HARDWARE Short Questions That Don't Deserve Their Own Thread - Reloaded

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > HARDWARE Short Questions That Don't Deserve Their Own Thread - Reloaded

«‹›11 »

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

This thread is for short questions that don't require much background or troubleshooting. This is NOT the thread for upgrade/purchasing advice or more involved troubleshooting:

Post in the PC Building & Parts Picking Megathread instead for advice on if you should upgrade or what parts or computer to buy.

If you need help troubleshooting an issue, post a thread in the Haus of Tech Support forum. Make sure you use the template in the sticky Rules thread!

Anything not about computers (cellphones, tablets, TV/audio) goes in the Inspect Your Gadgets forum.

Alereon fucked around with this message at 20:06 on Dec 2, 2014

# ¿ Sep 13, 2014 17:20

Adbot: ADBOT LOVES YOU

# ¿ Apr 26, 2024 09:47

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Stall_19 posted:

Not sure if this is the right place to ask but...

This is my current computer
http://h20565.www2.hp.com/portal/si...44892.492883150

I bought it in 2008 and have used it ever since but there are some annoyances that come up. Firefox can occasionally crash and at sometimes I get very slow performance when I first start firefox as well. It can also stall after it comes back from power-save mode, especially if I left a program open. Long story short I'm considering buying a new computer if it's not too much money. I'm thinking something like this that I found on Ebay

http://www.ebay.com/itm/Dell-Dual-C...=item4185a3ec77

Is it worth it? Is it better enough than the one I currently own to pay $175 for a new one? I don't necessarily need a new one but I'd like one if the price is right.

No, that is only marginally better than the one you have now, if not worse (they don't actually provide real specs, the ones they do provide are generic for machines that could be from half as good to twice as good). You really do need to get a new, decent quality machine if you want a good experience. That doesn't necessarily mean spending a lot of money, but if you cut too many corners you end up with a machine that gives you a frustrating experience with a short lifespan. Check out the PC Building & Parts Picking Megathread for advice on buying a new machine.

# ¿ Sep 15, 2014 06:15

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

BurritoJustice posted:

I've been having pretty weird performance issues with my PC.

I have a 3570k at stock currently with a TC14PE on it, a MSI 780ti Gaming, 16GB of 1600MHZ C8 Ram, and the only drive in my system is a 1TB 840 EVO. I've been slowly upgrading my system to remove every bottleneck I can, but I still get REALLY lovely performance in games. Games that my 780ti should be handling with ease are really chuggy. I get around 30FPS in Borderlands 2, 40~ in Far Cry 3, 200 down to 70~ in TF2 (!), less than 60 on occasion in minecraft for Christ's sake. Really inconsistent, droppy framerate too. The CPU runs at around 40C under load, the GPU 65C. I just can't see what could be causing such terrible performance. Could it be somehow related to my motherboard? I have a Asus P8Z77-V, and if I run a DPC latency checker it is over 1000μs which is like 10 times what reviews say it should be.

I'd suggest updating to the latest motherboard BIOS, loading default BIOS settings, and ensuring you have the latest Intel Chipset and Rapid Storage Technology software installed. Background software can also cause high DPC latency, so remove anything you're not using. If you still need help, a Haus of Tech Support thread would probably be most appropriate.

Edit: I'm not sure if the DPC latency is necessarily the cause of what you're seeing but I think it's related.

Alereon fucked around with this message at 14:14 on Sep 15, 2014

# ¿ Sep 15, 2014 14:04

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Fucknag posted:

I have the latest official drivers; are there, perhaps, unofficial drivers or something like that that might resolve my issue? I know it's a long shot for the ages, but there's no harm in asking, right?

Here's a modded beta driver you can try.

# ¿ Sep 17, 2014 04:39

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Titor posted:

I'm about to pull the trigger and buy a Western Digital 2TB Red to replace my current malfunctioning HDD. This is for storage purposes and Nvidia Shadowplay (1080p 60fps) only. Anything I should know beforehand or am I good to go?

