GTX 1080 Ti Will Be Landing Around PAX East (March 10th)

>wccftech
into the trash

Poor Vega
another DOA

>>58407824
They're in different market segments. AMD learned in 2008 they didn't need a Halo product. Of course Meyer hedged against that strategy with Hawaii and Fiji, and we all know how well Fiji went.

>>58407453
Vega will destroy the XP for $350. 1080clipcut stands no chance

>>58408723
obvious nvidia shilling is obvious

>>58407824
not everyone is richfag

File: AYYMD-on-suicide-watch.jpg (434KB, 1920x1080px) Image search: [Google]

434KB, 1920x1080px

>>58407453
RIP AYYYMD

>>58407453
does this mean price cuts for current cards? Will we see a 1060 6gb for sub $200? Will the 1070 finally come fucking down in price to a reasonable amount?

>>58407453
Isn't this going to cost like a grand?

Ok time for summary:

Vega is the biggest fundamental architecture change since they switched from Terascale to GCN 1.0
The new NCU means a massive boost to efficiency in the shader core and a much easier ability to fully saturate the GPU with work.
The biggest problem with Fiji(furyX) is that its 4096 shaders didnt mean very much because of large bubbles in the pipeline and large ammounts of shaders idling and doing no work each cycle.
Even with a low Level API like Vulcan which is best case scenario the FuryX is still not running with all cylinders firing because Vulkan only solved the Driver overhead problem. There is still bubbles in the workload because of the fundamental design of the GCN architecture itself.

AMD is switching from standard 4x16-bit blocks of a normal GCN CU(compute unit) to This new design of a variable 8+4+2+2 block and either four of these or a mix of these and standard 16-bit blocks in this new NCU(next compute unit)

They could not Do this in the past because they did not have the technology to build a Hardware Scheduler and Memory Scheduler Complexed enough to run this kind of architecture.

We have test samples of Vega with NCU's disabled and with very low clockspeeds and its beating a 1900mhz overclocked gtx 1080 in DOOM. Now there is a gain from using Vulcan over DX11 and OpenGL for AMD but thats on the Driver Overhead side not on the fundamental shader operations side. IT just means the CPU is feeding draw calls to the GPU at peak efficiency.

Alot of people dismiss the test because its vulcan but they dont know what they are looking at. The Vega sample is running FIJI 1.2 drivers Which means the instructions are running through the Vega NCU's like they are standard fixed 16-bit blocks like you have on a FuryX card.

So you need to look at it from this perspective.

This New architecture is more advanced than Maxwel/Pascal. You are going to see unreal performance from Vega when its finally released. Poor Volta

>>58409911
mmhmm yea

I understand some of these words.

>>58409911
Muh dik

>>58408723
>a $350 card from AMD will beat the $550-650 card from Nvidia
Let me guess 8c 16t ryzen is going to cost $250 too and destroy the 7700k right?
Why do you guys do this to yourselves?

>>58408746
So it doesn't matter because the Vega card that beats the current 1080 is going to cost a lot of money as well

>>58410036
ryzen is confirmed $200. cry about it

>>58408804
No.

Nvidia isn't going to drop prices unless AMD creates some serious competition with Vega.

>>58409911
>You are going to see unreal performance from [AMD Product] when its finally released. Poor [Intel/Nvidia product]
Every single time

>>58410066
If you keep throwing shit at a target, eventually some of it will hit.

>>58409911
>AMD didn't had time to make drivers for their in currently testing/fine tuning GPU because who needs drivers for testing when we have hype and lies

>>58410052
I would be very happy if this was true, unfortunately it isn't. So I'm gonna have to wait and be disappointed with everything AMD this year, again and just buy the previous from their competition during black friday
But feel free to prove me wrong

>>58410102
previous gen*

>>58410036
Ryzen will bulldoze the shit out of Intel and you know it

>>58408723
False flaggers fuck off. It's a nice joke, but think of the gullible anons.

>>58410199
You already made this post.
This is just spam at this point

How much better will this be than the 1060?

>>58410209
It's an irrevocable FACT friendo. go be a marketer somewhere else

>>58410227
it's poojeet marketing pasta

>>58409984
Let me simplify it for you. The New Variable NCU's will be capable of processing primitive 8/4/2 instructions side by side the larger 16-bit/32-bit operations and if needed they can pair up into 16-bit blocks to perform the regular ones.

