Vega

>yet another X is dead meme thread
Fuck off back to >>>/v/

>>60338640
What do you mean, The MI25 Card has to deal with the GV100 at its own HBM2 game.

>>60338634
Vega will beat titan in 2 weeks. Volta will beat vega in 2 months. That is how video cards have worked for years.

>>60338707
You think Vega will be dropped on the 31st? or just teased at? My thinking its Teased on the 31st, "Launched" in June then actually released in laaaaate June with actual stocks in mid to late July.

File: www.hardware.fr-2017-05-12-00-30-53.png (35KB, 595x469px) Image search: [Google]

35KB, 595x469px

>>60338680
>MI25
paper

>>60338777
Haha thanks :0, Basically gets crushed in all aspects, plus new ones for AI

File: 1484548311184.png (252KB, 595x654px) Image search: [Google]

252KB, 595x654px

>>60338634
Yes. Commence the wait for Navi.

>>60338923
This, I love this. Im also torn, Upgrade to the first 2080's or wait for cut down GV102 TI

>>60338932
wait for ti then buy the 2080 so you don't shit on by the launch pricing

t. got a 1080 when they released, few months later they're like 30% cheaper

>>60339178
Prepare for a die shrink of the existing Nvidia products! Nvidia backed themselves into a corner, you watch.

AMD BTFO

File: 1494487855012.png (1MB, 1228x1502px) Image search: [Google]

1MB, 1228x1502px

>>60338634
HOLD ME GUYS
I CAN'T WAIT FOR NAVI

>>60340016 Navi, 2019

>>60339204
I'm kinda ok with that. The 1080 is a nice balance of efficiency and performance so I wouldn't cry over getting more of the same.

Probably going to buy a Vega card though just to see what they're like (and claim a business expense by shoving it in my work pc for a while)

>>60338634
No, Vega already killed Volta since even 802mm^2 die from NVIDIA can deliver only 30tflops FP16.

These threads live in a bubble where people think that the average person has almost $1000 to drop on a video card. Normal people will pay a lot less for 15-20% less performance than the Nvidia equivalent

>>60340319
never mind that fuckhuge die size and the yields they'll get with it

each one will cost a fortune to make

File: 1487463821177.png (2MB, 1440x900px) Image search: [Google]

2MB, 1440x900px

>>60338634
>With Volta being on the horizon, is Vega pretty much fucked?
Holy shit did you actually miss the Nvidia conference yesterday? Yet you're posting this anyway?
How dumb are you, OP? How can you waste everyone's time posting such misinformed, ignorant drivel. You clearly don't follow the industry enough to question whether Vega is fucked or not.

They revealed that Volta was a complete lie. The performance target for Volta was a lie.
All it is is a refined 16nm process (that they're calling 12nm, even though it's not), and more CUDA cores on the enterprise 15TFLOPs card - which won't be here for another year.
There is no new architecture coming. Nothing.
"Volta" was really Maxwell+Pascal+Volta. Step by step increases. There is no big leap coming. The biggest leap Nvidia will have in the past 5 years and the next 5 years was Pascal.

>>60340837
Wasn't the conference basically about all the extra botnet they're implementing?

>>60338634
>Volta
>on the horizon
It's more than a year away and it's primarily designed for deep learning, not games.

>>60338634
>on the horizon
IF by on the horizon you mean at least 9 months away sure

>>60340372
Normal people don't want to deal with amd shitty drivers too.

>>60340856
>Wasn't the conference basically about all the extra botnet they're implementing?
Mostly. But they also did a paper launch of GV100.
The last time they did a paper launch, the GP100, it took a year to arrive.

If GV100 compared to GP100 is any indication, we're not looking at anything remotely close to the performance improvements from Keplar>Maxwell>Pascal. Likely less than 25%.
They might do better at 1440p and 4K where GCN has crushed them for the money for along time, with a higher CUDA to ROPs ratio and having faster GDDR6 (and maybe HBM2 on the Titan Xv and 1180Ti or whatever they call them).

But still seems like Q2 2018 for Volta just as I fucking expected 6 months ago and haven't changed my stance on since.
Q4 2018 for Navi. AMD would be smart to refresh Vega in Q2 2018.

but vega is going to beat the 1080 ti with an easy 1600 clock

>>60340941
At 1440p and 4K, yeah. But in some Nvidia optimized games, I think it'll still lose at 1080p.
(though anyone getting a card as expensive as either or 1080p is mind-numbingly retarded)

>>60340922
2012 wants its memes back.

>>60340922
>he unironically deals with NVIDIA's garbage drivers
You enjoying that botnet experience with 200 processes and a control panel that is still designed for Windows XP that bricks GPUs every second update?

>volta 800mm2 12nm
>only 30% faster than 600mm2 16nm chip

it's a dud, maxwell was great no question, but pascal and volta are boring

just 15 TFLOPS outside DEEP learning that you have to rewrite everything for and buy the card for $15 000
it's fucking nothing

Huan was selling them on stage, begging for pre order in q4 2018 for teslas, what the fuck
meaning consumer cards will be 2018 q2

>>60340980
That "12nm" is refined 16nm FF+. Just google feature sizes.

>>60338634
>Volta is a node refinement of Pascal
>Pascal is a nodeshrink of Maxwell
>Maxwell is a refresh of Kepler
>Kepler is a cut down Fermi
>Fermi is a Tesla with hardware scheduling
>NVIDIA is STILL reusing the 8800
Jesus fucking christ. They're just as bad as Intel regurgitating the Core architecture every year.