Also the Calibration Retry Count on that HDD rose from 164 to 317 in only 10 days. How much of a short lifespan would I expect if this trend continues?

Yes the WD Red is a fine drive, and there's no way to predict how long a failing drive will limp along but you should certainly copy off any data you care about before it does completely.

# ¿ Sep 17, 2014 20:47

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Pogue_Mahone posted:

Thanks for the answer, guessing the same. I am thinking a bad sensor should be nothing to worry about? I just stress tested the build with p95, and my CPU hit 79 degrees C before I quit - since I have a Coolermaster Hyper 212 I thought I wouldn't hit such a high temp? Was expecting 70C at most? Should I take the cooler off and put on new thermal paste?

Sorry for the newbie questions, I rarely keep up to date with PC build until I am building a new one (every 4 years or so)

It's not actually a bad sensor, the software you're using just isn't setup to read the sensors from your motherboard so it's essentially returning random numbers. Which Prime95 test did you use? Small FFT and Blend will generate extremely high temperatures, In-Place FFT is the correct test to check CPU temperatures, and you're looking at the CPU Package temperature (which shows the highest CPU core temperature).

Technical details: The Small FFT test uses the AVX2 instructions, which generate so much CPU load that it switches to a special high-voltage mode that is only used for AVX2. This drastically increases power consumption, well beyond what can be easily dissipated out of the CPU. The In-Place FFT test is more consistent with the maximum CPU load you will see in the real world. Blend alternates between them. Basically the Prime95 AVX2 tests are like Furmark for your CPU, it's valuable to make sure your CPU doesn't crash or power-off under maximum theoretical power consumption, but you don't need to worry about the temperatures you see. This also means that for overclockers you need to be careful when overvolting that the additional AVX2 boost isn't too high, or switch voltages to fully manual mode so the automatic boost doesn't happen.

Alereon fucked around with this message at 01:43 on Sep 26, 2014

# ¿ Sep 26, 2014 01:41

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Flipperwaldt posted:

I think since Vista a non-quick format zero fills already.

A quick format basically just wipes the FAT and marks all clusters as free, a full format also tests all sectors to see if they work. A quick format definitely doesn't remove data (you can tell it doesn't take long enough), a full format should be sufficient, but to be safe I'd do a single full overwrite pass before selling a drive.

On Windows 7 and later a quick format will send TRIM commands to SSDs and trigger them to actually erase the deleted dated, however this doesn't happen immediately and isn't guaranteed to happen at all. Before selling an SSD you should perform a "Secure Erase" rather than an overwrite, which guarantees that all data is wiped and the drive is returned to factory-fresh condition.

# ¿ Oct 3, 2014 22:03

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Flipperwaldt posted:

Sorry if this is the dumbest question, but testing involves writing to it in this case, right?

And thanks for clearing that up. Full format is what I meant; a lot less confusing than what I called it.

Actually I just mis-read your post as saying a quick format, sorry! :shobon:

So yes, you are correct, the full format should erase the entire partition one sector at a time, though I don't actually know if it tests the entirety of every sector on every drive (including modern Advanced Format HDDs).

# ¿ Oct 3, 2014 23:46

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Sir Unimaginative posted:

Because the Intel 7260 (now in desktop kits too!) is the best drat Wi-Fi card on the planet and it supports both Wireless-AC and Bluetooth 4.0 .

This is true, but be warned that Intel re-uses model numbers, so make sure you are getting the current Wireless-AC 7260, not the older 7260 models, like the N version. If you are using a good Wireless-AC router and adapter you really will get wired-equivalent performance.

# ¿ Oct 7, 2014 18:38

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Flipperwaldt posted:

I'm not going to type it all out again, I'll just link you to a post I made earlier, outlining how my personal experience with the 7260 makes me disagreee very strongly with that statement.

