Does anyone have any info about new firmware for vanilla/basic 840 drives? Only article I have seen is from PC Perspective, and that is from last October. It would be nice to know if Samsung is going to provide firmware updates for vanilla/basic 840 drives also. http://www.pcper.com/news/Storage/Samsung-Germany-...
There will likely be a fix for OEM drives as well, though it will take a bit longer since it has to go through the PC OEM's validation first and will be distributed by the OEM.
I second this request. I have a laptop with a 500GB 840 (not EVO) and it suffers from extremely similar symptoms of significantly slower performance with older data. Many others have noticed this issue but Samsung seems to be willfully ignoring the possibility that the issue affects all TLC NAND drives. This is unfortunately turning out to be a bad customer experience, but in fairness I have yet to just try to RMA the drive directly.
I third this request, it's really bad how Samsung just ignores the original 840, the issue with it is identical, here is a HD Tune graph of a 120GB 840 belonging to a friend of me who bought it on my recommendation:
Nah, you just have to deal with it. That's how Samsung rolls. I've learnt my lesson with Samsung Galaxy S (their first money maker) and they still treated them like trash after a year. You just have to wait until there is no major fault with the product (usually 6 months to a year), then you can buy with confident. Hey, another plus point of waiting is you get it cheaper :D
Here's my HD Tune graph. It's of a 480GB SanDisk Extreme Pro. You can use it to compare. I have an 840 in the file server. I'll see what it shows in a bit.
While I certainly wouldn't call it a solution, per se, what about booting into a linux live CD and using dd to re-write every sector on the drive, and then (if you're running Windows 8) optimize the drive to re-TRIM it? You wouldn't want to do it often, for obvious write endurance reasons, but it would get you back to decent speed for some months.
I've been using a utility called 'DiskFresh' which allows you to rewrite all of the data on your drive from within Windows. It is free and simple to use. Still, performance is not 100%. See before: http://1drv.ms/1H4JYL7 and after: http://1drv.ms/1CX3qnl
I hope this update fixes this issue for good. Between myself and my family members, I have 5 of these drives to be disappointed about. The reviews and the prices were both great... sucks to get burned like this.
Looks to me like the new firmware will have these drives running great for their useful lifespan, which is years. By that point they'll be well overdue for replacement anyway. I'm glad Samsung is fixing the issue rather than just telling you to buy an 850 Evo. 840 I'm not too sure about, some don't seem to have major issues others do - I would bet running DiskFresh every once in a blue moon would help those guys even more so than it did you.
I recommended the vanilla 840 for about a year, when it was the price champ. The 840 EVO was never price competitive with the Crucials, which had arrived in the mean time.
It's hypothetical worst-case scenario assuming a full drive refresh (in reality only some cells are refreshed). It's unlikely that the cells would be refreshed on a weekly basis because it generally took at least a month for the read performance degradation to occur. In any case, the data refresh won't have any impact on typical client workload endurance.
Until Samsung says what the real update rate will be, I'll just assume 26%. Seriously, if you bought any other product in the world, and then were told that it was going to die 26% sooner than you were expecting, I'll bet you would say there was an impact. I completely understand that most people are unlikely to see any effect within the warranty period, but don't most of us expect products to last far longer ? In order to make me whole, it seems fair to refund 26% of my purchase price. I'll bet those Samsung guys are really happy about your 26% hypothetical scenario. Maybe it will force them to release some actual info, so that we can see it's not that bad. Until then - I want my 26% back !
Is really an issue when that 26 percent reduction takes you from "Far, far, far, far more write endurance than any consumer class user will use over the course of the hardware's service life" to only "Far, far, far more write endurance than any consumer class user will use over the course of the hardware's service life"?
So, does anyone want to join me in contemplating a real class action lawsuit and recall of the entire product line?
I'm honestly fairly serious about this. If the drive has to constantly refresh (re-write) cells and the cells have a specific endurance, this drive can't possibly end up lasting as long as it's rated for.
Your B ("this drive can't possibly end up lasting as long as it's rated for") does not necessarily follow A (" If the drive has to constantly refresh (re-write) cells and the cells have a specific endurance")
See the SSD endurance test over at Storage Review that ran for a year or two and only ended a month or two back to see why I say this. SSD endurance is very, very conservatively rated. Very, very.
