Support NeoGAF

[PlayALLtheGames]

Fine tuning in AfterBurner and I got some new high graphics scores. This is at 2114/6039, 1.093v.

http://www.3dmark.com/fs/13242424:

http://www.3dmark.com/spy/2145079:

 

[Fess]

Fine tuning in AfterBurner and I got some new high graphics scores. This is at 2114/6039, 1.093v.



http://www.3dmark.com/fs/13242424:


http://www.3dmark.com/spy/2145079:

Oh wow, crazy scores!
Is it the CPU/RAM that is tanking my physics and combined benchmarks in Firestrike? I'm on an i7 4790K stock clocking and DDR3 1600MHz RAM

 

[PlayALLtheGames]

Oh wow, crazy scores!
Is it the CPU/RAM that is tanking my physics and combined benchmarks in Firestrike? I'm on an i7 4790K stock clocking and DDR3 1600MHz RAM

Not that crazy considering this is a water-cooled card getting almost 1.1v but, hey, I'm not complaining!

As for your question: what scores are you getting? What's your performance like in the CPU-specific tests? Have you confirmed they're abnormal for your specs?

 

[Fess]

Not that crazy considering this is a water-cooled card getting almost 1.1v but, hey, I'm not complaining!

As for your question: what scores are you getting? What's your performance like in the CPU-specific tests? Have you confirmed they're abnormal for your specs?

Yeah it doesn't look pretty at all. The graphics score is still okay but the CPU / physics score is thousands of points lower. The 3DMark achievements literally say that the GPU is ready to rumble but the CPU doesn't want to play lol XD
It's not like I thought the rig was fully future proofed, it's 4 years old except for the GPU, but I honestly didn't think this much had happened since then on the CPU side. Just look at that those physics and CPU scores...

Note: I haven't tweaked the system in any way so I probably have a million services running lol but I doubt that it's something like that causing a big gap like this. Is there some settings in Windows 10 that can be all messed up? I installed Windows 10 maybe 3 months ago and hasn't tweaked it from default settings.

 

[PlayALLtheGames]

Yeah it doesn't look pretty at all. The graphics score is still okay but the CPU / physics score is thousands of points lower. The 3DMark achievements literally say that the GPU is ready to rumble but the CPU doesn't want to play lol XD
It's not like I thought the rig was fully future proofed, it's 4 years old except for the GPU, but I honestly didn't think this much had happened since then on the CPU side. Just look at that those physics and CPU scores...



Note: I haven't tweaked the system in any way so I probably have a million services running lol but I doubt that it's something like that causing a big gap like this. Is there some settings in Windows 10 that can be all messed up? I installed Windows 10 maybe 3 months ago and hasn't tweaked it from default settings.

I mean you have an almost 3 year old processor but based on your scores I don't think there's anything wrong. When you say the CPU/physics score is "thousands of points lower", what are you comparing that to? I assume not my scores - I'm running a completely different setup - but you're better than 97% in FS, which seems to confirm you're running normally.

Perhaps more importantly keep in mind differences in these benchmarks typically don't translate to major differences in games performance.

 

[Fess]

I mean you have an almost 3 year old processor but based on your scores I don't think there's anything wrong. When you say the CPU/physics score is "thousands of points lower", what are you comparing that to? I assume not my scores - I'm running a completely different setup - but you're better than 97% in FS, which seems to confirm you're running normally.

Perhaps more importantly keep in mind differences in these benchmarks typically don't translate to major differences in games performance.

Aha. Yeah I was actually comparing to your monster, not the best idea maybe lol
It's interesting though, I admit to be a total noob when it comes to benchmarking but I honestly didn't think the CPU would make that much of a difference on the total score. You see reviewers posting benchmark scores everywhere when new graphics cards arrives but the info about what other stuff they have is not posted as clearly.

But as you say in the end it might not matter much. I'm now running Forza Horizon 3 between 80 and 120 fps on three screens at max settings. I'm okay with that.

