How did AMD succeed with Ryzen/Threadripper while monumentally

>>61963920
Two different divisions.

>>61963920
How did you brilliantly succeed with 4chan shitposting but can't run a profitable blog?

>>61963920
Priorities.

vega aint bad considering its priced lower

>>61963920
>Put a white man (Keller) in charge
>Zen is a success
>Put a street shitter in charge
>Constant failures

File: 1477314579257.png (151KB, 1000x630px) Image search: [Google]

151KB, 1000x630px

>>61963944
Try harder please

RTG is Raja's designated shitting division.

how is Vega a "monumental" fuckup? it performs on par with a 1080 and even has fantastic build quality and HBM stacks that will last forever and has excellent overclocking room and even better compute performance

>>61963966
>>>/pol/

>>61963990
Haha
Very well, MAD shill
Here's your 2 rupees

>>61963966
this

>>61963920
>>61963930
This right here my dude. You've got the main AMD division which handles CPUs and embedded stuff then you've got the "Radeon Technologies Group" which is the GPU division of AMD.

CPU side is doing good shit right now with Ryzen and Threadripper. GPU side is shitting itself in the enthusiast graphics card market right now with DOA Vega 64 and the barely worth noting Vega 56.

Can /v/ leave already? It's a compute monster. What's not to love?

>>61963956
>>61963966
>>>t_d

>>61963920
Bad cost-efficiency. Low yields, high cost components, and an architectural design that probably sounds great on paper but will probably require further iterations and a die shrink or two to a high-yield process before their production costs per unit are low enough to make it sellable at a more aggressive price point.

>>61964045
This.
Power companies love it too!

>>61963920
Both vega and ryzen were built for data centers.
Turns out that data center CPU's work well as normal consumer CPUs, but data center GPUs don't work that well as consumer gaymen GPUs

>>61963920
Gaymers BTFO
>>>/pcbg/

>>61963920
>monumentally fucking up Vega
Fucking up for who? The damn thing is sold out everywhere.

>>61964134
Because only like 100 were listed up for sale. Most of Europe still hasn't gotten a single card.

>>61964208
We can't, the new NPP's aren't finished yet.

>>61963920
Do we thing the GPU and CPU software teams are separate?

Obviously a team of software engineers are frantically working on unlocking the fine wine in Vega. I'm just wondering if that same team would be the ones working on the NPT bug that has plagued AMD processors all this time.

As soon as it's fixed I'm buying Ryzen.

File: zpnrqc[1].jpg (348KB, 1252x719px) Image search: [Google]

348KB, 1252x719px

>>61964208

Got mine in my tiny European shithole for launch price :3

>>61964225
hmmm

>>61964230

The npt bug is fixed friend. Though you'll need to apply the kernel patch yourself on most distros.

File: ZxF5ofd.jpg (108KB, 1198x1200px) Image search: [Google]

108KB, 1198x1200px

>>61963990
I'll take the b8.

Vega 64 is $600-$700.
A GTX 1080 is about ~$550.

With Vega 64, you'll get
>High power usage (~400W)
>High temps
>Thermal throttling because of said temps
>Disappointing gaymen performance for the price

A GTX 1080 will get you
>Reasonable power usage (~200W)
>Minimal to no thermal throttling depending on cooler
>Decent geymon performance for the price
>Anywhere from $50-$150 still in your wallet


Choice seems obvious to me ¯\_(ツ)_/¯

>inb4 go back to /v/
>inb4 nvidia shill

>>61964045
Cry baby bitches like you are funny. I buy a graphics renderer to render my fucking graphics.

>>61964231
>AMD RX VEGA 64™
>499$
>1 gallon of petrol and a box of matches
>3$
Not a good investment desu.

>>61964230

AGESA 1.0.0.6 fixes this.

>>61964269

maybe if you're a gamer manchild. It's easily the best option for my compute workloads.

>>61964134
Sold out because there were only like 16,000 in stock around the world lmao. Miners probably bought up 1/3 of them.

Raja Koduri is a snake oil salesman. And Jim Keller is a certified shit wrecker.

seriously speaking vega was in development for a long time in time that amd didnt had resources

the card itself is a beast no question about it they finally have a better triangle render than nvidia
BUT
given the lack of money basicly what i think went down is that they decide to put all of the time to develop a good pro driver(which we already saw that it shits on p6000) and let the gaming driver for later on
almost every new feature the card has is not enabled even the primitves arent working as intented its literally just a shrinked fiji on its current state

File: Nqhq83G.png (235KB, 500x488px) Image search: [Google]

235KB, 500x488px

>>61964299
Go buy a fucking FE Vega then, you dipshit. FE Vega is literally aimed towards rendering and neural network fags like yourself. It's marketed for that shit and it performs well in that shit.
RX Vega is marketed for gaymen. It doesn't perform so well in gayms. Therefore, it's trash.

>>61964372

Why the fuck would I spend twice as much for identical OpenCL performance? Are you retarded?

>>61963966 (checked)
kek

>>61964372
AMD doesn't gimp their consumer cards for most compute tasks. Buying a pro card is silly.

>>61964265
>render my fucking graphics
BLACKED 3d renders?

>>61963920
because unlike intel, nvidia was actually doing stuff the last 6 years

>>61964385
>RX Vega 64: $600-$700
>FE Vega - $1,000
Check your math buddy.

Can someone explain to a casual what went wrong with Vega? The hype was really high before it came out. So what exactly went wrong?

>>61963920
Gaymers BTFO
>>>/pcbg/

>>61964437
like what? there havent been a major hardware change in nvidia lineup since maxwell came in
all of their lines so far is just die shrinks and microcode updates
the good stuff will come once fp16 comes into play and nvidia will be forced to enable all the cuda cores on the gaming cards skyrocketing their tdp

>>61964443
In short: RX Vega is a workstation card with shitty drivers.

Basically, they were hyper focused on creating a strong workstation card. They ended up doing that and released that as FE Vega. RX Vega is literally FE Vega, but gimped with shitty drivers that don't even use the card's full potential in games.

>>61964441
Not to be that Anon, but the dude said he only paid ~$500 for it. That sounds like it'd be half of $1000. Obviously that doesn't hold true anymore.

>>61964443
the "gaming" drivers have nothing of the new tech enabled what so ever
on its current state its using fury x drivers since its almost identical

>>61964443
AMD lies about pricing, unfinished drivers

AMD lied pajeets died

>>61964486
Nah dude. $499 is the old MSRP for Vega 64.

>>61964441

Bought mine for 500. Not sure what you're talking about.

>>61964478
maxwell was massive though

>>61964255
Link? Citation?

>>61964285
No, the NPT bug is still present.

>>61964495

If he bought it at 500 he still bought it 500 and FE would've cost him twice as much you retard.

>>61964359
The gaming driver and the pro driver are one and the same. Just a few features enabled/disabled underneath.

It's not a totally different driver.

>>61964514
Sounded like you hadn't bought one yet. Lucky you for getting at MSRP.

