Recommended Posts

I have a friend who has been using SSDs for YEARS. He works for a retailer and therefore gets to play about with them all the time (also to note is he has 8x SSDs on his personal machine). After all of these years he's had ONE SSD fail him. Now try and compare that to the amount of HDD fails he's had and you suddenly realize SSDs are less likely to fail.

Exactly. I've upgraded my PCs through tons of SSDs, starting with the OCZ Vertex 1. Not a single one of them have failed. I've also built numerous PC's for a company I do consulting work for, and again not a single one has failed so far. 95% of those are OCZ Vertex SSDs, which a lot of people claim have a high failure rate, so I'd say it's pretty impressive. I've had quite a few HDDs die over the years, so in my personal experience, SSDs are a huge step forward in reliability.

In the end, there's simply no sense in thinking a HDD is more reliable than a SSD. HDDs bow to the physics gods, and there's nothing humanly possible to fight that. SSDs have no mechanics and only die due to controller or firmware defects. Both of which are getting exponentially better year over year.

It's also worth noting that in theory he's correct, since each block of the memory can only be written/erased a certain number of times. The controllers manage that though and therefore heavily prolong the life of them.

Saying you're not going to buy an SSD because you're worried about data loss is silly imo. You're missing out on a SIGNIFICANT performance increase and for what? Potential data loss? Lets say 4 years down the line it dies quite frankly if you don't have a backup of that drive, you've failed at computing 101.

There's no need to worry about read/write cycles. SSDs have more than enough cycles to last much much longer than anybody sane would use it for.

http://www.tomshardware.com/reviews/ssd-reliability-failure-rate,2923.html

There's no need to worry about read/write cycles. SSDs have more than enough cycles to last much much longer than anybody sane would use it for.

http://www.tomshardw...-rate,2923.html

That's why I noted that the controller would prolong, just in case there was any doubt. :p When I was learning about the theory behind NAND memory at university I did a lot of research into it, which is why I'm confused that there's a misconception that they have a high rate of failing. Without any controllers (and other bits keeping it going) I can see why people might think that would be the case but the reality is they're reliable thanks to those things.

"When a SSD fails, the failure is likely to be catastrophic, with total data loss. HDDs can fail in this way too, but often give warning that they are failing, allowing much or all of their data to be recovered". Also: wear leveling, bugged controllers with subsequent loss of data and some other issues I don't have time to research right now.

https://secure.wikim...ard_disk_drives

For me, right now SSDs are a no-go. Miniaturization will bring this very limited and controller-dependent storage technology to an halt. NAND Flash isn't for storing data, not for me.

The only way an SSD is going to fail catastrophically is when a physical NAND chip gives way (similar to if a platter in a HDD gave way).

SSDs use striping, ECC and (depending on the controller) a couple of other methods to maintain the data on them, in all cases they can recover from a dead cell (like a bad sector on a HDD) unless there is a physical chip failure.

The S.M.A.R.T. readings on a SSD are far more indicative of drive health in SSDs as the death of a drive is gradual due to degrading memory, thus you can see when a drive begins to die, where as most failures in HDDs are due to physical malfunctions such as dying rotors/heads or disk crashes.

I am bemused at your reference to wear leveling above, as that makes no change (negatively) to the lifetime of a drive, and dramatically improves the drives life by writing pages into areas of the NAND that have been erased the least. Bugged controllers can lead to lost data, however I am yet to become aware of a bugger controller that wasn't related to a specific firmware release, and am yet to see a loss of data from these bugs. This is in direct contrast to Seagates head parking firmware woes recently.

Add to this technologies such as write amplification management and disk based compression, and the drives are highly fault tolerant and capable of dealing with more erases than would be readily apparent.

If I may be so bold as to suggest, that making assertions as to complex technologies, and then stating "wear leveling, bugged controllers with subsequent loss of data and some other issues I don't have time to research right now." (which is nonsensical btw) would suggest that perhaps you ought not be commenting authoritatively on the subject?

The one thing you did get right however, is that miniaturisation (the shrinking of the die size on a chip) will indeed kill off SSDs in the long run. As the process for creating the cell shrinks, the sells become less capable of holding a charge with each erase. This situation is dramatically magnified by the MLC (multi level cells) which hold 2 bits of data by storing a different voltage inhibition (this is wrong, but without getting into the nitty gritty of it) in the cell. The erases make cells less capable of dealing with differing voltages on read, and thus are likely to fail more readily than simple on off assemblies like SLC (single level cells).

Bleh.

I thought an SSD burns out relatively quickly ... or is that wrong ?