Side note: Your comment about my CPU being almost 3 years old had me checking my receipts... I was certain I had owned this computer for 4 years... but nope. I bought this thing pretty much exactly 3 years ago. Which means I'm on my third graphics card in three years! I'm just a terrible terrible person! :S

 

[PlayALLtheGames]

Aha. Yeah I was actually comparing to your monster, not the best idea maybe lol
It's interesting though, I admit to be a total noob when it comes to benchmarking but I honestly didn't think the CPU would make that much of a difference on the total score. You see reviewers posting benchmark scores everywhere when new graphics cards arrives but the info about what other stuff they have is not posted as clearly.

But as you say in the end it might not matter much. I'm now running Forza Horizon 3 between 80 and 120 fps on three screens at max settings. I'm okay with that.

Side note: Your comment about my CPU being almost 3 years old had me checking my receipts... I was certain I had owned this computer for 4 years... but nope. I bought this thing pretty much exactly 3 years ago. Which means I'm on my third graphics card in three years! I'm just a terrible terrible person! :S

Yeah I run my 7700k at 5Ghz so our scores are definitely going to be different.

And I also purchase GPUs like every 1-1.5 years. It's an obsession.

 

[papersleeves]

Is anyone able to sustain his overclock while running Mass Effect Andromeda?

I know how GPU Boost 3.0 works (power-voltage-temps limits, speed bin, etc). On some games, I can hold 2113 mhz @ 1.093v without any issue while on others, I have to settle between 2025-2088mhz @ 1.025-1.081.

But ME:A is an asshole. The game barely goes above 2000mhz, fluctuating most of the time between 1850-1950mhz @ .975-1.025v.

Am I an isolated case or most have the same behavior?

Are there any other games that act in a similar way?

 

[PlayALLtheGames]

Is anyone able to sustain his overclock while running Mass Effect Andromeda?

I know how GPU Boost 3.0 works (power-voltage-temps limits, speed bin, etc). On some games, I can hold 2113 mhz @ 1.093v without any issue while on others, I have to settle between 2025-2088mhz @ 1.025-1.081.

But ME:A is an asshole. The game barely goes above 2000mhz, fluctuating most of the time between 1850-1950mhz @ .975-1.025v.

Am I an isolated case or most have the same behavior?

Are there any other games that act in a similar way?

I've run it at 2114/6039 at 1.093v but this is a water-cooled card staying in the mid 40C.

Not sure what's causing that with ME:A specifically but check your temps when the game is running and compare it to other games. Also, what's your memory running at? Have you tried to set a custom voltage curve?

Finally keep in mind the difference between even 1950Mhz and 2113Mhz on frame-rates is going to be minimal.

 

[papersleeves]

I've run it at 2114/6039 at 1.093v but this is a water-cooled card staying in the mid 40C.

Not sure what's causing that with ME:A specifically but check your temps when the game is running and compare it to other games. Also, what's your memory running at? Have you tried to set a custom voltage curve?

Finally keep in mind the difference between even 1950Mhz and 2113Mhz on frame-rates is going to be minimal.

Well my temps aren't all that great. The ICX cooler on my EVGA black edition doesn't do what I would call a proper job. The card sits around 64-65C at 100% fan speed so I guess by the time it hits these temps, the speed has been binned down 2-3 notches. Still, it's next to impossible to stick at static speed. It keeps varying for me. Mass Effect isn't the only game doing this. Doom does it as well, same for Dark Souls III, Witcher 3.

In fact, pretty much every new game affects the speed, whereas older games stick to 2100mhz without any issue.

I keep getting the power limiter showing on the AB OSD, but I never saw voltage or temp limiter. This is with the ASUS Strixx OC bios, with 330W TDP.

I do am using the curve, btw. Check my previous post, it has pictures of my AB settings.

 

[PlayALLtheGames]

Well my temps aren't all that great. The ICX cooler on my EVGA black edition doesn't do what I would call a proper job. The card sits around 64-65C at 100% fan speed so I guess by the time it hits these temps, the speed has been binned down 2-3 notches. Still, it's next to impossible to stick at static speed. It keeps varying for me. Mass Effect isn't the only game doing this. Doom does it as well, same for Dark Souls III, Witcher 3.

In fact, pretty much every new game affects the speed, whereas older games stick to 2100mhz without any issue.

I keep getting the power limiter showing on the AB OSD, but I never saw voltage or temp limiter. This is with the ASUS Strixx OC bios, with 330W TDP.

