-
Windows no longer obstructs my view.
Using Kubuntu Linux since March 23, 2007.
"It is a capital mistake to theorize before one has data." - Sherlock Holmes -
I would agree with the survey. If it's got to have a spindle I prefer Hitachi drives. Samsung used to be my #2 choice and I've never had either a Samsung or Hitachi drive fail, but Samsung doesn't make hard drives any moreOriginally posted by Snowhog View Post
we see things not as they are, but as we are.
-- anais ninComment
-
My research in the past shows that SSDs fail at a rate about equal to HDDs. The latest stuff I've read states the the total failure rate of SSDs is now less than HDDs but they're more likely to have bad sectors, and that the age of an SSD is more a wear factor than use. One vendor - Puget Systems - says Samsung SSDs are the most reliable component they've ever sold. Of course, vendors will inflate the facts to sell products too.Originally posted by GreyGeek View Post
I'm sure most of us realize that most people don't take the time to post comments or reviews if we're satisfied with a product, only if we have trouble. Therefore in my view, bad reviews from retail customers are taken with very large salt grains - and an exponential reduction in weight. IMO, the 87% 5-star on Amazon is much more significant than the 3% 1-star. Interesting comparison: The Western Digital (my preferred HD brand) 750GB "Black" (performance) model 2.5" drive has a review ratio (5-stars to 1-stars) of 82:4 (780 reviews) and the model closest to your current HD, the 320GB "Blue" drive is even worse at 76:5 (1300+ reviews). Therefore, one can logically conclude that if Amazon reviews are a primary decision factor; the SSD is the correct choice. When you factor in the "free" benefits: power savings (better battery life), reduction in heat (better overall component life), and reduction in accidental damage from dropping, the SSD stands out as the best choice.
My anecdotal experience with SSDs vs HDDs: I bought my first SSD (cheapest I could get) about 8 years ago to play around with it. It died in a couple months and was replaced under warranty. The replacement lasted several years of light use. During that time, I had 4 HDD failures at almost exactly 3 years - just past their warranty expiration. These were heavily used and the first one gave off warnings so I pulled the others (they were a set of 4 used for RAID) all off-line and re-purposed them as backup drives. They all died within 4 months of each other (that's what you call quality control - nearly simultaneous death just past warranty dates
).
Since then, I have purchased Adata (2), Samsung (3), and Kingston(1) SSDs and WD(4) HDDs, none of which have failed or had bad sectors. If it means anything, my sister works for a media distribution company that has servers all over the world. They use Samsung SSDs by the tens of thousands and no platter drives.
The point to all this: There's a lot of FUD about SSDs most of which is debunkable. Computers in general are a fast moving target and SSDs are developing at a higher rate than that. The only reason to buy a platter drive is the GB-to-$ ratio. All other factors have diminished in deviation as to be inconsequential.
EDIT: My oldest Samsung SSD is a 840 Pro model: Power_On_Hours = 24262, Uncorrectable_Error_Cnt = 0, Total_LBAs_Written = 21346150209Comment
-
Another interesting note: Samsung 850 PRO series drives have a 10 year warranty. Most HDs are 3-5 years. That says a lot.Comment
-
So, do you have to reparation them to eliminate the Windows exe's and security junk so you can use Btrfs? What about "trim" to maintain speed. Available under Linux?"A nation that is afraid to let its people judge the truth and falsehood in an open market is a nation that is afraid of its people.”
– John F. Kennedy, February 26, 1962.Comment
-
Nope.Originally posted by GreyGeek View Post
Yes, and there are a couple ways of doing it. I personally run 'fstrim -a' as a weekly cron job instead of adding discard to fstab. The difference in performance is probably academic, but I don't generate enough garbage to run garbage collection more than once a week.Originally posted by GreyGeek View Post
Personal preference and completely unnecessary unless you're just into saving all the disk writes you can, but since mine are single-user machines I also set 'noatime' in fstab.we see things not as they are, but as we are.
