+ Reply to Thread
Page 1 of 2 1 2 LastLast
Results 1 to 20 of 25
  1. #1
    Junior Member
    Join Date
    Nov 2014
    Posts
    15

    Angry New Graphics Card Crash. Need Help ):

    I recently bought a geforce 780ti ghz edition and now i can't play Tomb Raider for more than 3 minutes without the game crashing. I have been watching Temps and nothing unusual
    I was thinking that i need a bigger power supply. thanks in advance

    PC

    Power Supply - 800w
    CPU - i7 4820k
    Ram - 16GB DDR3
    GPU -
    nVidia GeForce GTX 780ti GHZ edition

    (recently had GTX 760 and ran fine)

  2. #2
    ...Don't Mind Me... ShyguyXPC's Avatar
    Join Date
    Sep 2004
    Location
    USA
    Posts
    19,361
    really? a bigger power supply? You're already running an 800W PSU, what do you think you need? A Nuclear reactor?

    Do you even know how many watts the video card really pulls?

    800 is MORE than enough, even to run a GTX Titan Z.




    Whats the brand and model of the Power Supply?

    What are the operating Temps of the GPU when it occurs, the CPU temps when it occurs?





    (On a side note, why'd you go from a GTX 760 to a 780 Ti? With the newer GTX 980's out on the market?)

    for what it costs for a 780 Ti, you could get a pair of GTX 970's and run SLI, assuming your board supports that. at 145W a piece, thats less than what a single GTX Titan Z uses, power wise, and would beat the pants off a 780 Ti.

    Hell, just a single 970 could take on a 780 Ti and come awful damn close on its own.

    AnandTech | Bench - GPU14

    Computer Hardware, Video Cards & Video Devices, Desktop Graphics Cards, Newegg, gtx 780 ti - Newegg.com

    Computer Hardware, Video Cards & Video Devices, Desktop Graphics Cards, GTX 970 - Newegg.com



    Back on topic, it could also be driver related, did you update to latest drivers for the card? Updates for Tomb Raider as well?

    GTX 780 Ti, though a GTX 700 series card is actually based on the GTX Titan, not the Kepler based cores for the GTX 770 and below.

    Maybe also try clocking the 780 Ti back on clock speeds to a standard 780 Ti, and see if its stable, it could be an out of the box unstable factory overclock card as well, which would be covered under warranty if it was new or had a transferable warranty like many EVGA cards have.

  3. #3
    Junior Member
    Join Date
    Nov 2014
    Posts
    15
    the PSU is a casecon adk-800-80 and the highest temp I've seen is 73 degrees Celsius (163F) and CPU is 36C (96F)

    I've removed all the old drivers and installed the new ones and then updated them.

    I do agree with you, but it was a present. Still good tho haha

  4. #4
    ...Don't Mind Me... ShyguyXPC's Avatar
    Join Date
    Sep 2004
    Location
    USA
    Posts
    19,361
    Never heard of that brand of PSU.

    looking at prices for it, that I've seen online via google search, its a bit cheap in price for an 800W unit, most good 800W, heck even 700W units, you can expect to pay $100 or more, not $60-70.

    GTX 780 Ti, just stock clocked is 250W for the card itself, Factory OC card at 1GHz or more, might be a tad more.

    but GTX 760 in comparison, uses 170W. 80W difference.

    How old is the PSU, considering the rest of the system hardware, guessing maybe a year or so tops?

    $70 for that unit here: Storm Computers Pty Ltd | Your 7 Day Computer Store, with High Quality Service & Sales, Repairs, Computers, PC Upgrades, Open Sunday.

    another site has the same model for $150.

    But a few other forums, members have stated that Casecom Power Supplies are just complete garbage.

    One went so far to comment, after going through 4 of them, just not worth it, once you load it up, 50% or more of its max load, they start having problems.

    on a 800W PSU, with a 130W CPU, 250W GPU, and rest of the system, that'd be real close, if not over 400W right there.

    It could still be something as simple as a part or cable in the case got bumped during installation, CPU Cooler for example.