The problems are several tiers high

Miners
Cost of production
Shit marketing department
Performance of product is over a late and uses too much power
Hype train then disappointment

>>61964529
it is cause you dont need hbcc so much on the workstation or culling tech
dont forget this is the first time amd has an awesome culling and rendering triangles perfomance (they stated it was on 11 but on the whitepaper they released it says they went up to 17..)

>>61963920
>Zen = Jim Keller = White
>Vega = Rajeesh = POO

Whatever happened to the people screaming that the FE wasn't for gaming and RX Vega and gaming drivers would fix everything?

>>61964568
Now they don't care about gaming or are waiting instructions from redteam+

>>61964514
like where? i've been spamming that F5 key since launch a few days ago and i've yet to se a single vega card below 650€
i guess im going to skip yet another GPU generation. that old R9 290° is still more than fine for 1080p.

File: 1502986440901.png (185KB, 817x460px) Image search: [Google]

185KB, 817x460px

>>61964568
so you are implying that the card itself is shit and not the drivers?

>sold out worldwide
>Vega a failure

Get a job and learn the meaning of success.

File: aX2Zmxx.png (3KB, 280x272px) Image search: [Google]

3KB, 280x272px

>>61964596
There were literally 16,000 shipped out worldwide.

Not to mention that it's literally shit for the price.

>>61964589
I have no idea why it has disappointing gaming performance, probably shit drivers given that it's AMD. It's just funny that so many people here and on /v/ said you can't judge Vega gaming performance with Vega FE since it's "not a gaming card." Whenever someone pointed out that Titans and Quadros get roughly the same performance as their GeForce equivalents in games, they just screamed "This is d-different!"

>>61964565
The hardware is essentially the same. So your low level driver is going to be the same. The higher level stuff is going to be programmed in the same way. Maybe a module that is focused on gaming will be disabled in the pro driver and vice versa.

But we aren't talking chalk and cheese here. It's essentially the same driver.

>>61964596
It's the Nintendo strategy of bragging about selling out of a low stock.

>>61964610
16000....
LOL this is just sad you dont even know what you are talking about

>>61964583

Just some local store. I figured I wouldn't get one for weeks. Just wanted to see if they even had listings yet and they had 2 in stock at 500. I almost cried when I fucked up my paypal login on the first time and figured someone else would snatch it.

File: 1501472048810.jpg (19KB, 275x253px) Image search: [Google]

19KB, 275x253px

>>61963920
Ryzen: Designed mainly by white people in America.

Vega: Designed mainly by Poo and Chinese

>>61963920
On the extreme offchance you want a serious answer, it's because RTG is way way understaffed for driver development and Vega FE and RX Vega hit hardlaunch with hilariously unfinished drivers.

Vega, being GCN 5, carries over Fiji's fundamental limitation of only being able to process 4 triangles per clock versus the 6 triangles per clock that Pascal can handle at the front end of the rendering pipeline. This limitation is not much of an issue in professional rendering and professional compute applications, where Vega already performs very well, but for gaming specifically this was the key bottleneck that prevented Fiji from ever being able to really stretch its legs in games, as it would be heavily front end bottlenecked.

Vega was designed to include primitive shaders and primitive culling to allow processing way more than four triangles per clock (up to 17 they assert in the whitepaper), and thus to be able to keep the rest of the rendering pipeline saturated at all times unlike Fiji.

However, their driver team is so understaffed that they hit hard launch with primitive shaders still not working and so Vega is still bottlenecking on trialngle throughput in games. This can be empirically demonstrated in that Vega 56 at the same core and memory clock will perform exactly as fast as Vega 64 AIr in games.

It's this front end bottleneck that explains why Vega FE can trade blows with full GP102 in pro rendering applications like Creo, Solidworks and 3ds Max while struggling to match GP104 in games. It remains to be seen whether RTG's driver devs can actually get it working any time soon, but if they actually do really succeed in substantially increasing Vega's triangle/clock rate then Vega will see big gains in gaming.

>>61964045
>It's a compute monster
It isn't just a compute monster. It's geometry performance trades blows with a P6000 on half finished non-certified drivers in pro rendering applications.

The only area in which Vega is underperforming right now is gaming.

https://www.asus.com/us/Graphics-Cards/ROG-STRIX-RXVEGA64-O8G-GAMING/
12+1 phase power

File: kfL98Dp.jpg (19KB, 482x424px) Image search: [Google]

19KB, 482x424px

>>61964640
http://wccftech.com/amd-radeon-rx-vega-limited-quantities-launch/

>>61964765
and yet oc uk said they expect way more than 90000 both aib and stock in less than a month..
hmmm
wccft being wccft as always

>>61963920
They dropped the ball on gaming since they realized that it is a niche market compared to servers and data meme market. Vega excels at compute heavy applications while it sucks for gaming.

File: mhm.png (60KB, 156x143px) Image search: [Google]

60KB, 156x143px

>>61963920
because AMD makes CPUs and ATI makes GPUs

>>61964941
2006 was a good year :(

>>61963920
Vega was developed by Pajeets and Chinks.

Ryzen was developed by White engineers.

Go figure out the rest, OP!

>>61966415
Literally came here to post this

edit: rip my inbox :)

File: Son im d...png (105KB, 255x345px) Image search: [Google]

105KB, 255x345px

No one post this yet

>>61963920

because instead of focusing on bruteforce and gddr5x, they gambled on HBM/2 and it failed. and then they thought that people might bitch about only having 8gigs of vram, they created HBC which further decreases the overall bruteforce of the gpu.

they could have just made vega run on gddr5x, have it be released a year faster, and be able to mix and match how much vram they wanted on the card. and we wouldnt have HBC taking 15% of the die and that way maybe increasing the CU's on the card.

Imagine that, we could actually have a competitor to a 1080ti a year ago or at least a competior to the 1080 then releasing a competitor to the 1080ti later.

File: 1448903258282.gif (412KB, 320x240px) Image search: [Google]

412KB, 320x240px

>>61966713

and it would have been much cheaper or at least help AMD make more money.

i believe this too be final nail in the coffin for HBM on consumer cards now that gddr6 is a few months away

>>61966713

HBM doesn't take up die space idiot. They're added on an interposer together, not on the same silicon wafer.

>>61966744

HBM is staying because it's ridiculously power and space efficient compared to GDDR.
The only reason Vega cards aren't less than half the size is because of the cooler.

