Bamda
Posts: 519 +285
This is the GPU I have been using since launch. I will not be upgrading till I see a major reason to do so and NO Ray Tracing isn't one of them.
How Does the GTX 1080 Ti Stack Up in 2020?
- Thread starter Steve
- Start date
harm9963
Posts: 128 +69
SilentMarket
Posts: 49 +33
Evernessince
Posts: 5,469 +6,158
And yet you've yet to address the fact that steam has and does still have ongoing issues with it's numbers. We are all well aware of their issues with Internet cafes in the past (and likely still continuing). You've also yet to address the Chinese language issue as well.
https://www.pcgamer.com/chinese-is-...ar-language-according-to-its-hardware-survey/
I've proven that the survey has had and continues to have issues. You need only google steam survey to see the number of articles written about the survey's issues.
What have you got to back your argument up?
FYI the question was never "Is steam a gaming app". It's "Does steam's data collection methodology work?". There is clear evidence it has problems.
Adhmuz
Posts: 2,343 +1,186
After looking at the results of the last few months of surveys Valve did some correction to their results. The 1060 has drop nearly 4 whole percent points since the December poll and Chinese has also dropped to just shy of 22%, the ladder I can't find a way to look back to December unfortunately.
Even Nvidia's market share has been reduced to 77.8%.
These look a little more accurate, and even if they aren't 100% accurate what are you saying the market shares should be?
As far as representing what gamers on PC are using hardware wise what else do we have to go by?
At least Valve admits they have issues and are actively resolving them, I can't see it being difficult for them to analyze the data and remove duplicates based on a devices hardware ID or mac address.
Evernessince
Posts: 5,469 +6,158
Valve doesn't collect hardware ID anymore after people got pissed at them for data harvesting. If they did still use more identifiable data the survey might be more accurate.
Admitting you have an issue is not a remedy to solving it nor is it a plus in any way, shape, or form. That goes double when that issue has been ongoing for awhile now.
Sales of course. You know, an actual representation of what gamers are buying and using.
Both Amazon and Mindfactory release sales data.
The fact that the steam survey needs to go through such massive corrections, gaining and then dropping 13% of an entire language base in a single month (this was before corona mind you as well) should throw a huge red flag. If any other organization did a survey like this, they would be out of the business of doing surveys. You can't just take your eraser and undo past results by your own admission were incorrect.
The 1080ti came out 3/2017. The 5700XT was released 6/2019. Not really a fair comparison. How about the 5700XT vs the 2070 Super? Now that five-year revisit makes sense.
As for how well cards age, AMD tends to release their gpus not "prime time ready", then take a year or more to get them in proper working order. But if I'm paying the money now, I want the performance now. The Green Team delivers right out of the box.
Last edited:
krizby
Posts: 429 +286
Well PC hobby is not that expensive and owning the best GPU can save you lots of time with fine tuning graphical settings for playable FPS, you pretty much set everything to Max with a 2080 Ti.
Not only that the 2080 Ti did not devalue at all after 1.5 years, new 2080 Ti are still selling for 1200usd+ and believe it or not 2080 Ti is still selling really well according to data from mindfactory.de (being the 4th in revenue behind 2070 Super, 5700XT and 2080 Super).
Last edited:
Alfatawi Mendel
Posts: 299 +497
Mithan
Posts: 135 +150
I own mine, almost 3 years old to the day. No plans to change it unless the 3080Ti costs the same at release and offers a 50-70% performance boost. Unlikely..
Im sure 4k gaming was even less prevalent when it first came out in 2017. Toms Hardware, Anandtech, pcMag and even Techspot had 4k benchmarks with its review years ago.
DAOWAce
Posts: 325 +64
Still on my 1080 Ti because there's zero 2xxx equivalent that's actually an upgrade. No, the $1,200 2080 Ti doesn't count.
Struggling to run 3440x1440@120Hz. 60FPS is doable in most titles, sure, but some do drop below it. However my panel is 144Hz and seeing 60Hz/FPS again after being at 120Hz for so long is.. pretty disgusting. G-sync does help, but it's still very noticeable when the framerate drops in the 80's. Unfortunately it requires exclusive fullscreen because windowed gsync is still broken after all these years. I don't typically run fullscreen because borderless is extremely useful to alt+tabbing, but if it's a very demanding title and doesn't really lend itself to alt+tabbing (single player games), I'll resort to it.
Maybe NVIDIA will actually release a proper GPU update this year and not double the price of it so those of us who will never do SLI again can have a good 21:9/4K experience.
Struggling to run 3440x1440@120Hz. 60FPS is doable in most titles, sure, but some do drop below it. However my panel is 144Hz and seeing 60Hz/FPS again after being at 120Hz for so long is.. pretty disgusting. G-sync does help, but it's still very noticeable when the framerate drops in the 80's. Unfortunately it requires exclusive fullscreen because windowed gsync is still broken after all these years. I don't typically run fullscreen because borderless is extremely useful to alt+tabbing, but if it's a very demanding title and doesn't really lend itself to alt+tabbing (single player games), I'll resort to it.
Maybe NVIDIA will actually release a proper GPU update this year and not double the price of it so those of us who will never do SLI again can have a good 21:9/4K experience.
Evernessince
Posts: 5,469 +6,158
It also has more memory too. It wasn't until more recently that Nvidia finally caught up price wise to it's 3 year old card. The 2080 launched at $800 while the 1080 Ti was selling used and often in warranty for $500.
amstech
Posts: 2,662 +1,825
That doesn't really prove anything.
But the discrete GPU market share is almost uniform with Steams results, and that proves more then anything you've said.
Tbf AMD are usually first to a new mode, Radeon VII was the first GPU to 7nm it's just ashame it was just s cooler overclocked Vega 64 with 4 compute units less. 5700 and 5700XT should be due a price cut, the 5700 although competes against the 2060 super and the 2070 in some games should cost a lot less due to its physical size as a chip. RX480 was one of the first GPU's to 16nm and that cost $200-230 when released but some how for a similarly sized chip AMD want to charge $100-150 more for these new cards because they beat Nvidia's stupidly priced cards for less than what Nvidia charges.
Evernessince
Posts: 5,469 +6,158
Steam has AMD at 13%, other's have AMD at 27%
AMD has sold more GPUs than Nvidia, according to this analyst report
Team Green is falling behind, though not by much
That's not "almost uniform". That's MILLIONS of PCs in difference.
Thank you for further proving my point that steam's numbers are completely inaccurate, if a massive swing in the chinese language alone weren't enough.
I don't see how a single correct number (which you are having trouble finding) could have undone all the mistakes valve is and continues to make either. Definition of cherry picking (or trying to at least
).
Since the 2080Ti is roughly 20%-30% faster than the 1080Ti (which I have too), it seems like a safe bet that the 3080Ti(?) will be 20%-30% faster than the 2080Ti- and so a 50%-60% upgrade over 1080Ti, plus RTX. Well worth an upgrade if you have the money. Rumor has it that prices will be lower this time around. Fingers crossed...
There's no doubt that 3440x1440p has way more pixels than 2560x1440. It's NOT just stretched out, it's about 34% more pixels to render! That's quite a difference. Not 4K, but still a huge pixel increase.
I have a similar setup to yours: 3440x1440p 120Hz monitor (Alienware). I often cap it through Riva Tuner to 96Hz to prevent framerate swings. I'm surprised you're struggling to hit 60fps.
Forget Ultra settings if that's your thing; go with High. The difference is imperceptible.
Last edited:
N/T
pencea
Posts: 351 +337
How does the Fury X hold up against 980Ti in 2020
Anyways a 700usd 2017 card took a dump on a 400usd 2019 one is a massive shame already. I still remember how the 250usd HD 4890 killed the 650usd 8800 GTX which released 2 years prior. How the mighty ATi have fallen...
His not wrong though. Take a look at the revisted article not too long ago. AMD's 7970 have aged much more gracefully than the GTX 680. It's now a good deal faster especially in modern titles like Doom Eternal.
Last edited:
krizby
Posts: 429 +286
HD 7970 was released at the time when AMD GPU division was still competencce, I reckon that level of competency fell right off after the release of R9 290 which I think was the best GPU of its time. Fury X, Vega 64, Radeon VII are living proofs of the incompetence that is Radeon Technology Group.
But anyways 5700XT will never reach the performance of 1080 Ti, ever. 5700XT just doesn't have any of the advanced optimization that come with DX12 Ultimate that would counter the raw throughput and VRAM size advantage that 1080 Ti has.
Last edited:
pencea
Posts: 351 +337
To be fair both the 5700 & 5700 XT are not that bad. Sure they don't feature the fancy RTX nor have top right performance but they do offer reasonable performance for what they are.
Given AMD being a much smaller company than Nvidia not to mentioned have significantly lower R&D budget than its rival. it's quite impressive what they can still do. Especially their CPU division.
QuantumPhysics
Posts: 6,306 +7,258
Exactly:
And I will continue to accept Steam's results - which techspot themselves has posted on several occasions over something that "some random avatar" on the internet says.