    But if this only happens with one game, or specifically games where the GPU is at load.

    It could be that the PSU can't feed enough power to the card when needed, and some sort of issue arises with the card and game in question.

    Does it happen with any other games too (crashes, at, before, or after 3 minutes), or just Tomb Raider?


    Is your CPU overclocked as well?

    How many HDD/SSD's in the system? DVD/Bluray?

    Any other expansion cards? How many case fans?

    Liquid cooler kit, custom H2O Loop? Aftermarket Heatsink?

    I'd take a look at this Power Supply Calculator and see roughly how much power your system really is drawing, as well as recommended Wattage Rating.

    eXtreme Power Supply Calculator - The most trusted PSU Wattage Calculator


    My system as is, shows it should be pulling about 575W, recommends 625W, and its probably a tad more, I have my two GTX 560 Ti's overclocked to 900MHz, sometimes 950MHz, up from factory overclock of 850MHz, which is up past stock clocks of 822MHz. So they should be drawing a little more power than the standard 170W per card.

    I'm running an NZXT Hale82 850W PSU (rebadged Seasonic M12 II unit), one of the best quality brands you can buy is Seasonic, also why many companies like NZXT, Corsair, XFX and even some Antec's and a few other companies, use their internals, and just slap on a custom casing. FSP or Fortron is another one thats good.

    a couple other OEM's also make good components used in other popular brands like Coolermaster, Thermaltake, OCZ/PC Power & Cooling, Enermax, Silverstone and even EVGA.

    I can't seem to find any relevant info on the wattage or amp ratings for the +12V rail/rails for that PSU that you have, so no idea if it is 800W total and just has a low amp rating for the 12V rail, or what...

  5. #5
    Junior Member
    Join Date
    Nov 2014
    Posts
    15
    I checked my wattage it was 507w and recommended 557w

    I played Grid 2 and saints row 4 for about an hour and they both ran fine

    then i played Dishonored and Bioshock infinite for about half an hour

    Remember i was playing Tomb Raider fine for around 6 hour periods with the 760.

    Could it still be the PSU? sounds pretty convincing (at least they aren't expensive)

    Thanks for putting effort into helping me.

  6. #6
    ...Don't Mind Me... ShyguyXPC's Avatar
    Join Date
    Sep 2004
    Location
    USA
    Posts
    19,361
    I haven't searched around, but its always possible its an issue with the card, remember the 760 and 780 Ti aren't the same GPU, but more than likely I'd normally say its the PSU thats the culprit, if the 780 Ti runs fine with other games.

    But since its only the one game, its hard to say for sure, if it is the PSU. I have Dishonored and Saints Row 4, and neither are really heavy taxing graphically wise, but SR4 does have a lot of lighting and particle effects, and such too, hard to say.

    Not sure how graphic Tomb Raider is, haven't played the new remake yet.

    Could be some driver related issues with the card and the game. GTX 780 Ti, is based on the GTX Titan, which was never a Gaming card anyways, the Titan was a converted Nvidia Tesla card used in 3D rendering, and GPU based Computational/Processing tasks for workstations and servers.

    They merely enabled video output capability (original Tesla derivatives had no outputs, the cards are slaved to the system and used to independently process data, sort of like Distributive computing, or bit coin mining, etc)

    Nvidia tweaked them a bit, and made them capable of being used as a dedicated Video card... as a result of its popularity with people with money to burn, it was considered a hit, at the same time we had the GTX 690, which was a dual 680 card. the 690 went for an easy grand, when Titan came out, it was a good grand and the 690 was dropped to around $700-800 range. The 690, though consumed more power and was an SLI hindered option (games that don't use SLI well or at all, limited to a single 680 chip), was still the better deal, even though Titan used less power, and was about on par, or a tad ahead, didn't warrant the $200 or more increase.

    As 690 went bye bye with debut of the GTX 700 series, Titan dropped in price, and then a lesser spec'd variant debuted, GTX 780. Then the 780 Ti, which is just a faster, FULLY Enabled version of Titan. Then the Titan Black which is the same as the 780 Ti but I think has 6GB memory buffer. Then the Titan Z, which is basically a dual GPU Titan Black, 12GB total.

    all 4 are Titan/Tesla based cards.

