RTX 2080Ti crazy prices ...

Surly said:
0% chance of this happening in first half of 2019. Zero.

A snowballs chance of it being released a year from now. But most are predicting xmas 2019 or Q1 2020.
Click to expand...
Gap from Vega FE to Vega 64 wasn't anywhere near that long. AMD is not the old Raja shitposting marketing fest it once was, remeber threadripper rumours? No you don't because they kept that shit tight as a drum. Even co-workers didn't know.
Don't think RTG could pull something like that with a bit of help from the Zen team?
 
Got my 1080 last November before the prices went crazy. I'll happily stick with it for a while longer. It handles all the games I play just nicely.
 
Every generation of the Titan has been >$1000. Hell the Titan V is a cool $3000. I don't see think anything has really changed here.
 
AltTabbins said:
I dont see a RTX Titan. I wouldn't be surprised if we didnt. I think they are going to keep this the top end and move that entire line to Quadro. Just speculation.
Click to expand...

So moving the Titans to Quadro and giving Ti's Titan pricing is equivalent to nothing changing?
 
RamonGTP said:
So moving the Titans to Quadro and giving Ti's Titan pricing is equivalent to nothing changing?
Click to expand...

Oh, I think things are changing for sure. Nvidia knows people are going to buy the Ti at the price point. Hell, I bought 2.
 
RamonGTP said:
wow... impressive. I think it's been about 2 decades since I've seen anyone reference 3D Mark as a true gauge of performance, and was serious about it.
Click to expand...

Time Spy for DX12 and Fire Strike for DX11 - what do you recommend other than those two?
 
RamonGTP said:
Games. Real ones
Click to expand...

Ahh...that was a snarky "I don't play benchmarks" post.

For me, it's easier to use benches to quickly assess performance / determine bottlenecks / find issues (versus other setups).

I use games to determine overclock stability. Since most games don't have built-in benchmarking tools these days so it is hard to compare apples to apples.
 
sk3tch said:
Ahh...that was a snarky "I don't play benchmarks" post.

For me, it's easier to use benches to quickly assess performance / determine bottlenecks / find issues (versus other setups).

I use games to determine overclock stability. Since most games don't have built-in benchmarking tools these days so it is hard to compare apples to apples.
Click to expand...

Well that's the problem, static benchmarks aren't an accurate form of measuring the performance difference of two different cards. They're good to asses performance gains from the same card before/after an overclock and not a whole lot more beyond that.

Has nothing to do with being snarky. If you don't believe me, just take a look at the reviews for 20 series cards. They have both static benchmarks and real game benchmarks. The performance difference between 20 and 10 series cards is never as high in actual games as it is in something like 3dMark.
 
RamonGTP said:
Well that's the problem, static benchmarks aren't an accurate form of measuring the performance difference of two different cards. They're good to asses performance gains from the same card before/after an overclock and not a whole lot more beyond that.

Has nothing to do with being snarky. If you don't believe me, just take a look at the reviews for 20 series cards. They have both static benchmarks and real game benchmarks. The performance difference between 20 and 10 series cards is never as high in actual games as it is in something like 3dMark.
Click to expand...

Wider gap in performance difference in static benchmarks doesn't mean it isn't accurate. If you do a complete apples to apples comparison the performance difference is accurate.

If you're worried about 20 series to 10 series, my static benches show that DX11 is better with the 1080 Ti and the 2080 is better at DX12. Game-wise, there's a difference because of the varying engines used AND the difficulty in completely replicating a bench in a real game.

I'm not a professional reviewer so I use something like 3DMark to get benchmarks of my system. I'm not going to run through 15 games and do compare and contrast because like you said - the reviews have that. I do run through the main games I play (ARK, BLOPS4) on my main box (the 2080 is on my guest box) - so I can tell you all about that but I only use two different resolutions (two systems) so that's not as interesting and using something standard like 3DMark which everyone can very easily replicate.

In my case, I'm using a 5820K which is at least different than most and perhaps some people will find value in that versus the reviewers who all have more modern CPUs. Hence why I shared. Isn't this why we're here?
 
RamonGTP said:
Well that's the problem, static benchmarks aren't an accurate form of measuring the performance difference of two different cards. They're good to asses performance gains from the same card before/after an overclock and not a whole lot more beyond that.

Has nothing to do with being snarky. If you don't believe me, just take a look at the reviews for 20 series cards. They have both static benchmarks and real game benchmarks. The performance difference between 20 and 10 series cards is never as high in actual games as it is in something like 3dMark.
Click to expand...

A problem with a lot of the reviews is they run into CPU bottlenecks with the 2080ti and are either too dumb to realize it or don’t mention it. I’ve seen some avg in 60Hz capped games...
 
Dayaks said:
A problem with a lot of the reviews is they run into CPU bottlenecks with the 2080ti and are either too dumb to realize it or don’t mention it. I’ve seen some avg in 60Hz capped games...
Click to expand...

It's literally every review out there, so no one knows how to benchmark 20 series cards? I doubt it.

sk3tch said:
Wider gap in performance difference in static benchmarks doesn't mean it isn't accurate. If you do a complete apples to apples comparison the performance difference is accurate.

If you're worried about 20 series to 10 series, my static benches show that DX11 is better with the 1080 Ti and the 2080 is better at DX12. Game-wise, there's a difference because of the varying engines used AND the difficulty in completely replicating a bench in a real game.
Click to expand...

