181°F is 82°C. That's a nominal temp for a blower-style card under full load. If Warhammer is working fine, but Wildlands is not, it's very likely that it's an issue with that game, not the card. 
 
 

Time to make a video. 

OK.
So, this is completely out of my range of knowledge.
I just played the same games I play (Warhammer Vermintide, TCGR Wildlands, and Mirror's Edge Catalyst), and got totally different results than I have been getting since I have had the card (but, I dont always have monitoring software going, so its possible, even likely, that these are things that have been happening on those very rare occasions when the video card or games dont crash.
 
http://s1378.photobucket.com/user/dave3d_bc/media/3rd_vidcard/10_18_2017_3rdrma_wtd_zpsjmx41u4y.png.html?sort=3&o=0
 
Note that it is the same card, same specs (I still dont OC anything), and these anomalies are happening:
The fan speed 100% changed to a higher state: 4833 rpm, instead of what I usually see is the low 3000 to high 2000's.
The memory controller is 68% instead of its usual 99-100%
The temp was 143F instead of the usual 181 to 185.5565 (I attribute this to the higher rpm of the fan this time, plus the lower gpu memory controller?)
 
So, how does this happen?
I have had the fans at 100% for the last week to week and a half, and the numbers just arent the same at all. Today they were excellent. Before today, every time I have monitored things, they have been gpu mem controller 100%, MUCH lower 100% gpu fan speed rpm's, and much higher temps. AND the room temp is 73F instead of the usual 71F, so its actually hotter in here tonight.
What could have made them SO much better tonight than before (oh, and I didnt get a single crash tonight, although that happens once in a while, it is very rare that I dont get a crash since I 1st put this 3rd rma card in  my system (I have witnesses, my friends I play with, and I drop out of games because of crashes, or the system auto reboots and I even lose them from skype).
I didnt install the new driver yet, I got the itch to play while I could.
Could NOT playing with my friends in coop mode affect the video card? Because that is THE only thing that has changed today, I didnt play coop with my buddies. But, that couldnt have an affect on how much gpu memory controller is being used, or especially what the video card THINKS is the max 100% rpm, right? How is it one rpm one day (at 100%) and 400 to 2,000 rpm more the next day, at the same 100% fan speed setting?
 
This is the kind of stuff that has been happening since (I am guessing) since 6 months after I bought my brand new card. Overheating and playing like dog poo, then playing fine, then playing like poo again, then stuttering all over the place like a slideshow, then crashing games, then crashing the machine (probably drivers, but who really knows at this point? Threads that evga says there is no way that it could be the drivers causing this and that, then saying its the users machine, then etc etc etc).
COULD it be the PSU is going out, and is affecting the internal preset mechanisms and bios if it doesnt get enough juice to it?
Could it be the card putting itself in low power states and not telling things to ramp up all the way when its supposed to, so part of the card is working at 100% and part is stuck in 30-40% still?
How can that happen when the bios, the code written, is supposed to do exactly what it was written to do? It shouldnt be able to deviate like that right?
Or, is the Titan X the 1st card to have a sort of AI, where itself monitors and turns on things that it 'thinks' should be the way it 'thinks' they should, but its wrong a lot of the time (I know, sci-fy stretch, lol).
 
But seriously, how can these wide variations be happening?
What more testing can I do to try to narrow it down (although it seems to be narrowed down to gsync and the 900 series to me) ? The tech today got me thinking that the power supply could be going wonky, but the only thing I see there is the 3.3v going down to 3.18 one time, and it usually stays at 3.31 to 2.89, which is well within specs (EVGA Supernova Gold 1300W psu is what it is).
Maybe if I remove the HDD's that are in there right now, and start fresh with no other drives at all from the previous (right now)  os drive (ie: clone the drive to another fresh drive and use ONLY that fresh drive, not the other 3 drives attached)?
Get rid of the M.2 drive, and that way the bios will assign ALL the pci-e lanes, and not cut off the 3rd x16 lane (which was for 3 way sli, which I am not using anyway, so it doesnt matter if its down or not).
What bios options 'could' interfere with the video card doing what its supposed to? What should I look for in the bios? What should the video card hard coded stuff be/need in order to run correctly? I have the pci-e lanes at max capacity, and the slot for the video card on gen 3. I think the power/voltage/thingy is set to 100. Should it be different than that? That is what it came up with, being everything on Auto. What bios settings should I report back to you guys and gals?
 
