2080ti 2080 Ownership Club

There was that one guy on OCN that did the shunt mod on a Titan V, and then the heatsink touched the mod, and the whole thing died when it shorted out and blew, 3k toast ouch.
 
Mine is on the way.. I just hope it fits. Most of these cards are over a foot long which is somewhat worrying.
 
Sup fellow 2080ti owners! I wanted to ask you guys...

Have you noticed higher than advertised boost clocks even when running stock with no overclocking? I keep seeing clock speeds of 1900-2030MHz, verified it in GPU-Z too. Just found it to be peculiar.

Yup. Mine settles in at 1985 after a couple hours of intensive gaming. The performance at 2560 x 1440 has been PERFECT and I'm finally satisfied. I think this is the longest I've ever gone without overclocking a new GPU... The need just hasn't been there yet.
 
There was that one guy on OCN that did the shunt mod on a Titan V, and then the heatsink touched the mod, and the whole thing died when it shorted out and blew, 3k toast ouch.

I only did the one shunt on the back of the card and also flashed the GALAX 380W bios. I am 99% sure my waterblock is acetal where the 2nd shunt is at the end of the card if I ever felt the need. I would have to run two leads out beyond the waterblock though so I can easily change resistance. It’d be crazy to be taking the block on and off dialing it in.

For you guys settling around 2 Ghz stock that’s really good. Most stable OCs are around 2.1Ghz so you’re not exactly leaving much on the table anyways.

This card is a beast. I finally have enough GPU to anti alias my favorite game (World of Warships). DSRing 6000x2500 with 4xMSAA down to 3440x1440. :cool:

shot-18.10.16_22.13.29-0030.jpg


Beat Saber also looked great at 4k per eye. Only at 7ms (11ms max for 90 Hz) IIRC.

To hold 2100Mhz / +1000 (8Ghz mem) I think I am around 400 watts lol. 28C idle/50Cmax on the waterblock with 25C liquid.
 
Last edited:
Nice!! If it’s a reference board throw the 380W GALAX bios on it!

Of course it’s your choice and all consequences / risks are yours. ;)
I'm not a risk taken.

Great card so far. Able to play 4k without any issue.

Also have a evga setup in progress. 1080 Ti will replace with another 2080 Ti.

Setup was under $400.
 
So, another replacement card came, and this one seems to be working with no issues whatsoever. The LED issue with the other card came back though in SLI mode. It's been turning on and off sporadically, don't know why. Also, whereas the new card lights up on system start, that one, even when it worked, didn't light up until Windows booted. And it was a different shade of green. Outside of the LED issue though, everything seemed to be working.

A couple things. Time Spy seems to work great when XMP is enabled (18045).

https://www.3dmark.com/spy/4752505

Time Spy Extreme was pretty much the same score with (10951) or without XMP (10934).

https://www.3dmark.com/spy/4752589
https://www.3dmark.com/spy/4689594

The other people with similar systems all have overclocks on their system though, so not too worried.

On game results, if you're not running 4k or above, don't bother with SLI. With or without SLI enabled, The Witcher 3 was pretty much at 120 fps. I'm guessing this is where the 2950x caps out. 4k, I've got a solid 60 fps though, with every option, yes, even hairworks, turned on. Monster Hunter seems to run at 80 fps min.

The one issue I did have, is that 3dMark didn't run visually. The screen pretty much froze before even loading came up, but the test ran, and it did collect data. I'm guessing this is a driver issue, as I can't possibly see how hardware would cause that problem. (I.e. Once the test collected, it went onto the next loading screen in a frozen state, but the system was still responsive).

*EDIT*

So, changing the drivers, 3DMark displayed. It was weird, because it never even displayed the loading bar. And Single Card mode worked fine, just not SLI. But now, everything seems to work on 3DMark with no issues.
 
Last edited:
So, another replacement card came, and this one seems to be working with no issues whatsoever. The LED issue with the other card came back though in SLI mode. It's been turning on and off sporadically, don't know why. Also, whereas the new card lights up on system start, that one, even when it worked, didn't light up until Windows booted. And it was a different shade of green. Outside of the LED issue though, everything seemed to be working.