Your issues were caused by your router crashing under load. You didn't see the issue with other adapters because they weren't good enough to place the router under significant load. You likely would have seen similar behavior with other decent-quality Wireless-AC adapters.

# ¿ Oct 7, 2014 19:01

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Flipperwaldt posted:

That may be an adequate explanation for the case where the whole router was taken off the air. Router crash seems less plausible for the much more frequent cases where all other connected devices can keep on working without issue during the time the laptop with 7260 AC thinks there's a problem. Intense bluetooth traffic reliably murdering wifi speeds also cannot possibly have a lot to do with that either.

Swear to god that I understand that it gets recommended because likely it works well for a lot of people in a lot of settings and that I'd love to have an alternative explanation that allows me to fix it.

Note that you can't use Bluetooth and 2.4Ghz WiFi simultaneously, you CAN use Bluetooth and 5Ghz WiFi simultaneously. In addition to using the same frequency band as 2.4Ghz WiFi, most dual-band combo adapters use the same radio hardware for Bluetooth and 2.4Ghz WiFi, so they have to "timeshare." This can work well for light usage, but once you start having heavier data transfers it quickly becomes apparent. Bluetooth 3.0+ High-Speed is able to actually send data over the WiFi connection, so I don't believe this timesharing is necessary, though it still would reduce the WiFi data throughput on the same frequency band. I think even Bluetooth 4.0 Low Energy might improve things since its biggest purpose is to reduce the amount of time the radio is in use, leaving more time for the WiFi.

I'm not saying Intel's drivers are perfect and you didn't encounter real issues, though I've had good experiences on the 7260-AC with both great and lovely routers, but it is pretty important to set reasonable expectations for the performance you'll get.

Bonus next day edit: Bluetooth High-speed was actually introduced in Bluetooth 3.0, not 4.0 like I said above. The reason impact could be very severe is that, for example, if you're running at nearly 100% utilization of the Bluetooth link, the adapter has nearly 0% of its time to devote to WiFi traffic, so you might get basically no throughput. I could see this getting even worse if the router is trying to send data (such as acknowledgement packets) while the adapter isn't listening because it's in Bluetooth mode. It doesn't take very many missed packets to completely gently caress throughput on TCP connections.

Alereon fucked around with this message at 15:57 on Oct 8, 2014

# ¿ Oct 7, 2014 20:57

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Imazul posted:

Hopefully I am in the right thread.

So I got a new 970 card and now I have an old 770 card collecting dust. Would I get any performance gain by setting it up as a dedicated Physx card? Can I actually set it as a dedicated Physx card even if my motherboard does not support SLI ?

You would see significant gains in games that use PhysX and support hardware acceleration...which is very, very few titles. Most games that use PhysX for gameplay (versus only graphical effects) don't support hardware acceleration.

# ¿ Oct 8, 2014 21:47

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

lady blue shanghai posted:

Sorry but is there a thread like this for software? I need a program to do something and I want goon advice but I don't know where exactly to ask.

Two possibilities:
The useful Windows Freeware thread (what free app does what I need?)
The Windows and Windows Software Megathread

A thread in the Haus of Tech Support might be a good idea if you need more individualized help on how to accomplish something.

# ¿ Oct 9, 2014 17:16

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Knifegrab posted:

Really stupid question here: My motherboard has 4 3.0 usb ports directly on the mother board, and my case has 2 3.0 usb ports on the front (which are in fact plugged into my motherboard). Is there any difference between plugging an item into the front of my case versus directly onto the mobo at the rear of my case (in terms of latency, speed, etc)?

There is no difference between internal and external ports pe se, however it is likely they are connected to a different controller chip on your motherboard so may give a different experience. On most systems four of the USB 3.0 ports are provided by the Intel chipset, the rest by another chip from a company like ASMedia, for example. The Intel controller generally works best.

# ¿ Oct 10, 2014 17:23

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Knifegrab posted:

Are we talking significant difference or just slight? As in should I make sure all my input controlls (keyboards/mouse/joysticks) are in my motherboard then to obtain the lowest latency?