I do am using the curve, btw. Check my previous post, it has pictures of my AB settings.

If you're using the same XOC BIOS others of us are (me included), there should be no power limit. When you flashed to that BIOS, did the temp limit and power limit sliders get grayed out in AB? They should.

XOC BIOS I'm using: XOC

 

[papersleeves]

If you're using the same XOC BIOS others of us are (me included), there should be no power limit. When you flashed to that BIOS, did the temp limit and power limit sliders get grayed out in AB? They should.

XOC BIOS I'm using: XOC

I must admit I am not using the XOC Bios, but rather the standard ASUS Strixx OC bios. Should I worry of using this bios on my ISX air cooled card? Do I have to unlock the TDP via a bat file like I needed to do with my current bios?

 

[PlayALLtheGames]

I must admit I am not using the XOC Bios, but rather the standard ASUS Strixx OC bios. Should I worry of using this bios on my ISX air cooled card? Do I have to unlock the TDP via a bat file like I needed to do with my current bios?

Well I was under the impression the XOC is the Strix OC (as opposed to the Strix Gaming). So, it sounds like that's your default BIOS? I could be wrong; there are so many fucking versions of these cards by every manufacturer it's hard to keep track.

 

[Aztechnology]

Ended up grabbing an EVGA 1080ti SC2 during the Newegg discount promo and with credit all said and done about $630. So I'm happy. Anyone else with this card? OC recommendations? Etc.

 

[PlayALLtheGames]

Ended up grabbing an EVGA 1080ti SC2 during the Newegg discount promo and with credit all said and done about $630. So I'm happy. Anyone else with this card? OC recommendations? Etc.

Find a max for OC/benchmarks, then try to see how close you can get to those scores with as much lower clocks as possible

-Install AfterBurner
-Unlock voltage
-Raise temp and power limits to max
-Start with high OC on core and memory or learn how to use voltage curve to set core OC
-Run Heaven benchmark on extreme but not quite at full screen to have access to AfterBurner for real-time changes
-If stable, watch for temps over extended time; potentially increase overclocks
-If unstable, reduce clocks as needed until stable and good temps.
-Once stable, run FireStrike, SuperPosition (4k optimized and 1080p extreme) to stress the card. Again reduce overclocks as needed for stability and decent temps.
-Find max scores under max, stable OC
-Save profile

Then do the same but as I said see how close you can get to your max benchmark scores with that profile while giving the card less voltage and lower clocks. Since a higher clock can actually reduce scores, you'd be surprised what you might be able to achieve.

 

[Aztechnology]

Find a max for OC/benchmarks, then try to see how close you can get to those scores with as much lower clocks as possible

-Install AfterBurner
-Unlock voltage
-Raise temp and power limits to max
-Start with high OC on core and memory or learn how to use voltage curve to set core OC
-Run Heaven benchmark on extreme but not quite at full screen to have access to AfterBurner for real-time changes
-If stable, watch for temps over extended time; potentially increase overclocks
-If unstable, reduce clocks as needed until stable and good temps.
-Once stable, run FireStrike, SuperPosition (4k optimized and 1080p extreme) to stress the card. Again reduce overclocks as needed for stability and decent temps.
-Find max scores under max, stable OC
-Save profile

Then do the same but as I said see how close you can get to your max benchmark scores with that profile while giving the card less voltage and lower clocks. Since a higher clock can actually reduce scores, you'd be surprised what you might be able to achieve.

Oh sorry I should have clarified I'm pretty well versed at GPU Overclocking and fairly experienced at CPU OC. Saying OC reemdations was way too vague. I think I was kind of just looking for what other people had reached with my particular card. Impressions, pitfalls/issues etc.

Or is it generally been like the 1070 where OC's have been realtively static across board partners within a percentage or two? Is there different memory types depending on AIB partner etc?

I do appreciate the write up though, it's good advice!

 

[papersleeves]

Well I was under the impression the XOC is the Strix OC (as opposed to the Strix Gaming). So, it sounds like that's your default BIOS? I could be wrong; there are so many fucking versions of these cards by every manufacturer it's hard to keep track.