A couple things. Time Spy seems to work great when XMP is enabled (18045).

https://www.3dmark.com/spy/4752505

Time Spy Extreme was pretty much the same score with (10951) or without XMP (10934).

https://www.3dmark.com/spy/4752589
https://www.3dmark.com/spy/4689594

The other people with similar systems all have overclocks on their system though, so not too worried.

On game results, if you're not running 4k or above, don't bother with SLI. With or without SLI enabled, The Witcher 3 was pretty much at 120 fps. I'm guessing this is where the 2950x caps out. 4k, I've got a solid 60 fps though, with every option, yes, even hairworks, turned on. Monster Hunter seems to run at 80 fps min.

The one issue I did have, is that 3dMark didn't run visually. The screen pretty much froze before even loading came up, but the test ran, and it did collect data. I'm guessing this is a driver issue, as I can't possibly see how hardware would cause that problem. (I.e. Once the test collected, it went onto the next loading screen in a frozen state, but the system was still responsive).

I'm surprised you even got the RTX 2080ti to work with your rig. XD
 
So I think one of my Ti cards is bad. Even though the system will work for maybe a day at a time, I continue to get crashes and blue screens sporadically.

I took out the second card, and with just 1 I still got a blue screen not even 10 minutes into Far Cry 5. I booted it up again, and another crash.

So I swapped the cards, so far the second card has been solid, but I want to give it some time to be sure (as I said, sometimes it would work for a whole day with no problems).

But it does look like that one card has some issues. Nvidia seems to be pretty good with RMA, so maybe by this weekend if I'm sure I will email them.
 
So I think one of my Ti cards is bad. Even though the system will work for maybe a day at a time, I continue to get crashes and blue screens sporadically.

I took out the second card, and with just 1 I still got a blue screen not even 10 minutes into Far Cry 5. I booted it up again, and another crash.

So I swapped the cards, so far the second card has been solid, but I want to give it some time to be sure (as I said, sometimes it would work for a whole day with no problems).

But it does look like that one card has some issues. Nvidia seems to be pretty good with RMA, so maybe by this weekend if I'm sure I will email them.


Yeah I had crashes with Black Ops twice in one day and had two crashes with Kingdom Come Derliverance the whole PC would lock up the only thing I could do was reset it.
 
Last edited:
Yeah some of the unstable RTX 2080 Ti reports I've been reading about have been quite unnerving. I've had The Division freeze like 2 times since I've gotten the card, but that game isn't really known for being 100% without glitches. lol. All other games have been solid.

I hope you guys with the unstable cards get good replacements. I'm seeing reports on Reddit about there being better supply on the 2080 Ti cards. Some folks are "pre-ordering" like a couple days ago and getting shipping notifications 1-2 days later. So that's good news.

It doesn't seem like Nvidia is intentionally doing some artificial supply drought this time. lol. Who knows for sure though right? XD
 
Well, my 5th card arrived today, and so far, everything seems good. Ran through the standard battery of tests and nothing's gone wrong. No artifacts (yet, and crossing fingers I don't get them), and lights are working.
 
Well, my 5th card arrived today, and so far, everything seems good. Ran through the standard battery of tests and nothing's gone wrong. No artifacts (yet, and crossing fingers I don't get them), and lights are working.

This is hilarious and also slightly frightening since I definitely voided my warranty bahahahha!! My card seems solid so far *knock on wood*.
 
This is hilarious and also slightly frightening since I definitely voided my warranty bahahahha!! My card seems solid so far *knock on wood*.

Man you full well knew you could afford to lose that once you took the soldering iron to it. XD
 
Yeah some of the unstable RTX 2080 Ti reports I've been reading about have been quite unnerving.
Yeah, the Nvidia forum is loaded with people getting crashes and blue screens. However, they seem to be sending out replacements like candy, so that's a plus.
 