Your input devices should be connected to the black USB 2.0 ports on the back of the motherboard. Only USB 3.0 devices should be connected to USB 3.0 ports, and preferably to the Intel controller first, though how well devices work with the Intel vs 3rd Party controller varies by device. In theory USB 3.0 ports should be functionally identical to USB 2.0 ports for USB 2.0 devices since the extra pins aren't even connected, but it does not actually work like that for some reason, plenty of people see issues with their mice and keyboards when using USB 3.0 ports. Mass storage devices do often run better though.

Alereon fucked around with this message at 18:14 on Oct 10, 2014

# ¿ Oct 10, 2014 18:11

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Knifegrab posted:

So what should I do if my motherboard only has two 2.0 usb ports, where my keyboard has two usb plugs and my mouse has one, and my hotas and rudder set up each have one as well (meaning I have 5 input plugs, but only two 2.0 usb ports)?

Your motherboard has eight USB 2.0 ports, two directly on the back of the board and three headers along the bottom of the board. If your case doesn't have USB 2.0 support and you don't have your own D-brackets handy, you can pick them up from a computer parts store for like...$5, if they don't just give them to you for free because they have a box of a thousand. These just go in one of the I/O slot panels on the back of the machine and have a cable running to the USB 2.0 header on the motherboard, and are used if your case doesn't have a header to plug-in there instead.

Using a USB 2.0 add-in card would result in a very similar set of compatibility issues because it's a third-party controller, the point is that you want to be using the chipset USB 2.0 ports for an experience that is guaranteed to be issue-free with your legacy USB 2.0 devices.

Alereon fucked around with this message at 18:44 on Oct 10, 2014

# ¿ Oct 10, 2014 18:41

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Knifegrab posted:

Hmm ok I'll look into that before I get a PCI usb thing.

Definitely do not buy an add-in card, it won't help you at all. If you can't get D-brackets use a powered hub (your keyboard has one right?) or just use the Intel USB 3.0 ports until filled, then the supplemental USB 3.0 ports.

# ¿ Oct 10, 2014 19:01

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Alan_Shore posted:

In terms of 7.1 gaming headsets, the Logitech G930 is number 1 on Amazon. Is that the one to get?

That sounds like more of a question for Inspect Your Gadgets. Note that 7.1/surround sound aren't real features when it comes to headphones, you only have two ears and your source will be able to generate stereo or do smart downmixing for you.

# ¿ Oct 14, 2014 15:49

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Factory Factory posted:

Not only has positional audio simulation via DSP been a thing since the 90s, but some headphones actually do have full sets of surround drivers in them.

We only have two ears, yet they are nevertheless capable of hearing position in 3D space. With control of the source and environment (e.g. in a closed earphone cup) you can do the same thing with the source that the receivers do.

To clarify, because human beings have only two ears, stereo headphones provide perfect 3D audio. Adding additional drivers or surround processing destroys the 3D positional audio effect, it does not enhance it. Essentially what you're doing in that case is removing the 3D audio information that exists in the stereo signal and then attempting to simulate it by adding delay and reverb or piping it out of additional drivers, which is only a pale imitation of the information you threw away that already told your brain where the sound was coming from.

Here's an example of the kind of excellent 3D audio you can experience with stereo audio through headphones, close your eyes and you should be able to tell where around your head the match or matchbox is: (sorry to link an ASMR video, this is the best example of the match demo I could find)
https://www.youtube.com/watch?v=C2TGd6LvHpg

The way this actually works is that the shape of your head and ears subtly alters the sound you hear based on the direction it comes from. These subtle alterations are called Head-Related Transfer Functions, or HRTFs. These are taken into account to at least some extent when audio is rendered or mixed to stereo, so if you're listening to a good quality stereo audio signal from headphones you'll get an EXCELLENT 3D audio experience. This DOESN'T happen with surround sound, because instead we use separate drivers to actually produce the sound from somewhere near where it should seem like it is coming from. This can work okay enough with surround speakers, but when you're relying on cramming multiple drivers into headphones or doing simplistic reverb+delay processing on a DSP, you're giving up a lot compared to the audio you started with.