I believe there are 2 Strixx bios. XOC.rom and ASUSStrixxOC.rom. I was using the latter. It's not my default bios since I have an EVGA Black Edition. I was using this one because it has higher TDP than what I had by default on my card.

Anyway, I just tried the XOC bios. It's working well as in the voltage/TDP clearly are unlocked, but I just cannot use this bios. The reason? Temperature. I keep getting temps over 70C with it and by the time it hits 72-73C, the card crashes.

It's too bad because with this bios, I can force the card to use a specific core frequency at all time while sticking to an exact voltage. But with the test I've done, my conclusion is that it's better for me to use my previous bios with fluctuating core frequency/voltage than to stick with static value with the new BIOS.

All of this makes me want to buy the AIO hydro kit from EVGA. I have no doubt I'd have much better results with it.

 

[PlayALLtheGames]

Oh sorry I should have clarified I'm pretty well versed at GPU Overclocking and fairly experienced at CPU OC. Saying OC reemdations was way too vague. I think I was kind of just looking for what other people had reached with my particular card. Impressions, pitfalls/issues etc.

Or is it generally been like the 1070 where OC's have been realtively static across board partners within a percentage or two? Is there different memory types depending on AIB partner etc?

I do appreciate the write up though, it's good advice!

Pascal comparisons using the same cooling have been within a certain margin just because of the inability to fuck with the BIOS (other than flashing). Silicon lottery matters as always but assuming you don't get an abnormally poor overclocker, disparity when comparing air to air or water to water has been low.

See how your 3DMark scores match up with other 1070s with an advanced search on 3dmark.com, if you haven't already.

I believe there are 2 Strixx bios. XOC.rom and ASUSStrixxOC.rom. I was using the latter. It's not my default bios since I have an EVGA Black Edition. I was using this one because it has higher TDP than what I had by default on my card.

Anyway, I just tried the XOC bios. It's working well as in the voltage/TDP clearly are unlocked, but I just cannot use this bios. The reason? Temperature. I keep getting temps over 70C with it and by the time it hits 72-73C, the card crashes.

It's too bad because with this bios, I can force the card to use a specific core frequency at all time while sticking to an exact voltage. But with the test I've done, my conclusion is that it's better for me to use my previous bios with fluctuating core frequency/voltage than to stick with static value with the new BIOS.

All of this makes me want to buy the AIO hydro kit from EVGA. I have no doubt I'd have much better results with it.

Yeah, unsurprisingly temps are your limitation. I would absolutely look into an AIO.

 

[PlayALLtheGames]

So I'm posting this because I honestly had no idea this thing existed until someone posted it in another forum. It's the ROG Poseidon Platinum Edition. Perhaps someone here might be interested in its unique features.

It has both air cooling and a pre-installed block if you go water down the road. The price is hefty but I'm not aware of any other cards like this (of course, I wasn't aware of this one either so what do I know?)

https://www.asus.com/us/Graphics-Cards/ROG-POSEIDON-GTX1080TI-P11G-GAMING/

 

[Marmelade]

I believe there are 2 Strixx bios. XOC.rom and ASUSStrixxOC.rom. I was using the latter. It's not my default bios since I have an EVGA Black Edition. I was using this one because it has higher TDP than what I had by default on my card.

Anyway, I just tried the XOC bios. It's working well as in the voltage/TDP clearly are unlocked, but I just cannot use this bios. The reason? Temperature. I keep getting temps over 70C with it and by the time it hits 72-73C, the card crashes.

It's too bad because with this bios, I can force the card to use a specific core frequency at all time while sticking to an exact voltage. But with the test I've done, my conclusion is that it's better for me to use my previous bios with fluctuating core frequency/voltage than to stick with static value with the new BIOS.

All of this makes me want to buy the AIO hydro kit from EVGA. I have no doubt I'd have much better results with it.

72-73°C isn't even hot, why would that be the cause of your crashes?

 

[papersleeves]

72-73°C isn't even hot, why would that be the cause of your crashes?

You tell me. What I know though if that my crashes only happened above 70C.

 

[PlayALLtheGames]

You tell me. What I know though if that my crashes only happened above 70C.

Marmelade makes a good point. Those temps will be limitation in terms of throttling clocks but seem low for crashing. These cards are made to be stable well into the 80s if needed.