It means the throughput in the processing core is going through the roof for variable workloads like you have in video games.

GCN architecture is designed for large compute tasks because its a Sever/Professional architecture that has been Re-purposed for gaming.

With Vega they have built a GPU that will be much more geared from gaming while still handling the heavy compute tasks in the Professional/Server market.

Vega is going to be very fast ESPECIALLY in 4K/5K resolution and beyond.

You are looking at a card that will keep pace with the GTX 1080 at 1440p and then BTFO it at 4K which is smart on AMD's part because 1440p/144hz is a boutique setup for a small part of the market. The majority of people buying this card want 4K, 60hz performance on Ultra settings and eventually 4K, 120hz performance going forward.

The ultimate performance of the card is going to depend on the Overclock Ceiling but even 4096 of these variable shaders running at 1500/1600mhz will match a TitanPascal and possible even beat it by a few points.

Since you have TitanXP pushing 100fps on some well optimized Vulcan/DX12 games in 4K that means a solid 4K, 120hz GPU at the high end for much less than the 1200$ price of a titan.

>>58409911
>We have test samples of Vega with NCU's disabled
>The Vega sample is running FIJI 1.2 drivers
Sauce me up.
Vega should be more advanced than Pascal given how Vega has their drawing technology now and Pascal is just shrunken Maxwell. They have their async compute plus new NCU. There's also the delicious HBM2. AMD, please don't fuck up.

>Literally shilling for botnet hardware.

>>58410281
agreed nvidia is trash from an evil corporation run by a now nothing business man

>>58410260
>>58409984
kek

File: VegaPerformance.jpg (400KB, 1200x928px) Image search: [Google]

400KB, 1200x928px

>>58409911
Slower than GTX 1080 OpenGL my man.

>>58410503
OpenGL is legacy software thats being phased out. The key difference between OpenGL and Vulkan is Driver overhead (CPU bottleneck). Finally the Vega card is running GCN 1.2 drivers which do not account for the the variable shader technology that Vega is built on.

>>58409911
I really want Vega to love up to this, but I'm not holding my breath. We'll see once it's released.

File: DOOM.jpg (75KB, 892x501px) Image search: [Google]

75KB, 892x501px

>>58410552
Are you retarded?

VEGA is running Vulkan in that image, while being slower than the GTX1080 OpenGL.
Also you can see VRAM for that VEGA model is 8GB and the Driver version is 1.4.1 running Vulkan 1.0.33, current Vulkan version is 1.0.35 which is just one month old.

>>58410630
Are you so dim that you dont even know the fundamental difference between Vulcan and OpenGL? or the difference between how Nvidia drivers function versus a hardware sample with re-purposed and modified Fiji 1.2 drivers hence why the driver is 1 month old because thats when they cobbled it together.

You have to do better with your bait M8

>>58410816
Hahah what?

Are you retarded for real or what?
AMd driver is 1.4.1

Vulkanb 1.0.35 is mostly Nvidia shit, no reason for AMD to update to it.

https://github.com/KhronosGroup/Vulkan-Docs/compare/08310a88df8c803ac916102e7a364fff912e3281...789549ff080dd8aece944b7c5261eb42f97b55c7#diff-92d0e5376f9e604e03fac86fbb661950

Embarrassing yourself rajah.

>>58410861
>Vulkanb 1.0.35 is mostly Nvidia shit, no reason for AMD to update to it.

Now you are just pathetic.

>>58410882
Read the changelog and source faggot.

>>58410552
>OpenGL is legacy software
only for gaymur /v/tard like you

>>58410036
>Let me guess 8c 16t ryzen is going to cost $250 too and destroy the 7700k right?

The 6700k already destroys the 7700k.

>>58410882
NVX are experimental Nvidia extensions, idiot.

is there a reason for threads like these? Why don't you want both companies to be successful?

>>58410100
drivers are literally the last thing you do

They've been using fiji drivers until recently, now the hardware is mostly or completely finalized and they're doing the software/driver side of things.

So on completely 100% unoptimized silicon without proper drivers they're getting 1080 performance? That sounds pretty fucking good to me.

>>58410908
Sure thing pajeet, now POO IN LOO

>>58409911
>AMD only shows performance in a game that has an API specifically designed for AMD
HMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM

>>58410930
Literally nvidia shills

l i t e r a l l y

>>58410978
LOWER performance WHILE running Vulkan and using HBM2.

Also driver is version 1.4.1

>lmao lets not make drivers to test this new hardware and make sure it works like it should because fuck quality insurance and fuck all those millions we invested in this new architecture if it doesnt works we only lost millions in RDR and years in development and probably the future of out company will be jeopardized too but who cares about that

>>58411018
That's not how drivers and GPU development work, you retard.

>>58411040
Explain how ti works.

>>58411018
Who is this idiot?

If the drivers were finalized the card would be on sale. It's still like 3-6 months out or something like that. It's literally still in RnD labs and isn't being manufactured. They don't even have finalized clock speeds for it yet.

This is some seriously desperate nvidia damage control / shilling holy fuck

>>58411015
>>58411040
>new HBM 2 new architecture overclocked ayymad card
>so overclocked CES hired additional firefighters just in case
>still loses to a gddr5 nvidia card
>on an amd made game

>>58411049
you summon amd drivers

>>58410978
I don't understand why AMD would leak any info about Vega performance until it's almost done though. All this does is build up unwarranted hype around their products. I get that they NEED positive buzz but they also need to just shutup and release a good fucking product.

People are going to be hoping that the software optimization is some kind of magic secret sauce that will unleash the hardware to its true potential, but if AMD comes up short in the tech press' mind then they are just going to fall into their own "AMD doesn't live up to the hype" meme. This is even further confounded by the API clusterfuck-war that is being waged behind the scenes right now.

They should just stay tight lipped about it, make no promises, and release it finished and ready. If its good, let the finished product speak for itself.

File: 1482505775942.jpg (306KB, 989x1881px) Image search: [Google]

306KB, 989x1881px

>>58410015

>>58411083
Do they come to you before or after the housefire?

>>58411104
housefires are copyrighted by nvidia

>>58409911
>blablablalbala next amd's shit will totally be amazing blablalbalbalbalblabla

Oh yeah, we have totally never seen this one before.

When are you guys gonna start pulling the history revisionism you always do when amd's next disappointment ships and you pajeets start pretending that nobody hyped it up and that it was always meant to perform like shit, just like you've always done every single time amd delivered their next turd?

>>58411117
thats funny because the 9590 exists

>>58411117
Ayymd using nvidia copyrighted shit again?

>>58411130
so do the titan x with their >100C VRMs and exploding VRMS on 1070/1080 cards

>>58411127
>AMD shills hype up next product that will TOTALLY shit allover nvidia
>Comes out
>Its shit
>"GUYS ITS SUPPOSED TO BE FOR BUDGET GAMING ITS NOT MEANT TO COMPETE WITH NVIDIA AT ALL!!!"

>>58411137
just amdshit playing catchup like always

>>58411094
It doesn't matter what AMD does or says, nvidia's marketing department takes it and twists it on social media sites - like this one - to be a bad thing.

Mark my words, if vega comes out, offers better performance at lower cost to the 1080 or even 1080 ti, there will be "people" here saying it's shit and you should just go with nvidia.

The nvidia shilling on this board is seriously unreal. Not even the actual nvidia subreddit gets it this bad. I see things like posters intentionally lying about the capabilities of freesync/gsync to "AMD drivers suck" when they are unequivocally better than nvidia's right now.

>>58411148
how are the only company actually innovating in the industry playing "catchup"?

pascal is just kepler 3.0, their engineers are incapable of anything better

>>58411155
Good thing that wont happen then

>>58411111

>>58411140
>put r(ajeet)x480 in mobo
>draws 600 watts through the mobo
>house catches fire
>still can't poo in loo

>>58411189
>make things up online
>realize nobody will believe you
>go back to masturbating furiously to huang's leather jacket in your razer dungeon paid for exclusively by your parents

>>58411177
>vega launches
>its shit
>will still lose to a 970
>"hahaha its meant to be a cheap card not to compete with nvidia guys"

Vega is already underperforming, on Vulkan, on 4k. Vulkan and 4k are the only things that Fiji was good at, and Vega is underperforming already in that mode. Even leaked Ashes of Singularity benches gave it a bad score.

