/w4v/ - Daily Wait for Vega Thread - On Tour In Budapest TODAY

Who's in for Budapest?

budapest is tickets thing, NDA confirmed
damn.
estimate for leaks this evening?

>>61434227
>NDA confirmed
Sauce? It'd be some bullshit if they made people sign NDAs. I'm not sure it'd fly at PDXLAN.

>>61434278
assumed that because it's not open doors(even though tickets are free)

>>61434517
The only reason they probably set it up is to keep track of capacity and nothing more. They've also done registrations for past Capsaicin events and those didn't have NDA funny business (unless you're The Media).

The event started. Is there someone doing a livestream or liveblog?

is anyone here NOT expecting
>no fps counter or fps cap
>hidden settings

>>61435758
vegavan makes zero sense then, they'd simply launch it at siggraph and be done with it

>>61435887
except they need to do something to try stall people from going nvidia.

>>61435901
Case A: It costs $400 and performs like a 1080ti

Case B: It costs $800 and performs like a Titan Xp

Any indication if it's the cut down or full version of Vega?

>>61436032
I think that they're talking about monitors.

>>61436127
they increased g-sync cost? it was $200 before

>>61435943
>try stall people from going nvidia.
anyone who wanted to do so already did.

>>61436200
It could be both, Vega has similar performance for $100 less and also the monitor cost $200 less.
Having to drag the monitor into the price/performance argument would be pretty weak though.

>>61436261
>similar performance for $100
kind of a big deal

>>61435901

Classic AMD. Nothing but lies and bullshit.

AMD won't show framerates until it's ready for release

if they do people are going to go "wow it's too low I'm getting a 1080 Ti"

if they don't people will just complain

the monitor price is relevant to me. g-sync tax is absurd

>>61436754
b-but it's PREMIUM sync man

>As expected, the additional power target of 25% moves the card to 440 watts of directly measured power…ouch!

AYYMD HOUSEFIRES

File: file.png (206KB, 586x375px) Image search: [Google]

206KB, 586x375px

Wanna bet the left one is the one with the 1080?

What a joke of a company.

>>61436767
It actually is. Enjoy your flickering black screens with no fix whatsoever.

>>61436929
yeah, same way here https://www.youtube.com/watch?v=ybF7r4rogHc&t

>>61436983
Holy shit that girl's chin.

File: 1489285810932.jpg (58KB, 1494x176px) Image search: [Google]

58KB, 1494x176px

>>61435901

I am buying RX Vega and nobody can convince me otherwise. Get freaking wrecked, nviggiots.

>>61437122
THISTHISTHISTHIS

>>61437122
Same. Unless the power draw is 300W, that's the only red flag for me.

>>61437122
I probably will too but that doesn't mean I'm not going to hold AMD's ass to the fire over this pathetic display. RTG honestly deserves all the criticism they get.

https://forums.anandtech.com/threads/vega-navi-rumors-updated.2486940/page-187#post-38989113

>Incredible. Waited 2 hours in the queue, but gave up. Very unprofessional organization.

>This is what I have been told:
>They had a FreeSync and a G-Sync system, one equipped with an RX Vega, the other with a GTX 1080, display was some 100 Hz ASUS.
>The "contest" was to tell which one performs better. You have 2 minutes with each system. Finito.
>Additionally, they have some ASUS gear to show.

>That's it. Don't know who is / was responsible for this. What a joke.


AYYMD IS FINISHED & BANKRUPT

AYYMDPOORFAGS CONFIRMED ON SUICIDE WATCH

>>61437661
>Very unprofessional organization.
RTG in a nutshell

File: aHR0cDovL21lZGlhLmJlc3RvZm1pY3JvLmNvbS9SL0EvNjkzMDQ2L29yaWdpbmFsLzA0LUNyZW8tUGVyY2VudGFnZS5wbmc=.png (33KB, 711x533px) Image search: [Google]

33KB, 711x533px

>>61437661
Lel, get rekt nvidiot.

>>61437687
>10% faster while using 2x more power

AAHAHAHHAHA

>>61437661
>Waited 2 hours in the queue, but gave up
didn't they have tickets ? why queue?