-- anais ninComment
-
Mmmmm ... I've already ordered an identical replace drive, which arrives tomorrow. Also, even though my HD is 640GB the Btrfs scrub checked only 58GB, which is what "show" also shows. I may get a 128GB SSD just to experiment with."A nation that is afraid to let its people judge the truth and falsehood in an open market is a nation that is afraid of its people.”
– John F. Kennedy, February 26, 1962.Comment
-
Highly recommended. The difference in performance is phenomenal.Originally posted by GreyGeek View Post
If I may offer one more piece of advice; I've seen significant improvement in a SATA 1 netbook with an SSD; since on a home computer sustained throughput generally isn't a requirement there's no sense in going for a SATA interface your machine doesn't support. F'rinstance, if your machine can only do 1.5 or 3.0 Gbps SATA there's no advantage to spending extra money an SSD that will talk faster than your interface will
Last edited by wizard10000; Aug 10, 2016, 09:22 AM.we see things not as they are, but as we are.
-- anais ninComment
-
I disagree as a statement of value. Good luck finding an SATA II SSD that costs much less than a comparable SATA III model. Since all SATA III drives are backward compatible, you're actually recommending buying an older (in terms of production date) than what's current. Using Samsung and Amazon again: The only available SATA I drive is 120GB at $60, their SATA II drives have been discontinued and are only available used, but the 128GB 850 Pro w/10 year warranty is the most expensive option at $90. You can get the EVO model (5 year warranty) for $65.Originally posted by wizard10000 View PostHighly recommended. The difference in performance is phenomenal.
If I may offer one more piece of advice; I've seen significant improvement in a SATA 1 netbook with an SSD; since on a home computer sustained throughput generally isn't a requirement there's no sense in going for a SATA interface your machine doesn't support. F'rinstance, if your machine can only do 1.5 or 3.0 Gbps SATA there's no advantage to spending extra money an SSD that will talk faster than your interface will
IMO, the added value of better firmware and more current mem chips would outweigh the additional $5-30 or so you'll spend. Furthermore, in this case - the SSD has a better than average chance of outlasting a 2010 model laptop.Comment
-
Sorry for not making my point clear. Let me try againOriginally posted by oshunluvr View Post
My point is that if your SATA port will only do 300MB/s on paper there's no need to spend *extra* money on a drive that'll run twice that fast. So with a SATA 2 controller if you're choosing between two drives, one that'll do 400MB/sec and one that will do 600MB/sec either will do because both of them will talk faster than your interface and there's no reason to make the extra speed a factor in your purchase decision.we see things not as they are, but as we are.
-- anais ninComment
-
My anecdotal contribution to the discussion:
In late 2010 I built a new desktop system and I wanted an optimum combination of performance and reliability. I consulted with oshunluvr who suggested a PCI bus SSD would be really fast and not use SATA connections, giving me good expansion capability for large hdds. In the succeeding 6 years I have replaced the motherboard, the graphics card, and the pair of 1T hdds that I use as a BTRFS data storage system, but I have kept the rest of the system including the SSD. Here it is at almost 6 years of running pretty much continuously. I did not remember that there have been 187 power failures -- not sure what that's all about, but I really like the uncorrected error count, out of almost 10 million. It's an ext4 filesystem running a Linux OS.
