/w4v/ - Daily Wait for Vega Thread

>>61359572
bugger off m8

>>61359587
Don't like it? Don't respond. Better yet, tell us how we're all in for Another Massive Disappointment.

File: 2 posters.png (116KB, 1350x544px) Image search: [Google]

116KB, 1350x544px

>3 posts
>2 IP addresses

got you now

Nice overclocks on gamer nexus' hybrid watercooled Vega FE.
Plenty of headroom for the RX to be higher.

>>61360043
Actual good news about wega? No wai.

If vega blows I'm going to buy threadripper

>>61360058
Kinda, a very stable and constant 100mhz+ on core and memory but also another 100w's. 400w average from the cables alone.

Water cooling also drops the leaking though, which means these might not be great cores/memory and a more mature process will gain a lot for this setup.

What would be the price ceiling for this?

>>61360287
Don't know pricing on RX Vega yet. FE Vega is $1500 for the watercooled and $1000 for aircooled. Keep in mind this is not "gaming ready" GPU at the moment. It's for prosumer market.

>>61360314
How much RX Vega must be for it to be "too much"

File: 272.jpg (24KB, 635x272px) Image search: [Google]

24KB, 635x272px

Reminder

>>61360343
You can already get 1080's for $500. I'd say they need to price it at that or lower.

File: poo on the mu.jpg (152KB, 1024x635px) Image search: [Google]

152KB, 1024x635px

>>61360043
You forgot to mention that he found that it was drawing 400W through the PCIe power cables alone, plus whatever it sucks from the PCIe slot (another 75W). Close to 500W total power draw to just about edge out a GTX 1080 drawing a third of that. Or to put it another way, more power consumption than a 295X2. For a single GPU.

Great job, Raja!

>>61360395
>4k ALUs running at INSANE clocks draw fuckton of power
No shit sherlock.

>>61360354
>480mm2 die with HBM2
>cheap

Is Vega going to be a loss leader?

>>61360422
Maybe it's time to rethink your architecture at that point, rather than shitting out another underwhelming housefire that will undo all the good will Ryzen has generated.

Wait for Drivers(tm).

>>61360480
>Maybe it's time to rethink your architecture at that point
They did.
And don't worry, every card will be a housefire till we reach MCM.

>>61360480
They can't afford to, there's a reason why all their video card releases since 2011 have been minor iterations upon GCN1.0.

Hopefully with Zen's success, RTG will have the budget to develop a new architecture at last and put graphics core next to rest.

>>61360594
Too bad Vega is not a minor iteration.

>>61360615
That was the lie AMD attempted to peddle, but evidently it isn't too much of a departure. Same old hot, loud, power guzzling GCN with disappointing performance.

>>61360643
>l-lies
Lmao.

>>61359572
Jesus man, no Vega yet? Holy fuck.

>>61360677
There's Vega FE.

>>61360685
Nah nah, I mean the one for video games.

>>61360721
That will be very very very late, hopefully early September for AiB cards.

File: 1492989943767.jpg (39KB, 599x555px) Image search: [Google]

39KB, 599x555px

>>61359908
Oh no, whatever are you going to do???

>>61360721
>>61360729
Reference cards will be out much earlier than that.

>>61360790
Maybe. AiBs get final bios for RX Vega on 2nd of August.

>>61360478
Isn't AMD the inventor of HBM? So they don't have to worry about patent use costs?

>>61360952
They are.

>>61360952
It is a joint venture of AMD and SK Hynix, but HBM is still expensive to manufacture regardless of who invented it.

>>61361083
You say that like 10 high speed GDDR5X modules are cheap.

You're looking at some $20-40 total BoM cost top for the 8GB Vega, 16GB is a different story.

>>61361156
10/10 made me chuckle

>>61361156
At least they met Q2 target.

>>61361156
>when real Vega was supposed to release Q3 2017(Q3 2016 tapeout)
>when real Vega will compete with Volta

>>61361225
Vega FE is real Vega.
The drivers are not there.
At all.
Hey at least ROCm works and seems decent.

>>61361225
>Q3 2016
This is what bugs me, tapeout was way too late for just a mere Pascal competitor.
And I personally know AMD didn't redirect funds from RTG to Zen until middle of 2016, which would not impact Vega's tapeout at all.


There's something way to fishy with Vega, and considering AMD has done literaly ZERO damage control for the terrible 'gayming' reviews of FE, they're hiding something

>>61361298
New shader stage is alone worth discussing, shame RTG gives us ~zero actual low-level info about Vega.
At all.

>>61361349
Of course because they haven't posted it yet, there's no press slides or presentation.

>>61361203
>met
>"""""""met"""""""

>>61361392
Maybe they'll tell us more at SIGGRAPH.
As for now, they are totally silent.
And they were for the whole Vega R&D cycle.