Relative to HDDs, you are entirely right. The thing is that you will (generally) be upgrading your computer a LONG time before the disk dies. The average disk life is projected at longer than 10 years. Mine is currently projected to live for another 9 years. and I've had my Vertex 3 for a couple of years and abuse the little ****** like crazy.

tl;dr - Yes, but it won't worry you unless you deliberately try to burn out the disk.

Edited by articuno1au

"When a SSD fails, the failure is likely to be catastrophic, with total data loss. HDDs can fail in this way too, but often give warning that they are failing, allowing much or all of their data to be recovered". Also: wear leveling, bugged controllers with subsequent loss of data and some other issues I don't have time to research right now.

https://secure.wikim...ard_disk_drives

For me, right now SSDs are a no-go. Miniaturization will bring this very limited and controller-dependent storage technology to an halt. NAND Flash isn't for storing data, not for me.

I don?t know who wrote that, but I take issue with it.

HDD are one of the few mechanical pieces of a modern computer. This opens the door to mechanical failures, especially as they operate at such high speeds.

Of course SSDs do experience electronic failures, but this would likely be due to a manufacturing defect or electrical damage. And the same is true of HDDs as well.

Generally, when an SSD, it is a planned failure due to wear. When a HDD fails it is a catastrophic, unplanned mechanical failure.

As for buying advice, you're in luck as SSD prices have begun to drop. Right now there's an amazing deal on the Crucial 256GB m4 for just $170! Ends today, I think, so I'd hurry.

That's why I noted that the controller would prolong, just in case there was any doubt. :p When I was learning about the theory behind NAND memory at university I did a lot of research into it, which is why I'm confused that there's a misconception that they have a high rate of failing. Without any controllers (and other bits keeping it going) I can see why people might think that would be the case but the reality is they're reliable thanks to those things.

I think some people take Newegg reviews too much at heart. haha :p

I don?t know who wrote that, but I take issue with it.

HDD are one of the few mechanical pieces of a modern computer. This opens the door to mechanical failures, especially as they operate at such high speeds.

Of course SSDs do experience electronic failures, but this would likely be due to a manufacturing defect or electrical damage. And the same is true of HDDs as well.

Generally, when an SSD, it is a planned failure due to wear. When a HDD fails it is a catastrophic, unplanned mechanical failure.

As for buying advice, your in luck as SSD prices have begun to drop. Right now there's an amazing deal on the Crucial 256GB m4 for just $170! End today, I think, so I'd hurry.

Personally I'd recommend him to wait a few days as Corsair are rumoured to be launching their new Neutron series soon.

http://www.corsair.com/neutron/

I think some people take Newegg reviews too much at heart. haha :p

Well the OCZ Vertex 1's didn't really help the reliability case eh? :p

Well the OCZ Vertex 1's didn't really help the reliability case eh? :p

Those Sandforce controllers did seem to be a problem for some people. My Vertex 1 has been running strong for the past 6 years or so, and probably has a million read/writes on it. I should do a life scan on that thing and see how much longer it thinks it will survive. :)

Those Sandforce controllers did seem to be a problem for some people. My Vertex 1 has been running strong for the past 6 years or so, and probably has a million read/writes on it. I should do a life scan on that thing and see how much longer it thinks it will survive. :)

Yeah, thank god for Marvel. <_< I'm going to do the same out of curiosity.

edit: 8 years, ntb.

Image%202012-07-23%20at%209.10.52%20PM.png

Mine is both bigger, and more used!

My penis SSD is way better than yours :p

(I'm just kidding. I'm sure you won't be offended, but thought I had better say so >.<)

LOLLLLLLL........

It's not how big it is, it's how you use it.. :shifty:

I find it curious you've powered it on 60x but it's been online for 5000+ hours. :o You really don't like turning it off eh?

  • Like 2

From your question above, I highly recommend you get SSD and Windows installed by a professional. Windows 7 takes about 20GB when installed.

Windows 8, surprisingly, takes up about the same amount of space. The real space eaters with Windows (XP and later) aren't the OS or even applications and games, but data files the applications and games access - precisely the features that should NOT be parked on an SSD.

The reason SSDs are such major performance boosters is the same reason thumb-drives eat optical drives for lunch when doing installation - utterly-ridiculous read performance.

I proved this to myself rather easily (and despite not having this in mind) - I bought two 8GB USB 2.0 thumb drives (one specifically for OS install testing).