I know I have the rma's already, but I am wondering what could be causing this, and I dont want to have to go through this whole thing again.
So, if you know anyone that knows how to troubleshoot weird stuff like this, then please ask them and let me know asap, because I will be pulling things apart and be shipping everything in less than 2 weeks.
 
The only thing I can think that could be wonky is the PSU giving up the ghost (but this rarely happens, especially when you have a 1300w psu and the system only pulls about 500-600w, so plenty of headroom to account for degradation of psu parts).
The motherboard is setting something about the pci-e lanes that is not compatible with the Titan X, like the power, the hz, frequency, or who knows what? Not me, thats for sure. (doesnt the Cpu also manage pci-e lanes? Could it be a cpu setting then?).
Or, is it what 'I' think it is, and there is just a whole ton of wonkiness between the 900 series and the G-Sync DisplayPort monitors? A lot of the people who have had issues with the 900 cards have had evga psu's, and g-sync monitors.
It doesnt seem to be the bios battery, because everything is still as it was when I left it the last time, and it doesnt have to re-address everything (I have had this issue in the past on seperate machine, but this is not what is happening).
The memory is always the same, it doesnt change, and I have run memtest+ and the memory is fine.
The cpu rarely gets to 100F, although when my friends came up and we had 3 beasts of machines going at the same time , I did see temps in the 123f or 132f (dyslexia, sigh) (the room has its own ac, but it just held it at bay, lol). The cpu is water cooled, dual radiator.
The HDD's arent overheating. The 4TB drive gets the hottest, at 109F.
The case has all the same pressure fans (except the Titan X runs much faster and pulls more air than anything, at 100% fan speed), so there arent really any dead spots in the system. Its a HAF-X Super tower, with plenty of room and airflow.
The system works flawlessly with ANY other video card in it (well, any that I have, which is a lot actually). It just doesnt seem to work with the Titan X well at all.
 
Thanks for any thoughts on this, and hopefully not things you come back with will be all physical, lol, like pulling stuff out, because I am hurting like the dickens just from getting the evga tech the p/n for the psu, and will probably be down for a day or 2 because of it. Remember, I am disabled (but not drawing Disability money, waiting for the hearing in another year, so 3 years not being able to work, 0 income in 3 years so I cant just go out and buy a new motherboard, or a new power supply, or a new video card, etc, and things that used to take minutes and didnt hurt now take 10x longer or more, and end up with me in bed with ice packs and/or heating pads for a day or 3 before I can get going again). :( And, I dont have anyone to help me, except my wife, and she is disabled also from leukemia cancer in her bones. So, she can only help so much. (sorry, not trying to sell a story, but my problems are much more than what I am saying, and I really cant do much at all, although I was healthy until 3 years ago, and I built ALL of my computers, from 1995 until 2015 (this current system, where the Titans dont work), and have NEVER had so many issues with 1 piece of hardware in my life. And, I take CARE of my stuff. I make things last a long time. Most of my computers and parts can be seen as new (as I rarely use them) or I keep them in such nice condition.
 
Thanks for reading, and hopefully someone will come up with an idea that will make it work. This card really is better than the other 2 I have had to rma, except for the crashes and the weird settings that keep changing (100% should BE 100%, not a real 40% sometimes and 60% others, and 100% rarely).

Yes, as shown on the link, I have both hw monitor and afterburner running, and they both show the same thing.
And, this 3rd card is the only one that has had the crashing and rebooting issue.
So, you think I it could be either one?
Its going to suck not being able to have a gaming machine going until I get them both back from rma, but if thats what it takes.
So all this time, it could have been the psu going out on me?
Should I contact support and only send the psu in, then test when the psu rma comes back in, and if its still crashing, send the video card in?
Or, should I just send them both in, and test them both at once?
They did rma both the card and the psu, so....
 
I do have a psu I could test with, but it is an oldrer one, like from 2008/2009. It is a 1200w single rail though, so should have enough for the Titan ?
 
Thanks.

Yes, 1200 Watt PSU has enough power
 
Rig 2 - in my signature - has 2Way SLI - EVGA Titan X SC and custom loop all running on 1200 W PSU - pulls about 650 - 750 watts depending on load
 
Max I've seen is 925 W, with CPU OC and very heavy loads

Sajin
....
Rebooting whlie gaming is usually a sign of a faulty psu or video card.

And since the computer never crashed or rebooted before I got this 3rd titan x, and as soon as I put in the titan x it started  rashing and rebooting, that means it HAS to be the titan x, right?

I just got the RMA emails yesterday from them, but havent sent out the psu or the video card yet.
I did, however, get the 1300w evga psu out, and the corsair 1200w put in.
I am typing on the machine now, with only the psu changed out, same 3rd titan x.
Everything seems 'snappier'. More responsive.
So, maybe it was the psu.
 
I am going to game w/o afterburner on, and see if it crashes or not.
If it does not crash, then I know its the psu.
If it crashes, I know its the video card.
 
Right?

Hmmmm.
I swapped out the evga 1300w psu, and put in the corsair 1200w psu.
I ran quake 3, Vermintide, The Crew, and Wildlands.
I ran the titan x w/o afterburner, and the card got to 185f/85c.
BUT, it didnt crash.
 
I am not saying that this is definitive, 100% the psu, but the psu is going back no matter what.
I think I need a couple days testing (probably a week until my friends can play with me online) to make sure the card is not crashing at all.
But, still, yesterday it didnt crash with the evga psu in either. So today is not concrete proof that things are better.
As the video card started heating up, everything seemed to slow down, like my mouse speed, my kb speed. When it was fresh, with only getting to 135f (quake 3 doesnt stress it much) it was very snappy and fast still, just like on bootup. But, as the card got hotter, everything seemed to get wonky.
But, the card did well.
So, we will see.
I may have to talk to that evga tech guy and just get the psu replaced for now? Dang I forgot his name though.
 
Also, I noticed that the gpu memory controller did NOT go to 99 or 100%, AGAIN, this day. I am thinking it might be the card only when that memory controller gets to 100%.
Anybody have any ideas on how that gets that high? Any way to try to replicate it, besides having to just play games?
Maybe a gpu benchmark app or something?
 
http://s1378.photobucket.com/user/dave3d_bc/media/3rd_vidcard/3rd_titan_newpsu_zpsxl1rgtoo.png.html?sort=3&o=0
Link to new HW momitor today.
 

You'll get better scores in the 3dmark stability test with your gpu fan set to 100%.

As your system become sluggish, what are:
 
The CPU core temps?
 
The MB - Northbridge temps ?
 
As the Titan X heats up under load, what Fan RPM ?
 
You mention changing TIM, ever change the thermal pads ?

@Cool GTX:
Temps are all posted on my links to photobucket. But, like I said, all the cpu temps never get over 108, EXCEPT when my buddies came up this summer and all 3 heavy duty gaming machines were going, and the room temp got to ~85f. The cpu probably got to 123 then. But, that was only for a week, then they were gone, and the temps went back to normal. I have the corsair GT cooler cooling just the cpu, and all the air coming from it is super chilly. Only the video card gets hot.
 
I am not sure how hot the mobo NB gets, I dont know which sensor that is on HWMonitor. bHWM only shows cpu temps and speed, voltages (which with the Corsair, the 3.3v is stable now, and not fluctuating like the evga 1300w supernova was doing. All the rails are rock solid now). But, my fans are running at 100% all the time. I have: 2 250mm(?) fans, one in front blowing in across hdd's, one on the side blowing in on the lower mobo/video card. I have 1 140mm fan right above the front 250mm fan, blowing into the video card and over the top of the video card and over the ram. I have 2 140mm fans blowing out the top of the case, pulling all of the heat from the case (these are slightly higher rpm fans, so it gets all the heat out). I have 1 pci-e slot cooler pulling any heat off of the top of the psu. The psu has its own cool air channel, and sucks air from the bottom of the case to straight out the back of the case, so there is very little heat from the psu.
So, yeah, the whole case is one big heat sucking and spitting out machine.
 
The titan fans are already stated many many many times.
If I DONT have afterburner running, the fan only spins up to 40-53% fan speed, which gets up to 85C/ 185F, and everything slogs down after about 20 minutes of gameplay.
If I DO have afterburner running (at 100%, I dont do the fan curve, its just always at 100%), then the temps only get as high as ~71.7C/161F, and games dont get that 'slogged down' feeling until about when the temps get to the 145f area and above, which could take as little as 20 minutes, or up to an hour.
But, it ALWAYS gets that 'slogged down' feeling. EXCEPT when I use my GTX780. Then, everything seems to stay as it always should be, snappy and quick.
 
I only ever changed the TIM 1 time, and that was my 1st titan, evga said to do it to see if the temps would go lower. The other time (the rma for the first card) everyone said to get water cooling for it, so I bought the H90 and the Corsair kit, and I used the paste that came with the h90 already on it, the temps went crazy low, but the card still had the same issues.
When I talked about having crazy amounts of paste, and very thin paste, that was what evga put on there, not me.
------
 
@Dr.Death
Yes, I have the latest bios. It has the extra Haswell-E option in it, to support more stable voltages, also.
Also, I only have 1 Titan X, not 2 or 3. Just 1. And EVERY other video card I have put in this system has been super good, not a single issue. ONLY the Titan X (on my 3rd card right now, so 2 rma'd cards) has a problem.
But, I think Sajin and the EVGA tech I spoke to on the phone were right. Since I put in my other power supply, this thing has been running good, no crashes (although I DO still have to run afterburner, with the gpu fan set to 100%). BUT, the card has NOT replicated what it WAS doing with the GPU MEMORY CONTROLLER going to 99-100%
I think that is what has been causing the Titan X to crash all the time.
 
 
But, since putting in the Corsair PSU, I have not had a single crash. And, I did some big time video card stress tests with 3dmark and my own games.
I want to know HOW to get the GPU MEMORY CONTROLLER to 100% again though, to confirm that is actually what is causing the crashes.
And, I would like to do some other testing, like prime95 or anything else anyone can suggest, to stress the system, to make sure I am not going to rma this video card, since it seems stable with the new psu in it.
:)

EDIT:
NM
It has base 1000 clock, with a boost clock when it decides to.
I see in the amazon description now.
 
But, I am supposed to get 7010 memory clocks?
All I see it getting to is 3506? Is this dynamic also? If so, how come its not going to full speed when the memory clock is? Or, am I using the wrong hardware monitoring software to be able to see correctly?
 
http://s1378.photobucket.com/user/dave3d_bc/media/3rd_vidcard/3rdtitanx_clock_wonky_zpsnb2ztjdc.png.html?sort=3&o=0
 
Anyway, still no crash.
:)

Sajin
3506 x 2 = 7012. 

DOH!
Dumb me.
On my behalf, I WAS up until 6am this morning, benching this card and trying to get it to crash.
Funny how a power supply can make the video card crash like that?
The only thing I saw funny was the 3.3v rail, coming in very low at 3.189 every once in a while.
I didnt even know that the video card used 3.3v rail at all? Or does the PCI-E lane use it, and maybe thats why the vid card crashed?
Its weird, not having the card crash anymore, since it crashed ever since I first got it, lol.

Hmmm, cool.
Learn something new every day.
 
But, GTA V just locked up on me.
I couldnt even move the mouse, use the KB, or escape the game.
I had to Ctrl + Alt + Del
 
I hope its just GTA V though.
Going to try some other games.