I didn't say there isn't a difference. I said the difference, in actual games (which is what you'll be using it for) is no where near what the static benchmarks would lead you to believe. So yes, it certainly does mean it isn't accurate. When static's show 50% improvement and games show 20 or less, that's pretty damn inaccurate.

I'm not a professional reviewer either, you don't need to run through 15 games. The work has been done for you by dozens of pros already. Including the ones that run this site.
 
RamonGTP said:
It's literally every review out there, so no one knows how to benchmark 20 series cards? I doubt it.



I didn't say there isn't a difference. I said the difference, in actual games (which is what you'll be using it for) is no where near what the static benchmarks would lead you to believe. So yes, it certainly does mean it isn't accurate. When static's show 50% improvement and games show 20 or less, that's pretty damn inaccurate.

I'm not a professional reviewer either, you don't need to run through 15 games. The work has been done for you by dozens of pros already. Including the ones that run this site.
Click to expand...

The reviews actually had huge differences. AdoredTV did a whole episode on it. A lot of review sites are fucking terrible.
 
RamonGTP said:
It's literally every review out there, so no one knows how to benchmark 20 series cards? I doubt it.



I didn't say there isn't a difference. I said the difference, in actual games (which is what you'll be using it for) is no where near what the static benchmarks would lead you to believe. So yes, it certainly does mean it isn't accurate. When static's show 50% improvement and games show 20 or less, that's pretty damn inaccurate.

I'm not a professional reviewer either, you don't need to run through 15 games. The work has been done for you by dozens of pros already. Including the ones that run this site.
Click to expand...

Yeah, that's why maybe you should use 3DMark. The results are nowhere near 50% gains and not 20%, either.

2080 Ti - Fire Strike Extreme 1.1 - 15,292 --- +9.4%
https://www.3dmark.com/fs/16712695

1080 Ti - Fire Strike Extreme 1.1 - 13,977
https://www.3dmark.com/fs/13179751

2080 Ti - Time Spy 1.0 - 13,578 --- +30%
https://www.3dmark.com/spy/4725718

1080 Ti - Time Spy 1.0 - 10,446
https://www.3dmark.com/spy/2112885

EDIT: full disclosure - the 1080 Ti is a MSI Lightning Z OC'd - so the gap is wider if both are equal (my 2080 Ti is not OC'd in these runs). Each result listed above has very detailed notes about the configuration - which is why I love 3DMark because it keeps a database!
 
Last edited:
Dayaks said:
The reviews actually had huge differences. AdoredTV did a whole episode on it. A lot of review sites are fucking terrible.
Click to expand...

Then we'll have to agree to disagree on what huge means. 980Ti > 1080Ti was substantial (around 70% iirc) 1080Ti > 2010Ti isn't even half of that while at the same time moving the pricing tier up to Titan levels. It's the best shittiest card on the market right now with the paltry performance gains over last gen while at the same time occupying Titan pricing.
 
RamonGTP said:
Then we'll have to agree to disagree on what huge means. 980Ti > 1080Ti was substantial (around 70% iirc) 1080Ti > 2010Ti isn't even half of that while at the same time moving the pricing tier up to Titan levels. It's the best shittiest card on the market right now with the paltry performance gains over last gen while at the same time occupying Titan pricing.
Click to expand...

I mean... some sites got 45% better and others 10% better for a certain game.
 
Dayaks said:
I mean... some sites got 45% better and others 10% better for a certain game.
Click to expand...

So what you're saying is... If you take the best case scenario you can find for a 1080Ti > 2080Ti, it's still SIGNIFICNALY less than the average performance gain we saw from 980Ti > 1080Ti, while at the same time occupying Titan pricing.

I think you just helped prove my point.
 
RamonGTP said:
So what you're saying is... If you take the best case scenario you can find for a 1080Ti > 2080Ti, it's still SIGNIFICNALY less than the average performance gain we saw from 980Ti > 1080Ti, while at the same time occupying Titan pricing.

I think you just helped prove my point.
Click to expand...

It’s been a while since I looked at benchmarks and I am not going back to debate site vs site with you.

I can say in my case (high IQ 60Hz) this card is noticeably better than a 1080ti, it’s basically the best case. All I was saying is under good conditions (not CPU limited) the 2080ti has been close to 50% faster than the 1080ti. (Dx12, vulkan especially) If DLSS ever makes an appearance it can be up to 100% faster.

A decent portion of the die is RT cores and tensor cores. If those upcoming techs mean nothing then the value would appear low.
 
Last edited:
Wait for the next gen. This one is a dud. It's not worth upgrading until they've got something that can really bend 4k/120fps over a log.
 
Last edited:
I think that's the only 2080ti currently in stock. I got the zotac amp ti one preordered for the same price but it sold out in about 15 min and now its listed for 100 more on the same site.
 
When I used to do benchmarks on video cars all the time, I got paid for it. I play games on my video cards and so far the 2080 ti isn't making a case for me. I also have way too many hobbies in my life and will probably divert funds towards those hobbies due to these crazy pricing levels
 
I think a used 1080ti is the best option out there right now for anyone not trying to spend outrageous money on a upgrade.
 
These are the special arsonist edition. They will not only spontaneously combust, but take a few of your rig components while doing so...:D
 
Furious_Styles said:
Him and 5 others scored big today.
Click to expand...

I actually had it in my cart about 1 minute after he posted and just need to press the buy button. Decided against it for now. $1350 for a video card just started to seem more and more crazy as I went through checkout.
 
You must log in or register to reply here.
Back
Top