>>60340989
that is even worse, it means they didn't do shit to architecture outside "tensor" which use is yet to be seen

>>60341002
they give real performance gains though

>>60341018
But all they do is increase the core count

File: 1367330432501.jpg (325KB, 852x939px) Image search: [Google]

325KB, 852x939px

>>60338634
volta is literally just moar coars + specialized hardware for google's deep learning algorithm
it's an 815mm^2 house fire with yields about as small as your dick and price tag that makes intel's products look cheap.
not that vega will be much better.
vega just implements rendering and fp16 shit that nvidia has had since maxwell, so it's gonna be on par with volta on that tech.
well they're also adding fp8 ops which might be useful for something
Supposedly should also have higher IPC, but that's probably only in some specific workloads. If I had to guess it's also probably more related to machine learning.

>>60341023
Oh and clockspeed, can't forget that!

Let's not forget that without ATI/AMD we'd all still be using DDR2 on our GPUs

>>60341023
well... intel doesn't.

>>60341026
GCN was competing with Maxwell without that breakthrough, can't imagine Vega being bad.
But hype is high, it's going to be retarded arguments "it's can't beat 1080ti even though it's withing 7%" like with ryzen nobody would notice the price tag $150-200 less than 1080ti

>>60341026
>specialized hardware for google's deep learning algorithm
Google uses ASICs for it. Irony, huh.

>>60341072
To be fair the Fury X had the same RRP as the 980 Ti and I don't see AMD going any lower than that, so at best Vega will undercut the 1080 Ti by only $50. Need to consider the cost of HBM2 and the larger die.

press conference on 31st.
16 cores incoming(rumors of ES started floating around), and vega launch

>>60341088
Vega's HBM2 setup is probably cheaper than 11gigs of GDDR5x with traces. But the die is bigger.

>>60341082
Their entire marketing gimmick is that you have the flexibility of a GPU with the Tesla that you won't get with an ASIC.

So they basically crammed an ASIC in the silicon and cranked up the die size to accomodate.

>>60341100
Not when you factor in the scarcity of HBM2 currently.

>>60341102
I want to see those yield numbers.

>>60340954
Their ogl driver is still single threaded and broken.

>>60341106
is it? samsung making HBM too now it's not one player anymore

>>60341102
Google's TensorFlow ASIC is as big as HDD so that 'flexibility' is a meme.
>>60341106
Wut? If you mean Samsung HBM2, then yes, but Hynix can produce fuckton of their HBM2.
>>60341113
>ogl
>in 2015+2
Why? Use Vulkan nigger, even fucking Khronos Group says that.

>>60341110
Whoa, 1.7% is back, baby. Rev up those memes.

>>60341072
sure but we have so little specifics about vega, it's hard to say where it will land overall.
A lot will depend on final clock speeds too.

>>60341136
>sure but we have so little specifics about vega
They literally said everything about arch features.

>>60341121
>vulkan
Vulkan can't replace ogl. I emulate and amd is no buy for me until they fix their shit, but looking in their last 15 years they won't be getting my money soon.

>>60341155
Vulkan can and will kill OGL as soon as you replace H1B pajeets with proper fucking programmers.

>>60341136
we actually know a lot, it's just hard to transfer into fps numbers for now

but look, if simple furyx at 1Ghz can be on par with 1070 just bumping the clocks might do it

>>60341155
Nobody cares about you stealling consoles games.
>>>/v/

>>60338634
it was fucked by default. nobody buys high end amd gpu. just look at how much the fury series crashed and burned. also it's out way too late. many people, including me, bought 1070's months ago. iirc the 1070 is in the top 5 most used gpu on steam.

File: 1488739966565.jpg (94KB, 1024x1019px) Image search: [Google]

94KB, 1024x1019px

>>60341187
>steal
Kill yourself
>>>/plebbit/

>>60338634
>800mm^2 chip
>12 nm

We wont see that for at least a year.

>>60341153
not really
they gave us a high level view about it, but not how it translates to real world performance and what kind of optimizations (if any) are needed

>>60341172
>if simple furyx at 1Ghz can be on par with 1070 just bumping the clocks might do it
well I imagine furyx would be better than 1080 if it had as many rasterizers as vega (11 triangles per clock vs 4 on fiji) and more schedulers since those seem to be the main bottleneck.

>>60341309
>what kind of optimizations (if any) are needed
They said something about gamedevs making use of primitive shaders directly instead of driver wrapper translating vertex shader into primive one. That's it. We'll se on Computex.

It's "on the horizon" just like Navi is, Hynix already said volume production of GDDR6 will be in early 2018, that's where Volta for consumers sits.

You might get another 750ti a few months earlier but nobody cares about $120 crap

>>60341457
Volta is simply bigger Pascal. It's now a new arch. Also when novideo will start using HBM2 in consumer cards?

>>60341469
Once it becomes cheap enough for them to slap massive margins on it.

Right now an HBM2 Titan X would probably retail for $2000 to feed their greed

>>60341482
I dont think 11 gigs of GDDR5x is much cheaper than 2 stacks of HBM2 and interposer.

>>60341494
Consider the yields, if a chip is bust all they can do is trash the HBM2 along with it.

Hence why the Radeon Pro Duos come with a full, dead chip in the box (including HBM dies)

>>60341469
It is a new arch, but not a new arch where we care, like increasing IPC, clocks and general performance.