>>61361405
Duh. They "met" their Q2 target.

>>61361405
Hey, silicon's out.
Deadline met, investors won't sue.

The driver or worst case a hardware defect are a entirely separate case.

>>61361427
I don't think they'd sell defective chips at ebin prosumer cards.
If that was the case they'd dump it to someone much like Intel did with Early Purley.

>still not /w4n/

Why was AMD trumpeting about Zen from the rooftops in the leadup to release, but are barely acknowledging the existence of Vega?

>>61361466
Navi is pretty far away and we've yet to see properly working Vega.

Here's a compilation of Vega FE issues that seem completely abnormal.


>pro and gaming mode don't do anything
>power gating/AVFS seem disabled
>geometry performance worse than Polaris
>stream binning rasterizer not working
>total usable bandwidth much worse than Fiji, leading to bottlenecks above 1400MHz
>texture performance FAR worse than Fiji


Assuming drivers are the cause for all of these, you're not looking at a mere 20-30% increase in performance from drivers, not even close, this is a gutted chip at the moment, the texture performance and bandwidth alone are fatal.

>>61361477
They were?
Zen had like 2 (or even one, the one with 3ghz ES) demos before New Horizon event.

>>61361492
>they disabled power gating
>even fucking Primitive Discard Accelerator from Polaris
WHY?

>>61361492
agree Itc clear that the chip wasn't finished at all make all the excuses you want this is just dumb form a company point of view.

>>61361492
Oh, I forgot.

>core voltage stays 1.2V stays regardless of clockspeed

>>61361477
If you haven't noticed, RTG has a different marketing team, and take a look at their previous release which they praised to the moon.
It's probably radio silence until they figure out when they can get the drivers working, I guess the AM4 APUs and late release of Raven Ridge(compared to Q1 Summit Ridge) are a pretty good indication of the driver situation

>>61361528
The silicon was finished like ages ago.

>>61361521
its because its still runni9ng with modified fiji and that chip didn't have those features.

>>61361505
we are not taking about cpu's dummy

>>61361521
If you look at the radeon source code, you'll know power management is the largest part of the code, and considering Vega is hinted to use a lot of Zen's power management, it's totally not ready.

>>61361561
>chip didn't have those features
What? Its full Vega10 XTX silicon.

>>61361548
the bugs where clearly not fixed on silicon and it looks rushed.
this is not a software bug its clearly just defective.

>>61361532
>>>core voltage stays 1.2V stays regardless of clockspeed
It gets better, you can achieve stable clocks (1400MHz) at 1.0V
Vega Fe is overvolted to hell, hopefully the same is not true for RX because there's no way they'll be able to meet the 12TFLOP at 225W they said last year.

If the reference RX VEGA can match or beat the Ti the AIB RX's will absolutely obliterate it.

>Please don't just be slightly faster than a 1080. PLEASE!

>>61361610
It it was defective it would never be sold as a prosumer card, dummy.
They'd dump the chips as a compute boards to chinks, Amazon or Google, much the same way Intel did with Early Purley.

>>61361625
GN Hybrid's got considerably closer to 1080 Ti, so not having Radeon Pro software is gonna make things even better.

>>61361492
Could FE be the ultimate sandbagging?

>>61361625
We know nothing about Vega right now.
Hell, DOOM demo from January performed better than Vega FE.

>>61361657
It's was the only way to meet Q2 target.
At least ref looks nice and PCB design is stellar.
>>61361655
On pure compute?
Hilarious.

>>61361492
If they turn on TBR and don't use retarded stock voltages that alone would lower the power draw significantly, leading the card to boost to 1600MHz all the time.
The others are just icing on the cake, the TMUs being like 40% slower than Fiji even at 1400MHz is really worrying though, could Vega not even use TMUs, some other transform stage?

>>61361747
>TMUs being like 40% slower than Fiji
What the fuck?
Like really, I've never seen launch this confusing.

>>61359840
With threads like that, you're setting yourself up for one.

Is Vega still GCN in anything but ISA?
Because all this shit working literally backwards is weird.

File: Screenshot from 14-07-2017 01:39:04.png (24KB, 1664x241px) Image search: [Google]

24KB, 1664x241px

>>61361863
http://llvm.org/docs/AMDGPUUsage.html#processors


Shit's very different, that can only explain the driver situation.
Pic related.

>>61361961
Oh, that explains Greenland/GFX9 being so late.
It's basically a totally new uarch that simply retains old ISA.

>>61361961
Someone on B3D forums was actually implying that Vega has a bunch more registers than Fiji, leading to the far improved compute performance even at the same clockspeed.
Guess that would make sense, but why isn't it doing anything for gaming?

>>61362062
Drivers?