# ¿ Oct 15, 2014 02:13

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

MrMoo posted:

I can only hear the 3D audio positioned behind me, is it supposed to work in front as well?

It's MUCH harder to get good center/front positioning because this is where the HRTFs have the smallest effect on the sound, the differences in HRTFs between individual people are the greatest, and the difference in signal each ear receives is minimal. It's actually hard for you to tell where real sounds are coming from without being able to move your head and body and notice how the sound changes.

# ¿ Oct 15, 2014 02:55

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Straker posted:

When one of my seagates was being flaky I did some research and saw they counted stuff strangely or perhaps used some smart fields for slightly different purposes than originally intended, but it ultimately all made sense and was internally consistent, and I couldn't find any evidence of the conspiracy theory type stuff I see posted here a lot.

Of course I love being wrong because then I get to learn more stuff, so anyone? please?

You can read more about how Seagate records SMART errors here. There's two issues: first, that the soft error counts are tracked as sector error rates, which is much less useful for reviewing the current health of the drive. However, these soft error rates aren't the most useful for troubleshooting anyway, so this isn't as important. The bigger issue is that Seagate drives don't count hard errors accurately, they simply don't report any until enough have accumulated that the drive is pretty convinced that it's failing. I seem to recall that someone did research into Seagate's bad sector behavior a few years back and I thought someone like Backblaze found that Seagate drive failure rate was less correlated with SMART error counts than other brands, but I can't find what I remember reading now. I will say that I've seen a LOT of systems with obviously failing Seagate drives that had healthy SMART error values, including drives that I was pretty sure were reallocating sectors in front of me (with audible read/seek retries) that immediately failed Seatools, yet showed no reallocations. This should be testable using drive scan software that reports per-sector latency, then comparing the number consistent latency spikes (which show the seeks to the new locations where the drive has remapped bad sectors to) to the number the drive reports.

# ¿ Oct 19, 2014 15:16

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Mr. Crow posted:

So I tried updating my BIOS and it seemed to go fine, but for some unknowable reason it decided to throw a password on my BIOS? The hell? It also screwed up my boot order I think as I get BOOTMGR missing error which I can't get into BIOS and fix so it's locked...

Clear the CMOS via the jumper on the board.

# ¿ Oct 23, 2014 18:17

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

shodanjr_gr posted:

Are there any pitfalls to avoid when doing RAID0 with SSDs? I've got a machine at work with an LSI 9750-8i RAID controller feeding two Samsung SSDs in stripping mode. Performance is HORRENDOUS even on trivial workloads (we are talking sub-platter disk levels of performance here) and I don't have the first clue about how to debug this. Any ideas?

Yes, this is actually a really complicated question, check out the SSD Megathread, or post a thread in the Haus of Tech Support with more details about your configuration, including what drives you're using, what operating system, and what kind of motherboard.

# ¿ Oct 28, 2014 01:22

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

rt_hat posted:

I've got a stupid question about USB memory sticks and security.

Or would it be safer to use memory cards like CF or SD through a USB memory card reader, assuming I trust the memory card reader ?

You're going kinda far down the rabbit hole here, do you worry about where you get your mice from? You're correct not to run random drives from unknown sources, but name-brand drives you source from a legitimate reseller (NOT some Amazon store) are fine.

Jan posted:

I'm pretty certain this is impossible.

You need drivers to interface at with a USB drive in the first place. If a USB drive had to provide drivers itself, it wouldn't be able to provide them since it would be unreadable for lack of drivers. Chicken and egg situation, basically. To break this, it's up to the OS to provide drivers to read the USB drive and whatever filesystem it is formatted with. Since the host OS has to provide its own driver, this is basically not an infection vector.