File: 1310429820535.jpg (13KB, 300x300px) Image search: [Google]

13KB, 300x300px

>>60341026
The fact that Linus is on that picture shows how retarded AMD fanboys are.

That guy has alot more AMD sponsored video's on his channel than Nvidia ones

>>60341509
Pro Duo's are 8 stacks with two interposers. Two stacks on one (also smaller) interposer are much cheaper. Basically novideo are greedy laze kikes to adopt something new.
>>60341521
It's as 'new' as Kepler Refresh.

>>60341528
He shilled Intel and NVIDIA for free when that image was made

>>60341532
You mean he reviewed them, and now he's actualy shilling video's for AMD (getting paid for video's of their products)

>>60341521
I love how Intel and NVIDIA have managed to change the meaning of what an architecture is like this

>>60341528
in case you couldn't tell from the gtx 480, that's a really old image

>>60341551
>disclaiming a sponsorship is now shilling
Kid..

>>60341552
Yeah it's not like AMD has marketed huge architectural improvements for their last 3 gpu generations, while they only turned out to be very small improvements on the same GCN architecture

>>60341110
1.7

>>60341565
You need to remember that AMD has much less R&D money than NVIDIA and especially Intel.

>>60341562
>it's fine to call him an Nvidia shill based on nothing, but calling him a shill on actual sponsored video's by a company is bad!

>>60341565
Architectural improvements are accurate, they aren't claiming it to be a new architecture.

Why do you think Polaris is still called GCN? Why do you think Terascale 2 was a thing? Because they aren't trying to pretend they are new architectures like their competitors do.

>>60341574
>was literally trying to hide the 'sponsored by Intel' disclaimer on his older videos before he got called out on it

>>60341575
>Because they aren't trying to pretend they are new architectures like their competitors do.

Really kid, are you this naive?

Look at AMD's marketing slides compared to Nvidia's. They are just as bad when it comes to marketing. AMD even compared CF cards to single ones, and showed performance gains compared to cards of 3 generations older.

They are both share holder driven companies that value profit over everything else, if you think AMD is some sort of good guy charity company you are mistaken

>>60341594
>AMD even compared CF cards to single ones
Oh yeah because a $400 CF config is unfair to compare to a $600 single GPU.

>>60341594
they will defend this

>>60341606
>CF config is unfair to compare to a $600 single GPU
It is, because CF configs come with a whole whost of problems, and don't work well in many titles

>>60341594
>and showed performance gains compared to cards of 3 generations older.
[citation needed]

>>60341521
But it increased performance by 40% while using the same power(assuming Nvidia isn't fucking with TDP), that's worthy calling a new arch, but every generation regardless of being a new arch or not does exactly that and we can't call every generation of GPUs a new arch.

>>60341594
>if you think AMD is some sort of good guy charity company you are mistaken
Compared to our favourite 3.5 green jews and TIM blue jews they are truly the charity of silicon industry.

>>60341616
Worry not, CF isn't quite as garbage as SLI

>>60341625
Yeah..no

>>60341620
>But it increased performance by 40% while using the same power
ASICs are low power, and the performance increase is in THEORETICAL compute performance.

>>60341620
12nm FFN apparently has 20% power savings so I'm far less inclined calling it a new arch.

>>60341620
It's a bigger fucking die and perfomance increase corresponds to die size increase. And no, GV100 is 300w TDP.

>>60341636
Except it's true, SLI is still as garbage as it was in 2005 and Crossfire has always been superior.

>>60341647
Yeah that doesn't sound like a unsubstantiated biased fanboy claim at all

>>60341653
>SLI still can't work over PCI lanes
>SLI has inferior scaling

>>60341661
So substantiate your claims with evidence

Lets see those charts

Is anyone impressed how AMD has managed to reign in leaks of Vega when they've already started printing retail boxes?

Besides the leaks of the 6-7 month old engineering samples, there's been NOTHING about it.
Leaks of Polaris and Zen have been numerous.

File: Nvidia-AMD-Multi-GPU-Benchmarks[1].jpg (324KB, 1657x526px) Image search: [Google]

324KB, 1657x526px

>>60341665
Can you just stop posting already you tech illiterate newfag NVIDIA fanboy

>>60341672
King Pajeet and his Kali cultists are killing everyone trying to leak info about Vega.

>>60341672
We have leaks of the :C3 ES which is from February, that's newer than the :C1 initial leaks we've saw in timyspy/firestrike which I'm assuming are the oldest of chips

File: image022[1].png (17KB, 450x448px) Image search: [Google]

17KB, 450x448px

>>60341677
Let's go old school

>>60341686
Or it's more likely anyone possible who can leak Vega has been promised a bonus if no leaks come out.
Hence, no leaks.

>>60341677
Nice, now with GPU's that aren't 2 years old and with an actual link to the article, and a wide range of games tested and an average used as a result as to avoid cherry picking

File: Untitled.png (66KB, 829x730px) Image search: [Google]

66KB, 829x730px

>>60340922
Even when this was a form of criticism, you didn't have to deal with them, see also: ATI Tray Tools for the oldfags

>>60341002
>>60341018
They give real performance gains because what he said isn't true.
Maxwell and Pascal were big architecture improvements that increased IPC and efficiency.
Volta, however, is simply a process refinement and more cores for the GV100.
They can't increase cores much on the consumer cards because that costs too much.

