Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
SlayVus
Jul 10, 2009
Grimey Drawer

King of Solomon posted:

OCZ1000PXS, never had any problems with it before. It certainly should be sending out enough power to the drive.

Why do you ask?

As power supplies age they can send incorrect voltages and amperage. This can wreak havoc on hardware.

I had a corsair 1000hx die about a month ago. I think it took my 980 Ti with it because the card died just a few days after replacing the PSU with a seasonic snow silent 1050w.

SlayVus fucked around with this message at 12:24 on Jan 6, 2016

Adbot
ADBOT LOVES YOU

SlayVus
Jul 10, 2009
Grimey Drawer
Here is a question that I have never seen asked. How often should I replace my AIO water cooling unit?

SlayVus
Jul 10, 2009
Grimey Drawer
If you want something that isn't likely to get damaged, Samsung's Portable SSDs. You can get a 1tb for $300 out a 2tb for $900. At that price though, you're better off buying three 1tb.

SlayVus
Jul 10, 2009
Grimey Drawer
It'll probably fit up to a 980 Ti. Going to go the safe route and say a 970 will fit without problems. I can't find any inside pictures though, so it's hard to say.

However, the power supply is only 180w. You will need to upgrade it to at least a 450w, 550/600w would provide give more headroom for a higher powered card.

The motherboard is 9.4"x9.4" so as long as the card is under 9.4" in length it'll fit. However, the motherboard is poo poo. 1 pci-e 16x slot, 1 pci-e 1x, 2 ddr3 dimm slot, and 1 m.2 slot.

SlayVus fucked around with this message at 19:49 on Mar 25, 2016

SlayVus
Jul 10, 2009
Grimey Drawer
Make sure you're applying thermal interface material and not thermal epoxy. I did that mistake several times. It works, but not like the actual TIM. You'll also glue your parts together and might damage something.

The reason I messed up is because my tube had no label on it.

SlayVus
Jul 10, 2009
Grimey Drawer
What processor does your computer have? Are you running any overclock?

SlayVus
Jul 10, 2009
Grimey Drawer

Shaocaholica posted:

Stumbled across this on eBay.

New old stock Supermicro P-III complete towers still in sealed boxes. The price is outrageous but just the fact that this guy has more than 10 of them still sealed is amazing.

https://www.ebay.com/ulk/itm/301950361620

Why would anyone pay $2,250 for one of those when you could spend that on a 6700k build with 1080?

SlayVus
Jul 10, 2009
Grimey Drawer
So I bought a 512GB X400 SSD for a friend to put in his system. He currently has an unpartitioned 1TB hard drive fit his windows installation that is probably 90-95% full. What's the best way for me to move his windows installation to the SSD, like software solution wise?

SlayVus
Jul 10, 2009
Grimey Drawer

FCKGW posted:

Most companies will offer some sort of cloing software. Sandisk has a re-badged closing software for free you can use.

http://kb.sandisk.com/app/answers/detail/a_id/17292/~/sandisk-ssd-concierge-service-cloning-guide

Problem is I can't clone 900GB to a 512GB SSD. I need some way to partition off the windows installation from the rest of the drive to clone it to the SSD.

SlayVus
Jul 10, 2009
Grimey Drawer
On my OS SSD, a mushkin 240 Chronos Deluxe, WinDirStat shows a giant 6.6GB file in the root of the drive. It labels the file as <Unknown>, I have enabled Show Operating System Files and show hidden folders and objects. This 6.6 GB file does not appear in explorer. I have a 2.4GB page file and hibernation is not enabled.

SlayVus
Jul 10, 2009
Grimey Drawer

Knobb Manwich posted:

Quick dumb question, can I forgo case fans if the gpu and cpu are hybrid coolers? I have a 160mm intake at the front of the case, vents along one side. I'm wondering if I can use the 120mm rear fan slot for the gpu and the 2x120mm slot on top for a bigger cpu cooler.