This actually isn't true if the USB device itself is malicious, as opposed to just being a USB Mass Storage device with a malicious file sitting on it. The device has full access to system memory even without the OS's cooperation, so even disabling ports or controllers in the device manager isn't full protection from a malicious device. Are you necessarily going to notice if another HID-Compliant USB Keyboard appears in Device Manager, using default drivers so Windows didn't prompt you? There's also the BadUSB thing from Blackhat, but we are kinda getting into crazy person territory rather than legitimate security threats for most people.

I strongly recommend that everyone read James Mickens' "This World of Ours" about realistic threat models. In summary, here are the three types of threats you face and how you mitigate them:

Casual attackers, like ex-partners and script kiddies: Use strong passwords and try not to make any seriously boneheaded mistakes.

Determined attackers, like criminals: Use really strong passwords and try not to make any mistakes, pay more attention to covering your poo poo.

The Mossad or the NSA: "YOU�RE GONNA DIE AND THERE�S NOTHING THAT YOU CAN DO ABOUT IT"

Someone handing out malicious USB drives at a conference is more like #2, so yeah don't do that. Slipping malicious USB drives into the channel under a real brand is more like #3.

# ¿ Nov 3, 2014 03:42

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Grawl posted:

It's this PC (Dutch site, sorry).

That board powers the USB from the +5VSB (StandBy) rail that remains powered even when the system is turned off. There may be a jumper that allows you to disable this (switching USB to the normal +5V rail), but customer reviews indicate there is not.

# ¿ Nov 5, 2014 00:23

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

NmareBfly posted:

Yeah, using software. I'll need to haul it into work to check with a multimeter, but that might be worth doing at this point.

The UPS is one of these which I only now notice is rated to 450 watts while the PSU is 550 (oops?) -- but I've never seen the load meter on the LCD screen going up past maybe half.

That's one of the "stepped-approximation to a sine-wave" models that aren't compatible with modern power supplies (that use Active Power Factor Correction). You could spend a bit for a decent true sine-wave model, but honestly I think having a UPS is less relevant these days. Decent Active PFC power supplies aren't sensitive to input voltage so you don't need a UPS doing voltage correction, and since computers (especially with SSDs) are so fast it's not very likely you'll have a long queue of important disk writes to lose if the power goes out.

# ¿ Nov 10, 2014 20:32

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

BurritoJustice posted:

I know all this. My fans are all PWM and are all plugged into PWM headers (using a PWM splitter for the back two). The fan curve on my ASUS motherboard does not allow me to set the fans lower than 20%, I want a way around this. Additionally I was curious as to whether I could run the thing without fans at all and still get fine temps for stock clocks (maybe with an undervolt if necessary).

Are you sure you can't do this on your board? In the Asus AI Suite I go to FanXpert, enable Smart Mode for the fan, and there's an Auto Fan Stop checkbox that stops the fan when it's below the temperature needed to trigger the minimum controllable speed. Fan control features do vary by board, but make sure you have the latest BIOS, AI Suite software, and have run the fan profiling tool if applicable.

E: Missed the second part of this post. This guy was able to complete benchmarks up to 4.4Ghz passive with good case ventilation, so I expect you'd be well-served. The TC14PE doesn't have fin spacing that's TOO narrow.

Alereon fucked around with this message at 16:03 on Nov 11, 2014

# ¿ Nov 11, 2014 15:09

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

PRADA SLUT posted:

Is there a specific type of USB drive that would be used for archival?

I'm looking for an efficient way to store about 20gb of data in a long-term backup in a safe.

You want to burn to an optical disc like a Blu-Ray for this. Data on Flash media fades with time, and may be unreadable after only a year. While they sell archival-grade optical media, I've never really seen issues with endurance that weren't due to things physically scraping against the disc and damaging the recording layer or UV light damage, and neither of those are issues for media stuffed in a safe. It would probably be a good idea to throw in oxygen absorber packets and make sure your safe is appropriately fire-rated though, if it's only rated for paper everything else will be melted when you open it after a fire. I think personally I would burn to single-layer Blu-Ray discs and put in at least two copies of every disc, as well as a decent USB flash drive just for convenience of access in the nearer term. Make sure to use file formats that are error tolerant, don't compress everything to a ZIP file first for example.

# ¿ Nov 29, 2014 00:31

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

dis astranagant posted:

I just cloned my system drive onto my new SSD. Do I need to unplug the old drive when I restart to be sure I boot off the new one? I'm gonna have to play musical cables anyway since I forgot that only some of my SATA ports are SATA 3 and naturally the unused one was a slow one.

Yes, definitely do this. If you have an Intel CPU and your motherboard has more than two SATA600 ports, double-check that your SSD is connected to ports provided by the Intel chipset, ports from other controllers will not work properly.

# ¿ Dec 2, 2014 01:26

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

a shameful boehner posted:

Post in the Parts Picking Megathread (despite the name it is also for deciding what PC to buy). It sounds like they have minimal needs so almost anything would work, but keep in mind if you buy a cheap, crappy PC it will die or just end up sucking horribly years before one that isn't garbage.

Edit: Late note, what I mean above isn't to buy an expensive computer, but just try to avoid the cheap crap.

Alereon fucked around with this message at 20:37 on Dec 2, 2014

# ¿ Dec 2, 2014 20:08

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Touchfuzzy posted:

So just to make sure, sticks of RAM rated at 1.6 volts (like the G.Skill Ares series at 2133mhz speed) are okay for Z97 boards, right?

Use 1.50v RAM unless you are buying high-end overclocking RAM. Since DDR3-2133 is a standard JEDEC speed, use standard voltage. You can get 1.65v DDR3-2400 Ares modules for the same price though if you do want to overclock.

# ¿ Dec 6, 2014 17:24

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Tin Tim posted:

I guess what it breaks down to is that I need to figure out if I lucked out and can keep it, or if I need to raise some cash to also get a new board+CPU

Post a screenshot of the CPU-Z window so we know what CPU you have. Also, what motherboard do you have? Overall, even if it can work it is likely much smarter to just pick up an Intel Core i5 CPU and compatible motherboard, since that gets you the full performance of a new machine.

# ¿ Dec 9, 2014 20:33

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Tin Tim posted:

Here's the CPU-Z

The board is "EliteGroup MB Elite GF7050M-M 2.0 AM2 mATX Phenom" which nobody probably has heard off

I do realize that I have to get rid of the few last old pieces in the near future, but the question just is do I have to spend big right the gently caress now, or can I split it over a month or two, you know? I had my PC built back in 09, and the board+CPU are the only old parts that are left from the original config. I kinda need to get a 64 bit system going asap, but if possible, I'd like to not burn too much money right away if I don't have to.

Your CPU supports Windows 8.1 but your motherboard does not. You are using an nForce 630a chipset from 2007, and nVidia shut down that unit years ago. You can try it with the built-in drivers but given how ancient that system is I can't imagine it working very well.

# ¿ Dec 9, 2014 22:12

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Hace posted:

Is there any reason not to use RAID 1 on a Z97 chipset w/ IRST, or would it be better just a buy a cheap PCI card and use that? It's just for two storage HDDs, the OS is on an 840 EVO.

The onboard RAID is far better than any controller card you'd pick up.

# ¿ Dec 14, 2014 18:28

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

lollybo posted:

So my N200 came in and I was disappointed with the build quality- the case seems very light and flexes a lot when I press down. I don't feel confident in it and am afriad that it can be easily dented. Is the Silverstone TJ-08E a good replacement? I can pick up one of those at Fry's.

EDIT: Finished my trip to Fry's. I saw the TJ-08E but was not blown away by the build quality, sure it is better, but I'm not sure if it is good enough to justify getting rid of the N200. I saw a lot of cases and actually think the build quality of the N200 is about average. I guess you can't reach Mac levels of sturdiness unless you are willing to pay >140 for a case.

