Monitor switches between HDMI and Analog

[Johnson Deng]

This has happened three times the past few months and basically what happens is that my monitor will stop displaying while gaming and switch back and forth between HDMI and Analog but it doesn't turn off. My computer stays on the whole time and if I'm on a voice program I can still talk and hear others. Also the game I would be playing would crash and I know this because my friends will tell me that I have left the game when it happens. The only way to have my monitor display again is to manually shut down my computer and turn it back on.

I have also recently replaced the case, CPU, motherboard, and PSU (4 months ago) so I do not think that they are an issue. I have recently overclocked my GPU and I suspected it may be the case so I have stopped overclocking since then. This issue does not occur often enough to where I can pinpoint the issue so any help is appreciated. I've been told a dying GPU would make my monitor display a message "no signal" but I have not received this message just the switch between "HDMI" and "Analog", is this message the same as "no signal"? The GPU is a little over 2 years old and the monitor is 4 years old.

PC Specs:
Intel Core i7 4790K Quad Core @ 4GHZ
ASUS Z97 Pro Gamer ATX LGA1150
EVGA GeForce GTX 970 Superclocked ACX 2.0 4GB GDDR5
Corsair RM750x Fully Modular Power Supply 80+ Gold
Kingston HyperX Fury 16GB Kit (2x8GB) 1866MHz DDR3
Crucial BX200 480GB SATA SSD
Western Digital Caviar Blue 500GB 16MB Cache 7200RPM SATA HDD

 

[TMTOWTSAC]

What monitor do you have? If it has some sort of auto source detection, when it loses signal it starts scanning inputs looking for anything connected to it. I know Samsung's do it, and a few other brands. Setting source to "manual" would stop the cycling, but either way it certainly sounds like a dying GPU. Dying monitor wouldn't cause the game to crash.

 

[Johnson Deng]

@TMTOWTSAC Hey I believe I have a Samsung 23.6 inch S24D300HL monitor. I'm at school and on mobile so its hard to reply, as soon as I'm home ill double check. I haven't had any artifacts or graphical glitches while gaming yet and isn't two years a little young for the card to be dying?

 

[TMTOWTSAC]

Ah, sorry. When I said "dying" I was referring to how you used it in your post. Ie, that the card had stopped sending the signal, as opposed to the monitor no longer receiving the signal. The first could account for a game also crashing, whereas the second generally would not. Didn't mean dying in the sense of the GPU being a goner.

Losing the signal could be a lot of other things. Driver issues, overheating, pcie errors, etc. But yeah, it could also mean the card is on its way out. One thing you can check is to try unplugging and re-plugging the HDMI cable. There's a handshake that occurs so it might reset the signal and give you a chance to see what's happening.

 

[Johnson Deng]

Ah, sorry. When I said "dying" I was referring to how you used it in your post. Ie, that the card had stopped sending the signal, as opposed to the monitor no longer receiving the signal. The first could account for a game also crashing, whereas the second generally would not. Didn't mean dying in the sense of the GPU being a goner.

Losing the signal could be a lot of other things. Driver issues, overheating, pcie errors, etc. But yeah, it could also mean the card is on its way out. One thing you can check is to try unplugging and re-plugging the HDMI cable. There's a handshake that occurs so it might reset the signal and give you a chance to see what's happening.

Hey, I'm home now and I can confirm that the monitor is actually a Samsung 23.6 Inch S24C350HL. Thanks for all the information so far btw. I have actually tried unplugging and re-plugging the HDMI cable, does not work only a hard restart works. I will observe temps and update my drivers to see if they are an issue.

Would it be a good idea to start planning on getting a new card now? Also, like you said I wouldn't imagine that my games would be crashing if the monitor was the issue, the only thing that bothers me is that the card is only a little over 2 years old and it might be dying, isn't that odd?

 

[TMTOWTSAC]

The fact that your voice program keeps running seems to imply that Windows doesn't realize anything is wrong. That definitely points towards something going on with the card. You've double checked that your overclock settings have all reverted back to stock values? You can also try reseating the card and pcie cables, according to some other users here a loose fit can cause power delivery issues. Seems unlikely if the card has been working fine until now, but it's worth a look.

 

[Johnson Deng]

The fact that your voice program keeps running seems to imply that Windows doesn't realize anything is wrong. That definitely points towards something going on with the card. You've double checked that your overclock settings have all reverted back to stock values? You can also try reseating the card and pcie cables, according to some other users here a loose fit can cause power delivery issues. Seems unlikely if the card has been working fine until now, but it's worth a look.

Yeah, that's what I was thinking, the last time it happened was two days ago and since then I've reverted everything back to the stock settings and I haven't encountered the issue yet but like I said before it does not happen often at all, the first time it happened was roughly 2 months ago and the second and third time was this week.

I just reseated the card and PCIE cables when I got home, so I guess all we can do now is wait. I'll prepare for the worst and look for new cards.

 

[Johnson Deng]

The fact that your voice program keeps running seems to imply that Windows doesn't realize anything is wrong. That definitely points towards something going on with the card. You've double checked that your overclock settings have all reverted back to stock values? You can also try reseating the card and pcie cables, according to some other users here a loose fit can cause power delivery issues. Seems unlikely if the card has been working fine until now, but it's worth a look.

Hey, I'm just checking back after reseating the card and cables, while also cleaning the dust and updating the drivers, it has once again happened while I was playing Diablo 3 with my friends. The increased frequency of this issue recently has me suspecting that it is my GPU dying. My game crashed and my monitor stopped receiving a signal from my graphics card but skype was still working for me like I said before.

Should I look into purchasing a new graphics card?

 

[TMTOWTSAC]

I hate recommending replacement if it turns out to be something else, but it does seem like the most likely culprit. It's happening in different games, the card isn't overheating, you've already replaced the PSU, and Windows doesn't crash along with the game. Unless you have another system you can test the card with, all that's left would be swapping your old components back in to see if it still crashes. And that would be a colossal pain that probably isn't worth it.

I found a thread on the EVGA forums here:

http://forums.evga.com/GTX-970-Black-Screen-Crash-during-game-SOLVED-RMA-m2248453.aspx

This type of failure seems fairly common for that card. Just about everyone in the thread wound up needing replacements. Some reported that they were able to get stable again by underclocking. But it only worked for a few people and they were taking as much as a 33% hit on their GPU/VRAM clocks.

 

[Johnson Deng]

I hate recommending replacement if it turns out to be something else, but it does seem like the most likely culprit. It's happening in different games, the card isn't overheating, you've already replaced the PSU, and Windows doesn't crash along with the game. Unless you have another system you can test the card with, all that's left would be swapping your old components back in to see if it still crashes. And that would be a colossal pain that probably isn't worth it.

I found a thread on the EVGA forums here:

http://forums.evga.com/GTX-970-Black-Screen-Crash-during-game-SOLVED-RMA-m2248453.aspx

This type of failure seems fairly common for that card. Just about everyone in the thread wound up needing replacements. Some reported that they were able to get stable again by underclocking. But it only worked for a few people and they were taking as much as a 33% hit on their GPU/VRAM clocks.

Hey again, just really want to stress the appreciation for all the help you've provided. I'm going to look into getting a new card because like you said it does seem like the GPU is the cause, nothing else would make sense. The increase frequency of the issue recently has me worried that it'll be fully dead soon. After I swap out the card I'm going to give it to a friend for him to use and see if he encounters any similar issues, if he does then we will know for sure.

Just to be sure that it isn't a temperature problem, my GPU is at 30-40C idle and 60-75 under load while gaming, these are safe temperatures correct? I clean the dust out of computer pretty often.

 

[TMTOWTSAC]

75 doesn't seem excessive from what I've heard. Lot of people here with OC'd cards were hitting 80. The 900 series cards ran hotter and were more power hungry to begin with. They went 2 generations on the same process and it was the only way to squeeze more performance out of them. It may be they'll have a shorter lifespan as a result, but too early to tell.