>>61361961
Some more GFX9 specific improvments

>For GFX9 the appature base addresses are directly available as inline constant registers SRC_SHARED_BASE/LIMIT and SRC_PRIVATE_BASE/LIMIT. In 64 bit address mode the apperture sizes are 2^32 bytes and the base is aligned to 2^32 which makes it easier to convert from flat to segment or segment to flat.

>The global segment can be accessed either using buffer instructions (GFX6 which has V# 64 bit address support), flat instructions (GFX7-9), or global instructions (GFX9).

>The M0 register is not used for range checking LDS accesses and so does not need to be initialized in the prolog.

In any case, Vega seems to be easier to program for at first glance

>>61360480
>Maybe it's time to rethink your architecture at that point

That's what Vega was supposed to be, but they had to put in a legacy graphics path so the card works without 1 year of driver development. Now they only need 6 months to get it to perform good.

>>61361621
>It gets better, you can achieve stable clocks (1400MHz) at 1.0V

How much power does it use that way? I wouldn't mind running it at just that, even if it gets 1070 performance, if it means that the card runs icy cool.

>>61362176
What fucking 1070 performance?
Vega11 is far away.
And they are yet to make Vega10 actually fucking work.
Like, read the thread.

>>61362186
Vega 11 is probably cancelled. We haven't heard shit about it.

>>61362210
Why would they cancel it?
It's a smaller die since you can't bin Vega10 into every niche.

>>61362186
>What fucking 1070 performance?

Wow, that alone triggered you so hard that you didn't even comprehend the rest of my post, didn't you?

Vega FE currently is between 1070-1080 performance.

If undervolting and underclocking to 1.0V 1400MHz means that Vega XL-TX-ABCDX silicon runs icy cool, I'd rather have that, even if it means the card is now reduced to 1070 performance.

I'm running a 3570k on stock reduced to 0.95V too (partially because it couldn't OC for shit).

>>61362262
You won't buy FE anyway and we don't know RX performance yet.
At all.

>>61362279
It's the same fucking chip, only board and drivers will differ. Performance will be, at minimum, Vega FE levels. If we are lucky, it might get more. But like I said, if it performs 1070 level and stays icy cool with some undervolting, that's already fine enough for me (assuming it is priced accordingly and not some bullshit $400 price tag because "lol premium card").

>>61360354
So basically a $100 undercut on price?

That honestly isn't enough. $200 would be, but I have to buy a bigger PSU for Vega and that's an extra $100 right there.

>>61362328
>$100
Where did you get this?

>>61361492
It sounds like the silicon design itself is fundamentally fucked.

There were rumours that Vega had some sort of last-minute design problem which required a new tapeout or whatever and Vega FE is the broken shit that they rushed to market to meet the quarterly deadline, which is why nothing works - the silicon is fucked.

>>61362380
Why would they dump defective chips as prosumer cards when you can give them to meme learning dudes?
Intel did the same with Early Purley.

>>61362380
Nothing that couldn't be fixed with a respin, which would take 3-4 months.

Question is, if the fuckup exists, when did they find out about the fuckup, and when did they start the respin?

>>61362374
Canadian money

so a $10 difference in american is $100 in canadian pesos

or close enough/whatever

>>61362394
>Why would they dump defective chips as prosumer cards when you can give them to meme learning dudes?

dunno, but if they did dump them it was probably to meet the quarterly deadline so they didn't get sued by their investors.

>>61362395
I would strongly argue they found out/knew about it at least around the time of that dual vega shitshow demo, and probably a few weeks earlier than that.

>>61362430
They won't dump defective chips as a. retail prosumer cards. That's lawsuit material.
There's no silicon bugs, since tapeout was ages ago, they had all the time in the world to do a respin.

>>61360354
So Vega will start at $700 if they fix the performance, $600 otherwise?

>>61362444
>They won't dump defective chips as a. retail prosumer cards. That's lawsuit material.

Yes they do, it's called binning.

>>61362456
Binning has nothing to do with selling defective (as in uarch features are not working) silicon.

>tfw you're essentially forced to unironically wait for Vega anyway

>>61362578
Just buy nvidia?

File: Cry - Horizon.gif (2MB, 360x203px) Image search: [Google]

2MB, 360x203px

>tfw MxGPU never ever
I don't want a billion VMs, just let me share between the host and one guest pls. I hate to dualboot, but I still like my games.

>>61362588
I was about to buy a 1080, but I don't feel to confident in buying a $540 GPU that's more than 1 year old

>>61362588
1060 and 1070 are out of stock, or largely inflated prices.
980 Ti and 980 are still pretty expensive, and often higher prices than the 10-series equivalents.
1050 (Ti) are lackluster.
1080 (Ti) are more than I'd want, and are more than I'd realistically like to spend on a video card.
As a 750 Ti user, there aren't many Nvidia choices that are really worth it, especially factoring in the fact that they're used, I'd ideally want more than 2GB buffer, and it offering a fair amount better performance than my current card.
Suggestions for cards are definitely welcome if you'd be willing to shed some light on choices.