>>61964685
>This limitation is not much of an issue in professional rendering and professional compute applications, where Vega
I hope this is a joke. A complex cgi scene could have dozens of millions of polys, maybe hundreds, because those cgi monkeys don't give a shit about efficient resource usage when designing assets. A complex game will have maybe a few million after tessellation

>>61963920
Budget

Lisa made is clear Zen is their focus

>>61964256

the HBM stack actually runs extremely cool and the card itself at load never goes much above 80c what the fuck are you talking about with "high temps" typical gaymen shitposter that belongs on /v/

the HBM stack is also power efficient on Vega UNLIKE on Nvidia cards where the GDDR5 runs quite hot and is power hungry in itself creating another failpoint, the wattage differences are also NOT 200w, in real world instances they won't be much more than ~75 watts from each other with both cards overclocked reasonably and if you live in the first world power is not expensive so who cares, again both viable options and like I said the pcb on the 56 and 64 Vega is actually the same as the 1500 dollar variant and extremely solid

>>61966713
Would've come out even worse because of AMD's godawful GDDR5X controllers. Supposedly the memory controller alone on the RX480 consumed ~40W of power, and to go with GDDR5/GDDR5X in Vega would've meant that ~80W of power would literally just be going to memory, for what's already a card that'll easily do 400w if you let it.
(Info is all from buildzoid's vega rant)

who AMD+nVidia here?

>>61963944
>vega aint bad considering its priced lower
But it isn't. That lower price was only for the first limited batch that went on sale on release day because AMD gave vendors $100 vouchers to artificially lower the price. All the other ones are now more expensive than their Nvidia counterparts. Vega is a fucking joke.

>>61963920
We won't be able to say definitively where Vega sits until the drivers are sorted out in a few months. Launch reviews all used drivers that had features still missing.
At present it looks like Vega56 at least is competitive with the GTX 1070 and nearly the 1080. It does appear heavily memory bottlenecked, and there seems to be a problem with effective memory bandwidth all together. That might get fixed with firmware, but at present a small memory OC can get a Vega56 at stock clocks equal to a reference GTX 1080. That isn't stellar, but its not a total trainwreck.

Vega64 however, at least right now, is 25% to 32% slower than the GTX 1080ti on average. While still drawing more power.
But when it comes to compute and pro use it seems Vega is pretty good.

Theres a lot of inconsistency when looking at performance, most seem memory bandwidth related, so we'll have to see how long it takes AMD to unfuck this with updates. Even a 10% uplift would at least make Vega64 somewhat less embarrassing, and it would make the 56 very impressive.

File: aHR0cDovL21lZGlhLmJlc3RvZm1pY3JvLmNvbS9SLzYvNjkzMDQyL29yaWdpbmFsLzAxLVNvbGlkd29ya3MtU2NvcmUucG5n.png (28KB, 711x533px) Image search: [Google]

28KB, 711x533px

>>61966845
Testing of Vega FE in Creo, Solidworks and 3ds Max shows it trading blows with a P6000 in those applications whereas the exact same GPU struggles against a 1080 in games.

From what I've been led to believe this difference results primary from Vega's front end bottlnecking much more on gaming workloads than those pro 3d rendering workloads.

>>61966999
https://videocardz.com/72123/amd-issues-official-statement-regarding-radeon-rx-vega-64-pricing
>Our initial launch quantities included standalone Radeon RX Vega 64 at SEP of $499 [...] We are working with our partners to restock all SKUs of Radeon RX Vega 64 including the standalone cards
>standalone Radeon RX Vega 64 at SEP of $499
>SEP of $499
>$499
MSRP hasn't changed, stop lying.

>>61967126
wasnt Nvidia gain almost 50% on those with the new driver

>>61966761

nobody is saying HBM is bad. On paper its a fantastic idea, but why have it if the card is still a normal length card and in the end it delays a product by a year. it doesnt give any benefits to the normal consumer.

>>61966905

Amd doesnt have gddr5x controllers. so you dont know that for sure. even if it was a shitty controller. they could have released it a year ago, and still it would have been better.

>>61967245
That was the Titan Xp, not the P6000 which is their $6,000 pro card that has certified drivers.

>>61963990
>>61963990
>Vega releases over a year later than the 1080 FE, more expensive and uses way more power
>performs on par or worse than 1080 in DX11,
marginally better in DX12
>turning on AA makes the card shit itself
>crossfire doesn't work
>consumer drivers still don't allow for proper voltage/power control
>some advertised features aren't enabled in the drivers yet
Did I mention it's over a year after Pascal's debut? Now you tell me, how do you consider Vega to be good?

AMD needs to hurry up with the Zen APUs

>>61966959
Ryzen and Nvidia?

>>61967003
How regular are AMD driver releases? Monthly still?

Can we expect next month a boost?

>>61967364
I think they're still monthly.

>>61967398
Guess I'll wait a month, decide between Vega and 1080

>>61963920
You're forgetting that Vega brings huge changes to the rendering pipeline. It can be as successful as Metal.

File: 1235.jpg (61KB, 604x499px) Image search: [Google]

61KB, 604x499px

Someone benchmark vega without AA.

File: 1502849893682.png (274KB, 1280x1440px) Image search: [Google]

274KB, 1280x1440px

>>61967280
>how do you consider Vega to be good?
Even with the drivers literally half finished and major new uarch features not implemented in drivers, reference Vega 56 clearly beats factory OC'd AIB 1070s, and when undervolted and overclocked, a reference Vega 56 is exactly as fast as a stock Vega 64 Air.

IF you can get it for it's MSRP of $399, Vega 56 is great.

Reference Vega 64 Air should be avoided, even if you can get it at its $499 MSRP, unless you are planning to watercool the card and willing to gamble on whether RTG can fix the massive fucking driver issues. For Vega 64 its definitely better to wait for AIB cards and see if RTG can get the drivers in a bit better shape by the time those AIB cards hit.

>>61963920

>succeed

They sure hyped all the fools and made everyone happy because it forced competition back on the menu. However don't be fooled, people are still overwhelmingly buying Intel shit.

That ridiculous price tag and tdp though...

I'ld rather buy a pascal gpu, it's going to be cheaper and at least it'll save me a few bucks on my electricity bill as well has lasting longer I bet.

Vega 56 is good.

>>61967474

Doesn't it consume an extra 80 watts under load as well as running way way hotter than a cheaper GTX1070? Yeah, no thanks.

>>61968284
compared to what, a stab in the gut?

File: 61e223389b1b0c65faa7c33719ff095c.png (674KB, 918x474px) Image search: [Google]

674KB, 918x474px

>>61963920
threadripper vs 350$ intel

>>61964596

Yeah because they can't even supply that small amount of cards you dumb fuck... Fuck they couldn't even plan so third parties would have time to design their own cards.

Fuck I don't even care, it's AMD trash anyway but I at least hoped they would bring the prices down for Pascal so I can upgrade sooner than later.

They can't even get that right.

>>61968342

Ryzen IPC is meh at best and their single thread performances sub par compared to Intel offering.

It's just not that good for gaming.

>>61964256
I'll take a bite of your bait.

>high power consumption
Yes, if you are a moron and run it at turbo mode, meaning you consumer 100W compared to power saver mode (and 65-70W compared to balanced) for a 3-7% fps boost, you deserve getting shafted at your electricity bill.

>high temps
As a result of above
>thermal throttling
As a result of above

Now on to Nvidia
>reasonable power usage
Only on the Founder's card, the rest go way over 300W. Even the Founder's card hit 307W consumption on Anandtech's Crysis 3 test.