I want AMD to get on top of their game again, but they have been piling disappointment after disappointment in the past decade. Their last good card was the 6970 which was mostly a beefed 5870 anyway (I had a 5850 and rocking an unlocked 6950 still). Vega is probably going to be 1080 tier in most games, pulling ahead in Vulkan or 4k benches, and Nvidia will steal their thunder by cutting the 1080 and 1070 back to sane prices and releasing a $900 1080 TI.

Also, having fuckload of superfast shaders won't matter when the card doesn't have enough ROPs to actually draw graphics.

>>58411155
Maybe, you're right. Maybe they can't just say nothing because that looks worse than at least trying.

Honestly, I have also noticed the Nvidia shilling has been ratcheted up a notch over the past year to obnoxious levels. I'm almost positive there is one paid Nvidiot because he is particularly dedicated with his "AYYMD housefires" posting style. It's psychological warfare, plain and simple.

>>58411215
i rate 3.5/4

even a 7970 beats a 970 now

>>58411177
What amazes me is that nvidia has completely squandered their market position and innovated NOTHING. All they've done is bought out companies to lock down new tech, and implemented gimpworks bullshit to make their old cards obsolete. Every decision has been monopolist anti-customer practices that only drive the cost up for anyone buying their product, rather than an actual attempt to innovate in the industry.

If nvidia gets BTFO they seriously only have themselves and their greed / hubris to blame for it.

>>58408835
stop posting your image, no one cares

>>58411242
nvidia can spend as much money as they like, it can only mask their incompetency so much

>>58411222
With that said, if they release a $300 Nano sized card, I'll get that even if it is worse than the 1070.

Sadly with HBM costing too much, this likely won't happen.

>>58411254
They're not wholly incompetent - they'd have never been successful with pascal/maxwell if they were.

But for all the money they can put into R&D compared to AMD, they sure have fuck all to show for it.

>>58411222
>won't matter when the card doesn't have enough ROPs
They couldn't possibly make this mistake again, could they?

>>58411242
>>58411254
Ayymd upcoming card already lost to mid range nvidia cards. Expect another low poo in loo price/ poo in loo performance card as always

>>58411296
>1070
>mid range
Kill yourselves.

>>58411280
They don't understand that they need to make a card good at pushing graphics, not just one that is good at bitcoin mining.

Their cards just don't word in HPC applications because their programming API is a joke.

>>58411276
they tried to make a competitive product to AMD with fermi

then they realized they don't know how to do power efficiency at all, so instead of trying to find a way to make it power efficient while retaining all the same hardware features, including the hardware scheduler, they just removed stuff from it until it used less power

so now all their cards are utterly uncapable of any sort of parallelism

>>58411327
Not everyone is a poor indian man

>>58409911
aww shit. and I just bought a 480 too.

AYYMD IS FINISHED & BANKRUPT

>>58411372
Rip

>>58411012
>API specifically designed for AMD
is this what retards actually believed?

do you not realize that the Khronos group also includes Nvidia, Intel and ARM?

dumb shitposter

>>58410630
>barely 60 fps in a game more than 20 years old

>>58411155
>le evil marketing bogeyman

Do you shills think people who currently rock Nvidia GPUs think like "I use Nvidia GPUs because AMD is evil!"

Do you AMD fanboys seriously think Nvidia owners have never bought an AMD GPU before?

I'm gonna let you in a little secret, most PC gaming newbies first build used AMD GPUs, because they are in a budget and AMD appear what it seems to be a good alternative. I did even before AMD bough ATI and so many of my friends too.


People don't buy AMD not because they think AMD is evil or because Nvidia marketing but because AMD is just plain horrible.

People who to invested money in a PC just to have problems with their AMD GPU don't want to buy another AMD ever again, they would rather pay $350 for a Nvidia GPU than having to deal with AMD ever again.

People(beside shillbots and retard fanboys) don't give a crap about muh open sores bullshit and the AMD is an angel send by God to save us from Nvidia.

That's reality, not marketing.

You kids who have only owned one single PC in their entires lives need to fuck off.

>>58411372
Polaris has great performance for its price.
Vega will have better performance but also much higher price. It'll just undercut the 1070 and 1080 by a little. I'm not even sure if they can manage that, since HBM is expensive as fuck. Maybe they'll have GDDR5 versions out, but those will use tons more power.

>>58411408
>i have no idea what i'm talking about - the post
ask daddy to buy you the 1080 ti when it comes out babby

>>58411343
1070 is basically "high-end but I don't want to pay the extreme premium" tier

If a card costs more than $400 it's not mid-range, it's high-end.