That target system is, in fact my test-bench system, with the following setup -

CPU - Intel Q6600

RAM - 4GB DDR2-800 (2GB x2)

HDD - WD AC1000 (1TB ex-MyBook converted to desktop use) SATA-3.0g HDD

ODD - Samsung SH-223B SATA optical drive (comparison)

KB - Microsoft Wireless Desktop 6000 V.3 (USB)

Mouse - Logitech V220 Cordless (USB)

OS - Windows 8 Release Preview x64

MB - ASUS P5G41T-M LX2/GB

The thumbdrive itself is a MicroCenter 8GB USB thumbdrive (the same ones that they have been selling for $5.99 in their stores). Thumbdrives, like SSDs, are based on flash memory - in fact, one could say that an SSD is a thumbdrive writ large. However. thumbdrives, unlike SSDs, are designed around the USB externally-hot-pluggable specification (SSDs are designed generally for SATA or the mSATA spec for desktops or portable uses). I configured the test thumbdrive using the freely-available USB/DVD image configuration tool from the Microsoft Online Store, and using the downloadable ISO image of the Release Preview.

Okay. We have our contenders - optical drive vs. thumbdrive. SATA-150 vs. USB 2.0. Internal vs. external.

Winner - USB, and it wasn't close. The thumbdrive took less than half the install time compared to the optical drive.

The why of it - While USB thumbdrives (in fact, flash memory in general) has somewhat of a write lag (exacerbated in the case of thumbdrives by the lower-quality flash compared to SSDs) compared to magnetic media (such as hard drives) when a flash drive is the source, it utterly lacks a *read lag* compared to an optical drive - thus a flash-based source has the best transfer rates to any target (non-RAID sources, that is).

Internally, an SSD is structured similarly to a RAID array of thumbdrives; just as a RAID array of hard drives has better throughput than a single drive in the array, the same is true of an SSD compared to a thumbdrive. (Scaling upward, the same is, naturally, true of SSD-based RAID arrays compared to single SSDs.)

Remember, I used older hardware in the *seat of the pants* test - not modern hardware. Optical-drive performance has, in fact, gotten NO faster since; the same can't be said of even thumb drives, let alone SSDs. That means that optical drives are even further in the performance doghouse.

He's right, except the read lag he is referring to is latency, and the latency is the big difference as opposed to the increase in read speed (throughput).

For ****s and giggles. USB -> SSD WIndows installs are BLAZINGLY fast :p

USB3 -> SSD installs will make you cream your pants :p

LOLLLLLLL........

It's not how big it is, it's how you use it.. :shifty:

I find it curious you've powered it on 60x but it's been online for 5000+ hours. :o You really don't like turning it off eh?

That, and certain firmware updates reset certain smart values (specifically powered on times. It's the only one I have ever noticed).

I work and play off of this machine, It doesn't get turned off much :p

SSD = HUGE performance gain, IF your existing HDD is not performing as well as you'd expect. My Vertex+ Sata II drive replaced my WD 420gb as a boot disk, and I notice a huge difference. If you go Sata III with a new generation chipset the performance gains are exceptional. And as for stability and crash-proof; just maintain a backup. Same rule applies with ANY drive future, past or present that you want to preserve the data of.

Those Sandforce controllers did seem to be a problem for some people. My Vertex 1 has been running strong for the past 6 years or so, and probably has a million read/writes on it. I should do a life scan on that thing and see how much longer it thinks it will survive. :)

OCZ vertex has been out for 3 years. My first SSD was an OCZ Vertex 60GB. I got it because it was cheap, and it failed within 4 months. I've seen a few articles on the internet stating OCZ is the worst in terms of SSD reliability. I believe it, my friend's OCZ also died. I've been rocking Intel since late 2009, and still going strong.

I have two SSD's in Raid0 in my (otherwise not so extremely well-spec'd) laptop and all I can say is that it is absolutely amazing. Windows boots in around 15 seconds and everything launches completely instantly. Getting an SSD is the single best upgrade you can do for pretty much any PC nowadays. Go for it!

Yeah, you're wrong. I'm particularly amused that no-one picked up on "faster on newer hardware" as that's just complete tosh.

Nand doesn't degrade when not in use where as magnetic mediums do; So for pure storage (i.e. plug in, copy, unplug), the only downside of SSDs is cost. Even on scratch disks the drives survive for longer than the lifetime of the computer and S.M.A.R.T. is more accurate in predicting its impending death.

For those of us that are interested in performance the SSDs are immeasurably better due to their extremely low latency. Whilst I agree that performance isn't everything, this is a worthwhile performance boost.

Technically untrue. As the SSDs shrink in die size, the NAND memory becomes less durable and slower.

I would however agree that the firmware is a million times better now than it was then.

Not once did I say that NAND wouldn't die. :p Firmware improves NAND writes by making sure all the cells are written to equally, this is why previous firmwares failed to fix since you would go through NAND writes pretty fast. Also, I did state that in my opinion, a SSD is likely to fail as a HDD so the whole argument which is more reliable is very mute as they both have issues of their own. A SSD or HDD will never be 100% reliable.

Sorry but the one who knows very little about NAND flash it's you: memory cells are a long-term nightmare for anyone caring for his or her data.

