothing relevant uses OpenCL and that benchmark is obviously worthless, if it gives a Fury X and a 1080 at the same score

File: amd2.jpg (52KB, 568x612px) Image search: [Google]

52KB, 568x612px

>>59334790
AMD pajeets on suicide watch

>>59335453
The Fury X is a monster at compute dumb shill. most OpenCL benchmarks aren't relevant for muh gaems, but AMD's intention with Vega is to rape Nvidia and Intel on coprocessors/GPGPU as demonstrated by the Radeon Instinct event

it's probably safe to assume AMD filling the price points at 100 increments
rx 580 - 250
rx 590 (?) - 350
Rx Vega smallest - 450
Rx Vega mid - 550
Rx Vega biggest - 650

>>59335453
Are you stupid? If Vega is behind the 1080 ti, the card is DOA

>>59335442
With a whole CCX disabled, the 1400X will be lower power for miniITX.

>>59335490
RX580 will probably be $200, though.
And RX590 won't exist.
You failed.

>>59335502
Are you stupid? The Benchmarks gives the Fury X and 1080 the same score.

>>59335453
too bad amd doesn't provide software support like cuda
they shit their turd and leave you to deal with it yourself

>>59335490
>rx 590 (?) - 350

RX 5xx are confirmed rebrands

>>59335490
or ditch the 590 replacing it with the smallest vega and cut the line at 550.

Last AMD gpu i had sounded like a vacuum cleaner, heated better then my oven.

Have they improved?

>>59335524
Jesus...look the T-Rex

>>59335401
>Shit rumors from currytech. Disregard.
Of course, I'll disregard the AMD demo with a 8+6 pin card too, because a fanboy on /g/ told me so.

>>59335415
Yeah, you do of course. Not to mention the performance we saw from AMD's own demos put it closer to a 1080 than a 1080 Ti/Titan XP.

>>59334790
isn't the larger die size due to hbm2???

>>59335556
sounds like you had a gtx 480 than a amd

>>59335572
No, nobody is talking about the size of the interposer or of the HBM2, it's about the actual GPU.

>its a /g/ speculates and gets it totally wrong again episode

Yawn

>>59335556
avoid reference designs, nvidia isn't much better with stock coolers either

>>59335534
But they do

>>59335582
7870

>>59335502
Vega is 45% more raw FLOPS than Fury X and that's before you get into all the architectural improvements.
A benchmark that shows it to only be 18% faster than Fury X is obviously wrong.

>>59335633
your kidding right 7xxx series is known for the low temps
in fact some HIS cards had a max temp of 62c at full load..

>>59335675
45-50 in idle with idle fan being around 40%
70+ easily

It died after about a year

Lent a 970 from a buddy of mine after that, felt fucking amazing having a card not using its fans until it hit 60c

In idle it went down to 30c

Now i got a 1080 which cools pretty similarly.

Nshitia btfo

File: aaaddd.png (141KB, 1070x601px) Image search: [Google]

141KB, 1070x601px

>>59335197
DEELET THIS

>>59336033
>Vega Nano - $399

good luck

>>59335146
Small one as in weakest cooling which would necessitate a lower clock on the core and possibly memory, it also might have a less clean power delivery which also hurts its clocks

>>59335197
>Vega Nano - $399
See, we know the 1080 is priced stupidly above of what it's worth, but there is no way a fat chip like Vega will be $400. Not only that, it also has expensive HBM2 so expect prices starting at $600. I'd love to get one, but please be realistic of prices.

>>59336450
>realistic prices
margins are inflated as fuck. there's nothing realistic about current prices for hardware. No one can find exact figures for the total production costs for cpus and gpus, since they're all guarded under trade secrets and the best we can do is extrapolate from the old production costs for the virgin wafers and wildly add this and that to come up at a guesstimation.

>b-but that tech guy wrote that it costs 398,972572 burgers and 3 dead chinks to produce a titan xp.

Guess what, he took that figure straight out of his smelly ass, apart from the chinks; he totally went to visit the mass graves.

The price finding mechanism for hardware manufacturers has no concern for supply, competition, logistics, they go straight to end user market perception of value focusing on how much they can rape you before you start to bitch - hint, your a big boi, you can take it real good and a lot of it.

>>59335534
They do now.
http://gpuopen.com/professional-compute/

>>59336642
Usually costs are high in R&D but Pascal cost almost nothing in R&D. Especially not the 1080Ti.

The cost of the cards themselves isn't very much. In the $30-$150 range, I'm pretty sure. There's no way even the most high end consumer card costs over $150.

And Titans are one of the cheapest cards to make since they're made from garbage chips that would have been trashed and high margins are already made on the server cards they are garbaged from. The only reason they cost $1000 are to keep them from being bought instead of Teslas and shit.

Looks like the same disaster as the Fury X back then.

>>59337210
B-b-b-b-b-but wait the new drivers

Yeah ok but how stronk is the 580

>>59334790
>>80% larger die size
why would this be a problem?

>>59337520

>>59335536

>>59337593
I know it's a rebrand, that doesn't tell me its actual specs

>>59337619
It's literally a 480.

>>59334737
It was between 1080 and 1080 Ti on a 1.2 GHz clock, but enterprise release clocks have already been promised at ~1.55 GHz due to the 25 TFLOPS fp16.

This thing will completely rape the 1080 Ti if the leak is accurate and consumer clocks are roughly in line with the enterprise parts.

File: NCU.png (224KB, 1657x1456px) Image search: [Google]

224KB, 1657x1456px

>>59337925
Needs 1436mhz to hit 25 TFLOPS half precision, no...?

>>59334787
>How is it relevant when Vega is a completely new architecture that has 50% higher fp8 performance which that bench isn't optimized for?

fp8 isn't a thing. you've thinking of fp16. there's also no real world application that uses half precision floats outside of a few niche applications for ai/machine learning stuff.

>>59336642
>>59337144
exactly.
Most people neglect logistics costs, for instance. The final impact on pricing is a big factor when you're shipping worldwide out of a single factory in Taiwan. But that becomes moot with these guys. We could very well have 500 dollars top tier cards (consumers, server grade is another story entirely) while "they" could still maintain very decent profits per unit sold. But why the fuck would they even bother doing the math, the focus groups or hiring consultants on price finding, when their target audience actually have a passtime of calling other folks too poor to afford their flagships?

the titan price finding probably went like this:

>...so there's that. How much can we charge for it?
>800?
>nah, I feel we can go higher.
>1200?
>hm... maybe, let's try it a notch down.
>a notch? 1000.
>k, let's do it.

straight out of Wharton guise.

>>59338283
Colors in every fucking game are fp8 or fp10 you stupid fucking piece of shit go kill yourself

>>59338255
No, it needs 1525mhz

>>59338303

that doesn't make them 8 bit floating point numbers.

>>59338333
That's up for the driver and shit to decide. The texture format is 8bpc. Normally the driver may convert back and forth 32bit floats but it sure as shit doesn't have to.

>>59338287
Nvidia fancies itself as luxury brand, which takes into account that its bottom line must be at least at such a height the perception of value amongst its wealthy consumers doesn't start to decline. They would lose customers if they started to put Titans on the bargain bin.

ps That's messed up, because they aren't selling 50k watches nor 20k purses or shoes. This goes against mass production of goods, it should, at least in principle.

>>59338493
of course, no need to even touch the inherent value of a 50k watch, or designer haute couture. The very nature of such items antagonize mass production mechanics.

>>59338566
Funny enough the Fury X was sort of an attempt at "luxury". It shipped with an AIO.
But it wasn't good enough quality or quite enough performance. But it's the sort of thing you'd expect with the absurd price and lack of performance in Titan cards.

File: 1488924169165.jpg (45KB, 680x505px) Image search: [Google]

45KB, 680x505px

>>59334737
Hope /g/ doesn't fall for the, year of the AMD meme again

>>59338627
That's my take on it too.
Monkey see, monkey do.
Although my criticism was glaringly aimed at Nvidia I never said that AMD, or Intel, as a matter of fact, don't follow the same rules. Well, AMD attempted to sevral times in the recent history and never managed to catch that halo (the last possible item to attain was maybe the 6990, but that's debatable for sure).

If that's the sate of the markets, the competitors tend to protect and reinforce the staus quo that is so sweetly beneficial for their final margins. Not to mention that Nvidia was protected (because of the nature of the market) from AMD even trying to perform a aggressive dump strategy, my opinion is that they would've shot themselves int he foot if they tried it.

>>59334790
The thing is going to run hot as fuck, isn't it?

>>59338915
No, the top end server vega is supposed to be 25TFLOPS fp16 and passively cooler. So 220w TDP or less, I'm pretty certain.

I've noticed AMD meme'ing crossfire a bit more lately. I wonder if this will be a good thing or a rerun from the Rx480 launch.

File: 1426868284195.jpg (166KB, 585x454px) Image search: [Google]

166KB, 585x454px

Absolutely fucking clueless retard here. Looking at the reviews of it people are saying that it tends to get quite hot while working. How big of a deal is? Would it cause problems and should I wait for the better cooling versions to come out instead?

>>59334737
>amd
>gpu
... Since when has amd branded their gpu's amd instead of ATI?

>>59334737
The RX 480 is 45%~ behind the GTX 1080, thats with 36CU clocked at a peak 1266mhz.
Vega 10 has higher IPC, 77% more CU, and is clocked over 20% higher. It'll slap the shit out of the GTX 1080ti.

The card in the OP has its max frequency listed as only 1200mhz, its a test sample not running at its full clock. To make 12.5TFLOPS Vega 10 is clocked at 1526mhz.

>>59339197
2006 i believe

>>59339166
Modern graphics cards have a target temperature you can set, it will basically run as fast as it can while obeying your temperature, power consumption, voltage and clock settings. By default it seems to run at 84C. You can easily increase fan speed to bring the temperature down if that's what you want to do.

But yes, waiting for custom cooled cards is often the best thing to do in general. They come with the downside of dumping heat inside your case instead of straight out the back, but that generally only matters once you get to using 2 or more graphics cards, doesn't matter much with just one unless you want to have an especially small and poorly ventilated case, in which case it may very well help.

>>59334737
Is this Vega 11 or 10??!?

>>59338942
Hardware made for a sever rack is almost always going to be "passively" cooled, genius. The rack has its own row of jet turbine fans to circulate fresh air throughout the whole thing in under 20 seconds, and is usually in a climate controlled room to boot.

AMD listed the TDP for their big Vega part has 300.

>>59335453
>Nothing relevant uses OpenCL
What is photoshop?

>>59339367
Nothing relevant.

>>59334737
another pointless hype..... as it was with zen processors

>>59339327
Alright, thanks senpai

>>59339197

Did you build that time machine yourself or are you literally living under a rock?

>>59339422
thanks for the insight goldstein

>>59339928
you welcome

>>59338255
That illustration is retarded.

Completely general warp/wavefront fragment combining, which the illustration suggests, creates a ton of cache capacity/locality concerns beyond the substantial scheduling complexity itself.

What is much more likely is that NCU has a limited capacity to schedule in parallel pieces of simple divergent branch cases, which 8+4+2+1+1 SIMD makes possible by supporting any single partitioning from 16+0, 15+1, ..., to 8+8. This has a much more modest implementation complexity increase while still reducing the time of executing branches A and B from A+B to max(A, B):

int foo(int b, int x, int y) {
  int temp = 0;
  // normal SIMT flow sequentially executes both paths
  // if different wavefront elements take different branches
  if (b) {
    // less work in this path
    tmp = x + y;
  else {
    // more work done here - path above can only partially overlap
    tmp = x*x + y*y;
  }
  return sqrt(tmp);
}

>>59338627
>Fury X
Fury X was the overclocker's dream.

But then came the GTX 980ti and AMD had to OC the Fury X to the limit.

>>59340222
Vega has some complicated hardware scheduling that could possibly pack and schedule opts over the FPU automagically.

After all, what would be the point of being able to split it up like 16+8+4+4 if your SIMD function is at best four 4bit ops per FPU instead of being able to do 2 on each FPU and leaving the 16bit op open?

Sounds far fetched, but we'll just have to see.

File: file-6090cd636e93a4a457.jpg (44KB, 800x450px) Image search: [Google]

44KB, 800x450px

I-Is this real??

Are we going to see dual GPU cards again?

>>59340222
>That illustration is retarded.
Its also not accurate since all CU in Vega are still 64 ALU. There is no variable size CU in the Vega architecture.
That is total guesswork from a patent filing, and never once was it even hinted that Vega would include those features.

>between
>not surpassing 1080Ti

Well, yet again, AMD gives us trash.

File: NO!.webm (3MB, 1920x1080px) Image search: [Google]

3MB, 1920x1080px

>>59340259
>Fury X
>overclocking

NO NO NO NO AND NO.

>>59339292
>Vega 10 has higher IPC, 77% more CU, and is clocked over 20% higher.
It also has a dedicated geometry pipeline, the same kind of tech that gave Maxwell such a huge jump in performance from Kepler.

>>59340513
>underwear suddenly disappear.
>disawear?

Vega looks great, but I feel it's going to udnerperform.
What I'm really curious about is Navi. Having one small chip and then scaling performance with multiple chips on a MCM sounds heavenly.

>>59340564
Over two times the geometry performance per clock is nothing to scoff at, especially considering Vega clocks higher than Polaris by a great deal.

>>59340483
>believing random /g/ poster that's making shit assumptions

>>59340564
>It also has a dedicated geometry pipeline, the same kind of tech that gave Maxwell such a huge jump in performance from Kepler.
Hey, someone actually gets it.

>>59340564
>dedicated geometry pipeline
wtf is the geometry coprocessor™ then

>>59340638
*geometic processor

https://en.wikipedia.org/wiki/Graphics_Core_Next#Geometric_processor

>>59340222
you might be correct anon, I have no idea really.
But an illustration is meant for retarded people, like myself. I can easily infer the metaphor from that - they increase the utilization, reduced the wasted unnocupancy. They probably did it using an even more complex sounding method than you yourself brought up since they're engineers working on that problem for years on end and you are just posting on a bora-bora audiophile board for people obsessed with braiding cat-6e cables for better performance.

>>59340618
>What I'm really curious about is Navi.
fanboy here, let that be perfectly clear.

but fuck off, that's too much already. No wait anymore. Vega or bust.

>>59335715
>my subjective experience of a shitty purchase means all other products from vendor/producer XYZ are also shit

You're a mature, stable and trustworthy person.

I have no idea why everyone acts like the Pro Duo doesn't exist.

>>59340997
because I only heard of it just now

File: 1478013217166.png (27KB, 247x140px) Image search: [Google]

27KB, 247x140px

>>59334737
>Can't mine bitcoins faster than a 1080 Ti

it's over

File: 1483950203215.gif (468KB, 500x282px) Image search: [Google]

468KB, 500x282px

>>59338303
>>59338476
You have no idea what you're talking about. 8bit floating point values aren't the same as 8bit fixed point values and sure as hell colours aren't encoded as fp8 values because theres no such thing as unsigned floating point according to IEEE specification meaning you wouldn't even be able to encode 8bit color textures in 8bit float. You're a perfect example of an uneducated gaymer who lacks understaning of underyling concepts yet is first to offend others. What an idiot.

>>59340450
yes, it's real. and remember that the Pro Duo (2x Fiji/Fury) already exists.

>>59340479
yeah, people are going off the deep end from that patent filing. however, GCN CUs are 4*16*fp32 SIMD clusters, and there's nothing know on way or another whether NCU is 4*16*fp32, 4*(8+4+2+1+1)*fp32, or whatever. it's just known that Big Vega is an aggregate of 4096 ALUs across 64 CUs and that it's "all new and enhanced".

>>59334737
Cool, lets hope that AMD don't fuck up the release like they did with Ryzen and the Rx480

>1000/1200

sniffing a 1080ti at those super low clocks

promising

AMD kidna confirmed the TFLOP number a while a go so these clocks shouldn't be final.

Should I buy RX480 now, or wait till vega?

>>59343382
> should I buy the Toyota Camry, or wait for the year model Lexus?

>fury tier 2.0

>>59343819
A duel fury card a year ago would have been sweet desu

>>59343851
dual*

>>59340513
get that nigger shit off my /g/

>>59344098
>my /g/
:3

>Nvidia card faster in bitcoin mining

I don't believe it, driver issues or something

>>59335582
The 480 is possibly the coolest chip they've released to date. Its time to hop off the meme bandwagon.

>wait for nvidia gpu
>wait for amd gpu
>wait for vega gpu

waiting is suffering
how much longer

>>59344313
yeah the 480 is pretty amazing. ill give ya that

kind of reminds the first time i got a nvidia riva TNT

sure there was faster but not much was truly better.

>>59339197
Its ok I was confused as well.

I built a little AMD rig in 2000

>>59339166
if its not loud and doesn't throttle performance down its not an issue
not a single partner AMD card throttles, well maybe except some asus

>>59340450
Yes. Unlike We Copy Content Forever Tech, VideoCardz doesn't post bullshit.

>>59340450
>Are we going to see dual GPU cards again?
they are niche for video editors that can't afford full rack

one vega under 250w though, that sounds promising

>>59335197
1700X is $399?

>>59340997
No one benchmarks with it.

Does the Radeon Pro Duo actually give 2x performance in games?

It does have the record on like Firemark and stuff years later.. but how well does it actually game?

>>59341969
Yes but fp16 is more than enough accuracy to do math on an 8bit fixed point value, or even 10bit, since the mantissa is at least 10bits. Yet the graphics driver usually use fp32.

>>59343043
How much did Nvidia pay reviewers to pretend the Radeon Pro Duo doesn't exist and to not include it in Benchmarks against the Titan X and 1080?

not bad for having dumped the other share of available resources into ryzen

>>59343851
You mean... the Radeon Pro Duo which is as fast as the 1080Ti on average but came out much earlier?

>>59346782
It's roughly equal to the 1080Ti.

In some games it's faster, in others it's between it and the 1080.

Why do people benchmark with the Titan X P but not the Radeon Pro Duo which was faster than it in some games?
Shit makes no sense. Sounds like a conspiracy.

>>59347299
Back in the days of the 780Ti all of the reviewers skipped the 295X2 while including the dual chip Ttian Z

>>59334737
>1200MHz

lol, that's like 340MHz downclocked from the Mi25 Vega card, this is a prime example of ES if I ever saw one.

>>59346975
0
Radeon Pro Duo was always targeting gamedev.

>>59334737
>between
come back when they are at the same level.

>>59348835
>1200mhz test
>Vega enterprise is 1536mhz
>consumer card will likely be even higher clocked
Are you dumb? Oh yeah, you're dumb.

>>59346975
>Radeon Pro Duo
Because dual GPU card are all terrible

>>59347299
Why do people benchmark with the Quadro p6000? This is not an argument...

>>59349383
>Why do people benchmark with the Quadro p6000? This is not an argument...
Radeon Pro Duo isn't purely a workstation card, it is intended at the prosumer market just like the Titan's are
The equivalent to the Quadro's were the FrirePro's and now the Pro WX series

>>59347299
because it's hard to find, expensive and youtube reviewers are slothfull hacks, not to mention that they all tend to be nvidia fanbois anyways why would they even try to do more heavy lifting in favor of a company they love to slander during their downtime?

AMD should've left a few review samples with the biggest channels or sites, but even, I think most of those reviewers would simply leave those cards collecting dust in their cluttered inventories.

Besides, AMD marketing team is still suffering from the PTSD from back when they were futilely defending the HD6990 as the highest performing card on every forum out there. Those guys sure earned their extended vacation.

In summary,
from the reviewers stand point: it's too much of a hassle, no one gives a fuck and I don't wanna;
from AMD stand point: not this shit again, it's there if anyone cares, and I sure as hell won't dive into that quagmire again.

>>59335453
>if it gives a Fury X and a 1080 at the same score

The Fury X is a fucking monster in GPGPU, its gaming performance is mostly bottlenecked by ROPs, which are completely irrelevant to GPGPU.

>>59350374
The lenght for the exegesis of the technicality of a dual gpu still being a single card would put every Ivy League doctorate theses published between 2005 and 2015 to shame.

and /pol/ thinks that they are weaponized autism, poor fellas.

>>59350374
Now the 6990 IS a card I would sometimes see in benchmarks 4+ years later, still stomping new cards.

If they price is the same as the Fury X it will be another disaster.

They need to massively undercut Nvidia for it to be a success.

>>59334737
I'd take it for the right price. Fuck paying out the ass for a g-sync monitor.

File: 1475787445156.jpg (489KB, 2000x1859px) Image search: [Google]

489KB, 2000x1859px

>tfw too intelligent to upgrade from 1080p
>tfw all I'll ever need to max out games is a $250 card
>tfw won't waste money on overpriced "high end" garbage

>>59350480
the 6990 has ressurected more threads than anything I've ever seen on tech forums, that they should've launched a Lich king version of it. Or any other necromancer theme.

>>59350557
It just really goes to show how shit AMD's marketing is.

People gobble up Titan's that are 15-25% more performance for twice the cost.
Then there's AMD dual GPU cards that cost less and perform better that no one bought.

I think I'm almost certainly going to get the Navi 2x or big die card or whatever it ends up being.

It looks like they're going to make a 2x Vega card for H2 2017 but I don't think it's going to scale as well as the Navi one that should be the first to incorporate "infinity fabric" with the GPU. Could be wrong, though.

>>59350537
You have to go back.

>>59350471
>2005-2015
>doctorate

recap, include undergrad monographies, every institituon in North, Central and South America, and readjust the period for 1985 to 2017 and the word count would start to approach parity.

>>59334790
>>80% larger die size

That's because of the HBM2.
Note that the card itself will be half the size of a 1080ti since it doesn't have to lay out so many gddr5 banks on the PCB. It's just a GPU + VRMs.

So it'll be a mini-itx sized superclocked 1080.

What I want to know is the price, and whether they'll have a mid-range version. Probably not, since they are planning to re-release Polaris...

>>59350767
Doesn't make sense for them to make any mid range since the 480 is still more than enough for 1080p.

They'd just be cannibalizing the 480 for no reason.

I don't think this was already posted itt

https://videocardz.com/67275/amd-vega-spotted-with-4096-cores-and-8gb-2048-bit-memory

only new info is 8gb @ 2048 bit bus (1024*2)

>>59350822
>Doesn't make sense for them to make any mid range since the 480 is still more than enough for 1080p.
This. The RX480 plays virtually every 1080p game at 60fps minimum, and fits the specs for VR.

There's not much point in an in between. The other card people are looking for is 60fps at 4K, or 100fps at 3440x1440.

>>59350850
Was already known the consumer version would be 8gb.
8GB of HBM2 is plenty. It, with the new cache controller, generally going to be like having 12-24GB of GDDR5X.

>>59350901
... for gaming, that is.
For rendering and such, more memory is useful.

File: SapphireRear.jpg (696KB, 2500x1064px) Image search: [Google]

696KB, 2500x1064px

>>59350767
I'd like to see Vega on a non-X Fury style setup, with a short PCB but still three fans, with one blowing right through.

>>59350767
>That's because of the HBM2.
No it's not, wtf are you talking about?

>>59350374
>they all tend to be nvidia fanbois

why is it be a surprise that pc enthusiasts would prefer the best technology...

evil bias youtubers always using the best.

File: 4.png (61KB, 237x263px) Image search: [Google]

61KB, 237x263px

>>59350919
>>59350767
Even without the HBM stacks its 473mm, that's fucking huge

>>59334790
>>80% larger die size

GP102 is also worse perf/area than GP104 by a fair margin, since you can't simply clock a ~500 mm^2 chip like a ~300 mm^2 chip and not get housefires.

>>59350822
>>59350901
The reason I want a low-end Vega is to have a 15cm card that costs $250 max, and can max out 1080p not just now but for a few more years.

>>59350944
i can create a fun argument
>the die is bigger so the heat is spread out while providing most surface area to transfer heat into a heatsink.