>>61361156
nice one

Vega is either going to be the biggest sandbag / surprise in history or it's going to be the worst new arch / most disappointing launch ever.

Both seem equally likely at this point, though I'm still holding out hope it's going to be the former.

>>61365936
I'm betting on the later. I have zero expectations of Vega performing as well as a 1080Ti, I don't think it will be priced low enough either. No SR-IOV is disappointing, I thought they would at least leave it on the FE but that doesn't seem to be the case. It's a shame, I wanted to upgrade from the 970, but I guess I'll stay cucked until Navi comes out.

>>61360354
Nope. Nvidia can counter any vega shit. they got per/watt

>>61366497
nvidia die shrinks are 80% performance

>>61361492
>Nvidia sells a 3.5 card
>Everyone is hollering like monkeys over a gimped card
>AMD sells a 1000 card that is clearly broken
>Everyone accepts it

>>61368334
it's not broken in pro applications for the price

>>61366497
>they got per/watt
by cutting out features, when they are forced to get into dx12 perf/watt goes to hell because they weren't working on improving it and simply die shrinked

>>61368614
Well, not exactly.
Kekler was that and perf/watt was still mediocre.
It's mostly due to TBR and better DCC.
And some nice power gating.

File: 1489981669274.png (2MB, 1920x1080px) Image search: [Google]

2MB, 1920x1080px

QUICK, /g/ , post pics like pic related, I want to print them on shirts and attend the Budapest event

>>61366497
>they got perf/watt
...by cutting hardware sheduler and redirecting saved power to CPU.

>>61368649
Oh boy the PCI slot power draw "issue" that never amounted to anything more than a hill of beans .. yep you're totally gonna nail them with that.

File: 1488865804127.png (113KB, 918x786px) Image search: [Google]

113KB, 918x786px

>>61368649

>>61373073
NVidya should be elated that someone wanted to buy their leftover mobile junk.

>>61373203
Man even random chinks design better SoCs than nVidia.

>>61360478
either that or the margins on GP102 and GP104 are fucking massive

>>61373244
They are, especially on gp104.
Both AMD and nVidia sold ~600 mm^2 dies at ~$650.

>>61365936
4D sandbagging

>>61373255
I understand the correlation between die size and production costs is related to the number of die candidates per wafer but wouldn't transistor count also have something to do with it?

sounds to me like an equivalent die size on a smaller process would be more expensive but I could be way off

>>61373285
>sounds to me like an equivalent die size on a smaller process would be more expensive but I could be way off
Hell no. More wafer candidates thus higher yields is always better than less masks thus simpler litho.

>>61373285
Performance still matters and die shrinks give a lot of performance.

>>61373275
But disabling discard accelerator from Polaris?
Why.

Given the IF architecture, would it be possible to make a die with one Zen CCX, some HBM and a bunch of Vega cores? I'm curious how Raven Ridge will work.

>>61373418
Yes?
RR is one CCX and 768 Vega shaders connected with IF.

>>61373426
that was pretty much telegraphed but I'd be interested in seeing the source on those specific numbers

>>61373443
SiSoft ES benchmarks afaik.

>>61373452
>>61373426
Oh neat, I hadn't seen that. Thanks bruh.

>>61373285
Amount of transistors doesn't make a difference. Only size does, due to silicon only being 99.999% or so pure. The impure parts cannot be printed on, so whatever they printed there is an unusable chip.

With larger die size, more dies on the wafer will hit an impurity, so you can make less working dies per wafer.

Transistor count does not matter because they don't burn them one by one onto the silicon, they use photolitography (they take a large 2d picture of the die and "print" it onto the silicon wafer). Some nodes might use something else but the principle is similar.

>>61373354
Vega probably doesn't have the Polaris discard accelerator, or it has a newer, different one that needs drivers for it.

Based on previous generations, will the RX Vega XTX be worth the extra premium because of the liquid cooling? Also, what kind of liquid cooling are we talking about here, hybrid? Do I have to worry about reconfiguring my Just Wait™ for Vega Build?
I'm thinking of slapping a 360mm radiator on the front of a Fractal Design S, and if I will go for the XTX, mount the GPU's radiator at the bottom.

>>61373821
>RX Vega XTX be worth the extra premium because of the liquid cooling?
Only if you like non-gaymen design.
PCB itself is stellar, especially compared to nvidia's reference junk.

>>61373880
nigga I don't care about aesthetics as my case has no windows. Performance-wise, will the premium be worth it, in your opinion.

>>61373915
In all honesty i don't know.
RTG is completely silent about Vega.
Like, zero attemps at damage control.

>>61373880
>PCB itself is stellar
where did you see it?

>>61373943
why the vega van tour though?

>>61373989
>where did you see it?
Buildzoid did the analysis.
https://www.youtube.com/watch?v=zeoiPVVUnl8
>why the vega van tour though?
I guess RTG is somewhat confident in drivers right now.
We'll see.
I want the whitepapers.

>>61374008
not sure RX going to have same quality PCB on reference

>>61374028
Radeons historically had good reference PCB designs for flagship cards, see 7970 or Fury X or whatever.

>>61374048
why is while right side is empty though? just to keep cooler screw holes?

>>61374071
Yes.
Also a hint for Vega Nano.

>>61374084
I just want a fucking awesome flagship card with 1180 performance so I can skip the next generation.

>>61374115
>skipping Navi
Mate?
If Navi truly is MCMed GPUs, it'll be the biggest jump in performance since motherfucking unified shaders.

>>61374144
>t. AMD shill

>>61374835
t. leather jacket man

>>61374144
And that is why Vega hurts me so much. I have a Fury X and have no actual need to really upgrade, and yet there I am, wishing I had enough reason and expendable money to do so.
Still, I'm left wondering; if I did decide to get Vega, at a massive loss since selling off my Fury X will be near impossible due to the fact I took the reference cooler off of it for a custom waterblock solution, and then Navi comes along and completely trounces Vega? I'd be in despair.
...and yet there's still not 100% guarantee that Infinity Fabric will even be as good as we hope, with early reports of TR saying that there's some latency issues for latency dependent programs... and crossfire/SLI has always been so horrible because of latency, so unless the latency is that much lower compared to normal crossfire that it isn't even a consideration...

>>61375191
>...and yet there's still not 100% guarantee that Infinity Fabric will even be as good as we hope, with early reports of TR saying that there's some latency issues for latency dependent programs...
What?
You mean a database that fits into L3?
Aka something that's basically nonexistant in real life?
> and crossfire/SLI has always been so horrible because of latency
Because there's a fuckhuge difference in latency going through PCI-E and it's IOMMU grouping and chips sitting on ONE FUCKING INTERPOSER.
>And that is why Vega hurts me so much.
Remember the last time ATi took a long break?
That was long ago...

>>61375229
I'm aware that the situations are completely different, yes, but its still unproven technology until it gets publicly released and we see how it handles in a wider range of situations.
I'm hoping that IF does very well, there's no reason to want it to do poorly unless you're an Intel shill; but I'm also not willing to just throw myself onto the IF train like I did the HBM1 meme.

As for ATi? No, I don't remember, sadly. At that time in my life, the only PC games I played were simple games that ran on my old Thinkpad T-20, or playing on consoles.
I didn't start paying any sort of attention to graphics cards and PC stuff until the Radeon HD7950 came along.

>>61373781
great fucking answer anon, thank you

>>61375444
>As for ATi? No, I don't remember, sadly.
R200 was barely competetive, R250 launch would be a disaster, so they focused on R300.
They took their time, and basically gave a whole generation for nVidia to establish market dominance.
Then crushed it with 9700pro.
Fuck, 9700pro served me so fucking well.
>I'm aware that the situations are completely different, yes, but its still unproven technology until it gets publicly released and we see how it handles in a wider range of situations.
IF is a protocol, it's transport layer agnostic.
And it scales up to 512Gb/s.
We'll see.

>>61375191
Same, I have a fury x and have no reason to get vega but still want to buy one

>>61375502
>And it scales up to 512Gb/s.
what exactly stops it at 512? could multiple dies in a single MCM each individually reach that?

>>61375535
I currently have a 480 and a 4K Freesync display

needless to say I'm a bit antsy.

>>61375563
CPU dies? Never, even the 4 dies and the most powerful skylake xeon can barely pull 300GB/s

File: XJE1VnT[1].jpg (416KB, 2560x1440px) Image search: [Google]

416KB, 2560x1440px

Le Reddit post. Still interesting though.

>>61376372
Apparently 1700MHz+ is possible.
Hopefully Vega is more than clockspeed and all those features actually start working soon.

>>61376453
Oh, it's much more than clockspeed.
It's GCN in ISA only.

>>61376453
>400w

>>61376609
That's impressive for not-Fiji.

Vega needs to be another R600-tier missstep.
Or else there's going to be too much win this year.

>>61376609
Even power managment is fucked at the moment, power gating and AVFS disabled, TBR not doing its job, auto voltage stays at 1.2V regardless of clock.

It's a fucking mess, who knows how it'll perform once all that's working.

File: Die+milkbags+https+wwwpixivnet+member+illustphp+mode+medium+illust+id+63813481+trigger+mentionlist+bewbdudes+trigger+mentionlist+smolhol+trigger+mentionlist+monmusumaniacs_3cb5bc_6329204.gif (293KB, 480x273px) Image search: [Google]

293KB, 480x273px

I want one glorious autist attend AMD press even in wojak ushanka. Even more points for attenting nvidia event and getting escorted from premise live.

>>61376695
>1.2V stock

Based AMD targeting the undervolting market.

>>61376735
>Even more points for attenting nvidia event and getting escorted from premise live.
While screaming about woodscrews, 1.7 yields and porkshoulders?

>>61376748
More like AMD disabling or not implementing power management features in FE drivers.

File: Mattdoggy+used+roll+picture+mattdoggy+rolled+image+_51c7a7f46aa24cb6bd8958490fab8628.jpg (11KB, 326x184px) Image search: [Google]

11KB, 326x184px

>>61376753
If you are capable of screaming about it after getting tased by security goons.

Come on pajeet, pull another R300.
Leather jacket man is getting TOO cocky and is in dire need of bitchslap.

>>61376779
That too, but AMD's GPU division has been churning out highly overvolted stock chips ever since GCN launched, it got really bad with GCN2 though.
Situations where you can undervolt Fiji, Polaris and Hawaii, increase performance due to less throttling AND make it use less power at stock are hardly rare.

On the other hand Nvidia bins a whole lot stricter, so they probably throw a lot of chip that don't hit a certain voltage level away.

I think AMD does pretty stupid here, the high leakage/low leakage chips should really be sold as different SKUs, because not all of those low leakage ones go to embedded/Pro stuff

>>61376836
They need to bin everything to supply the demand.
Vega10 covers like three markets, being cucksumer, pro and meme learning one.
They need to bin EVERYTHING.

Also why the fuck Vega FE is faster per clock in SPEC than Fiji but sucks at everything else.

>>61376886
What are drivers.
What is firmware

>>61376902
Why did they even release it?
First impression is everything for GPU market.
It's full of fucking crybaby (man)children with 0 techinal knowledge of GPUs.

>>61361480
Vega is DOA

>>61376940
Based on?

>>61376940
>>61376914
Because it does fine for pro stuff, its intended market, so might as well make some early adopter money.
Because shareholders could have sued them for not hitting the Q2 deadline.
Because sandbagging(TM)

>>61376959
Yes it does fine in pro stuff but we all know what GPUs are for.
They are toys.
For gaymen.
The first impression is that Vega is bad for gaymen.

>>61376988
never in all my (admittedly short) years following hardware releases have I known so little about a product with so much information.

hell, the thing is RELEASED and we still don't know shit. mind boggling.

>>61376988
Don't be retarded.

>>61377012
The hardware IS there.
We don't know how it works.
If it even works.
Why the hell RTG are so silent.
>>61377028
Oh come on nVidia's entire success is built upon shiny presentation and agressive marketing.
Their quaterly revenue is 60-70% gaymen cards.

>>61376735
Very nice gif, anon. Very cute!

File: 1499425461805.png (312KB, 595x654px) Image search: [Google]

312KB, 595x654px

>>61370717
>>61373073
cmon guys, I want some real /g/ material

>>61379052
Raja give me another R300 you fuck.
Also Vega whitepapers please.

>>61376914
>Why did they even release it?

To keep their promise to shareholders. "Q2 launch".

File: 1490292507275.jpg (171KB, 1010x568px) Image search: [Google]

171KB, 1010x568px

>>61379052
Get Raja to sign this.

File: assblast.png (2MB, 1103x826px) Image search: [Google]

2MB, 1103x826px

jealous? :)

File: [HorribleSubs] Gabriel DropOut - 05 [720p].mkv_snapshot_02.48_[2017.02.06_14.00.42].jpg (114KB, 1280x720px) Image search: [Google]

114KB, 1280x720px

>>61379565
>Vega FE Bling Bling Edition

>>61379605
I wish nvidia also sold Titan bling bling edition with some quality PCB since both board and their blower suck.

File: 1472377365329.png (70KB, 225x179px) Image search: [Google]

70KB, 225x179px

I need nonmeme answer.
Should I go for AMD or just bite the bullet and buy 1080ti
It also affects what monitor I buy.
Also will buttcoin miners fuck up the prices again?

>>61379934
We don't really know anything substantional about Vega10 and no since Vega is shit for mining.
Just way 1080ti if you want the ebin flagship here and now.

>>61361156
RIP Hammond

>>61379934
>Should i go for AMD or a GPU with a better performance and lower TDP?
Gee i fucking dunno mate.

>>61379155
At siggraph?

>>61360790
inb4 fiscally bankrupt miner fags buy up all the fucking stock again

>>61380905
Vega is meh at mining.
So no.