Rest is spot on though

>>61968342
>using a game to benchmark a processor with more than 4 cores
Are you fucking stupid?

>>61968558

Is TW WH even multi threaded? I thought it was kinda like Arma 3 sucking off one core.

>>61968342
WOW. AMD BTFO. buy intel processors

>>61963920

Intel stopped pushing the envelope and nvidia did not.

>>61964041
Vega 56 would be a good deal at "MSRP". specially
for those with combined gaming/workflow

>>61964567
Is rajeesh an actual R&D engineer? or just the spokesman

anyway, let's hope that Raven Ridge redeems the shitfest that Vega has been

>>61968389
>wider core than kabylake
>meh IPC

No no, that's not how it works

>>61968528

So what? Vega eats a great deal more power hungry than Pascal, that's a simple fact you stupid fool... Once you get the custom OC Vega cards it's only going to get worst...

>>61968389
Because Threadripper was not made for gaming.
Hell, anything but a 1700X is overkill for gaming. You'd look just as much as an idiot if you'd buy a 6900K for gaming as well.

>>61968698
Stop being a dumb nigger and inform yourself on power consumption between different modes: https://www.youtube.com/watch?v=bIGpvBrwXvA

>>61968811

Hurr durr I'm a fatfuck and a pathetic AMD shill.

Jesus just kys...

>>61968869
Yeah, you know you fucked up your argument and went to call me an AMD shill.

News for you, you petty moron, I don't care, I buy what's best on the market when I need it.

>>61968949

Called you for what you are trash.

Fuck off with your obvious shilling.

kys.

>>61968811

what the fuck disgusting accent this is?

Make me feel sick in the inside.

>>61969003
haven't even seen the video but i bet my ass this is Adored TV. Yeah, thick scottishaccent. it's odd.

>>61968304
Undervolted and overclocked Vega 56 pulls around 240 watts but is clearly faster than even a manually overclocked AIB 1070.

pls give a newfag a quick rundown on the vega debacle

>>61963920
x86 begin standards well know ecosystem,don't need to write driver, llvm or gcc makes all job compiler,standard memories,easy interchange CPU intel to CPU amd.

GPU need write very complex driver for each new architecture,support OpenGL,new type memory, because team fail on driver,AMD just overclock GPU expect begin equals to 1080 nvidia.

AMD don't had money to build good drivers.

>>61963930
JIM
KELLER

>>61969990
>Hyped for over a year
>Marketing pointing out "Poor Volta" Nvidias next generation GPU.
>Typical Raja Koduri hype with two Capsaicin events, two E3 events, one CES event, and one share holders meeting
>Promises of new technology like HBM, tiled rendering, improved IPC, and many more

Vega is released.

>Paywall package deals that add 100 dollars minimum to the already expensive GPUs
>400w power consumption when overclocked and reports of throttling at 60 degrees are surfacing
>Completely bottlenecked in graphics rendering
>Half the technologies promised in the hype are disabled, not functioning, or completely broken hurting performance
>Update to the drivers for improved mining performance on day two of release

AMD has completely screwed themselves for the forseeable future as Vega is the basis for Navi. We will now have two to four Vegas on a single die.

The architecture is so fucked because Radeon has pushed the limits of their software development teams that one year later and the card still isn't functioning as promised.

Raja is a failure and needs to be shoved back in to a lab where he excels, or be let go if he isn't willing to accept his responsibility for creating the Bulldozer of GPUs.

>>61970023
So why didn't Raja take that in to account when designing his new GPU? He is no longer at apple. He knew he didn't have the resources he once did. So why is he still employed?

>>61970397
>Raja is a failure and needs to be shoved back in to a lab where he excels, or be let go if he isn't willing to accept his responsibility for creating the Bulldozer of GPUs.
>trusting Poos

>>61970349

>Senior Vice President and Chief Architect of RTG
>Raja had nothing to do with Vega

Retard, just fuck off.

>>61964057
>reddit
You have to go back

>>61970397
I don't think it's fair to blame Raja for RTG not having enough money to hire enough driver devs to actually get RX Vega's drivers done in a timely fashion. RTG's budget is a decision made above his head, and trying to get Vega SSG, Vega WX Vega MI25, Vega FE and Vega RX out the door in the same year with such a limited number of driver devs was just too much to do in the time available.

It's also way too early to call Vega Bulldozer tier. Who the fuck knows what RX Vega is capable of if the driver devs can actually implement primitive shaders at the very least, because right now Vega is still front end bottlenecked at 4 triangles/clock like Fiji was.

>>61966845
>tfw you used to be able to count on screen polies in games with your hands

>>61966959
>tfw 1600 1050ti and comfy as fuck
Miners can go fuck themselves though.

>>61964627
Those retards were funny. Vega had been posting same-looking numbers for the previous 6+ months in leaks and official game demos, yet they though there would be some voodoo magic driver with RX which would boost it to 1080Ti levels on release.

>>61968389
>Ryzen IPC is meh at best
>clock for clock better than a 7700k
Try again, dipshit. Ryzen's only issue with single threaded performance is that no Ryzen chip can really get past 4GHz, whereas Intel sells many of their corelet chips with 4-5GHz clocks.

The trade off that's utterly phenomenal is that almost every Ryzen chip can hit 4GHz, even with all eight cores on and stacked with one or three other dies for 16 and 32 core configurations, whereas Intel typically clocks similar core count CPUs a little above 2GHz.

>>61963941
One requires an adept application of Poe's Law. The other is adept pandering.

>>61963930
This. AMD is still like AMD + ATI.

File: 4rzmVesQyGhEWR7LPDBq0_F-1cXSSqS95X0LgbxObi0.jpg (126KB, 1024x684px) Image search: [Google]

126KB, 1024x684px

Vega is ok. Eehhh just price them like 1070/1080 and ok. Will you?

Yeah. no one will give you that price EVER, vega is a fuck up.

t. amdrone

File: 2017-08-01_121254.png (50KB, 825x561px) Image search: [Google]

50KB, 825x561px

>>61966959

5 years, going strong
but I think 660 was a mistake, should've bought w/e radeon at that time and sell it to miners for the same price of a new card

>>61964443
There's nothing wrong with Vega as uarch, just RX Vega is an unfinished product.
Hence why they are staggering the launch as much as they can.

>>61964517
>we added TBR
>nice!
And that's it.
>>61967245
That was Titan Xp.

>>61970416
Fine wine, fanboys buy GPU and waiting to good drivers.

Real reason
GPU hardware was made in china top notch team
Driver in India pajeet still couldn't figure how makes good drivers.