Fwiw I'm in Australia, ambient temps are heading towards 40 C and my air conditioning sucks.

Case fans help with air flow around components that need it. The VRM on the motherboard need cooling, as well as any hard drives in the case.

SlayVus
Jul 10, 2009
Grimey Drawer

Buca di Bepis posted:

I found a brand new 4-bay USB 3.0 raid enclosure at Goodwill today

What do I do with a 4-bay USB 3.0 raid enclosure

Test each bay to see if it can read and write in excess to a drive. Then you wait and see if it kills the drive from faulty voltage regulation.

SlayVus
Jul 10, 2009
Grimey Drawer
I can't find any information on this subject, but I'm kind of looking at building a new PC. One feature I want to add to my PC is having a ThunderBolt AIC to rig for having two desktop locations. However, I'm trying to find out if I can use say a Gigabyte or ASUS TB3 AIC in a board that has a TB header like Asus or Gigabyte puts on their boards. MSI and ASRock don't make TB3 AICs.

SlayVus
Jul 10, 2009
Grimey Drawer
Friend is having this issue with his PC. It says it only happens every so often, like every two weeks and only when the computer is idle. I think it may be bad VRAM on his GPU, but I'm not entirely sure.

SlayVus
Jul 10, 2009
Grimey Drawer

Puddin posted:

Whats the fan look like? Could be graphical artifacts from overheating. If he can monitor the temps and see if it spikes around when that happens it may be the culprit.

Also check the cables.

I know it's a GTX 970, but it's happening at idle, not under load. Thats the windows login screen, nothing would be stressing the GPU at that point.

SlayVus
Jul 10, 2009
Grimey Drawer

Kibbles n Shits posted:

Is there any condition under which a GTX 970 would work in my PCIe x16 @x4 slot, but not the primary PCIe x16 slot? I'm trying to narrow down these instability issues. I thought I had a bad PCIe slot at first but I put an older card in that slot and it works fine now. Putting the 970 in the primary slot creates massive system instability, but putting it in the other slot works just fine.

Usually the 16 and 8x slots are connected directly to the CPU. 4x and 1x to the chipset. Could also be too much power draw on the PCI e slot, see if motherboard​ has a supplemental power connector for the slots.

SlayVus
Jul 10, 2009
Grimey Drawer
My friend has a Windows 8 machine that I believe is running secure boot. Is it possible for me to mirror the hard drive to an SSD are will I need to reinstall Windows all together?

SlayVus
Jul 10, 2009
Grimey Drawer

Blue On Blue posted:

Stranger and stranger

I turned off all poe features to a port and plugged it in there , same thing. As soon as the link is live ALL poe ports are shut down and the only active link is the one the slim pc is plugged into

Tempted to contact the manufacturer and see if they have ever seen this before

Weird thought, but is the switch a Gigabit switch? Are there other devices handshaking at a 100 Mbps connection and the pc on a Gigabit? Could you enforce 100 Mbps on all ports?

SlayVus fucked around with this message at 02:34 on Oct 20, 2017

SlayVus
Jul 10, 2009
Grimey Drawer

Kintamarama posted:

My motherboard is of a pre-RGB header vintage (Asrock B85m ITX) and the case is cramped enough that a hub/controller type thing would be a difficult fit. If I bought one and plugged it into a fan header only, would the LEDs not even switch on? Or does the RGB connecter only serve to control the LEDs, and not plugging it in puts the fans on some kind of default colour cycle?
Usually the fans have two headers one for RGB one for fan controls. You can get a USB RGB fan controller for sound $40-70 and pick up phanteks RGB fan frames for $10-$20 a pop and just stick them over your existing fans, depending on model.

