Swapped the 2070 gpu for a gt710 from my desktop. Installed the spare HDD we created. Problem goes away! Shutdown the computer several times and then powered on with no issue. Did several Restarts from inside Windows and no issues. Tried to awake from hibernate and no issues. Sounds like you may have found the culprit. Now is there a way to test the 2070 gpa further to identify why it is failing?
One thing, I had to switch the monitor cables from DVI to HDMI because of my gpu limitations.
Reinstall the GPU and see if the problem reoccurs. Odd way for a GPU to fail, but it kind of makes sense if you think about it. VRM problem with the card, basically it's dying when the driver tries to load and ramps up the power usage. Wake from sleep is heavily dependent on GPU drivers.
I did reinstall the gpu and at first everything seemed fine. Then I went to put all the old HDD's back and the problem came back. After working around some things I think I've down to this. The computer seems fine when only one monitor is hooked up via DP. The gpu has three DP ports and I need to try the two monitors in all the combinations. I also have a HDMI cable that I could try along with one DP cable. But it appears that something about the combo of two DP cables and monitors is not working.
BTW the last hardware change on the computer awhile back was to add two identical monitors instead of two different sizes/manufacturers.
I've submitted a ticket with Gigabyte (the gpu manufacturer) to see if they can add anything.
Good idea. It's possible it's entirely some type of read EDIC read failure on the monitor. Although, I'm surprised you don't get an our of range warning from the monitor. Changing the interface to HDMI would be interesting to test.
Another question. Could you list the brand, model and specs on both monitors?
Both monitor are a 27" MSI G273, 1920x1080, 165 Hz refresh, IPS, 1 millisecond response.
I have not been able to boot the computer successfully more than one time in a row with just one monitor. Doesn't matter which port on the gpu I'm using (gpu has 3 DP and 1 HDMI ports). The process has been 2 boots that result in blank screen and freezing, one boot to Auto Repair. The Auto Repair comes up and I select Continue to Windows and that has worked on two times. On the third iteration of the entire sequence, I have to go into Safe Mode. I'm always able to get into Safe Mode. Tried restarting from Safe Mode and the screen froze again.
Not sure where to go next. I thought that the one monitor thing would solve it for now, but that now doesn't seem to have any affect on whether the computer freezes or not.
Lets reinvestigate the card. Perhaps reseating the card created a temporary window of functionality. I'd try swapping back to the other card and verify the issue goes away entirely on the same OS.
Swapped out for old card, kept OS and all other components. One monitor using HDMI ( I don’t have the right cables to try 2).
started right up four consecutive times without issue.
Can we do the same test on DP?
No, my gpu only has HDMI, VGA, and DVI ports
Half of of the tests this morning were with the HDMI cable
I do have a DP to VGA converter cable but that would be introducing a new variable.
Understood. Seems we're looking pretty firmly at the card now, though I'm wondering if the respective power draws are a factor here. I think our next step is to swap out for an equivalent GPU to the 2070 or PSU and see where we are.
Nice idea, but unfortunately I don’t have access to either. I did submit a ticket to Gigabyte to see what they can offer and to explore a warranty claim. Problem with them is that the automated system told me to expect a response in 3-7 business days. My other option is to head to Chicago to the Microcenter store and turn it over to service for any additional testing.
The manufacturer warranty can be tough in this situation, without knowing their testing methodology or whether they customize it to account for your issue. I'd bring it, or it may worth the cost to buy a new PSU locally just to rule that out for now. I'd say if the issue still occurs, you know it's a GPU issue.
May just wait till I hear from the manufacturer before deciding to bring it in. I’ll explore the idea of getting a psu. I’ll have to look at what we have in there. I think it was a bit oversized for the needs. I’m assuming that it would be best to match the wattage of the existing psu.
Once you get it to boot up, it works fine. Just not letting it hibernate.
Just so long as it's adequate for the hardware. 550-650W. I get the impression this is a power issue with the card. It's just not initializing properly on the cold boots and the some wakes from S3/S4.
So far can borrow a 400 watt supply. I’ll keep checking to see if I can borrow something larger.
Pretty light for a 2070. I wouldn't put any load on the card, should work alright for a short term test.
Two things...found a 850 watt power supply that I can use to test the system with a different psu. Should be able to do that testing early next week.
Heard from tech support at Gigabyte. Basically, they wanted me to test the card in another computer. Thought that didn't make much sense since I had already tested a different gpu in the computer in questions and it worked. If I wasn't able to to that they suggested I contact warranty to determine if an RMA process should be started. If I have to send the card in for that I'm assuming that the RMA process could take several weeks or longer to determine if this is a warranty issue. Not sure what other options I might have.
Cards we're testing with are fairly far apart. It would be interesting to see if the issue transferred to another system with the card.
If this is an issue with the card itself, manufacturer warranty would have to be carried out by Gigabyte and the time table sounds accurate. Alternative would be if you had a replacement plan on it from Micro Center at the original purchase, that could cover it as well.
Update: Wasn't able to use the 850 psu, so went to local Best Buy and picked up a 600 watt psu to test with. Hooked up new psu and had to use cables attached to psu, so was testing both.
1st startup was good with no problems, 2nd startup hung. 3rd and 4th start was good with no problems. After #4 I did a restart and had no problems. Decided to hook up second monitor again. Hung on that startup. Disconnected 2nd monitor and from that point we were back to 2 system hangs and then Auto Repair. Got into auto repair and safe mode and was able to restart to a successful start up. Still is hanging though most of the time. Seems like the best shot at a successful startup is the first start of the day.
Since nothing has really changed with the new psu, seems like all things point at the graphics card. Any final thoughts before I contact Gigabyte and begin an RMA process?
New update...Haven't been able to boot the system at all since the last series. I can usually get in by doing safe mode and then restarting. That's not working now either. Keeps freezing in the same place...right after the logo and the swirling dots disappear. The keyboard lights go out and the nothing but black screen.
Did finally get in. The secure boot was Not set to Other OS. I don’t recall changing if back to the UE mode.
Secure boot shouldn't be a factor. You'd get a BSOD and the failure would be obvious. What is sounds like is the GPU freezes or stops responding, and you see the freeze. When you're seeing the keyboard lights go out, it sounds like you can't see it, but the system BSOD's and the GPU doesn't respond and you hang on the initial POST. I think we have to move forward with the card at this point. The only alternative we have left is the board. The GPU is easier to RMA/replace.
Submit photos and a description of your PC to our build showcase
See other custom PC builds and get some ideas for what can be done
Services starting at $149.99