>>61971058
>rtg Beijing
>top notch
AHAHAHAHAHAHAAHA

These are the same idiots that have been throwing in more and more alus since the 290x even though performance wasn't scaling up. They're responsible for the hot and slow messes that barely compete with Nvidia for several generations and now gcn, designed by a long defunct team, finally hit a wall and their bullshit hacks aren't doing anything even though Raja showed up and gave them the beefy ALUs that run 15% higher frequency than and have 15% higher ipc and wouldve been the first time AMD matched an Nvidia flagship on almost every metric if those retards could've fixed the frontend and memory controller. They couldn't figure it out like nvidia's hardware team did so they're just going with the "programmable" general shader meme and passing the work off to the driver team that's not even a tenth their size

>>61971344
Programmable shader stage is much more flexible than throwing moar fixed function units.
:^)

>>61970148
CERTIFIED

>>61971387
It wouldn't even be a problem if these monkeys knew how to design hardware properly. Now it's up to an underfunded and understaffed software team.

Meanwhile over at Nvidia their hardware team does an excellent job but their software team is just as good and large enough to have an extensive software suite that just werks

>>61971437
>Meanwhile over at Nvidia their hardware team does an excellent job
>throw moar fixed fucn units
>but it bloats the die
>MOAR
(You).

>>61970397
>all those rumors and non-issues as greentext

how much does Nvidia pay you?

>>61971449
>big pascal at worst ties Vega 10
>beats the shit out of it most of the time, including most Opencl compute tasks that weren't specially optimized for GCN
>king of CUDA because Nvidia actually invested in a real software team of American experts that don't need to rely on (((open))) standards

>>61971449
>implies nvidia is a bunch of retards
>still beating AMD
What does that make AMD buddy?

>>61968662
>I'm so new I have no idea who was the lead architect of R300
gb2/v/

>>61971574
>What does that make AMD buddy?
Bunch of underfunded retards.

>>61963920
they quarantined off the cancer into RTG to keep Raja from infecting the rest of AMD

>>61971801
Lisa brought Raja in

>>61968558
>ignoring that lower fps could be a reason to not buy a Ryzen CPU reguardless of muh coar count
Are you a fucking shill?

>amdtards now defend vega
literally stockholm syndrome

>>61972210
Her first truly bad decision.

>>61971489
Less than AMD pays you.

>>61968389

Ryzens are perfectly fine for gaming you /v/tard unless you like playing strategy games and dwarf fortress.

>>61963920
>succeed
Intel is sleeping and has only woken up. Once they are fully awake and aware they will demolish AMD.

Nvidia is different, they're like schoolyard bullies who won't relent until AMD switches schools.

>>61973101
>Once they are fully awake and aware they will demolish AMD.
Hahaha it's going to take them until 2020/2021 just to get their new architecture out, and there's no telling if it'll actually be any good or not.
>>61973101
>Nvidia is different, they're like schoolyard bullies who won't relent until AMD switches schools.
Polaris was very competitive with NVidia in the mid-range before miners raped everything. They had the high end on lockdown, though. Unfortunately it seems like Vega won't be doing much until they get the driver and partner GPU situation figured out.

>>61973360
Polaris really was pretty decent, but they bungled up the launch (again) with the PCIe power draw debacle, even though it was a non-issue in practice. Polaris also wasn't quite good enough to really rock the market, I mean it was basically equivalent to 1060 6GB while being slightly cheaper, which is good but not immediately impressive. They did well but in the end NVIDIA still sold more 1080s than AMD did 480s, if the Steam survey is anything to go by.

Are there any Vega 64 vs Pro Duo (Fiji) comparisons yet?

>>61963956
Came here to say this

>>61973836
Nobody but racists cares about the skin color.

>>61973854
Race is more than skin deep

>>61966894
If Vega was using GDDR5 total board power would easily go over 400W stock.

Even with HBM2 using less than 20W at 1.35V and pushed to 1050/1100Mhz Vega's TDP and power consumption are huge, it's worse than Fermi.

>>61963944
Gr8 b8

>nvidya

Enjoy your 1080 and nonfree cuck drivers /v/.

>>61975773
You can't use the "AMD is open" argument anymore, AMD locked their BIOS just like Nvidia did with Pascal.

>>61974777
They'd just downclock the core.
Anyway there's nothing inherently wrong with HBM2 besides volume, since barely anything uses it.

>>61975876

Nice try nvidicuck.

File: amdrone.png (524KB, 920x900px) Image search: [Google]

524KB, 920x900px

>>61976044

>>61963920
Vega is not meant for gaming.

>>61976209

Yeah the top-end seems to be a tiny pool of gaymen from /v/, and a huge number of blockchain miners. AMD probably did the smart thing concentrating on the market with scale and money.

File: vega64 liquid cooling.png (528KB, 534x401px) Image search: [Google]

528KB, 534x401px

>>61963966
wouldn't that be RDSD? need to be a little more creative man.
for instance,
>Raja's Tempremental Graphics

>>61963920
>implying it's a success

>>61963920
A white man created Ryzen
A Poo created Vega

>>61968968
Butthurt Intel shill, just bought a 1600x, the holocaust didn't happen either.

>>61963920
It's obvious that their GPU money went to their CPUs, and now that Ryzen is destroying intel in nearly every category and miners are still buying out Vega, the GPU department will have a real budget again.

>>61963930
They all share the budget though.

>>61976678
Yeah that's Ryzen is in the best selling CPUs on Amazon.

>benchmark ricer market vs crypto mining market

I would just ignore that fringe hobbyist market as well.

File: lisa-monkey.png (647KB, 605x603px) Image search: [Google]

647KB, 605x603px

>>61970950

>>61964481
>>61964488
So what you're saying is I should buy one when I can find it for cheap enough (which I know is another issue) because once better drivers roll out it'll blow my tits off?

File: 64131663ad8dd0f251d987d7f6b1c400.png (638KB, 843x442px) Image search: [Google]

638KB, 843x442px

>>61968589

>>61963966
LEL

>>61963944
Where have you been since 2016?

>>61968589
it is threaded, but the issue is you hit a ram bottleneck before you hit a cpu one.

>>61966713
seeing how vega wasnt even better than gtx1080 they shoudl have just ramped production on 570-580 and roll in the dough from miners

File: 1467302016288.jpg (52KB, 568x612px) Image search: [Google]

52KB, 568x612px

>>61966999
>>61967145
>https://www.kitguru.net/components/graphic-cards/matthew-wilson/ocuks-andrew-gibson-clears-up-rx-vega64-pricing-disaster/
>So what exactly is going on with Vega’s pricing? We got in touch with Andrew Gibson at Overclockers UK who gave us the full breakdown:

>Launch price was $499 with NO games for the Black card, as outlined to us by AMD as a launch only price. AMD allowed us to sell a set amount at this price, which was several hundred, clearly not enough as they were sold out in approximately 15 minutes. After this the regular price was $599 with FREE games for both the black and silver cards, $699 for the aqua card plus taxes.”