>>58411372
Vega's literally the reason I put off a 480 purchase. Glad I did with the news about the architecture coming out.

>>58411434
Same bullshit from ayymad as always.
>release a product
>always get btfo by nvidia
>"b-but guys it's cheap"

>>58411435
>Surprised when people in the gijutsu (that's Japanese for Technology) are actually technology enthusiast.


Would it trigger you knowing that just between in the past 3 years I bought 4 different high end CPUs even including a FX8350 just because I can?

>>58411464
For a poor man like you that might be true. High end is anything >1000

>>58411406
(you)

>>58411434
I would have held back. I understand the 480 is a great piece of tech but it was a bit lower end than what I originally wanted. simply, I saw the Vega demos and reasoned that it was nothing special and I might as well not suffer my ancient shit for another 6 months.

>>58411481
>I have no understanding of the word "budget" whatsoever
holy shit adulthood is going to be difficult for you

>>58407453
>curryniggertech
they just keep making shit up and people keep eating it.

>>58411481
The RX480 was a failure because it didn't beat the GTX980Ti amirite?

Stop moving the goalpost, it's getting old.

File: 1368227244406.png (732KB, 1300x417px) Image search: [Google]

732KB, 1300x417px

>>58411483
>gijutsu (that's Japanese for Technology
死ね

>>58411513
I won't get 3rd grade burns from the ayymd housefire so it will be ok

>>58411501
You're clueless and out of touch with the actual market and customer purchasing power in general.

>>58411483
would it trigger you knowing i'm sitting in a chair worth more than your PC?

nobody cares about your baby 'wealth', PC components are not expensive.

>>58411524
It didn't even beat the 980 and except for doom in vulkan, it lost everytime to the 1060

File: RE6_Mercs_Image_Chris_EX1.png (669KB, 1024x1024px) Image search: [Google]

669KB, 1024x1024px

>>58411536
>3rd grade burns

>>58411483
>High end
>Fx-8350
Pick one

You probably also think that a 6700 is high end also

>>58407824
>keep showing off Vega running at low clocks and barely matching 1080
>nvidia forced to launch 1080 ti to regain market attention
>launch Vega a week later at 1500+ core clock and slap the shit out of the 1080 ti
Calling it now.

>>58411556
I think you are triggered.

In fact this whole post made me smirk big time.

Thank you for the smile my triggered friend.

>>58411544
Or you're just a pajeet poorfag

>>58411557
The 480's performance is about equal to the 1060 bud

File: GTX-1060-UPDATE-100[1].jpg (66KB, 649x221px) Image search: [Google]

66KB, 649x221px

>>58411557
>it lost everytime to the 1060
how can one person spread this much misinformation?

1060 barely outperforms the 480 in dx11 and gets destroyed in any dx12 title

>>58411571
>vega overclocked housefire demo
>slower than a 1070
>actual release
>slower thab a 1060
>"but it's cheap guys look at price/performance"

>>58411582
And you're just an obvious (though incompetent) nvidia marketer

>>58411557
$250 card keeps up with a $500 card. Hmmmm....

File: 1430775192503.jpg (337KB, 867x848px) Image search: [Google]

337KB, 867x848px

>>58411592

nvidiots cant into fine wine

File: geforce vs radeon.jpg (73KB, 960x960px) Image search: [Google]

73KB, 960x960px

>>58411513
If they hold back, then they have 0 new products on the market for like 3 years in a row.

The 480 gave them back a ton of market share, and boosted their stock a fair bit, so it was very successful in that regard. Plus it was a huge dig at the 970 and cemented them as the good natured company and Nvidia as the evil one, so a big PR win as well.

Honestly, they'd just need a competing product with competing compute libraries.

>>58411571
>amd will btfo nvidia for sure this time j-just you wait
It's not even funny at this point.

>>58411621
they enjoy their cards getting worse over time

>>58411632
its really kinda sad to see so much denial.

>>58411639
CANT AFFORD A $600 CARD EVERY YEAR? YOU AREN'T POOR ARE YOU?

>>58411629
>paying 250 for a housefire

>>58411664
why would you even need to buy a new card every year? are you some sort of /v/ermin?