    Where as the 760, and 770, were revised versions of the GTX 600 series, the 770 is the same as the 680, has some massaged clock speed increases, a bit better power consumption, roughly same performance, or a tad better for much much lower price.

    760 was a reduced spec revised version of the GTX 670, was something between a 660 Ti and 670, in performance, (though the 660 Ti IS a 670, just has a few components disabled, but same clocks, same number of cores, etc)

    Its possible, Tomb Raider just doesn't play nice with Titan Based cards?



    I was just reading this thread in a google search, you might want to try disabling Tessellation and TressFX in game options and see if the 780 Ti runs better.

    does the gforce GTX titan work? :: Tomb Raider Algemene discussies

    I had something similar in Civilization 5 that was driving me nuts, during some CG cutscenes, the game would ALWAYS crash to desktop, and I couldn't figure it out. Game runs on max settings on a single 560 Ti, but I ran in SLI, disabled SLI, ran with and with out V-Sync, messed with various settings...

    Only to find out, that there was some sort of Video processing setting, that specifically had to do with the CG parts, if disabled, it no longer crashed.

    Sure enough, that did it. Had nothing to do with actual gameplay settings, just video/CG parts.

    Apparently if I was running an AMD card, or a newer GTX 600 or better, I could enable that and not have the problem, but as long as an Nvidia 500 or below was running the game, it would crash with that setting enabled.

    Common sense on developers part would analyze the drivers and GPU in use, and gray out or enable it depending on the card used, but apparently not.



    otherwise if that doesn't work, then its possible it could be the PSU... but its hard to say, since its only the one game thats doing this.

  7. #7
    ...Don't Mind Me... ShyguyXPC's Avatar
    Join Date
    Sep 2004
    Location
    USA
    Posts
    19,361
    In any case, I'd still consider replacing the PSU with a known quality/reliable one anyways.

    Corsair, Seasonic, etc, even Coolermaster or something like that, depending on what you can spend.

    A good 650W should be plenty with some wiggle room, though if you ever planned on going SLI, I'd probably look at a 750-850W at least, with 2x 250W cards...

  8. #8
    Junior Member
    Join Date
    Nov 2014
    Posts
    15
    it still crashed after turning Tessellation and Tress FX off, i'll try buying a new PSU and see how it goes.

    Thanks for your help.

  9. #9
    Junior Member
    Join Date
    Nov 2014
    Posts
    15
    I got my new Power supply i went with a corsair ax760, and it still crashes ):

    I check temperatures after the crash and it all looked normal.

    Could it just be nvidias latest drivers dont cooperate with tomb raider?
    (I finished tomb raider with my old card anyway)

  10. #10
    ...Don't Mind Me... ShyguyXPC's Avatar
    Join Date
    Sep 2004
    Location
    USA
    Posts
    19,361
    Could be, I have had that happen with cards in past, not work well with a game, down the road later at some point suddenly it works.

    (usually had it happen with ATI/AMD Drivers though, but only happened a few times, as I've only ran ATI Cards once or twice in the last 10 years, otherwise usually Nvidia Cards, at least for my main game PC).

    There's got to be something with the game, or the drivers, that seem to cause the crash. I know one game, years ago, that had a similar problem, was actually the games fault, not video drivers, or anything. Sid Meier's Railroads, was designed for Windows XP, and 32-bit OS's.

    Seems, running the game on a 64-bit OS, caused some sort of crash at some point when using more than 2GB RAM... Game would run for a specific amount of time, and crash almost right on cue, There was a tweak to modify the games Executable, and enable addressing beyond 2GB (even though the game never used more than 2-3GB anyways), but once that tweak was enabled, the game would not crash anymore.

    So its entirely possible it could be something with the game itself, possibly corrupted game files, or something.


    Does it happen at the same point when it crashes, or is it just crashing a few minutes into the game, no matter where you are, or what your doing?

    Have you checked CPU Temps? That could be a culprit as well?

