Are you buying a GTX 1070/1080?

[Maverick177]

Going from HBM to GDDR5X seems to be a step backward, so big Vega for me. Go big or go home.

 

[MustISO]

Currently running a very old (in GPU terms) HD 6970. Haven't been spending much time gaming lately but that may simply be because my card wouldn't be able to play newer games an higher res/settings. Current plan is a 1070 to go into my new build which is only 3 months old.

 

[DaveSimmons]

I'm not being a smartass here, but I fully expect Maxwell performance to sink like a lead titanic.

Someday, maybe. But I only replaced my GTX 680ti with a 980ti for Fallout 4 because I wanted to stay at 1080p Ultra, not because I really needed to.

For at least the next couple of years there might be a game or two where I need to move a graphics quality slider down a notch or even to lower the resolution. To me that's not a "lead titanic."

 

[tential]

Yeah Maxwell is going to be getting the Kepler treatment in short order. Launch day 1080 reviews will show a 1080 about 20-25% faster on average against a 980ti. Six months from now it will be 50% and a 980ti will be getting 30fps in Battlefield 1 ()

Nvidia cards have the worst longevity. Maxwell will be good for a little while yet. We'll see 1500 core overclocked 980ti scoring about as well as a stock 1080 for a while. Once enough new games release though, it's going to go down like the Titanic, just like Kepler did.

What's odd is the 980ti doesn't seem to fall off nearly as hard as the 970/980 do.

 

[exar333]

Going from HBM to GDDR5X seems to be a step backward, so big Vega for me. Go big or go home.

HBM to GDDRX is a sidegrade/more or less equal. HBM2 to GDDRX would indeed be a step backward.

 

[DA CPU WIZARD]

Waiting for next gen when they can really utilize the new process node to its maximum potential. Also, awaiting the utilization of HBM2. Finally, a card built 100% for DX12.

 

[mohit9206]

I can see there are two types of people who will buy the 1080.
Those who are using an old gen card from 2-3 yrs ago and looking for a good performance upgrade.
And another is those who absolutely must have the fastest and latest card no matter what. These are the people selling their 980ti's and 'upgrading' to 1080 as soon as it comes out.
I'm neither of course. Am just an outside onlooker envying those who will be purchasing these cards

 

[Midwayman]

The main reason I'd be upgrading is to support 4k gaming, so nothing interesting at least until vega at least. Possibly later. Nothing on the horizon that my r9 290 won't still run at 1080p anyways.

 

[tential]

I can see there are two types of people who will buy the 1080.
Those who are using an old gen card from 2-3 yrs ago and looking for a good performance upgrade.
And another is those who absolutely must have the fastest and latest card no matter what. These are the people selling their 980ti's and 'upgrading' to 1080 as soon as it comes out.
I'm neither of course. Am just an outside onlooker envying those who will be purchasing these cards

If you own a 980Ti and don't upgrade I don't get it. It's a $200 upgrade maybe $250 tops after you sell your Ti. That's not much to stay high end...

It's the people from 2-3 years ago that I don't understand. Your Kepler 780Ti is on par with an R9 290, and you want a 1080 now? So that you can again hold onto your card for 2-3 years and have it perform... well hopefully not like 290, probably more like a fury nano/furyx?

$200 a year isn't bad to stay on the bleeding edge, however, buying a $650 GPU, then holding it for 2-3 years as it drops to a mid range trash tier GPU makes no sense.
You're paying the same as the person upgrading every year, but they have 2-3x the performance you do, because not only is the card they're upgrading to better, old nvidia cards just fall off. I've seen the 7970ghz above the 780Ti in some games. You're literally punishing yourself for no reason holding onto Nvidia high end cards, when selling them gives you far more benefit as you literally get to hold the fastest single GPU card crown forever for about $650 every 3 years... which well, is what you were paying ANYWAY!

 

[RussianSensation]

You sold me. If I am going to hold onto a card for 4-5 years and not care about performance when purchased. Then AMD may be the way to go.

With NV, starting in 2012, it's a double hit -- pay more upfront for marginal gains (980Ti OC vs. Fury X the only performance exception) and upgrade faster.

$299 280X > $380-450 770 2-4GB
$400 after-market 290 is now faster than a $700 780Ti
$550 980 is barely faster than a $280-300 290X for those cross-shopping when 980 launched.

Nearly every Kepler and Maxwell card in NV's line since 2012 has aged worse than nearly every competing AMD card in similar or even lower price ranges.