>Gibson’s claims are corroborated by an additional source too, as Norwegian tech retailer, komplett.no, also spoke out, telling the site tek.no that the cheaper launch price for RX Vega64 was limited to just 275 cards to allow it to appear more favourable at launch. Here is the full quote translated:

>“Komplett.no today confirms to Tek.no that the price of just over 5,000 kroner was limited to a limited edition of 275 graphics cards, as the company – as one of AMD’s “selected” online stores – was allowed to sell at a favourable price at launch. The RX Vega 64 version we had for sale was in a limited edition of this price and will unfortunately not be put up for sale again. When these were sold out, we had to remove this product from our pages”.

Think again, AMDrone.

>>61978746
>think
>amdrone
pick one

>>61971603
>underfunded
Intel has more money than 2x Nvidia's AND AMD combined, yet Ryzen still BTFO out of intel.

Vega is a a turd. A big, fat, steaming TURD.

>>61978802
Calm down, /v/eddit.

Vega is a fine workstation card

You know, i just... kind of feel like it's pointless talking about vega, because who is even going to be able to get one.

I think the only Vega card i would get, if i can get one, is the 56 however.

>>61964269
Top KEK, made my day

>>61964443
The hype was all lies.
Clock for clock Vega is only 6% faster than Fiji.

If you took Polaris and scaled it to 64 CUs it would kick Vega's ass.

Vega is more than a step backwards, it's AMD stumbling over their own feet and falling on their head.

>>61979201
>Clock for clock Vega is only 6% faster than Fiji.
In gayms, yes, and that's due to primshader being disabled.

Is Navi going to replace both polaris and vega, or will AMD have two architectures again?

>>61979258
Vega will also replace Polaris.
Vega11 is presumably Q1 2018.

>>61963920
I wished amd didn't accquire ATi and spend money on only CPU and their south/north bridge platform alone. Wait....would it be worse if intel got ATi ? Nah, could be even worse.

File: maxresdefault.jpg (37KB, 2000x770px) Image search: [Google]

37KB, 2000x770px

>>61963920

>VEGA 128 NEXT WEEK $650

But srsly Threadripper 1900X is super attractive. I'm going RAID-0 nvme drives and 2x GTX 1080's because they're going to be dirt cheap shortly. Need those PCI-e lanes, nigga.

>>61980006
Retarded self manages to reply to thread instead of making new thread. Must be getting too old for this "internet" thing.

http://www.gamersnexus.net/industry/3025-amd-statement-on-vega-pricing-retailers-fault

kek

>>61980254
>curlyhair is still trying to be (((Kyle)))
This is silly.

>>61979223
>wait for the magic dust
never change amdrones

>>61980289
>uarch-defining features are magic dust
Never change, br monkeys.

Vega is not meant for gaming.

http://www.youtube.com/watch?v=RNXGr-8jcnE

Look at the HUGE gains on the RX Vega!

>>61980289

>uma delicia

>>61980304
when are they going to work, drone?
like actually work

>>61980334
It is means for gayming.
The uarch features are mostly graphics.
It's just culling in shaders is not enabled right now and DSBR probably needs tweaking.

WTF I love Linus now!

http://youtu.be/evaBde4MKDQ#t=4m3s

>>61980343
>sopa macaco

>>61979383
I heard that volta isnt until at least March next year . I need to sell my 2x1080s before then!

>>61980365
>amdrone can't write a coherent response and resorts to forced memes

>>61980385
>uma delicia

>>61980254
This jew boy trying hard

>>61980411
t. triggered pajeet

>>61980411
Hes probably one of the most neutral reviewers out there you retard.

Vega pricing explained for the mentally impaired.

AMD both directly and through it's AiB partners sells RX Vega 64 air (GPU only) to retailers for $499 with a SEP (Suggested.Etailer Price) of $499.

Retailers obviously want to profit from this as they are not an AMD charity.

As a special launch AMD had a pool of money that meant they could give a rebate to retailers to offset the loss of profit with strict guidelines to sell RX Vega 64 air at $499 with no games included. Only a limited stock of these were supplied to select retailers.

The packs are most likely sold to the retailer at $499 (unconfirmed) with a suggested retail price of $599 (as would normally occur with any other product).

What is confusing people is that people think the $499 SEP/MSRP is the price of the GPU without the games bundle. This is true but only to the extent they are 'special' limited stock versions and retailers are getting a kickback from AMD in order to sell them without a profit loss.

AMD are pushing the packs with the actual MSRP of $599. End of. They say that they will have more stock of the GPU only available but again my bet is they will be limited in number and I doubt it will happen again after that. AiB partner cards will then go on sale and the MSRP will change for those anyhow.

how did amd even manage thread ripper and ryzen anyway?

>>61980814
Time and money.
Mostly time.
4-5year R&D cycles are rare nowadays.

>>61980834
even though 14nm is a recent thing?

>>61980860
Process node has nothing to do with CPU uarch.

>>61980888
they dont have to design it around it? how does that work then since it will be physically smaller?

>>61966643
Looks like Alex Jones as a 40yo virgin

File: 1502824025458.png (85KB, 1024x970px) Image search: [Google]

85KB, 1024x970px

>>61976593
Radeon Technologies Group

>>61964765
>wccftech
I work for a tech site myself and we just fucking ignore wccf because they fuck up so much.

>>61978746
>e-tailers using the swirling rumors to try and pressure AMD into bribing them into selling at SEP
I wonder ((((((((((((who))))))))))) could be behind this?
|
|>
|
|3
|

>>61978802
RX Vega's drivers aren't even half finished.

Until we see whether primitive shaders and primitive discard can solve GCN's 4 triangles/clock limitation and actually fucking saturate the rest of the rendering pipeline, we aren't going to have a fucking clue what Vega is really capable of in games.

The reason RX Vega launched with such hilariously alpha drivers is simply that RTG didn't have enough resources to hire enough driver devs to actually implement this shit by launch.

>>61981397
PDA is working.
Shaders don't.
>The reason RX Vega launched with such hilariously alpha drivers is simply that RTG didn't have enough resources to hire enough driver devs to actually implement this shit by launch.
Well they do, everything works in Pro drivers.
Shame they focused a little bit too much on Pro side of things.

>>61979201
What result would you expect when Vega is still front end bottlenecking on triangles like Fiji did? Unless and until they actually get primitive shaders and primitive discard both working together in Vega's drivers you won't be able to see the clock-for-clock gains, because the bottleneck at the front of the rendering pipeline precludes benefits from later in the pipeline from showing themselves.

It's already empirically demonstrable that Vega beats the shit out of Fiji clock-for-clock in any test that avoids the 4 triangles/clock bottleneck.

>>61981433
It beats Fiji clock for clock even in pure shader math AFAIK.
Vega is bretty nice, uarch-wise.

https://www.phoronix.com/forums/forum/linux-graphics-x-org-drivers/open-source-amd-linux/969834-more-benchmarks-showing-how-gallium3d-with-rx-vega-smacks-amdgpu-pro-s-opengl-proprietary-driver/page4
Meanwhile on leenux.

>>61981426
I know PDA is working, but without primitive shaders working you aren't seeing the full benefit because the bottleneck is in front of PDA is reached.