Always ignore AYYMDPOORFAGS, they lie all the time

There is no AYYMD Fine Wine garbage when AYMMD intentionally GIMPS older hardware

https://www.reddit.com/r/Amd/comments/5gqm2u/async_compute_disabled_on_gcn_10_since_1692/

>>58411621
>hey look! my mediocre card is just as mediocre 4 years later! take that green team!

>>58411536
you should worry about passing 3rd grade mate

>>58411592
just goes to show, if you fuck up release, you fuck up, point. reviews at release are way more important than half a year after the fact.

>>58411629
I mean, I would have personally held back and waited for Vega

>>58411703
>waited for vega
Except 1060 to 1070 performance but cheaper and your mobo will explode and set your house on fire

Vega is outdated, obsolete garbage

May 2017, Volta will destroy AYYMD once again, brand new architecture every time just like Maxwell & Pascal and AYYMD can't compete with Nvidia at all

>>58411652
yeah amdrones refusing to learn that the company they shill for is terribly incompetent is very sad

>>58411688
feature was never supposed to be enabled on that gen. if they left it up you'd probably be bitching about GCN 1.0 cards crashing during gameplay in your buttmad desperation.

the mistake was to have that shit on the silicon at all.

>>58411719
jesus fuck I recognize you

same shitty memes, same inability to write simple words properly

why don't you get a real job?

>>58411737
>same shitty memes
I don't have an ayymd card

>>58411408
I first bought AMD and ATI because they were acronyms rather than intel and nvidia which sounded like gaymen shit.

>this time nvidia gonna totally btfod
>amd release comes out and suddenly all the amd shills are quiet while some sucker they tricked into buying it are raging

>this time intel gonna totally btfod
>amd release comes out and suddenly all the amd shills are quiet while some sucker they tricked into buying it are raging


Then it repeats and repeats

>>58411703
>I mean, I would have personally held back and waited for Vega

then your stock would've dropped to $1 and your market share below 5%.

>>58411759
HAHAHAHAHAHA

The difference between the 80ti and the 60 is not enough to justify the price tag. There is no game that requires this either.

>>58411759
Stupid poorfag, I bet you poo on the street

>>58411759
nvidia means "envy" - very true of envidiots!

>>58411408
I've had more problems with my one fermeme gtx 560 ti than any other AMD/ATi card ever. Driver would crash every now and then, not to mention it was housefire tier. AMD/ATi? If something didn't go right it was user error and easily correctable.

>>58411786
Yeah amd envies its succes ;)

>>58411759
>people this stupid exist

>>58411798
>ayymd drivers
>better than nvdia
Why would you lie on the internet?

>>58411773
>>58411778
>>58411786
all the professionals use acronyms

DHCP
NAT
XML
HTML
VPC
HSRP
GRE
CPU
NAS
RAID
PBX
FXS/FXO
WLAN
WAN

etc etc shall I go on

>>58411759
here's your (you)

>>58411767
I mean my own personal fucking purchase dude come on

I would have not purchased a Polaris card for my personal usage if I had more substantial info about Vega

>>58411825
Please, do go on.

>>58411242
>What amazes me is that nvidia has completely squandered their market position and innovated NOTHING.
This, pretty much exactly. Especially on the hardware side, holy shit. Nvidia is so painfully boring and predictable. I guess if you're into the software stuff then maybe you'd appreciate what they've accomplished.

>>58411829
thank (You)

poo in loo
coin in nvidia bank account
jewtel
alloo snackbar praise jesus amen etc.

>>58411822
I don't use loonix as my main OS. My ATi x1650xt never had any driver issues and I played WoW on it for years until I upgraded to that dreaded Nvidia card, never again.

>>58411855
RIP
EIGRP
OSPF
BGP
NTP
SNMP
SMTP
SSL
TLS
AAA
VLAN
NIC

oh shiet these are all networking related can you tell what I do for a living yet?

>>58411822
they're pretty much since relive. also nvidia has had some pretty fantastic failures with their driver updates lately.

if Maxwell killed any and all Nvidia housefire memes then the AMD no driver bullshit should also be dead. but alas, Nvidia fanboys are just evolved console fanboys.

>>58411759
Damn son, you really fucking triggered the nvidia shills

>>58411773
>>58411778
>>58411786
>>58411799
>>58411815

>>58410280
I think it says it at the end of that doom 4k trailer thing.