I'm seeing reports on Reddit about there being better supply on the 2080 Ti cards. Some folks are "pre-ordering" like a couple days ago and getting shipping notifications 1-2 days later. So that's good news.

Yeah I preordered a zotac 2080ti amp edition as soon as it was listed from B&H on 10/16 and it shipped yesterday, should have it Tuesday. I used nowinstock and the preorder link was sold out in like 20 min and the price was then raised 100 bucks. Three fans and 5yr warranty did it for me over the founders for the same price.


Edit: same card is in stock again on b&h

https://www.bhphotovideo.com/c/product/1431194-REG/zotac_zt_t20810d_10p_geforce_rtx_2080_ti.html
 
Last edited:
Can folks that have a more modern CPU and a 2080 Ti give Unigine Heaven 4.0 a shot?

Here is my result with my 5960X @ 4GHz and an RTX 2080 Ti (+90 Core Voltage, 123% Power Limit, 88C Temp Limit, +160 Core Clock, +400 Memory). Trying to see if I'm CPU limited at that resolution or if the 5960X can hang on for a while longer.

FPS:
218.9
Score:
5513
Min FPS:
9.6
Max FPS:
392.5

Here's the settings to use to replicate it:
unigine_heaven_settings.jpg


Inspired by [H]'s 9900K review. Thank you!
 
I was looking at stock 2080 and 1080s and 2080 and 1080 ti’s.

The cheapest 1080 is a Gigabyte one at £459, the cheapest 2080 is a zotac one at £698. I have never bought anything Zotact so I don’t know what they are like.

The cheapest 1080ti is £647 and up. The cheapest 2080ti is £1099 and up to £1500.

Are the 1080 ti or non ti still worth buying, or would it be better to get a non ti 2080 instead ?

The 1080 and 2080 both have 8 gig ram, the 1080 and 2080 ti’s both have 11 gig. Is 11 gig physically used in games or do most games only use a few gig and the rest goes unused ?

Be nice. As I haven’t bought a video card for 3/4 years, and don’t know the situation now.
 
I was looking at stock 2080 and 1080s and 2080 and 1080 ti’s.

The cheapest 1080 is a Gigabyte one at £459, the cheapest 2080 is a zotac one at £698. I have never bought anything Zotact so I don’t know what they are like.

The cheapest 1080ti is £647 and up. The cheapest 2080ti is £1099 and up to £1500.

Are the 1080 ti or non ti still worth buying, or would it be better to get a non ti 2080 instead ?

The 1080 and 2080 both have 8 gig ram, the 1080 and 2080 ti’s both have 11 gig. Is 11 gig physically used in games or do most games only use a few gig and the rest goes unused ?

Be nice. As I haven’t bought a video card for 3/4 years, and don’t know the situation now.

Most games don't come close to the 11G I find. 8 is still plenty. I've owned a zotac 1070 and I can say it was a quality card.
 
Can folks that have a more modern CPU and a 2080 Ti give Unigine Heaven 4.0 a shot?

Here is my result with my 5960X @ 4GHz and an RTX 2080 Ti (+90 Core Voltage, 123% Power Limit, 88C Temp Limit, +160 Core Clock, +400 Memory). Trying to see if I'm CPU limited at that resolution or if the 5960X can hang on for a while longer.

Here's my result. [email protected]. Same GPU settings except my 2080Ti has a power limit of 115% and +160 core crashed so I lowered it to +150. Similar to you but the glaring difference would be that Min FPS.

Unigine  Heaven 4.0 demo Screenshot 2018.10.20 - 11.40.15.31.png
 
How is gaming on these cards? Crashes during long sessions? Glitches, artifacts weirdness? I just joined the 2080ti club and my buds and I were planning some long OW, Vermintide 2 sessions tonight.
 
I’ve had zero crashes so it’s possible not to have problems.
 
I was looking at stock 2080 and 1080s and 2080 and 1080 ti’s.

The cheapest 1080 is a Gigabyte one at £459, the cheapest 2080 is a zotac one at £698. I have never bought anything Zotact so I don’t know what they are like.

The cheapest 1080ti is £647 and up. The cheapest 2080ti is £1099 and up to £1500.

Are the 1080 ti or non ti still worth buying, or would it be better to get a non ti 2080 instead ?

The 1080 and 2080 both have 8 gig ram, the 1080 and 2080 ti’s both have 11 gig. Is 11 gig physically used in games or do most games only use a few gig and the rest goes unused ?

Be nice. As I haven’t bought a video card for 3/4 years, and don’t know the situation now.

What GPU do you have now, and what resolution/refresh are you running on your monitor?

The 10-series will be better for your money, but if you have the cash and want the best, the 20-series can be worthwhile.

Most games will not use more than 8GB, but some do depending on settings. For example, Deus Ex MD at Ultra 4K settings can exceed 11GB. Also, if you run Surround, like at 7680x1440 with AA can exceed 8GB.
 
How is gaming on these cards? Crashes during long sessions? Glitches, artifacts weirdness?
Gaming has been great on 2080 Ti but I have had crashing and blue screens. I think one of my cards is bad, I removed the problem card and I'm just with one now and things seems to be working. Will have to RMA with Nvidia.
 
What GPU do you have now, and what resolution/refresh are you running on your monitor?

The 10-series will be better for your money, but if you have the cash and want the best, the 20-series can be worthwhile.

Most games will not use more than 8GB, but some do depending on settings. For example, Deus Ex MD at Ultra 4K settings can exceed 11GB. Also, if you run Surround, like at 7680x1440 with AA can exceed 8GB.

My card at the moment is shit. It’s a EVGA 680 GTX Classified 4gb, not the neutered version, it may even be worth some cash to someone ? I doubt it though,

I have a dell u2410, which I can’t remember the res of, but I think it’s 1920x1200 ? I also have a 4k tv capable of 4:4:4 next to it in my cave. I would love to use that for gaming, but I’m gunna buy a new monitor as this one is old, good but old. I will keep it for day to day stuff, as I use it’s sd card alot alot.

I’m thinking about a gsync monitor, I haven’t read up about them to find out if they are still relevent ?

The difference between the 1080 and 2080 cash wise is a minimum of £150. Is the actual difference performance wise between them worth spending an extra £150 and getting a 2080 ?
 
Last edited by a moderator:
I would recommend a 1440p high refresh G-Sync monitor. Or going 4K on the TV.

Either way, a 2080 Ti would drive it nicely. The 2080 is definitely better than the 1080, but the 2080 Ti is the best.

If you are sticking with your 1200p monitor, even a 1080 could drive that. But since it seems you don't upgrade often, I would go with the 20-series so it has more life.
 
Here's my result. [email protected]. Same GPU settings except my 2080Ti has a power limit of 115% and +160 core crashed so I lowered it to +150. Similar to you but the glaring difference would be that Min FPS.

View attachment 113436

Yes, my minimum at 9.6 and yours at 42.9 - how to interpret this? Is it the 4GHz (me) versus your 4.9GHz, is it the more modern, better IPC - combination of both? Am I seeing a bottleneck or is it just not as good? Seems like overall similar results but I know the minimum thing is important because you don't want those huge dips.

BTW - been running Time Spy 1.0 with my OC and playing with the voltage. I have gone down to +70 Core Voltage and haven't seen much variance. Also not much variance from +100 Core Clock to +160 Core Clock so that's strange...
 
Yes, my minimum at 9.6 and yours at 42.9 - how to interpret this? Is it the 4GHz (me) versus your 4.9GHz, is it the more modern, better IPC - combination of both? Am I seeing a bottleneck or is it just not as good? Seems like overall similar results but I know the minimum thing is important because you don't want those huge dips.

BTW - been running Time Spy 1.0 with my OC and playing with the voltage. I have gone down to +70 Core Voltage and haven't seen much variance. Also not much variance from +100 Core Clock to +160 Core Clock so that's strange...

I wouldn’t read into it too terribly much unless it’s an actual 0.1% minimum.

There’s gotta be processor reviews out there with something close to Haswell...

Here:

https://www.techspot.com/review/1526-intel-4th-gen-core-i7-vs-8th-gen/

And it’s worse case since it’s a 4770k
 
Last edited:
I wouldn’t read into it too terribly much unless it’s an actual 0.1% minimum.

There’s gotta be processor reviews out there with something close to Haswell...

Here:

https://www.techspot.com/review/1526-intel-4th-gen-core-i7-vs-8th-gen/

And it’s worse case since it’s a 4770k

Perfect - thank you. I also did some research and I guess Unigine Heaven is a "worst case"/bad culprit with minimum frames - https://www.pcgamer.com/why-minimum-fps-can-be-misleading/2/ -

"Of the five games/engines we're testing for this article, Unigine Heaven has the least reliable/useful minimum FPS reporting. At lower resolutions on Nvidia hardware, a few dips in frame rate during scene transitions skew the numbers, so our 97 percentile results are substantially higher."

I think I will wait out the 9900K and see what the new X series brings next year.
 
Yes, my minimum at 9.6 and yours at 42.9 - how to interpret this? Is it the 4GHz (me) versus your 4.9GHz, is it the more modern, better IPC - combination of both? Am I seeing a bottleneck or is it just not as good? Seems like overall similar results but I know the minimum thing is important because you don't want those huge dips.

it always happens with unigine heaven and valley in the first run regarding minimums.. run a second time or even a third time to hot the GPU enough to give a more somewhat "real" result and see if score are constant across all the test..
 
Yes, my minimum at 9.6 and yours at 42.9 - how to interpret this? Is it the 4GHz (me) versus your 4.9GHz, is it the more modern, better IPC - combination of both? Am I seeing a bottleneck or is it just not as good? Seems like overall similar results but I know the minimum thing is important because you don't want those huge dips.

Also check Guru3D's review of the 9900K and the games tests in it with a RTX 2080 Ti. They have 5960X in there too but of course at stock clocks. You need to ask yourself if at most 20 fps in specific games that already run at 100 fps on 1440p is worth the money and effort. I'm on a 6600K @ 4.6 GHz and I'll probably wait for Zen 2 before even considering upgrades.
 
Also check Guru3D's review of the 9900K and the games tests in it with a RTX 2080 Ti. They have 5960X in there too but of course at stock clocks. You need to ask yourself if at most 20 fps in specific games that already run at 100 fps on 1440p is worth the money and effort. I'm on a 6600K @ 4.6 GHz and I'll probably wait for Zen 2 before even considering upgrades.

Thank you - that’s huge. I feel better about not having any significant CPU bottlenecking worries at 3440x1440! I’ll save myself around $1k for a new board and 9900K. :)
 
Can folks that have a more modern CPU and a 2080 Ti give Unigine Heaven 4.0 a shot?

Here is my result with my 5960X @ 4GHz and an RTX 2080 Ti (+90 Core Voltage, 123% Power Limit, 88C Temp Limit, +160 Core Clock, +400 Memory). Trying to see if I'm CPU limited at that resolution or if the 5960X can hang on for a while longer.

FPS:
218.9
Score:
5513
Min FPS:
9.6
Max FPS:
392.5

Here's the settings to use to replicate it:
View attachment 113412

Inspired by [H]'s 9900K review. Thank you!

heaven_2018_10_21_14_41_03_724.jpg


Note that this is probably not useful for testing CPU performance. This benchmark is heavily GPU limited, CPU is probably only being utilized at 10 - 15%. In a real game the CPU could be running AI, physics, game logic, etc. and that would be a much better test.
 
Last edited:
Yeah I only got 208 fps on the 8086K in my sig. Didn't oc the gpu.
 
Thank you - that’s huge. I feel better about not having any significant CPU bottlenecking worries at 3440x1440! I’ll save myself around $1k for a new board and 9900K. :)

Yeah, I would definitely wait for the 7nm wars. The 2080ti you could at least very easily see the performance boost. 5960x vs 9900k side by side you wouldn’t be able to tell in the real world IMO.
 
Last edited:
Back
Top