And not having enough driver devs to finish the pro and gaming drivers at the same time is still not having enough driver devs. They triaged pro drivers ahead of gaming, but with enough staff they could have finished both.

>>61981445
I know. It's too bad the fucking complete madmen threw it out the front door without the most important new uarch feature implemented in drivers yet.

I am really looking forward to see what its going to be capable of once it's no longer choking on triangles at the front end. How insane is it that I have no idea what Vega is actually capable of in gaming and the fucking thing has already been hard launched?

>>61964269
>not a good investment
>AMD makes the longest lasting cards on the market

>>61981502
> has already been hard launched?
The launch is staggered, probably artificially.
You see, only ANUS showed their AiB offering.
No Sapphire or XFX (these are traditional AMD vendors) yet.
No MSI or niggabyte (prepare for another round of 3slot THICCness).

File: 1406035209403.png (17KB, 473x400px) Image search: [Google]

17KB, 473x400px

>>61964045
Lets see it compute this

>>61981537
Sure, but the reference cards are available to purchase and its still not possible to give even a reasonable estimate of how Vega10 will performn in games once the drivers actually implement all the uarch features. That's pretty absurd, really.

>>61981609
>That's pretty absurd, really.
That's ATi.
>Sure, but the reference cards are available to purchase
No one seriosly buys blowers, and CLC one is pack-only offering.
Anyway, 56 one will be some nice value (assuming one OC's HBM stacks).

>>61981634
>Anyway, 56 one will be some nice value (assuming one OC's HBM stacks).
56 is going to be more than a nice value if they get the drivers finished.

Undervolted, +50% power target and max memory OC'd reference Vega 56 already dumpsters an AIB OC'd 1070 and gives 100% of Vega 64 Air stock performance. That is a pretty good starting point when critical uarch features haven't even been implemented yet.

>>61981656
Does current drivers have decent AVFS checkpoints?
Or it's still 1.2v@everything?

>But AMD Quote that they can discarge 17 primitives per clock. Nvidia can only 8. So where Comes the Advantage from?
>That 17 was only 11 a few months ago, so AMD is still finding ways to evaluate or cull more. Best guess is the packed math and scalar being a bit more versatile. Backface culling with really low precision as that should remove more than half.
Vega was literally in active development PAST the tapeout.
Programmable geometry pipelines are nice.

>>61971023
Yea man my 7950's are still going strong

>>61981710
It's useless in gaming and by the time it's fixed in 2-3 years mcm Navi+ will be out on 7nm.

It's not even that great at anything else due to the stupid tdp of 300-550 watt

>>61981666
It's not 1.2v at everything, but it is always using 1.2v at DPM 6 and 7, when most Vega10's will hit full boost clocks on way less than 1.2v, so they can easily take being undervolted with a +50% power target to reduce throttling at similar power consuption to stock.

>>61981710
Now they just need to actually fucking get that shit implemented in drivers! Fuck!

>>61981753
>higher geometry throughput is useless for gaymen
/g/ - Technology.

I wonder if consumer Volta will implement packed math.
AMD is working rather closely with Bethesda and Ubishit right now.
Also fucking Sweeney on AMD keynote.
Weird stuff.

>>61981771
T. Vega shill

Have fun waiting

>>61981858
>reeeeeeeeeeeeeeeeeeee technology is bad
(You).

File: vega-design-team.jpg (147KB, 1024x768px) Image search: [Google]

147KB, 1024x768px

>>61963920
All their GPU R&D got outsourced to inexperienced chinks and pajeets after budget cuts in 2014. For Zen, AMD pooled all it's resources and put their most talented engineers on the project in one final gamble to save the company.

>>61981890
Chinks & Pajeets were doing grunt R&D work for ATi since forever, dude.
Juniper was developed by pajeets and Cypress by chinks, both at once. And that's eight years ago.

Anyway, kudos to Pajeet & his team for prim shaders.
These are the most interesting thing in GPUs since the introduction of Polymorph Engines.

>>61981753
Navi will need the same drivers since it'll be the same macroarchitecture

>>61981837

the problem with the card is that it cant save it from being a powerhog and expensive for AMD to make.

you cannot get around that.

>>61981890
Is the sole aspect of your life race baiting?

File: goats-trees.jpg (1MB, 1600x1005px) Image search: [Google]

1MB, 1600x1005px

Conspiracy theory: VEGA is an intentional flop.

AMD is being targeted heavily by crypto-miners. While you may argue this is a good thing because it moves cards, it's in reality a bad thing because it erodes what little of a base they have in the long term. Crypto is nonsustainable. It isn't something you can rely on. It's Chads buying up massive lots of cards that would otherwise go to loyal customers or would grow your base.

AMD is losing customers. They're losing customers that would keep coming back and come to hold a sense of brand loyalty. They may want to shake that.

can we just put vega behind us and admit that it's good workstation card and a complete failure of a gaming gpu.

>>61963920
Vega is p based for everything except gayming.

>>61984852
Why would they want an intentional flop?

>>61985087
To get crypto-miners to focus on NVidia instead.

>>61985076
Not at the power it consumes and heat it puts out.

>>61985104
Yeah, that still makes zero sense.

Miners want AMD for a reason.

>>61985133
>Miners want AMD for a reason.
"Da red wunz go fasta!"

>>61985104
What the fuck are you on anon

>>61985327
I already explained this, Anon.

>>61985112

>has never seen the juice the quadros can pull

>>61985474
Why stop there? Why not bring in Teslas for actual computational workloads that shit all over vega.

Why didn't they just put fury on a smaller node with boosted specs and hbm2?

>>61985551
Becsuse Raja is incompetent.

>>61963920
Vega's fine, you're just an Intel shill.

>>61986509

Intel doesn't care about Vega- they are being molested by Nvidia. AVX and its iterations exist solely to slow the anal ravaging Intel are going to suffer in the big data centres. Hell its why Nvidia has been taking a bit fat chunk out of their market share for years because for these sorts of workloads a gpu will crush a cpu no matter how much throughput it gets.

>>61985551
Because it wouldnt help. Fury was laid out using the densest library so even if they shrunk it it wouldn't be able to clock any faster unless they redesigned it like the vega, though it would be more efficient. It would also still be bottlenecked like the original fury x

>>61986684
Vega is still bottlenecked.

>>61963920
>tricking youtube shills into saying the card is a great value then revoking the $100 wholeseller rebate
Pretty intelligent marketing 2bh

>>61986826
Until they realize they just spent their last remaining enthusiast capital on dicking people in the ass.

Then the few remaining people who buy Radeon GPU's to actually use them instead of mine with them, evaporates further, and then they become the new Matrox.

>Brilliant

>>61986900
Haha, yea. Like shilltubers won't swallow whatever they're told to.

>>61985028
>failure of a gaming gpu.
The drivers aren't even half finished, Vega is going to be an overclocked Fiji in gaming until RTG can actually get all of Vega's uarch features implemented in the gaming drivers.