Look for cases optimized for low-noise, the ones you're looking at are optimized for size and weight (and price for the N200) so are necessarily going to be made of thin metal and not particularly sturdy. Or, you know, don't care about how resistant your case is to fall damage because you won't drop it

# ¿ Dec 21, 2014 17:39

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Mackers posted:

Hi folks, I was trying to do a quick upgrade to my girlfriend's dad's PC by adding new RAM.

In my ignorance I thought that upgrading RAM was a simple matter of buying new RAM and sticking it in the motherboard so I picked up 4GB of 800MHz DDR2, popped out the two old 1GB sticks and inserted the two new 2GB ones.

Welp, the PC wouldn't even power up fully. Fans would spin up a bit but then stop again and this would cycle over and over. Never got anything but a black screen on the monitor.

Now results on google are talking about things I've never heard of like RAM timing and voltage and I'm in over my head...

In short: Is there a way I can get the new RAM to work? In dumb person speak if possible

Post a thread in the Haus of Tech Support, we need to know what motherboard it currently has (or the computer brand and model) and what RAM you tried to put in.

# ¿ Dec 22, 2014 21:06

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

lollybo posted:

I am a bit disappointed in the sound that my PSU makes. I bought a highly rated PSU from a reputable brand, the Seasonic 550W G series, which had rave reviews mentioning how quiet it was, but in my system it makes a "whirring" sound that can definitely be audible even a few feet a way. I installed it fan up in my Silverstone TJ08-e, and if I cover the vent with a booklet the system is noticeably quieter. Is there something wrong with my unit, or my expectations just too high for a 85 dollar PSU? I estimate the fan to be spinning between 400-700 RPM, it's not LOUD, but definitely noticeable if I pay attention. I could deal with it I suppose, but I was wondering if anyone else had similar experiences with this PSU or from Seasonic.

This is why it's important to buy a power supply that's appropriate for your system. You're pairing a relatively high-end system with a mid-range power supply, which isn't really going to give you a very positive experience. In particular, the Seasonic G-series power supplies are rather loud under load, so they aren't really good choices for systems where they will be loaded beyond 50%. This isn't unique to these models, that's why you buy a large enough power supply that your actual load levels will be achieved at a low noise level on that model. I think it would have been more appropriate for you to get a higher-end semi-active model where the fan is completely off until moderate load, or at least a higher-capacity model with better noise characteristics.

# ¿ Dec 26, 2014 23:25

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Do you have the latest motherboard BIOS, as well as the latest chipset drivers from the AMD website?

# ¿ Dec 29, 2014 01:28

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

I should note that there's no evidence WD Green drives are meaningfully less reliable than others. I even have some of the early drives that are VERY aggressive about head retracts and they continue to work fine after around 5 years.

# ¿ Jan 2, 2015 20:40

Adbot: ADBOT LOVES YOU

# ¿ Apr 26, 2024 09:47

Alereon: Feb 6, 2004; Dehumanize yourself and face to Trumpshed; College Slice

Jolan posted:

Brother-in-law gave me his laptop because it had become absurdly slow and a complete reinstall of Windows didn't help. I checked everything out and I'm pretty sure it's the hard drive itself: when I connect it via a SATA-USB-bridge to my desktop and run CrystalDiskMark, it's performing only about half as well as other drives that I test that way. Couldn't find any errors or faulty clusters on any partition, so I removed the partitions on it and full-reformatted part of it back to NTFS, but the new partition is still portraying the same symptoms (similar benchmark performance and no errors to be found). Does anyone have some tips on things that I haven't thought of to try before I go out to buy a new drive?

Run Crystal Disk Info, if it shows Caution or Bad the drive requires replacement. If it's a Seagate drive it will likely not report errors even if they exist, so I'd just replace the drive based on the symptoms.

# ¿ Jan 11, 2015 18:10

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > HARDWARE Short Questions That Don't Deserve Their Own Thread - Reloaded

«‹›11 »