Something like this. http://www.cmstore-usa.com/rgb-led-controller/. I legitimately do not know if this would actually work with these specific RGB fan frames.
Then either of these.
$10 version

$20 version

Edit: I believe this route might be cheapest if you're looking at keeping it like under say $150. Two and three fan packs with controller can cost $100 on their own. For $40+how ever much for each fan. You could do 6 fans in RGB for $100.

SlayVus fucked around with this message at 19:51 on Nov 10, 2017

SlayVus
Jul 10, 2009
Grimey Drawer

Reality Winner posted:

I found and installed some grommets, but they didn't do much. I'll have to look in to those silencer cases, but I don't think thats the main problem.

I have an ssd for system files, and the hdd is for storage.

The issue seems to be that the hdd will not spin down. I've looked at the windows power options and a few 3rd party programs to regulate it, but aside from ordering it to spin down manually, they don't seem to work. Maybe because it's a slave drive? I don't know. I'll spend some time tomorrow updating drivers and see if that fixes anything.

Almost all drives will power down and park the head when not in use. If you have Windows 10, see what task manager says is using the hard drive.

SlayVus
Jul 10, 2009
Grimey Drawer
So my Threadripper recently started exhibiting some weird behavior of high temps at almost no load. I have my 1950x over clocked to 3.9GHz and run it with an Enermax LiqTech 2.140mm AIO cooler. Physically touching the tubing coming from the block I do have a hot side and a cold side so the pump is functioning. I haven't moved it or touched anything inside for over a month now. It only started showing these symptoms within the past few days. The system is currently drawing between 350w and 410w according to my UPS where as I believe I should be idling around 90-100 watts and the highest load I've ever had myself only used like 250-300w max.

I'm currently scanning for viruses, but so far the only thing that's been found by Malwarebytes is a single PUP registry. My CPU is running at 4% usage and an idle temp of 68-72C. I wouldn't think my thermal paste would give out after less than a year.

SlayVus
Jul 10, 2009
Grimey Drawer

Indiana_Krom posted:

You should never have a hot and a cold side of a water cooler, the coolant should be flowing through the loop quickly enough that the difference between the in and out sides should be indistinguishable (like it should be a fraction of a degree at most). The amount of coolant inside is so small that even a slow pump should circulate the entire volume in less than a minute, having a side that is noticeably warmer is a huge alarm to me.

I redid my thermal paste and saw no improvement. Then I undid my overclock, reset the bios to factory defaults, then went in and undervolted the CPU by about .095-0.105v(Stock default is ~1.305). In Windows I'm idling pretty high at 45-48c now. I'm still also using a lot of power, around 260w. So I think my problem is two fools in that something may be wrong with my cooler and that I need to go in and figure out why my CPU isn't power state changing. Because even in the BIOS it'll idle at 45c.

SlayVus
Jul 10, 2009
Grimey Drawer

Indiana_Krom posted:

Possible your pump is seized up, water should never be in the block long enough to get hot. Check RPMs and make sure it isn't set to some super slow PWM setting.

The motherboard sensor says that the pump is running at 2.8-2.9k RPM. Its a 3 pin fan header and I have it plugged into the dedicated Water Pump 4-pin header on my mobo.

Adbot
ADBOT LOVES YOU

SlayVus
Jul 10, 2009
Grimey Drawer
Does this seem like faulty VRAM on my GPU. Using MSI Kombustor I can select the FurDonut 6500MB and it'll crash before loading. I can select the FurDonut 5200MB and it'll crash before loading. I can run one instance of the 3200MB. I can run two instances of the 1700MB and use a total of 5100MB of VRAM. Right around 5130MB of VRAM usage(according to AfterBurner) it'll crash. But it's not a hard crash, Kombustor will just close and I can restart Kombustor and keep using my PC and it'll still load stress tests without issues.

GPU is an EVGA RTX 2080 Super. PCB designs show 8 1GB VRAM chips, it's crashing when it's loading more than 5GB, which puts it into 6 chip territory. So I seemingly have one chip that it'll crash when it loads into is my working hypothesis.

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply