COMPUTER QUESTION D:
10 years ago
Ok,
So a kernel goes nuts and consumes a core (13%): NT Kernel & System.
- There is no correspondence or pattern relating with any of my activity, clicks, processes, usage.
- Tempratures are good
- Software is fresh after a full recover/reinstall/reformat (had a bluescreen during Recover, but got in through MSOOBE operation).
- Hard Drive light is constantly illuminated during the hangup/overclock of my CPU.
What's your hunch regarding my 3 year old Solid State Drive? Could this be the SSD going bad? It's a 2 hour roadtrip to go get my OEM HDD, but thinking about trying it.
Thanks!
So a kernel goes nuts and consumes a core (13%): NT Kernel & System.
- There is no correspondence or pattern relating with any of my activity, clicks, processes, usage.
- Tempratures are good
- Software is fresh after a full recover/reinstall/reformat (had a bluescreen during Recover, but got in through MSOOBE operation).
- Hard Drive light is constantly illuminated during the hangup/overclock of my CPU.
What's your hunch regarding my 3 year old Solid State Drive? Could this be the SSD going bad? It's a 2 hour roadtrip to go get my OEM HDD, but thinking about trying it.
Thanks!
I would replace it with something else and see if the issue is still there. That is mostly what I do anymore to rule out things.
In linux, its easy to see to read the errors, windows i am not so sure on how.
http://www.furaffinity.net/journal/...../#cid:45632806
My money would be on a driver or OS file problem. If you've not done so, http://pcsupport.about.com/od/tools.....fc-scannow.htm If that finds nothing I'd nuke every single driver with Driver Fusion, the base software will do it for free.
The fastest way to tell if it is hardware vs software would be a genuine clean install though.
In any case, since you are using the m4 this may be your problem http://www.anandtech.com/show/5424/.....the-bsod-issue
Crucial released a software tool for their drives after the m4s had been out awhile: http://www.crucial.com/usa/en/suppo.....rage-executive and it may be able to update your SSD for you as well.
This 5184 hour runtime issue was a very unique problem that only the m4's had, I'm not aware of any other manufacturer that experienced this kind of bug. It's a fast and simple (and permanent) fix with a firmware update, so I hope that's the issue with your system!
Really was a bizarre edge case, but to be fair to Crucial every single SSD manufacturer has suffered from critical firmware bugs at some point over the last five years, that just one happened to be Crucials.
It's safe to say that we're all in the pioneering phase of SSD still. It's gonna be another 2 or 3 before we're even a decade into this - just think how long it took to get past dialup, lol. This SSD is really fast and roomy, and a good friend at MIT assured me that the devolution won't really hit for 8 - 10 years into its life. Someday we'll be like "remember when SSD's would just fuck up on you."
Ironically, I feel that HDDs have really arrived. I can't remember the last time Ive seen a head crash or gotten a bad click. It's like they've reached the HDD apogee.
Ah, well here's a scary/fascinating thought for ya then! An SSD is basically an entire self-contained computer. It has firmware (BIOS), DRAM cache (memory/RAM), NAND (storage), and a controller (2-4 core processor). The software required to mange and track the data is very complex due to so many indirect layers of abstraction. Maybe it's the complexity but every year another manufacturer has some major issue, currently the unfortunate one is Samsung and they are not handling it well.
SSDs are peculiar in that they do need firmware updates, often to fix performance related bugs that end users won't ever notice (unless they run benchmarking software), or more serious bugs like this one. But firmware updates also have proven very problematic or even been the cause of problems such as drive bricking in the past. It didn't help some models and brands used to require destructive (data wiping) of the drive during the firmware update process. And yet, unlike a UEFI BIOS, firmware updates were often needed. So while I generally do recommend for users to update their SSD firmware, it's loaded with caveats about the risk unless its a drive with an already older, established safe firmware.
I completely agree with you, SSDs are still in the pioneering phase. Common SSDs are capable of maxing out roughly a dozen storage interface types, some that were even explicitly designed for them no less. With the possible exception of the GPU (with gamers), change any system component in a computer and none will have the impact on system responsiveness that an SSD offers, they are a must have for systems even with the occasional problem they still run into. I'm probably already preaching to the choir, but yeah xD That said, the "untamed wild west" feel of the market is diminishing quickly. Controllers used to come in all kinds of different design approaches, today there's only a few core names and many of them use Marvel controllers or ARM cores in the controller. SSDs are slowly becoming standardized, and the industry move to PCIe drives that use NVM Express (instead of AHCI) will only help with that.