>>59350991
Just no

>>59350984
I think APUs make more sense there, though. An APU with the power equal to a RX480 and 2GB of HBM2.

>>59350767
>Note that the card itself will be half the size of a 1080ti
>>59350944
Don't you love it when delusional fanboys get their delerious fantasies proven wrong by facts

File: 1402065566623.jpg (43KB, 450x450px) Image search: [Google]

43KB, 450x450px

>>59334737
>>1080: 12.876 fps
>1080ti: 19.773 fps
>vega: 14.784 fps

That's not 'in between' the 1080 and 1080ti.

That's slightly above the 1080 and significantly below the 1080ti

>>59350642
Go back where? Don't tell me you took the atheism part seriously. First day here?

inb4 i was just pretending i trole u xDD

>>59335453
>worthless
>what is password cracking
>what is mining cryptocurrency
nice try though

>>59350923
>would prefer the best technology
first, they should at least try to appear unbiased but they can't even help themselves from talking about AMD without shit eating grins, smirks and snide remarks. They are /g/ tier of fanboys and they all will excuse that behaviour justifying to themselves that they are simply catering to the biggest audience, which they are, but they also forget that they are also performing a service reminiscent of journalism which no matter how informal the format might be there are still some principles of ethics that must still be observed, since they dare call themselves professionals.

>>59351141
I hate Ryan Shrout so fucking much.
The only good thing about him is that his face looks so deliciously punchable.

after the DOA on arrival ryzen now this

amd is done

Nice contribution, faggot.

>>59351025
>conflating PCB with chip size
lmfao. you're trying to insult people when you're the fucking moron.

>>59351100
at 1200mhz

>>59350850
what if those scores are for that clock and readings are correct?

please excuse my oO"

File: 89HANHg.gif (693KB, 339x279px) Image search: [Google]

693KB, 339x279px

>>59351337
>leet get
>oO"

>>59350923
AMD was the best for almost 6 years straight, you retarded fanboy.

From the 4000 series in Q3 2008 up until the 980 launched in Q3 2014. At some points in that time, they offered 65% more performance for the money, they were raping Nvidia that hard. (But Nvidia still sold more cards due to better marketing)

I have bought a MSI RX480, but I am having regrets now.

DX12 is largely overrated and still shit.
Overall the GTX1060 beats any RX480, both OCed.
On top of it, the GTX1060 still consumes far less energy overclocked.

I laughing my ass off at all the retards thinking that AMD is somehow a social welfare club.

Use your tiny brains and take into account of how much HBM2 alone costs.

>>59334751
>No CUDA
Into the trash

>>59350537
I've been to 4k, if you have the money it's nice. But it's not a must have tech since I went back to 1080p. In fps games it was awesome, especially dice games.

>>59348982
>Believing AMD after Ryzen
Man you faggots are fucking delusional

>>59354952
It's going to be different this time!

>>59354952
>>59354979

You also forgot the hbm1 hype kek

>>59335675
I am still running a 7870xt and a 7950 which are both tahiti chips and they run both at least 80°C+ at full load.
The lower end cards ran a lot cooler but there were also significant differences between the manufacturers.

I am currently deciding if i should grab a 480 or wait for vega. The 7950 runs at 70°C idle because it does not clock down when using a >60hz screen and multiple screens which is a problem many cards had for a long time.

>>59354952
Ryzen is great. Far better than anyone expected.

>>59351850
>no CUDA
it's called APP (and it's better).

>>59354952
>guaranteed replies.
what a sad pathetic life you live, son.

>>59355169
That is a particular problem almost all cards still have, to a lesser extent.

>>59335675
my asus 7850 was pretty hot and loud, i had 82-85c all the time at full load

>>59355307
at being piece of garbage vs i7 7700k

>>59356243
Is this bait, or are you just that stupid?

its pretty simple

1. if it cost in the 450 range for 1080 +20% performance it will do just fine

2. if it has that same performance and is over $600 its DOA end of story

3. once volta comes out nvidia will nerf the 1080ti drivers while vega will get only better and a year from now they will be about equal

>>59357234
AMD has more cards in the Vega family, and supposedly Vega GPUs will be their first 7nm parts.
Vega will still compete with Volta.