That dude is spinning half truths though.
The driver isn't JUST a fiji driver, it's modified to suit the engineering sample. The question is, why use a driver for GCN 1.3 when polaris should be closer architecturally. Why not specific drivers? Because that video was made months ago for a testbed design that was far from being finalized. The drivers were rigged to suit that particular sample.

Also Fiji's pipline saturation wasn't due to "lack of a complex enough hardware scheduler," it's a hardware bug of sorts. An architectural flaw particular to the fiji. I'd wager it made it to finished parts because fiji's development began with TSMC's 20nm process in mind, which ended up not happening. That and fiji is an example of the absolute limits of the 28nm process. It's a massive die packed to the brim with transistors. So AMD did their best to "convert" the design supposedly late in development. I've seen claims they got as far as early prototyping (more like familiarizing themselves with the new silicon than building a chip) before TSMC dropped the ball.
So 28nm fiji was not only a compromise, but also rushed as AMD had an entire product line of rebranded cards with memory upgrades.

It's sad, actually. Had things been different, the fiji may have been much better, and had it been worth it we may have gotten a hardware refresh to sort out some of the issues in the design, but it ultimately would've just packed more money into sinking ship.

I'm looking forward to what AMD can do under better circumstances with Vega, but doubt it'll be as incredible as this idiot says.

File: 15698260_700357133462931_2418443324419401728_n.jpg (19KB, 414x289px) Image search: [Google]

19KB, 414x289px

>>58411920
"I buy inferior hardware based on naming conventions"

>>58411822
Nvidia's recent driver updates literally fuck up my shadowplay and corrupt a third of my recorded videos

And that's on top of the mandatory spyware and geforce experience + social media login crap.

Oh and nvidia is introducing a subscription service and you goddamn well know for a fact they're going to put shadowplay behind that paywall. (lol wat u r 2poor?)

How the fuck can anyone seriously defend this shit. Seriously, fuck nvidia and their stupid shitty drivers, bloatware, crapware, and spyware.

>>58411920
i am proud of this pooinloopost i honestly expected nobody to pay any god damn attention to it,

That's a true story though. On my first build that's honestly one of the factors that swayed my decision, I didn't know that much about computers and the Phenom 940 and HD 4850 were cheap (woo), hot (who cares), and could play anything I could throw at them so I went that route. 9 years later I'm hoping they'll resurface. I'll admit even as an AMD guy their current offering is OK at best, great for what it is if you actually consider the product and aren't a mouthbreather but certainly not best in class like some of the older chips and GPUs they did were.

I'm hoping for a glorious return, the Phenom II was a fucking beast and so were the 48XX series, and not many people will deny that. I'm just hoping for a rebound.

> The question is, why use a driver for GCN 1.3 when polaris should be closer architecturally.

why should Polaris be closer architecturally? I'd wager AMD's HAL is more or less the same across all GCN drivers with the latter ones having more architecture-specific optimizations.

that, or they forked the driver when they started working on Vega, and all the improvements from upstream weren't really of any use to them.

>>58411786
Envy is a sin tho

File: 1417586048100.jpg (84KB, 720x633px) Image search: [Google]

84KB, 720x633px

>Getting hyped for an ayyMD product

Why do you let yourself be disappointed every time

>>58412134
because the other alternative is permanent disappointment

>spend hundreds of dollars on a graphics card
>play hearthstone and league of legends with it

>>58412214
>tfw it's planetside 2 and even modern hardware can't run it at 144 fps reliably at 1080p

I'm permanently awestruck at how poorly optimized that game is.

dx9 is shit

SHIIIIIIT

>>58408804

They're gonna start at 800usd.

>>58407453
is the 1080ti supposed to be volta? arent they skipping the 1080ti and making the 2080ti?

i heard the 1070, 1080 will all get a refresh with volta later this year which will bring a 15%performance boost. i also heard that the 2080 will cost 500$

damn 1080 ti soon out

AMD keeps gettin btfo no ryzen cpu no vega in sight

Gefore Now gets subscription model

Nvidia shield bestseller on amazon

rip amd

>>58412214
Yeah but on 4k m8.

>>58410552
OpenGL isn't going away anytime soon.

>>58407453
>>>/v/

>>58414805

>>>/lgbt/

>Really thinking the RX490 will even beat the GTX1070

Oh boii, remember everyone buying the 970 even with their 3.5 meme because it was better than the R9 390?

Why Nvidia keeps winning did AMD finally quit ?