Since you bring it up, HDD reliability is an interesting topic in its own right. I agree in that they have the potential to last incredibly long under the right conditions. The issue is that there are an incredible number of factors that affect them, even something as simple as the user changing the spin-down timer (I disable mine) can let a drive last a decade. And yet, on the flipside warranty lengths on consumer HDDs are the lowest they have ever been. It's unfortunate as well that there used to be over a hundred HDD manufacturers at one point, today there are only Western Digital, Seagate, and Toshiba. Any other brands are directly owned subsidiaries.
HDD: ~8ms
Velociraptor: 4ms
SSD: 0.04ms
my OEM hdd of my old laptop had issues with the reader head and it started making clicking noises[i had to force disable docking of the head, which meant constant writing and reading] and the check came up perfectly fine
if the sectors are faulty, you're most likely going to get errors from software and bad sectors will come up in the check
if there are issues with the ram, most of the time you're going to see a blue screen of death
if there are issues with the videocard[dedicated], windows would either force use your integrated, or if -that's- faulty, your computer would not run at all
some more detailed information would clear it up though~
ram failure is very rare unless you're using a desktop or in cases of laptops, putting new ram in that doesn't match the same type that's being dual-channeled
once you get your old HDD back, put it back in and do a ramtest from the BIOS [and let it run on powersaver mode for an hour if it's a laptop], if it still occurs, it's most likely your ram, if it doesn't, it's the ssd
ssds as far as i know, don't make clicking noises during bad docking[which is among the most common issues with old drives], so you want to back up everything asap
another way to check if it's a docking issue is to move large files back and forth keeping the drive busy and see if it still occurs during that[that's how i saved one of the HDDs back then until i could back it up before i started using an SSD]
another common bsod is bad videocard drives and old videocards overheating[only really occurs in desktops, since they have their own fans and a lot of them as far as i know can't shut down the PC via bios to prevent heating damage, so they just shut down themselves]
I'm not a Windows user, so I can't really test any of this, unfortunately :/
While not entirely related, I know that my friend had issues with her USBs and I think she tried running recoveries and reformats but it took a reinstall of the drivers to fix it. It might be the same case here where the driver/Kernel have been failing and are only now showing symptoms.
Swapping to the HDD could also either fix or help find the issue.
The last option is to manually backup your data, do a complete reinstall and reconfigure everything manually. Not the best option but if it's a software issue a clean slate should fix everything and if it's hardware then it'll have to be done anyway.
This is an excellent test of popular consumer drives write capabilities: http://techreport.com/review/26523/.....-to-a-petabyte
Used to have a first gen i7, so was capped at Sata 2 speeds, 250mb/sec read/write... But after building a brand new i7 4790k this past winter, happy to see 450+MB/sec read and write (and 550MB/sec read/write from my Samsung), games load way quicker, as do any and all applications. Photoshop, under 3 seconds, as opposed to 7-9 seconds back when I had it on my regular drive.
Had a bad SSD experience myself with my first gen i7 920 system about 4 years ago, think it was a Corsair Force SSD, worked amazing, windows started fast, apps loaded fast, then I started getting 'the stutter' and occasional complete system lockups, and upon doing a hard reset, the drive wasn't detected by the motherboard, so at that time, I was put off from using SSD's from that point on until they were more reliable, up until I got my first Intel SSD back in 2013, due to being one of the best brands with the highest reliability rate.
The hard drive usage on my initial Intel 240gb, is sitting at 13.41 TB of data written, which I suppose is approximately 60 full re-writes of the drive, out of the estimated 3000-4000, and that's in just over 2.5 years.
Anyway, that's my story!
I mean Linux is a kernel that actually works well enough and tells you log information through dmesg/syslog (or, at least on boot-up I guess journald, depending on if systemd is installed.)
In any case, try it. Or look at the Event Viewer for windows for errors. That's another really handy tool. There's also ways you can take a core kernel dump when your system gets a BSOD, but since I have only used Windows up to 7 and not 8 or 8.1, I have no idea if the functionality still exists. Yeah, I don't even use Windows anymore.
Best of luck if you haven't solved it already~ It's probably hardware related, and the software doesn't know how to handle the problems, hence a freeze-up. Even my laptop reboots when the bottom gets too hot, so I use a metal caddy thing under it to distribute heat. I suspect the BIOS is doing it, and I gotta try to apply more thermal paste, but I don't feel like opening up my T410. It's kind of my main PC right now.
Anyway, you might also want to note the age of the RAM, and even if it's new, especially if it's new, try running Memtest, just to make sure that's not the problem. Bad RAM often causes not only bad hangups, but it could even damage your system files and data, even with modern-day ECC memory, so check 'em!