Just like 780Ti looks awful now against 290/290X/390, today a $550 980 is barely outperforming an after-market $280 290X (i.e., those prices remained in effect from October 2014 until 390 launched). That means the much hated on Hawaii chip managed to hang in there with 2 generations of NV x80 cards. Where is that $500-650 780 now? Ya, barely beating 7970Ghz/$299 280X.

Before 2012, NV wasn't any better. My 4890 cost less than GTX260 216 at the time of purchase and 4890 OC competed with 275 OC. During GTX470/480 era, HD5850 OC delivered stellar value. Then just dump that and get a faster card. My HD6950 cost $230 and unlocked into a 6970. At that time when I got the 6950, 580 was $450 on a great sale, while the cheapest GTX570 was $330 US.

At no point in time since I had 8800GTS did any NV card I was cross-shopping for an upgrade ever outlasted the ATI/AMD equivalent. Over time I started asking myself the question what is the NV premium for and I couldn't figure out an answer for myself.

Do you think NVIDIA will actively work to degrade performance in future drivers?

By shifting priority to Pascal drivers, by extension it means Maxwell may no longer get 100% optimization in the newest games. It's not about specifically coding drivers to cripple Maxwell but rather NOT coding the drivers to ensure that Maxwell performs as best as it possibly can over time.

If you own a 980Ti and don't upgrade I don't get it. It's a $200 upgrade maybe $250 tops after you sell your Ti. That's not much to stay high end...

It's the people from 2-3 years ago that I don't understand. Your Kepler 780Ti is on par with an R9 290, and you want a 1080 now? So that you can again hold onto your card for 2-3 years and have it perform... well hopefully not like 290, probably more like a fury nano/furyx?

$200 a year isn't bad to stay on the bleeding edge, however, buying a $650 GPU, then holding it for 2-3 years as it drops to a mid range trash tier GPU makes no sense.
You're paying the same as the person upgrading every year, but they have 2-3x the performance you do, because not only is the card they're upgrading to better, old nvidia cards just fall off. I've seen the 7970ghz above the 780Ti in some games. You're literally punishing yourself for no reason holding onto Nvidia high end cards, when selling them gives you far more benefit as you literally get to hold the fastest single GPU card crown forever for about $650 every 3 years... which well, is what you were paying ANYWAY!

This post is very good. It's either buy $550-700 cards and resell them quickly and get newer cards (so $100-200 cost per year) or buy $300-350 cards more often instead. Once 1070 launches, used 980Ti will probably drop to $350 or lower or almost half of its value a year ago. By summer 2018, I doubt 980Ti will be even worth more than $275 because with rebates, it should be possible to find 1070 for $325-330 in 12 months from now.

Buying a $700 high-end AMD or NV card and holding it for 3-5 years is a pretty bad strategy (unless you make $ with the card via professional work, rendering, mining, Photoshop, require it for research, etc. then you don't care).

Once enough new games release though, it's going to go down like the Titanic, just like Kepler did.

NV is also in a tricky spot. To showcase just how amazing Pascal is and push Vulkan/DX12/VR gaming, it's in their best interests to move on to Pascal ASAP. The greater the performance delta becomes between Maxwell and Pascal, the more impressive Pascal looks. This forces people who waited out on Maxwell to upgrade since they now feel so much better having skipped a generation that fell apart in latest modern games. I hope Kepler was more of an exception rather than the rule but recent benchmarks of 290/290X/390X vs. 970/980 aren't promising.

 

[topmounter]

Waiting to see what AMD announces. Regardless, I'll have a hard time buying a card that doesn't support FreeSync.

 

[digitaldurandal]

It begins...
980 Ti aftermarket 475 with 25 rebate.

http://slickdeals.net/f/8731567-evg...gddr5-video-card-475-after-25-rebate-shipping

 

[Maverick177]

HBM to GDDRX is a sidegrade/more or less equal. HBM2 to GDDRX would indeed be a step backward.

As of now, the only downside of HBM is its capacity, and after almost a year wit a Fury, I'm confident to say having 4Gb @1440p AS OF NOW, is a non issue.

 

[DaveSimmons]

If you own a 980Ti and don't upgrade I don't get it. It's a $200 upgrade maybe $250 tops after you sell your Ti. That's not much to stay high end...

I don't care about benchmarks or bragging rights.