>>61986947

I do hope this turns out true. My 290x is rocking fine for 1080p but I wish to make the move to 4K and a 1080ti is too much money for the performance in my opinion. A vega chip delivering that sort of performance for less would entice me - especially given freesync screens tend to undercut their equivalent g-sync screens by a fair bit.

>>61986947
>until RTG can actually get all of Vega's uarch features implemented in the gaming drivers.

Which will happen when?

>>61986928
Like people give a fuck about shilltubers anymore, they have all been exposed for what they are. Infomercials with out Billy Mays.

>>61987076
Who knows? I didn't think they'd be fucking crazy enough to hard launch the reference card an paper launch the drivers, so what do I know!

>>61963920
Poo in the loo

>>61963920
>monumentally fucking up Vega
they are selling those GPUs like crazy, at high prices...

>>61984852
nice """logic""" you tard

Is Vega really a fuckup when it sells like hotcakes? Seems to me that appeal to miners was the best plan

I mean, yeah, it fucks over people who wanted actual graphics cards for actual tasks that you need a graphics card for, but it makes plenty of sense from a business standpoint, no?

>>61987284

>Is Vega really a fuckup when it sells like hotcakes?

Everything AMD does is a fuckup because - inherently - when they make a good product it exposes just how badly you get jewed by competing products (especially from Intel) and this makes the consumer whore side of /g/ (as in, all of /g/) violently angry.

>>61984852
AMD should shutter RTG for a year. Stop ANY production of GPU's which would basically tell miners to FUCK OFF. Imagine all the butthurt miners and watch the mining craze (at least on GPU's) fucking take a nose dive.

When AMD deem that mining is dead enough open up their GPU division for business again with a complete revamp of their strategy.

If only...

>>61988621

Nyet - all that would mean is miners would go buy Nvidia since mining is vastly more effecient on a gpu (though for lols threadripper isn't half bad at it).

>>61987329
Well they did target 1080 level performance as was previously stated.
They never officially stated they would reach 1080Ti level performance as some here were speculating.
Still a massive disappointment that AMD cannot compete with NVIDIA's top dog after all this time.
Ideally they should be leapfrogging one another performance wise not play catch up with their competition for a sightly lower price a long while after.

Crypto currency mining only complicates matters considering it might not be sustainable but they do benefit from it in the short term until the demand for mining plummets and an influx of dirt cheap cards floods the market.

>>61988819

>They never officially stated they would reach 1080Ti level performance as some here were speculating.

True but there is nothing - that is publically available - that inherently limits GCN to reach such performance. The most obvious point is AMD's resistance at increasing ROPS and simply bolting on more compute units (and related solutions) to attain throughput. Now naturally 128 ROPS with the same ratio of compute units and whatnot that hawaii has would lead to a massive die which is likely why AMD don't do it but on the geomrty front they are lagging. If you want to crunch compute tasks GCN is an absolute beast and vega is no exception - it fights Nvidia cards thats are considerably more expensive and have notably larger dies.

My own theory is that AMD banked on either 1) GPGPU or 2) igpus basically invalidating discrete cards outside of niche accelerators (i'm looking at you xeon phi and tesla) when the Vega design was initially instiagted. That and a combination of zen basically sucking up all the major talent and resources in the company leads me to suspect that Vega is - in practical terms - merely a hot-clocked GCN chip rather than being the GCN successor everyone hoped it would be. While it is cliche Navi is looking to be that radical shift but I severely doubt it can meet the expectation as (effectively) MCM gpus are the holy grail and much like HMS Dreadnaught will instantly invalidate everything before it.

>>61988922
Nvidia also seems to be working on their own interposer as well, wonder if Navi will beat them to the punch.
I hope AMD does manage to compete with Nvidia otherwise we will end up in a similar situation to the stagnating CPU market which only managed to pull its shit together recently.

>>61988819
>>61988922
Actually, Vega was specifically designed to overcome the limitations of GCN. Most of the new uarch features in Vega like primitive shaders, primitive discard and draw stream binning rasterization are all aimed at addressing GCN's previous limitations.

Unfortunately, the driver team did not succeed in getting primitive shaders implemented by launch for RX Vega, so at present Vega is still front end bottlenecked in gaming by the 4 triangles/clock throughput limitations Fiji struggled with. This is why Vega 56 performs exactly the same as Vega 64 at the same clock speeds right now. Until primitive shaders get implemented in RX drivers there is no way to know what Vega is actually capable of in games.

>>61988773
I'd be cool with that.

File: SSG_678x452 (1).jpg (20KB, 678x487px) Image search: [Google]

20KB, 678x487px

>>61963920
>While fucking up Vega

Only gamers think they fucked up.
AMD is going to be swimming in money selling Ryzens and these Vegas

>>61989909
Good bye 30 billion dollar a year business growth.

Good bye enthusiast market.

Good bye gaming in general.

>B-b-but consoles!
Powered by a molested Jaguar core and a Polaris.

>B-b-but GPUOpen!
Used by precisely one company.

Good job AMD. You just blew your enthusiast capital. Life as the new Matrox must be pretty damn depressing.

>>61990134

> You just blew your enthusiast capital.

Hardly - even during the bulldozer days the gpu side of the business has been a black hole - its the cpu division that keeps the company afloat.

>>61990704
It's funny that you think so.

The past three or four years AMD has been pillaging GPU profits to get Zen out and not be a complete cluster fuck. And they barely managed to do that.

It's time to bring some people back from their culling under Rory Read. They can't continue as they have been.

>>61966894
Wow, you're a fucking retard, Did you watch Linus's video? He directly compared GPU only power draw. The water cooled edition was sucking a grand total of 388W from the 2 8 pin connectors + PCIe port.

If you're that fucking retarded, please don't talk.

>>61963920
They are literally releasing miner focused cards, including relevant model name and all.

At this point they are probably just handing the gaming market to Nvidia until next gen (or the one after it), once they have enough budget to come up with cards that are good at something else than just GPGPU.

How come when Intel and AMD first made multi core CPU, they cut the idea to use multiple die on one processor? Well it's back now, probably because they can't make monolithic cores efficiently.

>>61992829
It's not just about size, it's more about the number of transistors. Even on old, large dies you couldnt expect an uncut wafer to have no flaws because there were just so many transistors and there will always be errors. So you design your chips so that you can maximize how many chips you get per wafer while predicting the rate and distribution of error

>>61990134
They can't even produce enough chips to meet demand, you really think that's the most ofbtheur worries? It's not like they benefit from retailer pricing so there's no reason to control supply in a way to look like sales are doing well. That's a trick for companies only out to protect image, like sjw marvel, it's clear AMD is out for only DINEROS

>>61986793
Yea, but less so. Still a big fuck up in design though

>Hit render on Nvidia
>Go to bed

> Hit render on vega
>Get a sandwich.

>>61975899
>They'd just downclock the core.
And by then the performance would be Polaris level

>>61963920
AMD a shit.

>>61994885
t. pootel shill