>>61381087
>Vega is meh at everything

FTFY

>>61381087
>Fury X
>Fastest and most profitable GPU for Ethereum
>Vega is meh at mining
[Citation needed]

>>61381134
>mining
>profitable

Show me one person who's even so much as broken even on their $10,000 mining rigs.

>>61381184
I don't know any people who mine.

>>61381096
Hello Jensen! How's V100 yields?

>>61381289
All V100s are going to be sold to the US Govt.

So how good are the liquid coolers on the AMD cards? Are they quiet? Are they effective?

I seem to remember that the Fury X liquid cooler was incredibly loud. I've been trying to build a very quiet PC, and currently the GPU is the weakest link in my system when it comes to noise.

Tempted to get the RX Vega liquid cooled if it's any good, or if you can switch out the fans on the radiator (which I doubt you can unless they renew the design).

>>61379565
shouldn't it be end of q3 for water cooled FE?

File: w4vega.png (486KB, 1380x1706px) Image search: [Google]

486KB, 1380x1706px

how legit is this "leak" result? the video explains why this is the real deal (same code used by the doom presentation by amd last year, etc.)
if rx vega performs as good as 1080 there is no way im not buying that shit. even at >600$

here is vid:
https://www.youtube.com/watch?v=_Vll4cOeHgU

>>61383018
>Fury X liquid cooler was incredibly loud
faulty pump supplier, it wasn't incredibly loud it was a bit above later iterations

>>61383635
why would you buy 1080 for >$600?
sure it will live longer than 1080, but too many cons for that price?

>>61383681
i said i would buy the vega not the 1080. the equivalent 1080 is 640$ to 770$ on amazon...

>>61383864
Nvidia's GPUs are going to become cheaper once Vega gets released.

Might as well invest the $50 more and get a 1080TI instead. At least then you'll get top performance from day one and don't have to hope the poo in loos miraculously manage to develop a working driver.

>>61361625
RX Vega will not beat the Ti.

>>61361655
>GN Hybrid's got considerably closer to 1080 Ti
It was still 20-30% off in most titles.

>>61383903
50$ more? try 100$ to 300$ more. also fuck nvidia

>>61361625
Your tears, once RX Vega gets released, will feed Raja's loo witch.

>>61383018
AIO coolers just about always have noisy pumps unless you can control their speed.

If you're going for a quiet PC you don't want any watercooling

>>61370508
Seems to be working quite well for them. At higher resolutions your CPU is mostly sitting on its ass waiting for the GPU so might as well give it something to do.

Who unironically pays almost 2 grands for a GPU that's going to be outdated in less than 3 years?

>>61384068
Not really the pump I am worried about, it's the speed of the fan and the size of the radiator. The Fury X was cooled by a very small radiator and a single fan that went full retard under load, which made noise. So what I would be hoping for would be for a bigger radiator and two fans that you can swap out for better ones.

Given Buildzoid is gushing over the PCB i've got no doubts - one way or another - Vega is going to be a monster. It will either be a volta competitor or its going to be the second chernobyl.

The real question is how much does AMD pay you mister mod to contain all shitposters in one thread as a form of damage control.

>>61360043
AiO cooled GPU can oc its clock by 6%? Truly incredible AMD quality.

>>61384810
Vega is literally Fury X all over again with one exception which makes it so much worse.

>Draws tons of power when compared to Nvidia alternatives.
>As a result it runs so fucking hot that water cooling it is the only reasonable solution
>Still 20-30% slower than Nvidia's Ti GPU which Nvidia releases few months before AMD launch to hold the title of having fastest GPU in reasonable price when compared to $1000+ Titans.
>Still struggling with unreleased unprepared drivers that take years to be as good as Nvidia's.

The only difference now is that Vega seems to be way more expensive in production which could mean it's a DOA pos that won't be justified even by AMD fanboys and their price/performance charts.

>>61385442
Wait for Drivers(tm).

>>61385467

Why wait? /g/ is full of software engineers and gpu-related experts so they KNOW that Vega will be shit. Buy Nvidia™ to-day!

>>61383864
That's because miners fuck up with prices everywhere. If you think Vega will release at its MSRP you're delusional as fuck. Either way it's going to be more expensive than 1080 while having 1080 performance without 1080 efficiency. Gl finding a non-ref RX Vega that runs below 70 degrees under load.

>>61385502
>t. AMD fanboy

>>61383864
Even now I see bottom/mid shelf 1080s for $500-$600 in USA. With the retarded mining craze that makes 1070 gtx cost $500 and RX 580 $700 those prices will probably stabilize at $500. Vega would literally have to cost less than $500 to be worth considering.