To be clear: you don't get the crashing at those temps when on your stock BIOS, right? Only when on XOC?

Could just be unstable OC or perhaps power draw? Use HWInfo and start a log file then benchmark and see what you're pulling in wattage. Not sure what PSU you're using.

 

[papersleeves]

Marmelade makes a good point. Those temps will be limitation in terms of throttling clocks but seem low for crashing. These cards are made to be stable well into the 80s if needed.

To be clear: you don't get the crashing at those temps when on your stock BIOS, right? Only when on XOC?

Could just be unstable OC or perhaps power draw? Use HWInfo and start a log file then benchmark and see what you're pulling in wattage. Not sure what PSU you're using.

I understand. It's not rare to see FE cards in the 70-80C range, though my card doesn't appear to like temps. To make this clear, I also had crashes that happened on the original and the ASUSStrixxOC bios (not XOC), but that was with different settings. For one, my card never went into the 70's C with those bios, only with XOC due to the unlimited power draw.

With the XOC bios, everything appears extremely stable, but then I hit 70-72C and it crashes. Every single time.

Do you think it could be my PSU? It's a corsair TX850 but it's a bit old. I got it at the same time as my previous 2500K in January 2011. Always seemed rock solid to me though. Back in the days, I was powering my 4.8ghz 2500k and 2 overclocked GTX480 with it.

One weird thing with the XOC bios is the way the voltage works. If for example I chose a maximum speed of, let's say 2038mhz and equalize that speed from 1.031v to every other voltage value past it, the card will stick to that speed at 1.031v without ever moving the voltage further up. Same thing happens if I start to equalize from 1.045v, 1.062v or even 1.075v. But if I choose to equalize that speed (2038mhz) from 1.081v then the card will start at 2038mhz 1.081v then it will jump to 1.093v then 1.1v then 1.112v then 1.125v and so on. It's like it cannot stop adding voltage to the core. I don't know why this is happening.

 

[jambo]

http://www.3dmark.com/3dm/21403766

http://www.3dmark.com/3dm/21403446

Didn't tweak or change anything, pretty happy with the new system.

 

[tuna_love]

http://www.3dmark.com/3dm/21403766


http://www.3dmark.com/3dm/21403446


Didn't tweak or change anything, pretty happy with the new system.

what ram do you have?

 

[SolidSnakeUS]

I don't remember if I asked this recently, but are the new cards still on track for the middle of next year? Are we getting anything better than the 1080 Ti this year?

 

[terencecah]

Any Volta news

 

[lucebuce12]

Does anyone know when the Zotac 1080 Ti Mini will be available for sale?

 

[NeoFaff]

Does anyone know when the Zotac 1080 Ti Mini will be available for sale?

It released in Europe last week, I'm using one. One of the UK pc sites put up a preorder page about a month ago with no delivery estimates and I preordered it.

There hasn't been any press cover of zotac mini cards since their reveal months ago for some reason and stock is very limited - sellers seem to only have 3-5 at a time

Have a look at the US websites and see if they have any listings I guess?

 

[jambo]

what ram do you have?

16GB of Corsair Vengeance 3200, but I feel like the motherboard has defaulted to a much slower speed.

 

[Aztechnology]

Starting my runs to see where my card can get. I'm used to the bigger OC gap on the 1070, so barely being able to eek out OC's feels odd. But the card is great.

First test was 1080ti +80 to Core only.

http://www.3dmark.com/3dm/21420766

Edit:

http://www.3dmark.com/3dm/21421300?

Bringing up to +100Mhz core, +450 Memory got me these results, seems stable, but not much past 2025Mhz on the core seems stable, even with TDP adjustment. Seems I'd have to use voltage control which I'm not sure I really want to do (Probably isn't worth it for me).

What I really need to do is resolve the memory issues I'm having. My DDR4 is only at 2133Mhz and I'm only using two sticks on a quad channel recommended setup (Computer won't post with 4 sticks no matter what I do/combination, leads to a bunch of finagling just to get it to work with 2 sticks again. I've had both the CPU and Mobo replaced with RMA already after doing rigorous testing, so I'm still not sure why my computer won't post with four. I've tried a few different sets of RAM now. 2 sets of my own, and a friends. I don't really have the money to keep messing with RAM, especially as the costs are rising, instead of falling for DDR4.

 

[PlayALLtheGames]

I understand. It's not rare to see FE cards in the 70-80C range, though my card doesn't appear to like temps. To make this clear, I also had crashes that happened on the original and the ASUSStrixxOC bios (not XOC), but that was with different settings. For one, my card never went into the 70's C with those bios, only with XOC due to the unlimited power draw.

With the XOC bios, everything appears extremely stable, but then I hit 70-72C and it crashes. Every single time.

Do you think it could be my PSU? It's a corsair TX850 but it's a bit old. I got it at the same time as my previous 2500K in January 2011. Always seemed rock solid to me though. Back in the days, I was powering my 4.8ghz 2500k and 2 overclocked GTX480 with it.

One weird thing with the XOC bios is the way the voltage works. If for example I chose a maximum speed of, let's say 2038mhz and equalize that speed from 1.031v to every other voltage value past it, the card will stick to that speed at 1.031v without ever moving the voltage further up. Same thing happens if I start to equalize from 1.045v, 1.062v or even 1.075v. But if I choose to equalize that speed (2038mhz) from 1.081v then the card will start at 2038mhz 1.081v then it will jump to 1.093v then 1.1v then 1.112v then 1.125v and so on. It's like it cannot stop adding voltage to the core. I don't know why this is happening.

Not sure what's going on with the voltage but that and the crashing may just be a byproduct of using a non-native BIOS.

2038 at 1.031v is great though. I'd make that my permanent OC for gaming.

I assume your PSU is fine.

One test might be to flash your original BIOS and then do whatever's needed to get it into the 70's and see if it also crashes.

Non-native BIOS may be the simple cause. If your card really is crashing because of 70 degrees, that's probably grounds for a RMA.

Bringing up to +100Mhz core, +450 Memory got me these results, seems stable, but not much past 2025Mhz on the core seems stable, even with TDP adjustment. Seems I'd have to use voltage control which I'm not sure I really want to do (Probably isn't worth it for me).
.

Well it's not technically worth it for any of us but I'd mess with voltage just to see what you can hit on benchmarks. You've got a flagship card, feed it!

Your permanent/gaming OC will end up being lower obviously but I always find pushing GPUs to see their potential fun.

 

[lucebuce12]

It released in Europe last week, I'm using one. One of the UK pc sites put up a preorder page about a month ago with no delivery estimates and I preordered it.

There hasn't been any press cover of zotac mini cards since their reveal months ago for some reason and stock is very limited - sellers seem to only have 3-5 at a time

Have a look at the US websites and see if they have any listings I guess?

Haven't been able to find anything on the US so far. I guess it doesn't really matter since I'm going to buy it at the end of the year anyway but still.

Btw, how is it? The cooling particularly, since that's the one thing I'm slightly worried about. My setup does have good cooling but I'm still worried a little bit.

 

[MANGOD]

Whos rocking a 1080ti with an ultrawide? Thinking of picking up an x34a and wondering if the ti was a little overkill? Would a 1080 do the job at max settings and hitting 100hz or there abouts?
Edit this answered my question perfecty for anyone else wondering the same https://youtu.be/akJ0Hpf-wCQ

 

[teiresias]

Whos rocking a 1080ti with an ultrawide? Thinking of picking up an x34a and wondering if the ti was a little overkill? Would a 1080 do the job at max settings and hitting 100hz or there abouts?

I have the Asus equivalent monitor and Witcher 3 at ultra settings without hairworks out in the world my 1080 Ti sits at the mid-80s, so it's not maxing out the 100hz refresh on the monitor. Skyrim SE with some mods is closer to 40fps. Doom maxes out the monitor refresh. Take that for what you will.

 

[MANGOD]

I have the Asus equivalent monitor and Witcher 3 at ultra settings without hairworks out in the world my 1080 Ti sits at the mid-80s, so it's not maxing out the 100hz refresh on the monitor. Skyrim SE with some mods is closer to 40fps. Doom maxes out the monitor refresh. Take that for what you will.

Cheers for the reply just found a vid on utube with plenty benchs. Ti it is!

 

[jim2point0]

I have the Asus equivalent monitor and Witcher 3 at ultra settings without hairworks out in the world my 1080 Ti sits at the mid-80s, so it's not maxing out the 100hz refresh on the monitor. Skyrim SE with some mods is closer to 40fps. Doom maxes out the monitor refresh. Take that for what you will.

For me, the 1080TI is the only card on the market for properly driving 3440x1440 without compromises. And I still have to make compromises in some games (Quantum Break.... what an asshole that game).

 

[PlayALLtheGames]

Whos rocking a 1080ti with an ultrawide? Thinking of picking up an x34a and wondering if the ti was a little overkill? Would a 1080 do the job at max settings and hitting 100hz or there abouts?
Edit this answered my question perfecty for anyone else wondering the same https://youtu.be/akJ0Hpf-wCQ

Not overkill.

Ti is the only way to go for 3440x1440. Keep in mind some games are already giving this card an issue at max settings (Watch Dogs 2, Ghost Recon: Wildlands as examples).

 

[NeoFaff]

Haven't been able to find anything on the US so far. I guess it doesn't really matter since I'm going to buy it at the end of the year anyway but still.

Btw, how is it? The cooling particularly, since that's the one thing I'm slightly worried about. My setup does have good cooling but I'm still worried a little bit.

I have a Mitx case where the front fan is basically blocked by PSU wires. The card hit a max of 85C during 4 hours of playing Ryse at 1080p120fps with everything at max. Take that for what it's worth

I'll do a few hours of benchmark stress tests today and post the results later.

One thing I will say about the card though is that the packaging didn't make a good first impression - the box was obviously from a larger model with an insert to stop the card moving and empty holes where DVI to VGA and other accessories would've been. The yellow sleeve was fine.

 

[Aztechnology]

You know I'm not exactly sure what's going on as as clocks on my 1080ti SC2. On firestrike and Benchmarks with Unengine I'm able to do +100mhz which spikes up to 2045mhz or so at times. And +450mhz memory or 5950mhz whouthot crashing. However any games I play (Metro Last Light Redux, TW Warhammer, or PUBG at this time) at those clock speeds will spike over the 120% power threshold and just close. I'm not sure if it's he display driver crashing or what exactly though. The game just closed to desktop. (No notification about display driver crashing or anything, no artifacting etc.) I can get like +70mhz and +350mhz for gamea without crashing and in the low 70 temps. (Need to set new fan curves).

Is this normal unless you flash the bios and push the voltage up? Just seems like it has more to give but won't let me.

 

[thuway]

I have similar problems with my gaming x 1080ti. Using Afterburner I can get up to 2020mhz and than it'll throttle viciously to almost half performance.

It's like the card just wants me to use the auto oc feature in the gaming app and not touch afterburner .

 

[PlayALLtheGames]

You know I'm not exactly sure what's going on as as clocks on my 1080ti SC2. On firestrike and Benchmarks with Unengine I'm able to do +100mhz which spikes up to 2045mhz or so at times. And +450mhz memory or 5950mhz whouthot crashing. However any games I play (Metro Last Light Redux, TW Warhammer, or PUBG at this time) at those clock speeds will spike over the 120% power threshold and just close. I'm not sure if it's he display driver crashing or what exactly though. The game just closed to desktop. (No notification about display driver crashing or anything, no artifacting etc.) I can get like +70mhz and +350mhz for gamea without crashing and in the low 70 temps. (Need to set new fan curves).

Is this normal unless you flash the bios and push the voltage up? Just seems like it has more to give but won't let me.

So you're stable at +70/+350 on air which seems perfectly within the scope of normal.

I have similar problems with my gaming x 1080ti. Using Afterburner I can get up to 2020mhz and than it'll throttle viciously to almost half performance.

It's like the card just wants me to use the auto oc feature in the gaming app and not touch afterburner .

Try setting a more aggressive fan curve.

But ultimately it may not be a "problem." Some cards don't overclock as well as others and the disparity can be significant at times.

 

[Aztechnology]

So you're stable at +70/+350 on air which seems perfectly within the scope of normal.



Try setting a more aggressive fan curve.

But ultimately it may not be a "problem." Some cards don't overclock as well as others and the disparity can be significant at times.

What's wierding me out is that it's not a thermal issue, nor does it show any other normal signs of an unstable OC. Just a few mhz too far and instantly closes. Seems to correlate with power limit going beyond 120.

 

[ZOONAMI]

What's wierding me out is that it's not a thermal issue, nor does it show any other normal signs of an unstable OC. Just a few mhz too far and instantly closes. Seems to correlate with power limit going beyond 120.

That's the silicon lottery. Sometimes it doesn't really have anything to do with Temps, the card just doesn't wanna go where it doesn't wanna go.

 

[thuway]

But ultimately it may not be a "problem." Some cards don't overclock as well as others and the disparity can be significant at times.

Oh trust me I've set the fan curve as aggressive as I can get - this makes no sense. Ultimately, I've read alot of other users having similar problems with the Gaming X 1080ti. It really sucks knowing in the back of my mind there's an extra 5-10 frames left that's possibly locked behind a super aggressive thermal throttle.

 

[longdi]

My frustration with the curve is, you can set a perfectly tuned one, only to have it turn into a jagged mess on reboot.

My 1080ti also seem to have a mind of its own. Perfectly stable for a long gaming session, but will ctd on the next. Old games like coh2 runs into power limit at 1600p, but nier runs the clocks smooth as silk. Even in same benchmark, superposition, 1080p extreme has the clocks hover around max 20xx, while 4k optimised hovers around 19xx.

 

[Aztechnology]

Gat damn, I put my used Nvidia GTX 1070 gaming 8G up on Amazon for $480 and it sold. After fees I'll stop come out like $50 over what I paid near launch on it. This helps with my 1080ti cost .

Mining baby. I wouldn't be buying a 1070 for mining right now though given the significantly better options that are here or coming soon.

Now I just need to sell the unopened 1080 I have I got before I impulse bought the 1080ti on sale lol.

My frustration with the curve is, you can set a perfectly tuned one, only to have it turn into a jagged mess on reboot.

My 1080ti also seem to have a mind of its own. Perfectly stable for a long gaming session, but will ctd on the next. Old games like coh2 runs into power limit at 1600p, but nier runs the clocks smooth as silk. Even in same benchmark, superposition, 1080p extreme has the clocks hover around max 20xx, while 4k optimised hovers around 19xx.

I'm in the same boat. Stability is always fine at base but depending on the title, even older titles, stability varies wildly and is causing frustration. Some newer more demanding games are fine on higher clocks vs old titles that seem to draw more power even when not using the higher clocks.

 

[PlayALLtheGames]

What's wierding me out is that it's not a thermal issue, nor does it show any other normal signs of an unstable OC. Just a few mhz too far and instantly closes. Seems to correlate with power limit going beyond 120.

Yeah that's one of the things about these Pascal cards: they want power but BIOS TDP seem excessively restrictive.

You might try flashing a different card's BIOS. The XOC BIOS has no power limit at all. That may just mean you'll run into thermal throttling/issues instead but possibly with a higher ceiling.

Oh trust me I've set the fan curve as aggressive as I can get - this makes no sense. Ultimately, I've read alot of other users having similar problems with the Gaming X 1080ti. It really sucks knowing in the back of my mind there's an extra 5-10 frames left that's possibly locked behind a super aggressive thermal throttle.

That sucks but if you're running in the 2020s then realistically even hitting 2100mhz won't translate to 5-10fps.

Still, maybe think about trying another card's BIOS.

On overclock.net, the "1080Ti owners" and "How to flash a different BIOS on your 1080ti" threads are good resources to check with other owners of your exact cards for their results.

 

[Branson]

I'm Still debating between the ASUS or EVGA cards for my build. ASUS I've read is better but you can't beat that EVGA service.

 

[jim2point0]

I'm Still debating between the ASUS or EVGA cards for my build. ASUS I've read is better but you can't beat that EVGA service.

If fan noise isn't a concern, then the EVGA cards would be fine. Otherwise, I'd say the Asus is the better buy. Support is only really useful if there's a problem with your card and you bought it from a place that won't let you return it (so don't do that).