1080 ti release date

I am hoping for an october/november release.

When do you think it will be released?

Other urls found in this thread:

youtube.com/watch?v=InX5nyVH4aI
microcenter.com/product/455495/Radeon_R9_Nano_4GB_HBM_Video_Card
twitter.com/NSFWRedditVideo

October.

Wasn't rx480 going to be released October last year? Wasn't that a thing?

Anyway Fuck off op

y tho?

hi

January

Why? No competition.

Didn't you even read the news, Nvidia is gimping the 1080ti to use GDDR5 instead GDDR5X so people will still buy the Titan X pascal.

Stupid fucking chink jew

why not release it before there is competition so they can get autists like me to buy it before we might buy vega instead?

i don't care if they use GDDR5 instead of GDDR5X

that makes no real difference in games

>that makes no real difference in games

what?, do you even know what you are talking about

youtube.com/watch?v=InX5nyVH4aI

yes

But that ti suffix makes ALL the difference

no the cuda cores and 10.8 tflops do

Is there is replacement fot Titan Z(best card ever made)?

>2016
>Using CUDA instead of OpenCL

>posted a speculation video on rumours

clearly you don't, its a gimped card that's going to run worse than a 1060

A 375 watt space heater

Jen Hsun Huang said they don't love/want dual gpu anymore even if the efficient is better

There's zero chance of it being released this year. It'll be saved for next Spring to rain on Vega's parade. Volta isn't coming any time soon, since Nvidia have stopped mentioning it.

Hopefully soon, I'm due for an upgrade.

>all the richfags buy the titan x pascal
>then cuck all the richfags with the 1080ti

Business as usual

They're going to wait to see what Vega's performance is like so they can either release GP102 or GP100 w/ HBM as the new flagship. I don't think they can get away with releasing GP102 (current Titan X) as the 1080 Ti if Vega ends up being better and with HBM.

Sloow ram, why ?

Otherwise you dont need an upgrade, just gloating or what ?

how often does your house catch on fire?

What? CUDA cores != programming with cuda you double nigger. OpenCL basically abstracts your plataform so it can execute on ATI or NVIDIA or even FPGAs. In fact nvidia helped the OpenCL creation so the compiler probably translates almost the same as nvcc.

Excited for this, didn't quite get the level of performance I wanted from my 1070 so I will be looking at upgrading to 1080 or a second 1070 mid next year.
Hopefully this pushes prices down along with AMDs new cards.

...

How are we supposed to know if you have the shitty old maxwells or not though

>titan sli
>not 4k monitor

Bitch why?

Could be the Maxwell titans.

Also Refresh Rate>Resolution

hopefully soon

~6 months. Few weeks before vega launches.

I still laugh that we laughed bigass, overheating, power-sucking video cards out of the industry around 2000 or so and now it's the fucking norm.

The shit happened to people? Seriously, did terrorists flying into buildings ruin the common sense and logic of everyone around the world or something?

those are maxwell titans. pascal titans dont have gtx branding.

>either release GP102 or GP100 w/ HBM as the new flagship

GP100 is garbage for gaming, famalam.

It blows huge swaths of its ~600 mm^2 die budget on shit that makes zero difference for consumers, like 1:2 FP64:FP32 capacity (would be fused off to protect enterprise sales), 14MB of register files, and a bunch of proprietary NVLink controllers.

Vega 10 is going to be a 4XX mm^2 chip sitting somewhere between the 1080 and the TXp in performance.

Nvidia will only make a slightly different binning of GP102, unless they have another secret smaller HBM2 card in the waiting.

>Sloow ram
what is DDR

>tfw no longer a NEET so I have disposable income to get a 1080 ti
i did it bros...

Duuuude, don't blow it on a 1080, ti or otherwise.

that's bullshit, why the fuck would they downgrade the memory from the 1080 which has gddr5x?

The PG279Q is more responsive on 144Hz versus 165Hz.

The people talking about HBM2 on nvidia cards are fucking retarded. HBM2 is only on a Professional GP100 chip which has completely different architecture than gaming pascal cards.

you are only getting GDDR5 and GDDR5x on Pascal and maybe at the Top end of the Volta cards will have HBM 2 if your lucky.

