it is mid range, if i am correct gtx 1080ti, XP and V are high end,
as gtx 680 and 690 are still high end and it wont change

@David988 gtx 680 has a 298mm^2 die... it was meant to be called gtx 660 back in the day, but since AMD's HD7970 was a 352mm^2 die GPU that performed on par with it and AMD overpriced it at 550$ MSRP, while the hd6970 a single year earlier a 389mm^2 die GPU was 350$... And since Nvidia is actually good at marketing, they knew that they had to name their 550$ product a GTX X80, in this case, gtx 680 to be at the same price as the gtx 580.

So the bottom line is the gtx 680 was meant to be called a gtx 660 and cost 250$, but it ended up as a gtx 680 costing 550$ just like that... -_-

For as long as I can remember, only thing die size determined was how many transistors you can pack in (a given surface). Never ever have I seen any GPU rating being determined by die size. Companies determine value. My RX 580 dishes out slightly lower performance in most games than my friends 1060 does (-5 fps), while costing 190$ less. I've only ever heard you use die size as a tier chart. Even my professor in Computer architecture said that performance is what determines tier, crossing out the enthusiast GPUs (like titan). All that info you just dished out means squat and is completely illogical.

If anything, you compare the same generation GPU and determine the tier based on that. From 1030 to Titan. I'd even exclude Titan since it's for really rich people (a perverted business). Fabrication process stands by itself and is usually lower as generations go, until the lowest possible is reached before quantum computing. Don't see how entry/mid/perf tier has anything to do with die size. What I meant by illogical is that there can be the same performance for different sizes while you can have different performance for one size, so using die size to compare performance is not really definite.

I don't tend to agree with Psychoman very often here, but he is right in this case. Chip size is what determines the tier of the GPU.
But due to the lack of competition on the GPU front and Nvidia's damn brilliant marketing department, they have managed to effectively persuade the public that performance is what matters and what determines the tier standing of GPUs.

Arguing about the GPU tier is as useful for the common person as arguing about whether the earth is flat; in so far as its impact on our day-to-day lives.
As Psychoman has himself said multiple times, when he buys a GPU, he considers cost to performance. Ergo the tier does NOT matter unless you want to be bitter that you have to pay too much money for a mere "mid tier" card.

Same goes for profitability. Oh this company is selling its product at a loss, it must be a great deal.. I should buy their product. And by this logic, no one should ever buy things like bottled water, air filters, and hdmi/usb cables.
Yes, its a great idea for money to flow to the inefficient companies instead of the ones who manage to keep their costs down.

the gtx 1080ti/titan Xp is a 471mm^2 die size GPU and a GP102 chip, the GP100 is a 601mm^2 die chip that is unreleased.

The gtx 1080 is thus mid-range as it's half of what Nvidia can produce, well 51.9% to be exact. And die size is not about performance as I said many times and tiers aren't about performance... at 40nm a 300mm^2 die size GPU would perform worse than at 14nm 300mm^2 die size GPU if you just added more cores, without optimizing the architecture for it or improving the architecture, which would lead to better performance, but huge overhead if just added more cores.

@Penoys absolutely, but it shows how much it costs to them and thus how much more they are charging you, also shows how much better there is and/or can be. the GP100 would have been 80-90% faster than a gtx 1080... but there just wasn't a sane price for them to sell it at for a 100%-134% profit margin... I'm sounding like a broken record, but there is ignorance I can and ignorance that I can't pass by... this I'm tired of letting it pass by since 2011... -_-

Also, money flowing to high profit big and mega business is bad... they rarely use it and that money disappears from the circulation... Companies that sell at low-profit margins usually do due to actually using the extra money to expand their current divisions and/or make new ones... That's why many investors have withdrawn from Apple, even though Apple is set to soon be a trillion dollar company, they just don't have enough sources of income for their huge capital.

But why would you rate the GPUs by their die size today? Die size was a major contributer in the early 2000s but today, it's mostly cost to performance ratio and in Croatia, AMD wins on this front, mainly because 1060 is -200$ more expensive than RX 580, while providing a bit more FPS in gaming and decreased single core performance (mining/gaming/processing/editing/solving in Wolfram/...).
So the question still stands, why would you rate them by their die size when you can rate them by their performance in programs/games, from benchmarks, over analytical examinations in physics, to "Can it run Crysis 3?"

When I say cost to performance, I'm looking at my country, not the MSRP.
So, either rate it by cost to performance, absolute performance or don't rate it at all, because rating it by the die size is plain obsolete.

i canot remember where i readed it canot find it again but it was about gpu hierarchy and theyr classification entry, med, high end level, and there they sayed that 680 is high end, as i understood it for example for 10years from now gtx 1080ti will be usless and people will call it entry level or something but it will be actually still be high end.
there is also chance that that site was full of crap and that i am completly wron bec of it

i think its perception more than anything like current gpus most people would see them as gt 1030 as low end gtx 1050 low mid 1050ti 1060 mid 1070 and 1070ti high end and 1080 and 1080ti very high end (thats just an example people dont get your balls in a knot) anyways so if they were talking about that same series when it first launched like i mentioned then yes it makes it perfect sense they would say that but the title doesnt carry over generations like the 290x for example still a damn excellent card is arguably midrange now even though a couple years ago it was flagship for amd

@David988
and you are right a 471mm^2 die chip is high-end even with the rtx 2080ti being this huge.
the gtx 680 on the other hand was 298mm^2 and was never high-end.

This useless discussion started when you misunderstood the original commenter and responded with the wrong idea of what the person actually said.
These requirements ARE extreme for such a game. The 1080 wasn't called an extreme range card, rather than its need for this damned game. It's like requiring anything above an AMD APU or Intel HD for a Tetris game.
Not gonna go any further, since this comment is at risk of getting misunderstood as well. :)