>>59351850
CUDA isn't better than opencl

It's not 2009 anymore

All that power to run Counter-Strike: Source

>>59351850
>CUDA
kill yourself, you moron

>>59357257
Vega 10 will be 14nm

Vega 20 next year will be 7nm

>>59334737
I have a feeling it's still going to use a lot of power and run hot.

>>59357641
i have a feeling you're a fucking cuckboy

>>59357641
Tiled render should improve perf/watt

>>59357714
but Polaris

>>59341353
>mining bitcoins on GPUs
>mining non-ASIC coins on NVIDIA

I prefer Nvidia but when it comes to mining, AMD wrecks NVIDIA

>>59354952

Have you seen AMD's stock recently?

File: huurff.png (21KB, 529x343px) Image search: [Google]

21KB, 529x343px

>>59358290
99.9999% of this board lacks the mental faculties to understand the most basic aspects of trading and investing. Not to mention the income and necessary agency to do so.
Shitposting children and NEETs will never make a penny off of the market, so they just post memes to cover their crippling depression.

Minor dip on product release news, already rebounding. Q1 earnings report will show a solid boost to revenue. Q2 will show even stronger revenue, and will include enterprise parts, and new GPUs as well. Their stock is going to hit $20. With long term debt gone, being back in the S&P500, having their credit rating lifted, the intermediate term outlook on AMD will be very positive.
They'll be one of the only organizations ever to raise their credit rating after it had fallen so low.

So they still continue with vega 10 , vega 20 ? Or just vega then next archi ?

File: AMD-VEGA-10-VEGA20-VEGA-11-NAVI-roadmap.jpg (222KB, 1562x855px) Image search: [Google]

222KB, 1562x855px

>>59358487
They did have a sever roadmap mention Vega 20 for 2018.
Picture is super, super shitty though.

>>59357641
The Fury nano was one of the most efficient GPUs for it's performance

>>59358468
I don't get it at all

NVIDIA's stock is nearing $100, yet all they do is make GPUs and dominate in some very small markets with a few ARM chips and tesla/quadro cards

>>59358746
but muh revenue

THE AMD CYCLE BEGINS ANEW
>HYPE
>DENY
>DAMAGE CONTROL
>HYPE FOR NEXT GENERATION

File: amd.jpg (183KB, 1127x1231px) Image search: [Google]

183KB, 1127x1231px

>>59358746
because nvidia makes money.
Out of the last ~5 years ( 20 quarters) AMDs only had 2 profitable ones I believe.


AMD's stock already mostly reflects its future potential, why else would it have jumped by so much so quickly ? its not like sales quadrupled or anything close to that.

>>59358830
>Out of the last ~5 years ( 20 quarters) AMDs only had 2 profitable ones I believe.
JUST

>>59358746
Look at Nvidia's yearly revenue vs profitability.
>the company’s gross margin rose from 55.8% in fiscal 2015 to 56.8% in fiscal 2016.
>Its operating margin improved from 20.4% to 22.4% during the same period.
>This is because revenues rose faster than operating expenses, which rose by 4% YoY.

Nvidia is consistently, dependably profitable. That led to their stock being hyper inflated. They were only in the $20~ range until about November of last year where they had a steady upward trend until they topped $110 earlier this year. Having very strong market presence with pervasive marketing does wonders for a company. It lays the perfect ground work for big fish to multiply their wealth. Everyone knows its a super big bubble, these things are intentionally created. The market is a game.

>>59358821
>HYPE
>DENY
>DAMAGE CONTROL
>WAIT THE DRIVERS IMPROVEMENT
>HYPE FOR NEXT GENERATION

ftfy*

>>59356280
You are in a thread about graphics cards he is a retard from /v/

>>59355612
>>59357480
>>59357630
Tell that to any research org. All tooling is CUDA first. Nobody gives a fuck about APP or CL (or they are an afterthought at best). There's a reason K80 is the most popular GPU for machine learning on server farms. For local debugging with a consumer-grade GPU going with any AMD solution is pure retardation.

And they brought that on themselves by forgetting that software, SDKs, drivers, and support are important too.