    I'd see about using a Tool or app, and monitor CPU/GPU temps in game.

    Run something like EVGA PrecisionX or MSI Afterburner, they both can display GPU and CPU temps on screen in most games (some, very few, don't allow in game DirectX Overlay's, so you won't be able to see it, but most games do, I rarely don't have that running, except when I'm recording in game video with Open Broadcaster Software, which doesn't like to record when other programs are using Overlays).

    Then maybe just see what Temps the CPU and GPU are at while in game, leading up to a crash.

    Its usually easier to do that than enable and set up logging software, which many programs can do, like CPUz, and GPUz, as well as CPUID's HW Monitor program. As well as others like Core Temp.

  11. #11
    Junior Member
    Join Date
    Nov 2014
    Posts
    15
    I've Been playing Dragon age inquisition for long periods at a time with graphics on ultra, and have had no problems at all. (besides the fact that i could cook an egg on top of my pc case)
    And all Temperatures are normal and the game runs smoothly.

    and i ran tomb raiders bench mark a many times. (it used to crash after the second time in a row)

    I think Nvidias new drivers fixed the problem.

    thanks for your input.

  12. #12
    ...Don't Mind Me... ShyguyXPC's Avatar
    Join Date
    Sep 2004
    Location
    USA
    Posts
    19,361
    Hopefully that fixed things. I know there was new drivers recently, had Geforce Experience utility nag me about upgrading recently, though since I'm not running latest higher end games, and not experiencing much in problems, and on older GTX 500 series cards, figured I don't really need to, unless there's a specific fix I need in newest releases.


    With the recently filed Countersuit that Samsung just filed against Nvidia, trying to get the International Trade Commission, to block the sale of any Nvidia Products in the US/North America, I'm seriously thinking about dropping the cash on a newer GTX 970, or even a decently priced GTX 770 in a couple weeks if my Pay Check is big enough.

    Rather than wait till income tax return season.

    Though there are some decently priced Refurbished Radeon R9 290 and 290X cards on newegg for that same price range as well, though from benchmarks, the 290/290X trades punches with GTX 970, with the 970 pulling ahead on lower power consumption and peak load temps.

  13. #13
    Junior Member
    Join Date
    Nov 2014
    Posts
    15
    Something annoying just happened to me, I was downloading watch dogs and looking at more steam games when my PC just randomly turned off and then on again for 2 seconds. It done that 3 times without me doing anything. And now when I try to turn it on it crashes within a second and won't even try to turn on for at least 5 minutes. I've had no problems all week with my new power supply and graphics card all week until half an hour ago. I checked all my cables and nothing unusual... My PC was only turned on for about 30 minutes prior to this and is plugged directly into the wall.

  14. #14
    Junior Member
    Join Date
    Nov 2014
    Posts
    15
    And there's a green light on my mother board that stays on so there is power

  15. #15
    ...Don't Mind Me... ShyguyXPC's Avatar
    Join Date
    Sep 2004
    Location
    USA
    Posts
    19,361
    Quote Originally Posted by ShyguyXPC View Post

    Have you checked CPU Temps? That could be a culprit as well?

    I'd see about using a Tool or app, and monitor CPU/GPU temps in game.

    Run something like EVGA PrecisionX or MSI Afterburner, they both can display GPU and CPU temps on screen in most games (some, very few, don't allow in game DirectX Overlay's, so you won't be able to see it, but most games do, I rarely don't have that running, except when I'm recording in game video with Open Broadcaster Software, which doesn't like to record when other programs are using Overlays).

    Then maybe just see what Temps the CPU and GPU are at while in game, leading up to a crash.

    Its usually easier to do that than enable and set up logging software, which many programs can do, like CPUz, and GPUz, as well as CPUID's HW Monitor program. As well as others like Core Temp.
    Sounds like it could be CPU related, or could be a bad stick of RAM.

    Maybe take the CPU cooler off, clean off thermal paste, apply some new stuff and resecure the heatsink on again, and make sure its on tight, see if that helps. Otherwise, if its a bad stick of RAM, you could try running something like Memtest86 or Memtest86+.

