[SOLVED] Graphics card not being recognized

[Swordkd]

Here's the issue: Shut down my pc using the standard Windows shut down and when I turned it on, my video card wasn't working. Reinstalled the driver and it says that it can't install because no supported hardware is found.

MSI Vega 56, Asus Motherboard and Rosewill Hive 650w PSU(low tier B i believe.

I'm not thinking that the video card crapped out on me, but what are the chances that I lost the pci-e slot? I have a secondary that I could try. The PSU is older and I suppose that's a possibility too. Any other thoughts?

 

[Darkbreeze]

The chances the PCIe slot is to blame are pretty low in most cases. Much more likely is the fact that your PSU is not supplying the necessary power to support higher driver function. Hive series are very poor quality units. The fact that it is ALSO old, just adds to the pile. I'd start there.

Motherboard (And thus, PCI slot) are certainly possible, but not AS probable as PSU. How old is your motherboard and what is the model?

 

[Swordkd]

Asus P8Z77-V Lk . The version of the Hive PSU I bought wasn't bad quality, I did my research on that on. Guru3d lists the 750w as a recommended buy actually, and the 650w is from the same batch. The later versions weren't as good though. The video card has been working, but stopped after that shut down.

 

[Darkbreeze]

If you bought that unit BEFORE 2014, then it was an "ok" unit, but is too old now to be trustworthy and/or is a prime candidate for age related failures.

If you bought it AFTER 2014, then it is not the same internally as what you saw in that Guru3d review. And, any review from Guru3d I'd take with a grain of salt anyhow because their reviews are not very in depth, unlike those like we are accustomed to seeing from Hardware secrets, HardOCP, Techspot, Tom's hardware and JonnyGuru.com.

Also, the ONLY review I see for a Rosewill Hive unit through Guru3d is for a 750w model, and it is VERY common for companies to offer anything from mildly to wildly different internals on units in the same series when comparing models with different capacities. A 550 or 650w unit for example, from any given series, might have a COMPLETELY different internal platform, build quality, and capacitor selection from another model, whether 450, 750, 850 or 1000w within the same series. A good example of this is the (now) older M12II 620w Seasonic units that are a solid but outdated group regulated design that lacks SOME modern protections, while the 750 and 850w models of that same M12II series are not group regulated and have a DC-DC design that does have one of the primary protections, Over temperature protection, that is missing from the lower capacity group regulated models.

Just about every PSU brand has something similar, and from one capacity to the next within the same series they may not even be manufactured by the same company. CWT might make the 450-650w models while FSP, Great wall or another manufacturer might make other models in the series. Very common. Just because one model in a series is good or even mediocre doesn't mean other models in the series aren't terrible, and that is in addition to changes to a series that might happen down the road as has been known to happen in both directions, good to bad and bad to better, such as with the Hive and with Thermaltake's TR2 units, just as examples.

SO, you might be right, that it was not originally a terrible unit if you know for certain it was purchased prior to the platform or manufacturer change, but if it was, then it is certainly much older than the point at which it was ever intended to continue running beyond. Which still, admittedly, doesn't necessarily mean that it IS to blame, but it certainly means that it would be a really good idea to replace it no mattter what.


The problem is, your board is likely at or near 7 years old also, and might definitely be a candidate for some age related failures, so I agree it can't be ruled out.

I think the only way you are going to positively identify which of these, or none, is the problem, is by replacing one of them. Unfortunately when it comes to motherboard failures there is really only one way to absolutely identify that it is to blame if there is no clear visual evidence that something is wrong with it, and that is to eliminate all other potential culprits. Swapping in a different power supply or a different graphics card are probably your only actual paths forward, and are not terribly helpful, I know, in helping to determine what is wrong without throwing some parts at it. If you have spare parts or somebody willing to loan you parts, that might be a really good first step.

 

[LinuxDevice]

Sometimes it is as simple as the CMOS/BIOS settings changing. Especially if the button battery is low. Make sure your BIOS enables the slot. Maybe go into the BIOS and tell it to set to defaults, and then examine if any options need to be changed for using integrated versus discrete GPU.

 

[Darkbreeze]

That is a very good point. Given the age of that system it would be HIGHLY likely that the CMOS battery has run it's course and the BIOS is not retaining it's settings.

Have you tried removing the graphics card and using the motherboard video outputs from the CPU to determine if it's JUST a board failure, or if the failure to display is actually related to the graphics card at all?

 

[Swordkd]

That is a very good point. Given the age of that system it would be HIGHLY likely that the CMOS battery has run it's course and the BIOS is not retaining it's settings.

Have you tried removing the graphics card and using the motherboard video outputs from the CPU to determine if it's JUST a board failure, or if the failure to display is actually related to the graphics card at all?

I'm currently running off the on board video. I have another PC I can swap parts out of, just trying to narrow things down first.

Yes I bought the PSU after 2014, but it was a pre 2014 model. I know they went to crap afterwards. I agree that it's old, but I've not been near the wattage rating and I know that helps.

Anyway, thank you all for your thoughts. I'll start by swapping that card into my wife's system and go from there. If it works I'll assume the PSU because that does seem the most likely.

 

[Darkbreeze]

Sounds good, and right.

 

[Swordkd]

Update:

So, finally swapped my video card out with my wife's. I had to use the 2nd pci-e slot on my motherboard for her video card due to about a 2mm size issue, but it came up just fine. I put my video card in hers(she has a 750w psu) and nothing. So I swapped back using the 2nd pci-e slot for my video card, and it works.

So, I'm really confused. I don't know if it's an intermittent issue with the video card, or the primary pci-e slot on my motherboard. Regardless, it's running now without any issues so I'm just going to leave it alone and upgrade the mobo/cpu/ram like I've been planning anyways and go from there.

 

[Darkbreeze]

That's a good idea, and it sounds to me as though the card might have issues BUT it could be related to the CPU not having enough lanes available for the graphics card when used in the primary slot. There is mention of running at slower speeds, and PCI 2.0, when used with a CPU that is not 3rd gen, so perhaps it's just an incompatibility. I don't know your wife's machines specs though so that could totally not be related but it seems weird that it would not work in her machine and then work in your second slot. None of this seems to eliminate or convict anything as the culprit. An upgrade seems like a very good idea though and will likely cure the issue. I'd look at doing that before any warranty runs out on the card so that if there IS a problem with it, you can get it replaced, IF it is still under warranty which at three years I believe, it should be.