>>61437717
P6000 is a full GP102 (3800+ shaders) you moron.

https://www.reddit.com/r/Amd/comments/6o0kku/vega_equals_or_is_close_to_gtx_1080_performance/dkdlntu/

>the AMD rep guy was asked and he said it's a GTX 1080 non Ti against the RX VEGA

AHAHAHA OH WOW

AYYMD IS FINISHED & BANKRUPT

AYYMDPOORFAGS CONFIRMED ON SUICIDE WATCH

>>61437765
1070 levels we all knew this. not a surprise

>>61437687
Workstation perofrmance isn't that important compared to reliability

>>61437908
Then get one of those $500 Quadros you faggot, who needs performance am i rite

>>61437783
>clearly says 1080 levels
>"it's 1070 levels we already knew this"
hmmmmmmmmm

>>61437953
>or close to

so any dx11 game it will be 1070 levels because nodrivers

>>61437953

>the AMD rep guy was asked and he said it's a GTX 1080 non Ti against the RX VEGA

"1070 levels of performance? Pathetic! How can they compete with a card that's only as good as a 1060!? AMD is finished and bankrupt with their poolet 1050 clone!"

>>61437987
KEK.

>>61437122
>tfw got 144hz 1440p freesync monitor running on a rx480
Me too senpai

>>61438623
that's my goal in the coming years. I dont care how strong it is as long as it offers better than rx580 at better price than NVIDIA.

>>61437717
>>61437765
>>61437783
Reminder to report shitposters.

Do you think Rx Vega will be able to do 4K gaming?

File: mifhxs.jpg (199KB, 1599x956px) Image search: [Google]

199KB, 1599x956px

>>61437908
>reliability
KEK
Is that the new shill trick?
First efficiency matters, now reliability does.
Ahould i reind you which company had a history of unreliable products, dead chips, drivers burning the chips to death and causing system failure?

>our product is inferior while being more expensive and having proprietary blackbox features
>but
>but
>eehm, it's efficient yes

>>61439723
Actually a 1070 can do you 4k gaming.
Just set the shadows on medium, and tone down post processing a little bit.

So what's the PSU recommendation on this shit?

>Ryzen 5 (+ cooler)
>Vega RX

Assuming ~ 100% utilization on both on stock, that'd be around ~300w+~100w for these two alone. Then there's cooler(5w) + hdd/ssd(10w each) + 4-6 usbs (5-10w). This puts us at ~ 470w. If we overclock the CPU/GPU, ~130w + ~400w? So around 600w minimal + ~50-100w for extra threshold as to not put 100% stress on PSU.

700w is what I'm calculating as safe PSU requirement.

>>61439811
>So what's the PSU recommendation on this shit?
We don't know.

>>61439811
I use a heavily overvolted Fury with a 600W PSU and have never had any issues so I'd be surprised if stock Vega wasn't fine.

>>61440326
Mine's more of a theoretical safe zone when system is loaded 100%.

When you game, you don't use 100% cpu or 100% gpu power. When you stress GPU with things like furmark, or cpu with prime95 thats where you see the true limits. Fury has a calculated average power usage of around 275w, but when its unleashed/tested in furmark, it can go upto 400w (using all available gpu resources). Similarly with CPU, TDP (for intel) is average for stock power usage under gaming loads, but with max load, it goes ~50% more.

Lets assume your setup is like this.

Intel 4770K
Fury OC
2 HDD/SSD
5 USB
3 fans

4770K TDP is ~84w and that's accurate in gaming on stock. With bit of OC and load, it can go up to ~120w.
Fury OC'd is around 275 average and ~300 gaming with OC, and ~400 full load.
USBS/HDD/SSD/Fans ~75w

130 + 400 + 75 ~= 605. Thats within acceptable minimum (for 100% load threshold).
50-100w more to offset the 100% PSU usage for more smoother run. Ideally you'd want to go for that sort of safe mark.

But in a minimal standard,

4770K = 100w
Fury = 300w
USB/HDD/FAN = 75w
= 475. That works well when you're not fully stressing the system and only playing games/cpu load in normal conditions. The 600w would then be quite good since its a 125w extra on the load so your PSU isn't 100% all the time.

2 different marks to understand when buying PSU. The theoretical mark and the normal gaming condition mark.

>>61436983
>the only systems with hiccups were intel's
interesting...

>>61439811

if you have a single gpu 500-550w will be fine even with heavy overclocking

I have an r9 290x, an r9 285, and a 5820k, and even with the cpu and the 290x overclocked while running prime95 and two furmark instances (one for both gpu) I barely hit 630w~ with some other minor stuff plugged into the same socket, and while I'd still rather have the 750w for that extra bit of leeway the rig idles around 130w which has some incredibly shit efficiency out of the gpu

Does Vega compete with its technical power or its price:quality ratio?

I guess i'm asking can it beat 1080ti/TItan XP

>>61441185
We don't know.

>>61441185
Probably price:quality, I don't think Vega RX is going to compete in power:performance efficiency.

Vega is a pile of shit.

>>61441252
t. leather jacket man

so it's safe to assume that 1080p@60fps Vega cards will be in the low end of the spectrum?

>>61441252
I agree.

t. amd owner

>>61441311
These will be much much later.

>>61441311
1080p 60fps is already given for these modern >$200 cards.

>>61441311
>4
just get 390x or something lol

390? 390x

>>61441346
>>61441366

yeah, but given that the 480 was already capable of that for $200 or so, it makes me wonder if we'll se the same performance for something like $100~150 in the near future.

>>61441311
>1080p@60fps
is it 2011 again?
580 for 250 is already that.

>>61441467
The next somewhat big leap in performance lies in 7nm dieshrink.
Even bigger one is in MCM GPUs but it's like 3 years away at the very least.

File: 1408867300830.png (386KB, 1586x1114px) Image search: [Google]

386KB, 1586x1114px

now, seeing this trainwreck, I'm more than glad that I pulled the trigger instead of waiting™

I mean - the whole situation kinda reminds me of R9 290 premiere, when everyone was shouting "OMG 6+8 power pin HOUSEFIRES"
but it was actually a solid performer for the price
now with vega... doesn't look like it'll be this good in any way

File: 10110000 01100001.jpg (37KB, 306x306px) Image search: [Google]

37KB, 306x306px

The AMD's CPUs department is giving tremendous pride, but the GPUs department is giving tremendous disgust!

Is this R300 2.0?
Except they actually demo the cards with *some* gaymplay.

>>61441566

It's just a Fury 2.0.

>>61441614
Doesn't look like it.

>>61441620

If we take into account that it does not compete with 1080 Ti, you're right.

>>61441641
You don't know whether it competes with GP102 variants.

>>61441660
There's a Vega FE variant out already. It will be roughly a 1080 competition or an overclocked 1070 competition.

>>61442166
FE is not indicative of RX Vega performance.
In fact, Vega features are simply disabled on FE.
Everything, even discard accelerator (it discards less tris than Polaris).

>>61442203
You can also look at the 780ti to see what happens to cards without driver support.

>>61442244
I have 770.
I know THE pain.
Kill me please.

>>61442203
We know this, but for gaming purposes, this is not going to be a 20% increase in performance. At best, ~10%. That's still only at 1080 level.

>>61442377
>We know this, but for gaming purposes, this is not going to be a 20% increase in performance.
Wut?
Do you know what TBR does?
I am not talking faster discarding with primitive shaders.
Only TBR.
Do you know what it does in what happens when you implement it?

I'm starting to panic about the reported performance. Help me anons.

>>61442527
Wait™.
And go read some Maxwell whitepapers and listen to Kanter.

>>61442527
Wait for benchmarks, wait for drivers, wait for Raja's final desperation attack.

So... 1080 performance in Vulkan/4k and 1070 performance in dx11/144hz/1440p confirmed? Also why AMD only uses two 12 and 18 months old games in all their benchmarks?