Anyone caring for his/her data does not keep it in one place. You BACKUP and then BACKUP THE BACKUP (and once again if you are paranoid).

He's right, except the read lag he is referring to is latency, and the latency is the big difference as opposed to the increase in read speed (throughput).

For ****s and giggles. USB -> SSD WIndows installs are BLAZINGLY fast :p

USB3 -> SSD installs will make you cream your pants :p

That, and certain firmware updates reset certain smart values (specifically powered on times. It's the only one I have ever noticed).

I work and play off of this machine, It doesn't get turned off much :p

Agreed - which is, in fact, the other half of the argument.

The motherboard used for my *seat-of-the-pants* test is based on the consumer-stable/corporate-stable Intel G41 chipset with the ICH7 southbridge.

No AHCI support. No RAID. SATA support tops out at SATA-3.0g. USB support is only 2.0.

For that reason, I configured my *second* 8GB thumbdrive as install media (Windows 7) - even USB 2.0 speeds kick the crap out of optical-drive speeds for installation - and that's just to hard drives.

Installing to SSD (even from the same speed USB 2.0 thumbdrives) will naturally be faster, with USB 3.0-based thumbdrives faster still.

Even more telling, 8GB drives are far from expensive - MicroCenter is still selling the same thumbdrives I bought two weeks ago (Fairfax, VA) for all of $5.99 each - USB 3.0-rated drives (same size) are $3 more each.

Buy *at least one* when you buy the motherboard for your build; if you're still trying to decide between Windows 8 and Windows 7, buy two - one for each. Build PCs for others? Include a system-restore thumbdrive.

This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.
  • Posts

    • Cjam 2.5.0.0 by Razvan Serea Cjam is a lightweight and fast MP3 editor for Windows that lets you cut, join, and edit MP3 files without re-encoding. This means your audio quality remains untouched, and edits happen instantly. Cjam is ideal for quick, lossless edits—whether you're trimming music, combining tracks, or preparing audio for learning tools or podcasts. It features batch processing, scripting support, cue and playlist file handling, and a simple interface. Cjam is perfect for anyone who needs efficient MP3 editing without the complexity of full audio suites. Cjam requires a PC running Windows 10 or later and Microsoft .NET 6.0 or later. Key features for Cjam: No Re-encoding: Edit MP3 files without losing quality. Cut and Join MP3: Easily cut, trim, and combine MP3 tracks. Batch Processing: Edit multiple files at once for faster workflows. Scriptable Interface: Automate tasks with a custom command language. Cue and Playlist Support: Handle CUE and playlist files for seamless audio management. Fast and Lightweight: Quick processing with minimal system resources. Lossless Audio Editing: Ensure your edits don't affect audio quality. Simple User Interface: Clean, intuitive design for easy navigation. File Format Support: Works with MP3, Cjam-specific file formats (CJAMC, CJAMJ, CJAM). Cjam 2.5.0.0 changelog: Added clipboard-based import/export support for mp3DirectCut Added clipboard-based export support for REAPER Added support for naming IMP3 elements Changed the Reset behavior to preserve Undo/Redo history; use Shift key + Reset button to clear it Added a new command parameter (qcp) Added 8 new entries to lang.txt (main_c124-126, main_d150-151, main_m082, vme_c014, vme_d005) Fixed a bug where the il parameter was incorrectly applied when pasting VMP3s into the main list Fixed several other minor bugs Download: Cjam 2.5.0.0 | 1.4 MB (Freeware) Links: Cjam Home Page | Cjam Manual | Screenshot Get alerted to all of our Software updates on Twitter at @NeowinSoftware
    • WTF! Nah, not this one. If you were back then using computers in the 90s or you had a clue, you'd know this was huge and was seen and overused absolutely everywhere and you'd take this article as no more than a retrospective story. That's all it is.
    • these "editors" get paid by the amount of articles and not the quality of content; hence why we get this random stuff from the same "editor" time and time again.
    • 256 times the quality of an audio cd? I think audio cd's sound good so even twice as good as that would be something.
    • It’s not about the NOx. It’s about the electrical field.
  • Recent Achievements

    • Dedicated
      JuvenileDelinquent earned a badge
      Dedicated
    • First Post
      DrWankel earned a badge
      First Post
    • Reacting Well
      DrWankel earned a badge
      Reacting Well
    • Week One Done
      Supreme Spray LV earned a badge
      Week One Done
    • Week One Done
      Genuinetonerink- Dubai earned a badge
      Week One Done
  • Popular Contributors

    1. 1
      +primortal
      504
    2. 2
      +Edouard
      163
    3. 3
      PsYcHoKiLLa
      91
    4. 4
      Steven P.
      75
    5. 5
      Michael Scrip
      72
  • Tell a friend

    Love Neowin? Tell a friend!