AMD is going to have the First HBM2 gaming cards with vega10 and dual vega and they will have HBM2 completely replace the Polaris line at the mid range with Vega 11

The only GDDR5 card left for AMD in the 500 series will be the lowest end card and it will probably use GDDR5x modules on a 128-192bit memory bus (memory bus adds power draw and manufacturing cost so upgrading to the now cheaper R5x modules to reduce the bus size will even out the cost and reduce the overall cost to produce)

The big Kahuna coming out sometime in 2018-2019 is Vega 20 which is essentially Vega 10 but on a new 7nmFF process with Massive clock-speed increases and upgraded HBM2+

There is no way Vega arch is a bust because AMD is banking on it to be maintained and Refreshed until almost 2020 so it has to pack enough punch to smash pascal and be close enough to Volta also.


Also the Vertex rendering is different on AMD card and people notice improved visual effects and texture quality on the same game with the same monitor using an AMD gpu over a Maxwell/pascal gpu.

FPS isnt everything especially if your Textures look fuzzy and undefined and your color palate is washed (AMD is overwhelmingly preferred among hardcore RPG players because of this.)

The 390x was released around then.
Because It's Nvidia. The way you're meant to be played.
Ayyy

>mfw actually want to buy a high end AMD card
>there literally isn't one
give me a $300 nano at least

>only on a Professional GP100 chip which has completely different architecture than gaming pascal cards

That hasn't stopped them in the past

Is this now richfag speccy thread?

>only 16gb

Guess not

No need for more than that senpai. Not yet at least.

Microcenter had it for $280 today.

Only something a poorshit would say

they dont have them where I live, fuck

The man has six fucking monitors, user. I don't think lack of money is the issue here.

As in outside the US? Because it's one of those rare deals where you can order it online.

That seems to be a bug with speccy because the same thing shows up on mine

can't find it. maybe im retarded

but fuck, neither that nor the Fury even offer a 50% boost over my current 290x. I would spend $400 on a 1070 tier AMD card if it fucking existed

btw, I'm not a fanboy. I just have freesync

microcenter.com/product/455495/Radeon_R9_Nano_4GB_HBM_Video_Card

But if you have a 290x I'd just wait for Vega unless you want it for an ITX build or something.

yeah exactly, I'm waiting for fucking Vega
thanks for the help though

Sapphire Nitro Fury is usually up fopr $300 its $290 atm on newegg

oh, that's a lot better
hmmm... fuck, not sure how much resale I can get for my current 290x. It has a custom cooler. I hate Ebay and people are really shitty at buying local

Custom coolers seem to go way lower than what they should. Not sure why, gotten a couple good deals that way. Unless you have a Tri-X you're fucked on resale.

Also I got fucked selling on eBay once. You basically have zero protection. Guy just says it didn't work and making you pay all the shipping back, even with it getting damaged, eBay doesn't give a shit. I'd go local if you can.

>Also the Vertex rendering is different on AMD card and people notice improved visual effects and texture quality on the same game with the same monitor using an AMD gpu over a Maxwell/pascal gpu.
>FPS isnt everything especially if your Textures look fuzzy and undefined and your color palate is washed (AMD is overwhelmingly preferred among hardcore RPG players because of this.)


Do you have any sources to back up those claims, because I remember AMD following after Nvidia's steps and doing the same style of texture optimizations; billinear when trillinear is not needed et cetera with a driver update years ago.


Are you perhaps confusing texture quality with texture LOD, because unless it happens in every game then is a problem on the game developer side and not Nvidia's.

One perfect example of this case would be Wolfenstein The New Order which uses an extreme and efficient texture streaming on Consoles, once ported to PC the game suffered from terrible texture pop in because PC hardware wasn't fast enough to do it, you have to remember consoles have shared memory and the CPU/GPU can (or not) share the same bus plus PS4 hardware(not sure about Xbox One) can bypass its own L1 and L2 caches and read/write directly to memory with speed up to 20 gigabytes per second, making it faster than PCIe 3.0 at 16X lanes.

That's to make sure the 1080ti won't be faster than the more expensive titanX.

Why would you have a cheaper GPU be faster?

Meh it's called price anchoring.

>will still be shit in dx12 and vulkan and still no hardware async
good thing it can still brute force it's way through until it's time to get gimped like the rest of the other Ti

So because some guy who named himself "tech" and post on youtube you believe this?
For a start hes running on total assumptions.

Holy shit you are one thick cunt

>5930K

nice meme cpu

yeah, reselling computer parts that aren't memes is essentially impossible

but fuck, looking at benchmarks I cant even find a place where a fury gets more than ~5fps higher than a 3/290x. What a fucking waste of an entire GPU generation. I can't believe AMD fucked them up so badly