    That or try the system with one stick of RAM in the main slot and see if it will boot with each. Basically test each stick and see if one is bad.

  16. #16
    Junior Member
    Join Date
    Nov 2014
    Posts
    15
    I put new thermal compound on and it worked for about 2 minutes and I got pretty excited I plugged everything back in and got ready to watch game of thrones... Theeen... Crash.

    Then I messed around with the ram and still nothing.

    Its crashes are pretty consistent it runs for 1 second then stops for about 8 and just repeats

  17. #17
    ...Don't Mind Me... ShyguyXPC's Avatar
    Join Date
    Sep 2004
    Location
    USA
    Posts
    19,361
    Thats usually one of two things, either the CPU is overheating, or the RAM is installed in wrong slots (for example my motherboard has 6 slots, 3 and 3, for Triple channel RAM, if the three are installed in the wrong slots, it will do that).

    When you connected the new PSU, you connected the 24 pin ATX, but did you also reconnect the extra 8 pin (4+4) pin Auxiliary/Supplementary Power connectors as well?

  18. #18
    Junior Member
    Join Date
    Nov 2014
    Posts
    15
    It don't think it could be the ram is in the wrong slots considering they haven't been touched all year.

    If that's the extra 8 pin connector that's connected to the 24 pin atx that only goes on the psu side then yes that's plugged in. If not please elaborate thanks.

    I must remind you that played roughly 20 hours of dragon age and small portions of other games over a week with the new power supply. And then suddenly yesterday or the day before it shit itself.

    Could the CPU overheat in a matter of 2 minutes? I could imagine if it does then that seems plausible considering it keeps trying to turn on by itself then cuts out within a second after the first 2 minute crash

  19. #19
    ...Don't Mind Me... ShyguyXPC's Avatar
    Join Date
    Sep 2004
    Location
    USA
    Posts
    19,361
    yes, the 8 Pin from the PSU, should of been labeled 12V ATX or Aux 12V or something, not the 8 Pin PCIe for the video card, have seen people accidentally connect to that one, and connect the other end to the motherboard, usually had to forcefully connect them, when using Modular PSU's, but for a wired PSU, should be no mistaking connecting those.

    Some PSU's provide a 4+4 pin and a full 8 pin, either one should work, they're just for convenience, as the 4+4 can be a pain in the arse connecting to a board sometimes, although come in handy for boards that only use a single 4 pin Aux connector.

    But otherwise, typically from my own experience on high end Intel systems like ours, the symptoms were either CPU overheating (for whatever reason) or the RAM was installed in wrong slot.

    Could possibly be some sort of issue with the motherboard, What was the Mobo Brand and Model again?

    I know my EVGA boards, have that dorky stylish Aluminum heatsink, looks cool, but its passive, epoxied to chipset, and heat up way too much (just piss poor cooling over all).

    otherwise, no idea what else it could be.

  20. #20
    Junior Member
    Join Date
    Nov 2014
    Posts
    15
    It is a p9x79 le it's heat sink is aluminium.

    2 red LEDs turn on as well, A DRAM led and the CPU led

Quick Reply Quick Reply

If you are already a member, please login above.

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Similar Threads

  1. New Graphics Card, now games freeze and crash
    By Mattio55 in forum Technical Support
    Replies: 17
    Last Post: July 20th, 2012, 06:13 PM
  2. Graphics card crash with Half Life 2
    By Nbbear in forum PC Hardware
    Replies: 4
    Last Post: April 7th, 2009, 11:48 PM
  3. Video card crash
    By Tanis1978 in forum Technical Support
    Replies: 1
    Last Post: February 13th, 2009, 02:04 AM
  4. Graphics crash to desktop
    By SiLVerBuRn in forum Technical Support
    Replies: 10
    Last Post: May 2nd, 2004, 02:30 PM
  5. Crash crash crash creash AHHH
    By shampoo in forum Technical Support
    Replies: 11
    Last Post: December 7th, 2002, 09:19 PM

Posting Permissions

  • You may post new threads
  • You may post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Copyright 2014 All Enthusiast, Inc