My 980ti is working well, why should I make the effort to replace it? I could spend that time reading a book, watching a movie, playing games, posting here.

There's also a chance I'd get a lemon card like my first EVGA 560ti and have the hassle of doing an RMA. I ain't got time to bleed!

I should get around to selling my 680ti eventually though. I won't get much but someone on a budget would benefit.

 

[xthetenth]

If you own a 980Ti and don't upgrade I don't get it. It's a $200 upgrade maybe $250 tops after you sell your Ti. That's not much to stay high end...

It's the people from 2-3 years ago that I don't understand. Your Kepler 780Ti is on par with an R9 290, and you want a 1080 now? So that you can again hold onto your card for 2-3 years and have it perform... well hopefully not like 290, probably more like a fury nano/furyx?

$200 a year isn't bad to stay on the bleeding edge, however, buying a $650 GPU, then holding it for 2-3 years as it drops to a mid range trash tier GPU makes no sense.
You're paying the same as the person upgrading every year, but they have 2-3x the performance you do, because not only is the card they're upgrading to better, old nvidia cards just fall off. I've seen the 7970ghz above the 780Ti in some games. You're literally punishing yourself for no reason holding onto Nvidia high end cards, when selling them gives you far more benefit as you literally get to hold the fastest single GPU card crown forever for about $650 every 3 years... which well, is what you were paying ANYWAY!

Flipping a card is getting worse on the NV side. Now you need to sell before the reveal and then either suck it up and eat the founder's extra charge or spend even more time without a card. Time without a card sucks.

 

[jmartin357]

I ordered a pair of 1080s from falcon nw in a fancy new fragbox. first new pc since 2009*, first upgrade since I crammed a 7970 into that ancient beast in 2012.

hard to express how excited I am!

(*excluding the 4 month ordeal I endured with cyberpowerpc last year. for the love of humanity, never buy anything from that company)

 

[badb0y]

I really don't know yet but I bought XB270HU and I really don't like low FPS (basically anything under 75 FPS) so I want to see the performance increase. In all honesty though I hope AMD comes out swinging because if they can offer me similar performance at a lower price I will sell my monitor and go to the AMD camp. $699 for a mid-range GPU nVidia can go **** itself.

 

[mohit9206]

If you own a 980Ti and don't upgrade I don't get it. It's a $200 upgrade maybe $250 tops after you sell your Ti. That's not much to stay high end...

It's the people from 2-3 years ago that I don't understand.

$200 a year isn't bad to stay on the bleeding edge.

But why this addiction to upgrade every year just to stay on the bleeding edge?
Although the $200 a year is good excuse to do so. You should also know that by buying overpriced Nvdia flagship cards every year , you are also contributing to the increasing flagship prices because Nvdia can see that people are buying their cards regardless of the price. Although to you it just cost $200 to upgrade, it gives Nvdia the confidence to price their next year's flagships even higher which keeps driving the prices of all tiers of card upwards. So people who upgrade every year to the newest fastest card are just as responsible for the current state video card prices.
So you guys have the power to stop this. Don't upgrade unless you absolutely need the additional 25% performance. Nvdia will notice lower sales and drop prices accordingly.

 

[Azix]

next upgrade is going to have to destroy the 980ti and fury x completely.

 

[tential]

I don't care about benchmarks or bragging rights.

My 980ti is working well, why should I make the effort to replace it? I could spend that time reading a book, watching a movie, playing games, posting here.

There's also a chance I'd get a lemon card like my first EVGA 560ti and have the hassle of doing an RMA. I ain't got time to bleed!

I should get around to selling my 680ti eventually though. I won't get much but someone on a budget would benefit.

It has nothing to do with benchmarks or bragging rights at all. Like I said you're still spending the same amount of money whether you buy the 980ti and hold it for 3 years or get a new gpu each year and sell the old one.

Over 3 years you can either
A) hold into 980ti
B) get a 980ti sell it, 1080 sell it, 1080ti sell it.

Same price, but you get to avoid performance degradations that are now part of owning Nvidia gpus.

Look at 780ti owners. Why would you hold onto a card that performs worse than the midrange card it was demolishing at launch? Instead you could have sold your 780ti and had a 980 which was still faster than the 290x,then have gotten a 980ti and still be faster than the 290x.

Without making regular upgrades to Nvidia hardware your performance regresses. No one buys a 780ti saying "gee I hope this performs like a 290 or worse soon!" you expect great performance.

Even if you upgraded to the 980, guess what? The same gpus (290/290x) are at your performance level. And now if you're a 980ti owner? Well pascal is coming up so we see the 390x begin to creep up on the 980ti.

If you want Nvidia flagship, the buy and sell them it's the most efficient way of spending the SAME amount of money but retaining performance.

Otherwise, why buy a 780ti for 3 years when you could buy an amd midrange and have the same average performance over the lifetime of ownership for far less money.

 

[tential]

But why this addiction to upgrade every year just to stay on the bleeding edge?
Although the $200 a year is good excuse to do so. You should also know that by buying overpriced Nvdia flagship cards every year , you are also contributing to the increasing flagship prices because Nvdia can see that people are buying their cards regardless of the price. Although to you it just cost $200 to upgrade, it gives Nvdia the confidence to price their next year's flagships even higher which keeps driving the prices of all tiers of card upwards. So people who upgrade every year to the newest fastest card are just as responsible for the current state video card prices.
So you guys have the power to stop this. Don't upgrade unless you absolutely need the additional 25% performance. Nvdia will notice lower sales and drop prices accordingly.

Uh, these are people who buy flagship gpus anyway and buy Nvidia products anyway. Why would I want them to suffer if there is a way they don't have to? No matter the gpu company you want to buy if you can get a lower total cost of ownership for higher levels of performance I will of COURSE recommend that to people.

Holding on to a 980ti for 3 years just isn't a good decision. If you really want a longterm gpu I'd have to recommend amd, but when you can get the same thing from Nvidia and if you prefer Nvidia, then I have a strategy for you to keep purchasing Nvidia flagship even and not be price gouged buy reselling your gpu while it still retains the vast majority of its resale value.

 

[maddie]

Uh, these are people who buy flagship gpus anyway and buy Nvidia products anyway. Why would I want them to suffer if there is a way they don't have to? No matter the gpu company you want to buy if you can get a lower total cost of ownership for higher levels of performance I will of COURSE recommend that to people.

Holding on to a 980ti for 3 years just isn't a good decision. If you really want a longterm gpu I'd have to recommend amd, but when you can get the same thing from Nvidia and if you prefer Nvidia, then I have a strategy for you to keep purchasing Nvidia flagship even and not be price gouged buy reselling your gpu while it still retains the vast majority of its resale value.

Unfortunately, this only works with active used component markets. Not for most of the world.

 

[BHZ-GTR]

I am waiting for a more powerful graphics card, Like VEGA

 

[railven]

But why this addiction to upgrade every year just to stay on the bleeding edge?

Stay bleeding edge? No, because I can - yes. I always upgraded my AMD/ATI card when the refreshes hit. My hobby is gaming. Compared to other people's hobbies, I have a dirt cheap hobby, in my opinion. People drop thousands of dollars into a sound system, or their car, or even their wardrobe. I drop my expendable cash (that never comes out of necessities) on gaming. Consoles/PC hardware, etc.

(I admit it also helps that I get paid well, and the only debt I have is my mortgage )

Although the $200 a year is good excuse to do so. You should also know that by buying overpriced Nvdia flagship cards every year , you are also contributing to the increasing flagship prices because Nvdia can see that people are buying their cards regardless of the price. Although to you it just cost $200 to upgrade, it gives Nvdia the confidence to price their next year's flagships even higher which keeps driving the prices of all tiers of card upwards. So people who upgrade every year to the newest fastest card are just as responsible for the current state video card prices.

I don't think that's how it works. There is a ceiling. NV will hit it. But people continue to cry crocodile tears citing the inflated Founder's Price. If we have a slew of non-Founder's cards on the market, I hope every single person here that keeps claiming $450/$700 eats crow. If the AIB's gouge us, I'll serve my crow (I've had plenty of it in my life time haha).

The inverse, when I kept paying AMD's increased prices, I kept reinforcing them to up the price. So, having bought way more top tier AMD cards in my life time - I apologize ahead of time if the Fury X successor is >$650.

So you guys have the power to stop this. Don't upgrade unless you absolutely need the additional 25% performance. Nvdia will notice lower sales and drop prices accordingly.

Don't upgrade? Why? NV is going to fleece where NV can. If Vega 10 is >$650 are you going to tell us not to buy it either?

 

[biostud]

I'm waiting for dp1.4 monitors until I decide if I'm going nvidia or amd.