>>61385593

/g/ knows even less about gpus and their inner workings than /v/ (which is an impressive feat) what Vega is right now is suddenly being taken as gospel despite there being many, many factors that allow performance to change between now and when the RTX versions ship. /g/ is convinced hardware is all that matters and if that was remotely true AMD would've crushed Nvidia long ago. This hasn't happened precisely because Nvidia invested more heavily into their software ecosystem.

>>61385634
Hello AdoredTV. Shouldn't you be making another shilling video atm instead of shitposting in here?

>>61385660
AdoredTV shat on Vega and thought AMD was retarded for putting out rebranded GPUs.

>>61385765
Catch you later guys

>>61385502
Thank you leather jacket man!
Gotta buy 1080ti.

File: rytdht.png (7KB, 577x194px) Image search: [Google]

7KB, 577x194px

>>61362417
what that's not close at all

>>61383903
>Nvidia's GPUs are going to become cheaper once Vega gets released.

Only if no cryptocurrencies are still worth mining with a GPU, otherwise it will be impossible to buy one and Nvidia cards won't change in price.

I would die laughing if it turned out Vega hasn't come out because AMD is making a fortune Ethereum mining.

The biggest issue is the 1080 is only $500 which doesn't gove AMD a lot of breathing room. Is anyone honestly expecting 1080 performance with HBM2 at $400?

>>61388244
>Is anyone honestly expecting 1080 performance with HBM2 at $400?
No, because it's not going to be 1080 perf.
Think of Vega as of another R300.
That was like the last time ATi conceded high-end to nVidia and took a long break to cook a new uarch.

>>61385765
adoredtv is one of the most incompetent channels that is sometimes right(dx12 ryzen debacle)
he is always wrong when optimistic
he is always wrong when pessimistic about something

basically youtube equivalent of wcctech

>>61388278
>Think of Vega as of another R300.

vegavan is kind of a tell of something, they won't make audience to sign NDA right? what would be the point?

>>61388334
It's either they finally have a non-crashing driver.
Or there'll be an NDA.

Fucking RTG.
Why the fuck are they so secretive about Vega's low level design.

>>61360395
No one cares about power draw. Only about the performance. Can it beat a 1080 or 1080ti?

>>61388896
What matters is perf/watt.
Also you know nothing about Vega.

>>61360480
Navi is one year away, vega was always a stepping stone architecture.

>>61388909
>perf/watt
Nope.
perf/unit of currency

>>61361298
Yeah, vega was supposed to competell with the 10xx series. As it stands they are months away from competing with the 20xx series. This is bad for amd because nvidia will have them by the balls once volta drops.

>>61388912
>vega was always a stepping stone architecture
(Almost) complete GCN redesign is in now way a stepping stone.

>>61388956
>Yeah, vega was supposed to competell with the 10xx series.
>late 2016 tapeout
>compete with Pascal
?
>volta
Wow bigged dies thank you leather jacket man.

>>61361625
Oc pro vega is equal to non oc 1080.

At best a 10% improvement will mean it is wlequal to a oc 1080. No way this touches current ti, let alone volta ti.

>>61389019
Read the thread.

>>61388561
Scenario 1:

AMD posts arch slides, plenty of new shit but it's performing even WORSE than Fiji: Everyone laughs at them

Scenario 2:
They're waiting until everything is ready.

>>61389059
>performing even WORSE than Fiji
They would can the design ages ago then.
ATi/AMD was never afraid of shelving/canning the designs.

>>61362447
No, top of the line consumer vega will compete with 1080, but be $100 cheaper, as per amds usual strategy.

>>61389079
You don't know anything about Vega to make any assumptions about performance or pricing.

>>61374144
>>61374144
Yeah navi is less than 16 months away and with vega looking like this I may skip it as well. Especially since my 380 still does fine and 4k isn't quite the yet imo so I game in 1080.

Probrably just plan a navi 4k build.

>>61389143
We don't know about Vega's performance right now.
What if it's another R300-tier launch?

>>61388909
Gamers (gpu biggest market) don'tcare about price per wat. They care about absolute performance levels at a given price point.

>>61388982
>bigged dies
Amd tards complain about big gpu but big cpu is fine.

I care about the performance of the gpu, not it's size. My full tower will fit it just fine. Micro atx tards can suck a dick for all I care.

>>61389239
Vega is supposed to bring further perf/watt improvement over Polaris.
>>61389279
I am not talking PCB size you mongrel.
Bigger die = less wafer candidates = lower yields = higher prices.

>>61389093
Wen was the last time amd tried to compete against ti? 2014? And they lost a shit ton of money with the fury.

>>61389298
How's that relevant to Vega?
You (and me) know nothing about it.

>>61389286
Who cares about candidates? Wtf is this a presidental ellection?

>>61385624
$500????? the lowest trashiest 1080 is $550 on amazon.

>>61390090
Neck yourself

>>61390127
i got a "refurbished" one from amazon for 500 bucks and it's working and oc without any issue