>>60340837

File: image028[1].png (17KB, 450x593px) Image search: [Google]

17KB, 450x593px

>>60341701
Witness the NVIDIOT as he stumbles to find a counterargument

Oh the humanity

>>60341717
>ask for cards that aren't 2 years old, so we have up to date results
>he shows me cards that are 5 years old
>ask for link
>no link
>ask for an average across a wide variety of games as to avoid cherry picking
>he gives me one cherry picked benchmark.

Yup it's a fanboy

>>60341734
Still waiting for a valid counterargument, why don't you find a single piece of evidence showing SLI having superior scaling?

Oh wait, you can't even find a modern example because nobody uses SLI because it's common knowledge that it's 100% trash

>>60341715
>Pascal
>IPC

Ehh... efficiency and clocks, yeah, but that's largely due to 2 node jump to finfets

>>60341745
>why don't you find a single piece of evidence showing SLI having superior scaling?

Because I never made that claim, I said both CF and SLI have a host of problems and don't work in many games.

You made the claim that CF is superior to SLI so you should provide evidence.

So far you haven't, only old as fuck cherry picked results, exactly what I would expect from a fanboy

>>60341715
>Maxwell and Pascal were big architecture improvements that increased IPC and efficiency

Dude, pascal is maxwell on a 16nm with more cores and higher clocks. No ipc gains at all.

File: Multi-GPU-Scaling-Crossfire-SLI-Fury-X-Titan-X[1].jpg (32KB, 585x449px) Image search: [Google]

32KB, 585x449px

>>60341761
How much evidence do you need?

Are you just frustrated because there is not a single piece of evidence on the internet to suggest SLI has superior scaling?

>>60338640
You could say... V is dead

File: av_av[1].png (36KB, 1696x1183px) Image search: [Google]

36KB, 1696x1183px

>NVIDIA shills
>2005+12
Not even once

File: mankind1080-avg[1].png (51KB, 1826x962px) Image search: [Google]

51KB, 1826x962px

>>60341784
If only NVIDIA cards had DX12 support

>>60341772
You keep posting old as fuck shit, cropped out to the point where people can't even make out what the graph is saying.

>Are you just frustrated because there is not a single piece of evidence on the internet to suggest SLI has superior scaling?

No, because I already said both SLI and CF are turds for plebs, which one of the two is a bigger turd is not interesting at all, a turd is till a turd

>>60341801
>old as fuck shit
Just because nobody tests SLI (which hasn't changed in a decade) on your glorified 8800 GTs (see >>60341002 and enjoy your countless shitty refreshes) doesn't make the results you can't accept 'old as fuck shit'

>>60341812
>glorified 8800 GTs

You mean those glorified 8800's which produce vastly superior performance at the same power draw? (1070 vs 480)

>>60341755
>Ehh... efficiency and clocks, yeah, but that's largely due to 2 node jump to finfets
no it's not.
Maxwell introduced tired rasterization binning that greatly increased efficiency. That's why its power consumption was so low.

>>60341826
>at the same power draw?
Yeah NVIDIA tried hardware scheduling once, Fermi was the result. They aren't capable of a truly efficient GPU like AMD.

>>60341784
wtf I'm getting 3 furies now

>>60341838
> They aren't capable of a truly efficient GPU like AMD.

Considering the 1070 is vastly more powerful than a 480 at the same powerdraw, your statement just shows what a blind fanboy you are

>>60341849
Sorry what's that, I can't hear you over that lack of DX12 support you have because NVIDIA decided to shift all the work to software and DX12 doesn't work like that.

>>60341868
ok kid

>>60341849
>dumb gaymen babby posting on /g/ about technology he doesn't understand
Miss me with that Reddit tier shit

>>60341888
>ignoring reality
>b-but m-muh card is more energy efficient on p-paper!
KEK

>>60341910
The reality is you may as well still be running Windows XP because your outdated GPUs don't even support modern APIs

File: 1.jpg (78KB, 685x505px) Image search: [Google]

78KB, 685x505px

>>60341921
Cool, now lets take a look at reality

>>60341849
the 480 has comparable powerdraw to a 1080, not a 1070. the 580 draws as much power as a 1080 ti. amd are a literal joke.

>>60341946
Well, it's in between the 1070 and the 1080
>>60341938

When will novideo implement hardware scheduling again? I want another Fermi.

>>60341956
When they learn how to design an efficient GPU

>>60341938
This is fucking embarassing.

AMD wtf

>tfw undervolted Fury Nano is the most power efficient GPU on the market

>>60341973
Maxwell is pretty efficient.

>>60341994
>software scheduling
Yeah no shit it's efficient it's literally making the CPU do the work for it

>>60342001
I mean even if you add hardware scheduling it'll be pretty efficient.

>>60342014
Uh no, it won't. Fermi is what happens when NVIDIA have hardware scheduling

>>60342001
Then why is total system power consumption lower on a 1070 than on a 480?

Anyone else making screenshots for more spicy oc of delusional AMD drones about to get burned again just like they did with Poolaris, Failzen and Rebrandeon?

>>60342022
Fermi was mainly a TSMC fuckup. 500 series was decent.

>>60342025
Because CPUs don't draw that much power?

>>60342039
So shifting the work to the cpu is efficient then

>>60339204
Meh, my 1070 is the first GPU I've ever had that I'm satisfied with. Runs everything 2k 60fps ultra. If games progress further theres like 4 cards already above mine to pick from.

>>60342026
Rebrandia stop using G80 and make something new pls.
>>60342042
No, because it adds latency.

>>60342038
>500 series was decent.
LOL are you serious?

>>60342050
>No, because it adds latency.
Alright, let's see some evidence for that

>>60342060
Anything DX12/Vulkan.

File: 01-Power-Consumption-Voltage-Steps[1].png (21KB, 600x471px) Image search: [Google]

21KB, 600x471px

Maybe if AMD stopped recommending overkill voltages stock then their power draw numbers would be better on review sites.

>>60342061
Okay, provide some evidence

>>60342069
Use google, my dear redditor.

>>60340964
What are you even talking about, I didn't even have to install drivers. Windows did it for me.

File: sin2.jpg (193KB, 1280x720px) Image search: [Google]

193KB, 1280x720px

Why are ayymdpoojets constantly butthurt and defensive?

>>60342071
>use google
that's not evidence

Opinion discarded

>>60342084
VLC user detected, sasuga novideo

>>60342062
This is nothing more than a budget problem, AMD can release GPUs that don't need retarded voltages like the Nano, but that's costly.

>>60342094
Its madvr dumb ayyymdshit
also
>what is jpg
kys poojet

>>60341594
>showed performance gains compared to cards of 3 generations older.
considering how much they rebrand their shit, this shouldn't come as a surprise

>>60342120
>madvr
>2017

No wonder, Nvidiots love old technology.

>>60342120
it's porn, please DELID this

>>60342120
>madvr
Thanks for confirming you're completely clueless when it comes to video playback.

File: 4.jpg (157KB, 680x823px) Image search: [Google]

157KB, 680x823px

>>60342071
>>60342061
>anything DX12
>AMD has superior latency
Yeah...no

>>60341800
Oh you mean the thing that AMD has been telling you that you need for the last 3 years?

>>60342149
Oh look a wrapper.

>>60338634
Not really, Volta doesn't improve IPC at all over Maxwell even...

AMD is promising actual IPC improvements with Vega too (who knows if they'll pan out)

>>60342150
>NVIDIA have had 3 years to support DX12 and they still don't

>>60342131
>retarded memepv poojet
>ayyymd poojet
like pottery

>>60342158
Oh look a fanboy in denial

Oh look that retard derailed the thread... again.

>>60342164
>shitposting like a typical redditor
No surprises there

>>60342162
Actual performance is more important than IPC.

If you can clock your cards 500 mhz higher against the same powerdraw, which is what effectively happened with Maxwell and Pascal, that will net you a much bigger improvement than 5-10% IPC gains

>>60338903
Except it's all rampant speculation.
128ROPs is a big ask on a small power budget.

>>60342163
Am I missing out on something?

>>60342185
Modern technology

>>60342177
But what if you get huge IPC gains AND higher clocks?
You're telling me a Fury X with 20%+ IPC and 1600MHz wouldn't be impressive?

>>60342199
Well ofcourse it would

We don't have that though

>>60342199
It would have been much more achievable if Vega were on TSMC's 16nm which is designed for higher clocks.

Samsung's 14nm is designed for low power which is why Ryzen is so incredible at 2.1-3.3GHz as far as efficiency goes

Amdtards areabsolutely flummoxed and resort to shitposting and butthurt.

>>60342231
Bibeocardz are still far away from hitting LPP limits. And Vega has longer bibeline which means higher clocks on average.

>>60342210
How do you know Vega won't be that? It's the same core count as Fury X but has 50% more transistors, that clearly indicates a IPC improvement, 10, 20%, 30%? How much though?

And we already know it'll clock over 1520MHz

>>60342194
Man you gotta buy this doohickee because it has this new feature that software makers arent using yet but they might one day.
I'll buy it when its an actual thing.

>>60342244
>longer bibeline
Lest we forget the last time Intel tried that

>>60342246
>1520MHz
You people are really conservative on Vega clocks, you do know that the Mi25 you're using for reference clocks at 1530MHz at BASE clocks, not boost clocks?
From AMD's linux Vega commits we know Vega will have a new boost version, can it get to 1700MHz? Who knows.

>>60342246
>How do you know Vega won't be that?
I don't and neither do you

>>60342266
I mean reasonably longer bibeline. Like, Northwood was sorta decent, but Presshot. Oh well.

>>60342266
Longer pipeline is fine when you can get a notable increase in clockspeed, lets say from 16 to 18 stages.

Now when you're going from 16 to 29 stages.

Oh look he vanished as soon as everyone stopped giving him replies.

>>60342302
Because people starting talking about technical aspects and the NVIDIOT brain isn't capable of understanding such things

>>60342286
Also it's a common misconception, but theoretically longer pipelines can achieve more IPC than shorter ones, because they have more stages to optimize what's going through them.

Theoretically though, no one has managed to bridge the latency gap longer pipelines need yet.

>>60342271
Mi25 has a TDP of 300W. Consumer cards are more likely to have 250W TDP.

>>60342347
MI25 is rated <300W TDP. It could be everything from 225W to 300W.

>>60342347
<300W, that's with enabled 2:1 FP16 and other datacenter features.

But I really want there to be a 300W SKU for consumer cards, because more performance is always nice even at the cost of 50W

>>60342362
>that's with enabled 2:1 FP16
Haven't consumer AMD cards had this for a long time?

>>60342372
Polaris is 1:1 i think.

>>60342372
No.
It's half rate

>>60342362
>300W SKU
Probably the watercooled version at nutty clocks.

>inb4 get the aircooled one and OC it to the same level
WHAT DO YOU THINK I AM POOR

File: 4563548343253.png (34KB, 653x726px) Image search: [Google]

34KB, 653x726px

Volta sure got the Pajeets angry.

>>60342421
>Volta
aka Pascal

>>60342421
>bigger 8800 makes someone angry
Why? It's fun.

>>60342429
Architecturally, Volta is a bigger chance than Pascal was compared to Maxwell.

They're still Maxwell 1.5 and Maxwell 2.0 though

>>60338634

Vega has always been irrelevant. AMD doesn't provide proper competition to nvidia. It's a literal "products for poorfags" company with shoddy marketing and that's it.

I bet whatever you want that this heap of trash isn't even gonna touch the 1080, let alone the 1080ti and titan. AMD pajeets always keep promising outstanding things and NEVER fucking deliver. Anyone remember how the 480 was supposed to be a godly card? The thing is a glorified 390 and gets smoked even by a fucking 970 when compared OC vs OC.

Anyone expecting anything out of this company that does nothing but provide blunder after blunder is delusional.

>>60342441
You won't see any of the changes in consumer models.

>>60342445
I can expect 30% more perf/watt compared to Pascal, but that's the baseline increase every new generation brings.

Maxwell was a real fucking beast, 50%+ higher perf/watt on the same fucking node.

>>60342441
>They're still Maxwell 1.5 and Maxwell 2.0 though
And this is a bad thing how?

>>60342084
This is a blue board

>>60342455
>I can expect 30% more perf/watt compared to Pascal
Ha ha no it's not even a dieshrink.

>>60342465
It is, Volta is on 12nm while Pascal is on 16nm

>>60342473
'12nm' is slightly tweaked 16nm FF+, just check feature sizes.

>>60338923
NVIDIA JustWait™ technology

>>60342483
Doesn't matter, it will produce performance/watt increases

>>60342488
Maybe some 5%.

>>60342456
Because some people on technology forums have more interest in IC design than just add moar corz.

Maxwell = interesting arch
Vega = interesting arch
Evergreen = interesting arch
G80 = interesting arch

Even Fermi as an arch was interesting, but had too many issues with some not directly being arch issues.
So seeing the same shit being slightly changed again and again is not interesting.
We, as in the non gaymur retards, have a different way we look at GPUs and CPUs, usually not through the glasses of gaming.
Personally I'm still using Intel integrated GPU, I won't be upgrading to Vega, Volta or Navi because the most gaming I play is 10+ year old games.

>>60342487
Nvidia?

Pascal has been on the market for almost a year, while we are still waiting for Vega AMD's answer to it

>>60342496
Yeah...no
>We, as in the non gaymur retards
Ah you mean autists okay

>>60342512
Back the the ribbit with you.

>>60342512
Back to /v/ and leddit underage.

>>60341844

>>60342520
>>60342521
Epic memes lads

>>60342512
>someone far eclipses my general youtube-tier IC knowledge
>a-autist

Cute, don't let the door hit your inflamed ass on the way out.

>>60342499
>G80 = interesting arch
Even then ATi was first to unified shaders.

>>60342535
Don't you have some incredibly mature production work to do?

>>60342541
Xenos was a weird thing, but consoles are consoles and consumer GPUs are not consoles.

>>60342553
Still counts. Also first to GDDR3/5.

>>60342555
It counts but it's not practically important.

>>60342535
>eclipses youtube knowledge
You literally just stated your opinion that you care more about the architecture than about actual performance.

Well, nice opinion

>>60342575
>you're not allowed to have interest in IC on technology board
Maybe you should go back to lebbit?

>>60342597
>you're not allowed to have interest in IC
I never said that, you were the one who started namecalling like a little kid because someone else was interested in performance

>>60342608
You said EXACTLY that.

>>60342575
Architectures usually correlate with performance, but there are more ways of getting performance than new architectures, but as mention, they're BORING, dieshrinks are not fun, they've been fundamentally the same for over 20 years, ignoring stuff like HKMG, EUV/DUV, SOI which have their own interesting properties, it's been largely the same.

So yes, it's a nice opinion since I care more about the chip itself than its performance, if it's a good chip it's also gonna have good performance.

Also your interest in only the end effect of the chip design screams like you don't belong on the technology board but on the video game board.
Get back there.

>>60342623
Eeuh no?

Anons on cyka blyat board compare their set ups:
7700k+GTX1070 gives higher fps than [email protected]+GTX1080.
Ryzen is simply not enough to utilize full potential of 1080 - it downgrades it to 1070 level.
https://www.youtube.com/watch?v=lY9xghPy-uQ&feature=youtu.be [Embed]

Vega will be same shit, losing to 1080 and anons here will cry that it is not for gaiming but for those 2 muh graphics professionals.

>>60342652
Oх блять, шкoлoпacкaль тpeд и cюдa пpиeхaл. Шизики, вы пoнимaeтe, чтo нa paзных ceтaпaх в игpe бeз cтpoeннoгo бeнчa пипиpкaми мepятcя бecпoлeзнo?

>>60342634
>Also your interest in only the end effect
Your ability to read isn't very good.

I am interested in the architecture, but I am more interested in the end result, like any sane person would be.

In the end I don't buy a gpu to put it on a shelf and jerk of how good the architecture is on paper when it's crap in reality.

>>60341623

It's always funny when people point out AMD is a profit driven corporation as if that's some big reveal.

AMD is the best thing for the consumer to happen to the silicon industry.

>>60342668
Why would I need to buy the chip? I already mentioned I'm still using Intel integrated. Your /v/ is showing.

>>60342693
>why would I need to buy the chip
Again, learn to read, I didn't say you should.

>>60342706
>In the end I don't buy a gpu to put it on a shelf and jerk of how good the architecture is on paper when it's crap in reality.

You were also rudely implying that I'd buy any kind of shit architecture, I mentioned Maxwell, G80, Evergreen, which all were revolutionary designs.

In conclusion I think you need to reign in your /v/ mentality.

>>60342735
>I mentioned Maxwell, G80, Evergreen, which all were revolutionary designs.

Well, they were, and I never said I'd disagree

Where we disagree is where you would say Pascal is not interesting, while I would say it is since you're getting much higher performance at a much lower power draw.

Also as much as you want to shout from your mature high horse with your IGPU, we are talking about technology here that is mostly bought for the purpouse of consuming games.

You trying to tell me to go to /v/ because I take gaming performance into consideration is funny and it shows your disconnection to reality.

>>60342763
Pascal is boring fucking dieshrink of Maxwell. Also get back to /v/.

>>60342773
Maybe you should stop getting upset because people like things you don't, you sounds aspergers as fuck

>>60342791
Just get fucking back to /v/.

>>60342798
>reeeeeeeeeeeeeeeeeeee!

File: 1478812587698.png (101KB, 192x209px) Image search: [Google]

101KB, 192x209px

>>60342808

File: 1270066026663.jpg (9KB, 226x166px) Image search: [Google]

9KB, 226x166px

>>60342812

>>60342763
From a performance standpoint the Pascal cards are interesting, much higher performance than their predecessors, but 80% of that is from the dieshrink, Pascal itself as design is not interesting.

Mostly for the purpose of consuming games? You should take a look outside of the gaming market and see what kind of figures those markets make in revenue yearly, also who's the main buyer is completely irrelevant to me, that's Nvidia's problem since I do not have any stock invested in any technological company, why would I care who buys it?

You seem to take ONLY gaming performance into consideration, which smells of /v/, and /v/ smells bad.

File: 1365336557194.jpg (54KB, 640x479px) Image search: [Google]

54KB, 640x479px

never trust amd to deliver

>>60342825
Ok, kid

>>60342835
>R300
>R700
>Evergreen
>Tahiti
>Hawaii
What.

File: rusemaster Raja.jpg (15KB, 544x615px) Image search: [Google]

15KB, 544x615px

>mfw Nvidia can never beat Vega if we never release it

>>60342840
I don't want to hear that from someone who's only got gaming on their mind.

You should go back to /v/.

File: 1310483412100.jpg (36KB, 413x395px) Image search: [Google]

36KB, 413x395px

>>60342848

File: Fuhrer.png (20KB, 174x162px) Image search: [Google]

20KB, 174x162px

>>60342848
Genius

>>60342868
>You should go back to /v/.
You should go back to /aspergers/

>>60342868
>lets discuss the merit of gaming cards for buttcoin mining
ok kid

>>60342916
>>60342884
Calm down child, there's plenty of games out to relieve your stress.

My biggest issue with Novideo is Gsync. When I can buy a 4K Freesync 2 monitor with HDR I'm soooo gonna go out and buy a Vega+ GPU.

>>60342916
#rekt

>games

You need to be over 18 to post on this site.

>>60342944
What? You don't like our magnificent proprietary feature? Buy goysync, goyim.

>>60342931
I relieve my stress with ur mum

>>60342949
Look at this guy being all grown up and shit

What an example for the rest of the world

>>60342962
It's in the rules, you need to be over 18 to post on this site.

>>60342994
Yes and?

>>60342998
It's pretty self explanatory.
If you have problems comprehending I believe being underage is the least of your problems.

>>60343010
>resorting to namecalling in an attempt to make the other party look immature
Oh the irony

>>60343014
I don't have to make an effort at making you look that way, your gaming tunnelvision does that for me.

>>60343028
Ok kid

>>60343042
Once you grow to 30 you'll be missing the fun times you had as a kid, cherish those years.

>>60343089
Or, you could continue to have fun times

>>60343097
It's a different version of fun now, a good amount of time now is spent on making money, as a kid you had no such worry.

>>60343122
I actualy have a full time job and continue to have fun on the side of it.

You sound like you lost your ability to have fun when you grew older, I'm sorry that happened to you

>>60343122
>shitposting on an anonymous image boards for hours on end
>yeah I'm so busy making money I have no time to game!

>>60343137
What are you talking about? I already told you my fun switched from gaming to something else, and it's a fact that with a full time job you don't have enough time you had as a kid for enjoying yourself, now if your hobby was only gaming then fine, but what about DIY and books? Way less time spent on gaming, I still play games, but scarcely.

>>60343201
I have a full time job, play games, read books and do plenty of other stuff.

Just because you're interests might have shifted doesn't mean other people are immature, which is what you seem to think.

>>60343245
Your singular focus on gaming up in the thread implies quite the opposite, now this is the internet and everyone can bullshit as much as they want, so lets end this here.

>>60343252
>singular
It isn't singular, I already said I'm also interested in the technology behind it, but ultimately more in the performance itself.

Since you don't actualy buy the products I can understand why it might be opposite for you

>>60338634
Volta won't be out for consumers until sometime in 2018. Vega still has some time for some sales and profits if the performance/cost ratio is good enough.

>>60343278
Maybe I'd buy if there's something actually worth playing, for fucks sakes I'm still replaying Baldur's Gate, Gothic and HoMM3 for decades.

>>60342084
your settings suck. may as well run movies and tv w10 shit

>>>60343252
Witcher 3 is actualy an rpg that is worth buying

>>60343323
>>60343338

I've got a quick question

How is a HBM2 stack of 8gb in total going to compare to the 11gb of a 1080ti?

Better? Worse? Is the 3gb difference in VRAM going to be an issue in games? Theoretically speaking, since afaik there are no games that use 11gb of VRAM.

>>60343338
I played it, liked W1 more.
It's a okay game, I guess.
The last RPG worth playing was Divinity: Original Sin, and that's isometric.

>>60343361
It's going to have more bandwidth but less memory in total, which might be a problem in the future

>>60343361
There's no difference, but it's a fact that HBM2 has a few things very different from GDDR that directly benefit AMD's HBCC that GDDR does not.

>>60343362
I'm also having alot of fun with planet coaster, but that's a completely different style of game.

>>60343362

Same, for some reason Witcher 1 had me engaged from start to end, the other two I'd play for a while then come back to now and then, still haven't finished 3...

>>60343323
>Baldur's Gate, Gothic and HoMM3
Slavcore

>>60343361
no difference, they designed special cache to offset not enough VRAM problem

they ran MD on 2Gb VRAM with and without that cache, it ran 50% better with cache on avg and 100% better on minimums

>>60338903

again it depends on price/performance, if it gets crushed by 10% but is 40% cheaper, then its still worth it.

unless AMD goes full "overclockers dream" on us again and price it out of range.

>>60343366
So it might do more with less, so to speak? That's what I got from that anyway.

>>60343384
>>60343419
What is this special cache? Got an article I can read somewhere? I just don't want to buy a card only for games to hit a VRAM limit a year later and make my card into 970 2.0

Although the 970 had a different type of problem I guess, so maybe it's not the same thing.

>>60343407
Damn fucking right.

Those are just my favorite games though, Planescape, Arcanum, Morrowind, Fallout 2, FFT, Ar Tonelico, Grandia, Xenogears , etc etc are still close to my heart.

File: jigglypuff laughing.png (53KB, 398x399px) Image search: [Google]

53KB, 398x399px

>>60340922
>amd shitty drivers


>He doesn't know about shit's been reversed.

>>60343453
not the same thing, HBM is a different beast

They basically made additional IMC for HBM to leverage it's giant bandwidth

>>60343459
Do you play anything remotely modern or just old games?

File: mpv-shot0037.jpg (97KB, 1280x720px) Image search: [Google]

97KB, 1280x720px

>>60343758
Bandwidth is irrelevant in this discussion, HBM2 2 stacks don't have more bandwidth that GDDR5, but that's not AMD's intent, their intent is to leverage HBM2's unique characteristics for HBCC.

https://youtu.be/m5EFbIhslKU?t=1132
Video explains it better, at 18:50

>>60343768
Currently playing Nier Automata since I'm a large fan of Drakengard/Nier and before that I mentioned Divinity Originial Sin.

So for your question, I do play modern games but not in any notable quantities.
There's probably 1-2 modern games a year I want to play

>>60343471
na amd drivers still blow

i cant upgrade regularly because they ruin freesync so frequently you have to stay with a working builld

>>60343857
It seems like we're looking at a future where midrange GPUs have a single HBM3 8GB stack

>>60344100
That shouldn't be any more expensive than 6+ highly clocked GDDR6 modules

>>60340922
It's the opposite now. Nvidia got so cocky being in the lead for so long that their driver side has been slipping in quality.

>>60342442
desperate autists build up the hype and expectations far more than amd's marketing morons could ever do

>>60342455
>Maxwell was a real fucking beast, 50%+ higher perf/watt on the same fucking node.
They ditched most of its FP64 hardware, among other things

>>60343459
AT3 was shit

>>60344285
It was tiling that made it's so efficient.

>>60344411
Just like Gothic 3, Dungeon Siege 3, Sacred 3, Diablo 3.

Why does everything go shit with 3

File: nvidia telemetry and how to disable it.jpg (141KB, 975x499px) Image search: [Google]

141KB, 975x499px

>>60340922

BOTNET
O
T
N
E
T

>>60340837
We know they always lie, but who lied more?

>>60344475
And removing most of the FP64 stuff gave them a healthy transistor budget in which to implement it (and more FP32 pipelines or whatever else they wanted)

>>60342487
Seriusly as an amd and nvidia user, you are fucking retarded.

>>60342456
Maxwell to Maxwell 1.5, to Maxwell 2.0 is a slippery slope to Intel Core i. Series.

File: 1486848060711.jpg (22KB, 351x351px) Image search: [Google]

22KB, 351x351px

>>60338737
I don't want it to be real but it's the most likely outcome

>>60338737
JUST! Wait

>>60341672
There are so few leaks because nothing is leak worthy. Meaning it will be absolutely mediocre. So 10% over 580 at best.

>>60347625
What an idiot.