Code:smartctl 6.4 2014-10-07 r4002 [x86_64-linux-3.16.0-4-amd64] (local build) Copyright (C) 2002-14, Bruce Allen, Christian Franke, www.smartmontools.org === START OF INFORMATION SECTION === Model Family: SandForce Driven SSDs Device Model: OCZ-REVODRIVE Serial Number: OCZ-0C2Z27B5QEE4L0H2 LU WWN Device Id: 5 e83a97 f60df6247 Firmware Version: 1.37 User Capacity: 60,022,480,896 bytes [60.0 GB] Sector Size: 512 bytes logical/physical Rotation Rate: Solid State Device Device is: In smartctl database [for details use: -P show] ATA Version is: ATA8-ACS T13/1699-D revision 6 SATA Version is: SATA 2.6, 3.0 Gb/s Local Time is: Wed Aug 10 15:18:25 2016 EDT SMART support is: Available - device has SMART capability. SMART support is: Enabled === START OF READ SMART DATA SECTION === SMART overall-health self-assessment test result: PASSED General SMART Values: Offline data collection status: (0x00) Offline data collection activity was never started. Auto Offline Data Collection: Disabled. Self-test execution status: ( 0) The previous self-test routine completed without error or no self-test has ever been run. Total time to complete Offline data collection: ( 0) seconds. Offline data collection capabilities: (0x7f) SMART execute Offline immediate. Auto Offline data collection on/off support. Abort Offline collection upon new command. Offline surface scan supported. Self-test supported. Conveyance Self-test supported. Selective Self-test supported. SMART capabilities: (0x0003) Saves SMART data before entering power-saving mode. Supports SMART auto save timer. Error logging capability: (0x01) Error logging supported. General Purpose Logging supported. Short self-test routine recommended polling time: ( 1) minutes. Extended self-test routine recommended polling time: ( 5) minutes. Conveyance self-test routine recommended polling time: ( 2) minutes. SCT capabilities: (0x003d) SCT Status supported. SCT Error Recovery Control supported. SCT Feature Control supported. SCT Data Table supported. SMART Attributes Data Structure revision number: 10 Vendor Specific SMART Attributes with Thresholds: ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 1 Raw_Read_Error_Rate 0x000f 105 099 050 Pre-fail Always - 0/9806339 5 Retired_Block_Count 0x0033 100 100 003 Pre-fail Always - 0 9 Power_On_Hours_and_Msec 0x0032 100 100 000 Old_age Always - 48801h+39m+19.650s 12 Power_Cycle_Count 0x0032 100 100 000 Old_age Always - 651 171 Program_Fail_Count 0x0032 000 000 000 Old_age Always - 0 172 Erase_Fail_Count 0x0032 000 000 000 Old_age Always - 0 174 Unexpect_Power_Loss_Ct 0x0030 000 000 000 Old_age Offline - 187 177 Wear_Range_Delta 0x0000 000 000 000 Old_age Offline - 0 181 Program_Fail_Count 0x0032 000 000 000 Old_age Always - 0 182 Erase_Fail_Count 0x0032 000 000 000 Old_age Always - 0 187 Reported_Uncorrect 0x0032 100 100 000 Old_age Always - 0 194 Temperature_Celsius 0x0022 030 030 000 Old_age Always - 30 (Min/Max 30/30) 195 ECC_Uncorr_Error_Count 0x001c 105 099 000 Old_age Offline - 0/9806339 196 Reallocated_Event_Count 0x0033 100 100 003 Pre-fail Always - 0 231 SSD_Life_Left 0x0013 100 100 010 Pre-fail Always - 0 233 SandForce_Internal 0x0000 000 000 000 Old_age Offline - 2048 234 SandForce_Internal 0x0032 000 000 000 Old_age Always - 3008 241 Lifetime_Writes_GiB 0x0032 000 000 000 Old_age Always - 3008 242 Lifetime_Reads_GiB 0x0032 000 000 000 Old_age Always - 10048Comment
-
I think I understood your point, I just disagree with it - or at least partially. My point was there may be a reason to spend extra dollars on the latest technology SSD because there are other factors (as I listed before) than just interface or drive speed. Assuming all things equal (firmware, memchips, warranty, reliability, etc.) except IOPS - and assuming you didn't care about future use of the SSD in another system that might have a higher interface speed - of course, take the most cost effective option.Originally posted by wizard10000 View PostSorry for not making my point clear. Let me try again
My point is that if your SATA port will only do 300MB/s on paper there's no need to spend *extra* money on a drive that'll run twice that fast. So with a SATA 2 controller if you're choosing between two drives, one that'll do 400MB/sec and one that will do 600MB/sec either will do because both of them will talk faster than your interface and there's no reason to make the extra speed a factor in your purchase decision.
Additionally, my quick Amazon search shows you may benefit from these other factors without actually spending more, thus negating the dollar argument. Two years ago, I might have agreed with you, but the prices of SSDs have fallen faster than (to quote a Kansas old timer) "a stream-lined sewer lid."
I agree with the gist of your comment, but IMO the reality would likely be one would might up with the faster drive anyway because of the other factors. I guess my comparative retort to your point would be: There's no need to buy a slower SSD just because your interface won't do SATA III speeds.
I think we'd both agree that dollar-to-performance is what we're talking about. I just think the "performance" is not interface speed alone, but also longevity, future compatibility, and modern firmware - all of which are often enhanced with a faster speed rated over a slower model - regardless of interface.
I also now think I've flogged this dead horse enough. It appears we are not far apart in opinions, just viewing it from a different angle. So, I will allow you the final word as we close this discussion <bows gracefully>
Comment
-
Unfortunately for most us, we don't have Don's big "computer-bucks account" to spend on our upgrades!Originally posted by dibl View Post Teasing of course...
If I recall, those PCI/SSD boards were rather "spendy" back then. Sounds like you've gotten your money's worth tho...Comment
-
I think we're about 90% on the same page. Close enough for meOriginally posted by oshunluvr View PostSo, I will allow you the final word as we close this discussion <bows gracefully>
we see things not as they are, but as we are.
-- anais ninComment
-
Yes, I did part with some serious money. But, look at the reliability and durability that I bought! And I now remember why all the power interruptions -- the first motherboard started puking graphical garbage and then rebooting. I thought the problem was the graphics card, so I bought a new one ( that was not cheap, either). And it did not touch the problem, which is when I knew it was the motherboard that was the problem. So out went more $$$ in 2014 for a new motherboard. But it's been fine ever since. Here are details:Originally posted by oshunluvr View Post
Code:em: Host: Hibiscus Kernel: 4.7.0-towo.1-siduction-amd64 x86_64 (64 bit gcc: 6.1.1) Desktop: KDE Plasma 5.6.5 Distro: siduction 14.1.0 Indian Summer - kde - (201411230337) Machine: System: ASUS product: All Series Mobo: ASUSTeK model: Z87-WS v: Rev 1.xx Bios: American Megatrends v: 2004 date: 06/05/2014 CPU: Quad core Intel Core i7-4770 (-HT-MCP-) cache: 8192 KB flags: (lm nx sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx) bmips: 28023 clock speeds: max: 3900 MHz 1: 3400 MHz 2: 3399 MHz 3: 3399 MHz 4: 3400 MHz 5: 3400 MHz 6: 3404 MHz 7: 3405 MHz 8: 3404 MHz Graphics: Card: NVIDIA GM107 [GeForce GTX 750 Ti] bus-ID: 05:00.0 Display Server: X.Org 1.18.4 driver: nvidia Resolution: 1920x1200@59.88hz, 1920x1080@59.94hz GLX Renderer: GeForce GTX 750 Ti/PCIe/SSE2 GLX Version: 4.5.0 NVIDIA 361.45.18 Direct Rendering: Yes Network: Card-1: Intel I210 Gigabit Network Connection driver: igb v: 5.3.0-k port: d000 bus-ID: 07:00.0 IF: eth0 state: up speed: 1000 Mbps duplex: full mac: xxxxxxxxxxxxxxx Card-2: Intel I210 Gigabit Network Connection driver: igb v: 5.3.0-k port: a000 bus-ID: 0a:00.0 IF: eth1 state: down mac: e0:3f:49:e6:85:c4 Drives: HDD Total Size: 3120.7GB (26.8% used) ID-1: model: OCZ ID-2: model: OCZ ID-3: model: WDC_WD1001FALS ID-4: model: WDC_WD1000DHTZ ID-5: model: WDC_WD1000DHTZ Info: Processes: 273 Uptime: 1:38 Memory: 2052.3/7927.3MB Init: systemd runlevel: 5 Gcc sys: 6.1.1 Client: Shell (bash 4.3.461) inxi: 2.3.0Last edited by dibl; Aug 11, 2016, 03:59 AM.Comment
- Please do not use the CODE tag when pasting content that contains formatting (colored, bold, underline, italic, etc).
The CODE tag displays all content as plain text, including the formatting tags, making it difficult to read.
Comment