>>61442708
>Also why AMD only uses two 12 and 18 months old games in all their benchmarks?
AMD showed benchmarks of RX Vega?

>>61442527
I'm more worried about that 300+ watt power draw.

>>61442818
It's not 300+watts even on overvolted FE.

>>61443414

>>61443414

>>61443414

>>61443453
No thanks.

>>61442253
660 Ti, AKA 970 Jr. How do I into noosing?

>>61446880
680 here. brb offing myself.

>>61442818
I thought it was 285 watts on the aircooled RX Vega. Well, that's still 100 watts more than your average 1080. Which is kind of stupid. Drivers will be better in a few months. It's going to end up like Ryzen's launch where several months later things have improved quite a bit.

>>61437987
fuck i love this meme

>>61439811
Have 2x 1080s and 7700k 5ghz fuck runs cool less than 600w.

I mean 2*1080 equals 1 rx (or half) depending on OC.
Having crossfire rx would be a real crossfire on psu.
Just imagine 7900x+2 rx
You would need dual psus

>>61437987
/thread

Vega will require both driver AND game updates to perform at its best. Just like Ryzen did.

File: Navi.jpg (197KB, 585x768px) Image search: [Google]

197KB, 585x768px

Just wait for Navi

>>61449685
And literally everything else.

Not related to Vega but I have an AMD-related question. I still use AMD legacy card and legacy-Crimson drivers don't have OpenCL. Does anything even use OpenCL nowadays? I know Photoshop can use OpenCL acceleration but does it make a noticeable difference? Any other popular software that makes use of OpenCL? MadVR doesn't need it, right?

>>61451754
madvr is directx

File: AirplexRad.jpg (207KB, 1323x743px) Image search: [Google]

207KB, 1323x743px

>>61433763
The perfect setup for Vega

JUST PLAY ASHES OF SINGULARITY AND FUCKING DELETE THIS THREAD

>>61451879
http://media.gamersnexus.net/images/media/2017/GPUs/vega/fe/charts/vega-fe-aots-4k.png
P6000 > Vega by 30 fps in Ashes

>>61437084
>set up two rigs, your product vs. the competition
>introduce random lag in the competition's rig
>which one do you prefer?
>oh, it's our product? I guess it really is superior! make sure to tell all your friends!

>>61451987
it's not intel event, anon
don't be so spiteful for no reason, hard numbers coming soon anyway

>>61452004
Don't confuse realism for spite. I really don't care about the Ayymd vs. Novideo war, but anyone falling for lies this blatant needs to be taken out back and shot.

>>61451754
Handbrake uses opencl, but it's not worth it.

>>61452491
Oh, right, forgot about it. I don't encode videos so whatever. Thanks.
>>61451776
thanks.

>>61451858
And a fire extinguisher. Mandatory if paired with Skylake-X.

>>61436210
There's not a finite number of people wanting to buy Nvidia instead. I'm waiting for the release of Vega since I would rather use my freesync monitor... But if this is a flop I'll buy a 1070

>>61433763
>waiting to get your house burned down
wtf

What's with the "poor Volta" if it isn't even comparable to a 1080ti?

>>61453164
AMD's marketing team is full of retards. AMD themselves knew this was only going to compete with the 1080.

>>61453173
Honestly I'm fine with a 1070 card performancewise. My 290x is a little loud and hot but there's nothing out there for me to upgrade to around 3-400€ . Damn freesync

File: 8bse41rideaz[1].png (794KB, 2048x1152px) Image search: [Google]

794KB, 2048x1152px

Logo is literally a space heater

>>61453173
Poor Volta when they have nothing on the P6000 in any game.

>>61453211
AMD GPUs are such a clusterfuck of bad you might as well jump onto the Gsync ship.

>>61453582
Vega+Freesync could still be a pretty good value. We'll see when pricing comes out.

>>61453582
Perhaps I'll do but gsync is expensive shit

>>61453382
burning red

>>61453669
FreeSync 2.0

My main monitor died this week (My beloved HP LP2475w). I was going to wait for Freesync 2, 4K with HDR but I need it now so I have ordered an Acer XF270HUA for some 2K Freesync IPS goodness (Hope I win the panel lottery though).

Gonna run my overclocked RX 480 @ 1400Mhz on it for the time being. I might get Vega to tide me over depending on price.

If it beats a 1080 then for 1440P gaming it should be 'good enough'.

This thread is still alive? The fuck.

WHEN WILL AMD LINUX DRIVERS COMPETE WITH CLOSED NVIDIA DRIVERS

FUCK

http://www.guru3d.com/news-story/amd-rx-vega-shown-against-gtx-1080-at-budapest-event.html

Hmmm...whats amd mean by this?

>>61457081
That they'll probably launch some Vega10 SKU at whatever is 1080's MSRP?

>>61457006
>lincucks
ahahaha

>>61441506
KEK
underrated

>>61459361
..or not.
either way entertainment
my bet though that 1080 competitor will cost $300.

>>61459391
>484mm^2 die
>1080 competitor
Nah.
Especially considering it trades blows with full GP102.
>>61459361
That's pretty obvious, considering it's workstation perf.
The RTG marketing team needs a gas chamber though.

new video from AMD pope boyz
https://www.youtube.com/watch?v=HyyOP3ZI6qc

>>61459547
God, he's so fucking clueless.
And his voice is disgusting.

>>61459547
Who?

File: AMD-Radeon-RX-Vega_Budapest_2-740x494.jpg (65KB, 740x494px) Image search: [Google]

65KB, 740x494px

>>61459431
they found right girls though

>>61459571
elaborate

>>61459588
>3DPD
Eww.
>>61459571
>God, he's so fucking clueless.
Most tech youtubers are fucking clueless.

>>61459547
please, stop taking him seriously, want optimistic outlook backed by at least knowledge listen to this guy

https://www.youtube.com/watch?v=owL_KY9sIx8

>>61459625
Vega has 4 shader engines though.

>>61459625
What reasons do I have to stop taking him seriously? Everything he said so far made perfect sense.

>>61459649
Lol

>>61459391

B&C have been right about everything since ryzen. Hell they were calling AMD's implementation of SMT superior to Intel's last year when /g/ denied it was even possible.

>>61459649
He has a scottish accent so dumb americans don't understand him, nothing else wrong with him.

>>61459658
>nervous laughter

>>61459644
at a time of the video it wasn't confirmed if I remember correctly

>>61459688
It was confired back in January.
Vega does 11tris/clock with 4 shader engines (guess how).

>>61459649
Did you only start watching his videos yesterday? This retard literally claimed Nvidia had nothing after Maxwell and that the fury x would be some beast (I.e. what the 980 ti turned out to be). He also said amd will win the market so hard that Intel and Nvidia will become near bankrupt in his master plan video. Yes he literally said that.

>>61459710
Care to link the video where he says that?

>>61459710
>Nvidia had nothing after Maxwell
Ehh that's sorta right.
Both Pascal and Volta are Maxwell dieshrink with some tweaks for GPGPU applications.

>>61459649
well for one his technical prowess is null
I can pick up and analyze his videos where he is absolutely wrong on multiple occasion

he is wcctech of youtube, sometimes he hits it right sometimes misses

>>61459722
>reading comprehension

>>61459728
No, he's not. He predicted a situation for Nvidia similar to what Intel is going through now with 0 performance and efficiency gains. He claimed that rx 480 was 2 gpu stuck together as well.

>>61459729
>i can pick and analyze his videos
Then fucking do it.

>>61459797
>He predicted a situation for Nvidia similar to what Intel is going through now with 0 performance and efficiency gains.
That's kind impossible for GPUs since you can throw moar hardware.
Dude needs to kill himself.

You people need to seriously start doing your own research instead of taking the words of relatively clueless youtube celebrities as gospel, it's quite disgusting.

>>61459830
Well rx Vega at the same clocks as the fury x is worse and uses more power so it is possible I guess. He thought single gpu would hit a brick wall like 2 years ago and no more improvements can be gained without his retarded idea of stapling 2 gpu together.

It will hit a wall at one point, that's just common sense but not like he said it would 2 years

>>61459886
>Well rx Vega at the same clocks as the fury x is worse and uses more power so it is possible I guess
Could you please show me RX Vega benchmarked in some ~30 gayms?
Also include clock per clock benches against Fiji.

>>61459905
>lies Vega STRONG

Don't get your hopes up buddy, better wait (tm) for Navi.

>>61459926
Show me the benchmarks please.

>>61459930
I'm afraid you have to WAIT for them to surface for a tad longer.

>>61459905
Ask one of these people to post the benchmarks. They've been around for a couple weeks. I'm phone posting right now and don't have the links.

>>61459948
People have working RX Vega samples on hands?

>>61459905
http://www.gamersnexus.net/guides/2977-vega-fe-vs-fury-x-at-same-clocks-ipc

>>61459926
Oh, they're strong.
At least against Quadros that cost 5 times as much.
And that's not even the WX Vegas

>>61459970
>FE
I said RX.

>>61459985
>rx vega gaming driver available

ah yes, the classic denial. what are you going to claim next, the rx vega has superior hardware and isn't comparable at all?

>>61459979
That's cool if you're into rendering, not so much when you're buying it for the games. RX 580 has same FLOPS as a GTX 1070, sadly not the same performance in tasks outside mining.

>>61460004
>*shitposts loudly*
(You).

>>61459985
Are you fucking retarded? IPC won't change between the rx mainstream card and the higher end FE one. Neck yourself.

>>61460032
>using the term "IPC" relative to GPUs
Fucking necj yourself.

>>61460011
Did it occur to you that AMD might care more about professional performance than gaming with Vega?

>>61460032
Got any source on that juicy claim?

>>61460048
Why are we discussing it so eagerly then? Do we have a thread full of graphic artists and 3d animators? Are you one?

>>61460064
>people are discussing new hardware on technology board
WHOA

>>61460044
>deflecting to semantics

even the article refers to it as IPC in the sense its a clock-for-clock comparison. damage control harder.

>>61460053
prove that the mainsteam one has totally different hardware to the point the clock for clock comparisons are moot.

File: wtf.png (32KB, 808x594px) Image search: [Google]

32KB, 808x594px

>>61460089
Hmm...
really makes me think

>>61460064
The starting post was "Vega can't compete"
Someone said "It can against Quadros"

Now, if it can fight and beat a full GP102, it most fucking certainly can compete, making the above statement false.
Your issue is that it can't compete in the gaymer market.

File: vega-v-furyx-grw-1080.png (21KB, 805x331px) Image search: [Google]

21KB, 805x331px

>>61460107
Hmm...
really makes me think

still have to prove that the rx mainstream version is sporting totally different hardware to the point these initial tests are all moot.

>>61460184
>hardware
Oh boy.
Hint: it's not about hardware.

>>61460202
>get's blown the fuck out
>i-it's not about h-hardware
>d-drivers need optimizing
>proceeds to go into another thread and claim how amazing amd drivers are

every time.

>>61460241
Are you fucking retarded?

File: 1463089344906.jpg (31KB, 250x251px) Image search: [Google]

31KB, 250x251px

>>61460250
are you?

thanks for all this overwhelming evidence to prove otherwise. i'll accept my defea-
oh wait..

>>61460274
(You) tried.

>>61460274
>reaction images
die

File: 1484886074045.jpg (20KB, 480x322px) Image search: [Google]

20KB, 480x322px

>>61460284

>>61460311
Have a (You).

>>61460184
>Vega works just like Fury
This is the most legitimate argument supporting that the drivers jost dont work and there must be a massive improvement when they arrive.

>>61460781
It already competes with GP102 (and sometimes smokes it) where it works.
It's just shitposting.

>>61460184
Let me give you a hint, you clueless dumb fuck:

The real question anyone trying to read the tea leaves on how RX Vega going to perform has to be able to convincingly explain is how Vega FE can go toe to toe with a Quadro P6000 in pro applications while being at 1080 levels in games.

It should not be possible for both to be true simultaneously, and so the whole question comes down to coming up with a convincing technical explanation for how both of those things can be true at the same time on Vega FE, and therefore what we can infer from that about RX Vega.

Anyone who doesn't even bother to TRY and explain how both of those things can be true simultaneously and what that implies is just low effort shitposting (e.g., you) and doesn't deserve to be engaged with seriously.

>>61460781
its embarrassing that a whole year of supposed "innovation" results in this. one minute amd drivers are supposed to be better than nvidia, next minute i'm told i have to wait for good drivers to see the real performance. i'm willing to bet the rx vega will take at least 6 months to reach its peak performance as drivers get further optimized. keep in mind volta is out in 6-8 months too though so this is pretty sad on amd's part.

>>61460896
Quality and stability are not the same as performance and optimization.

The fact that you can form semi-coherent sentences should tell us you are smarter than what you are saying yet here we are.

>>61461000
excuses as usual. you will defend rx vega's lacking performance for its first 6 months, right?

>>61461044
Are you from the future?

>>61461054
>it's illegal to use the past as a basis for the future in my thread!

yeah i forgot the 290x didn't take 6 months to become competitive and start beating out the crock of shit that was the 780 ti. i forgot the 480 didn't actually release with a 15% performance disadvantage compared to the 1060 and it took 6 months for them to even be competitive. i'm going crazy amirite?

>>61461131
AND WE HAVE A TIME TRAVELLER!

>>61461155
failed to answer a simple question once again. this is too easy lel.

>>61461173
So if you're from the future, how's V100 yields?
Charlie is laughing again?

>>61461181
>the past doesn't exist

kek

>>61460896
>>61461044
See
>>61460862

>>61460896
>6 months

lmao

It took years for the 200 series to get properly utilized and you think they're gonna salvage this heap of shit in that short amount of time?

>>61461516
Samefagging is bad.

File: file.png (28KB, 349x309px) Image search: [Google]

28KB, 349x309px

>>61461554
It's also not a thing, but whatever makes you sleep at night, shill. You seem desperate.

>>61461516
the 200 series was alright if you bought it used for like $100 but the 300 series was a colossal fucking ripoff and if anyone bought the 300 series rebrands or the fiji gpu they should neck themselves

>>61459547
My favorite tech guy is AdoredTV and my favorite comedian is Frankie Boyle.
It's like Scot's are not afraid to tell the truth.

>>61433763
Putting this on my filter list.

>>61436967
Worksonmymachine®

>>61437122
>not enjoying good 1440p and great 1080p performance with Fury while waiting on the true shitwrecker that will be Navi @4k.

4k freesync monitors will be cheaper by then, too.

>>61436967
FreeSync is great, what're you even talking about? The biggest issue I've had is that it doesn't work in boarderless windowed mode for certain games, which means you have to go fullscreen mode. BW introduces a little bit of input lag, anyway, so it's nbd. Never realized just how much tearing I had before I got a freesync capable card.

>>61459588
>they found right girls though

That's not really hard in Budapest.

>>61460896
>one minute amd drivers are supposed to be better than nvidia, next minute i'm told i have to wait for good drivers to see the real performance

It's a new card that uses a placeholder driver. It's a fucking beta test. What's difficult to understand about that?

>>61461567
>I have both an internet capable phone and a PC

>>61462677
>It's a new card that uses a placeholder driver.

are you retarded? AMD themselves have said otherwise

>>61462840
Gaymen part of the drivers was a placeholder.

>>61459886
>his retarded idea of stapling 2 gpu together
It's not like even nvidia are obsessed with MCM now or anything...

>>61463220
>nvidia
>doing NUMA
I hope I won't die of laughter.

>>61463488
They have a whitepaper. AMD will still probably beat them to it.

Hello friends. I had computer. Police took it away. After 13 months they returned it today. (I was innocent!!!) In that 13 months I made other PC. Now i have 2 PC...

TLDR; Which one is better ? What should I do with the worse one ?


i5 4690
msi r9 280x
8gb ram

FX8350
asus r9 290x
16gb ram

I just use it to meme and play games online, occasionally I record some gameplay with OBS, and emulate retro shit (snes, PS1 etc)

>>61463548
Put R9 290X and 16 GB RAM in the i5 build
Sell whatever's left

Vega will underwhelm people since it's gonna use a ton of die space on new features that won't get used in any older and most newer software.

Beyond the HBCC, most people have zero idea what tier 3.0 conservative rasterization actually implies. (Drawing only fully covered pixels = on-card Umbra-style occlusion culling = scene setup essentially fully offloaded from host)

It'll be years before mainstream games use most of this unfortunately.

>>61463528
Maybe because AMD's had patents detailing such concepts approved since like 2011

>>61463569
ok why do you recommend that? thank you

>>61463592
i5-4690 beats the FX 8350.

>>61463573
>It'll be years before mainstream games use most of this unfortunately.

More like they'll never use these, since Nvidia doesn't support it.

>le sandbagging maymay xDDDDDD

Someone put a bullet in this guy's skull already.

https://twitter.com/BitsAndChipsEng/status/887736810625019910

>>61463801
It'll happen after PS4/BoneXX, which are all but guaranteed to be full Vega variants.

>>61459361
>AMD is just sandbagging!
I wish this meme would end and people would just come to accept that Vega isn't going to impress.

>>61463746
No it doesn't.

>>61462840
>>61463573
>>61463801
>>61463855
>>61463940

See

>>61460862

Unless you are going to put in the effort of even trying to give a convincing technical explanation of how you can reconcile P6000 level performance in geometry heavy pro workloads with 1080 tier gaming performance you can fuck right off with your low effort shitposting.

Remember it should not be possible for both to be true simultaneously.

File: 1496560092058.jpg (107KB, 540x514px) Image search: [Google]

107KB, 540x514px

>>61463974
what did he mean by this?

>>61464203
Raja dropped the real drivers into the sun.

>>61464236
He meant that FX-8350 outperforms i5-4690. i5-4690k can outperform FX-8350, but only once overclocked past at least 4 GHz.

File: smells like summer.png (25KB, 930x378px) Image search: [Google]

25KB, 930x378px

>>61462816

>>61433763
>yfw one rig had a 1070 in it, the other had a 1080

>>61439795
but he's right. reliable calculations is more important for workstations, that's why they have double precision and improved drivers for error correction.

>>61464300
Vega FE is targeted towards those that don't need certified drivers or ECC memory. That's what Radeon Pro is for.

>>61459729
Please do it. I'm curious. I mean unironically.

>>61464266
4690 has a slight edge but i'd still sell it and wait for ryzen 2

>>61464498
A slight edge in specifically single threaded workloads like playing CS:GO, and only playing CS:GO, maybe. Overall FX-8350 is a better processor than a 4690 that you can't overclock.
Bulldozer may have been a fiasco, but that doesn't change that non-overclockable i5s aren't about as useless in comparison as an i3.

https://twitter.com/AMD_UK/status/887653978049318914

Is this the new "Poor Volta"? I feel like there's a lot of opportunity for snappy comebacks here.

>>61464808
If they don't pull off a miracle they are dead.

>>61460004
>>61460032
>>61460044
>>61460089
AMD Drones

>>61463929
It was already anounced nextbox will be releasing with vega GPU, not guaranteed Sony will follow suite though

>>61453582
>Y-yeah buy nvidia goyim, its just 200$ more for monitor! b-but you will also be able to buy overpriced top 1080ti!
nnnnope

>>61463855
>put a bullet in this skull
>bitsandchips
Sorry, mr. random anon, these guys earned much more credibility than you in their time of existence.

>>61467060
>steam survey
>indicative of anything other than how many poorfags with laptops in the third world play dota2

>>61463746
Everything beats the utter trash that is the fx 8350. Even a shitty locked skylake i3 btfo's it in multithreaded games like bf1

>>61468493
Fuck Peru.

>>61467060
Weren't there some benchmarks that showed that if you spoofed AMD's CPUs as "GenuineIntel" you got increased performance?

>>61469520
GFXBench.

>>61436929
left was confirmed vega