Storage Review has only a few samples. Hardly what I'd call representative of the population. Excursions do happen, and you only need a few of those that barely make the cut to get out in the field to snowball into a PR nightmare But if Samsung guardbands their qual process enough (which i'm pretty sure they do), a reduction in NAND cycles will not be felt by the user before the warranty runs out.
and it taken them a year and a half of constant writing to make them fail (not something you normally do ) the Samsung 840 Pro was very impressive it never skipped a beat (until it did and died very silently) but the lack of any windows warning about pending failure in windows or the Samsung SSD toolbox is not very good but 2.4PB of data thats insane (but the 840 mostly went bad only after 300TB but did contune on to 700TB before doing other issues)
"We know that Samsung's 19nm TLC NAND is rated at 1,000 P/E cycles, so if the drive was to refresh all cells once a week, even that would only consume 52 cycles in a year. In five years time the total would be 260 cycles, which leaves you with 740 cycles for user data writes (for the record, that's 52GB of NAND writes per day for five years with the 120GB 840 EVO). "
This is a bullshit response and justification. Did Samsung feed you that? If it's rated for 1000 p/e cycles and it dies after I send it 740, then it didn't survive what it's rated and warrantied for, plain and simple. Samsung needs to strongly consider preemptively cutting checks to owners for at least 25% of the MSRP of the drive or just announce a straight up recall/replacement program and replace them with non-TLC drives.
Real world there's always write amplification; so some of the cycles the flash is rated for on paper will be consumed by the firmware not by user writes.
Also real world is that the number of GB of writes a consumer SSD is rated for is a small fraction of the total number of rewrite cycles. That's partly to make sure that even worst case amplification doesn't cause a drive to fail while still under warranty; and so they can sell drives with much higher rated write amounts at enterprise markups.
Samsung doesn't have any official endurance spec for the 840 EVO -- the 1,000 figure is based on our own tests. It's fully warrantied for three years, meaning that if it dies during that time frame you are eligible for a new, working drive, even if the failure was due to exhausted NAND endurance.
The cycle endurance rating NEVER refers to the user's activity - it measures the actual writes to the NAND. Whether you "send" it 740 cycles worth of data or 1000 is irrelevant. There is always going to be some write amplification just from wear leveling, if nothing else.
Anyway, it's highly unlikely that a user that had heavy enough usage for this to be a real problem would go anywhere near a consumer SSD, much less a TLC drive.
You're in a tizzy over theoretical numbers. Why? Why does it matter whether or not it actually hits 1000 P/E cycles when the fact is that unless you plan to use the drive for a decade and just absolutely hammer it with writes the entire time, you'll never get there anyway? (And ya know, if you are a person that really plans to hammer a drive with writes -- maybe in a database scenario -- the odds are very good that you won't be using a last generation TLC device anyway.)
Bit old but these drives have expected 1000 p/e, but reality it's more like 10x more then that (you can write over 1PB of data before it becomes an issue so at the time people are splitting hairs over it)
None samsung drives can fail around 300-600gb of written data
Samsung made trade off to bear those extra 260 P/E cycles to achieve performance by refreshing mechanism. This is not a perfect solution I would say, if some extensive user finishes his 760 cycles within 3 years then soon after 850 cycles product would fail (156 cycles have already completed by refresh). What samsung would say in this case ? Technically it has completed 1000 cycles so they might say this is not under warranty. But why would user want to trade off when he has paid full money. User wants both the things 1000 cycle endurance for HIS USAGE and performance mentioned in spec. Issue has been just patched but not resolved completely.
Do any of you complaining read the previous answers in this thread? Everything you're asking here has been asked before and has been properly answered.
all drives perform a refresh to make sure data is still accessible, but i guess the trigger for refresh was set way way to low on the 840 and 840 evo forcing the drive to do muti reads and ECC
I don't own an 840 Evo but feel compelled to comment. It certainly is starting to look like Samsung sold a bill of good with the 840 Evo's that they can't deliver on.
Also, Kristian I realize that you are trying to be impartial but it seems like you are defending Samsung. While the "letter of the warranty" doesn't say Samsung much do anything as long as their drive lasts 3 years that does not bode well for customer service.
I for one can tell you I won't touch a Samsung TLC drive with a 10 foot pole. In fact I'm going to stay away from all Samsung products in solidarity for the 840 Evo's that in my opinion are getting screwed here.
The 840 EVO is gonna in the end turn out to be a mistake I think. We had a fix, and now there's a fix that fixes what the first fix didn't fix. People in the know just aren't gonna use this particular series of drives. Maybe it took 3-D V-nand for TLC to really be ready for prime time?
I know that I personally will not be buying any more Samsung drives until the issue is conclusively fixed on not just the 840 EVO (of which I own a 500gb mSATA version) and the 850 EVO. I wouldn't mind getting a Pro series ifbit weren't for this debacle. I have also made sure friends in the market for drives are qware of how Samsung is mishandling this situation.
The problem is that from the minute the performance optimization is completed, the performance is degrading until it is run again! Just like it is now after runinng disk fresh or whatever to restore performance.
I just can't trust another Samsung drive right now. This issue has been on going for a very long time now, way too long to inspire any confidence.
The fact that the 840 is being ignored just makes this worse. And what about the Samsung PM851 OEM drive that come in laptops? That's TLC, will that get the fix?
Did you actually read the article? The new firmware is designed to detect which blocks need a refresh and to then refresh them during idle time, without user intervention.
I did thanks! But regardless if the refresh was started manually or without user intervention, as soon as the refresh is complete the data on your drive begins to degrade. And quite quickly at that, within a month there is a noticeable drop in performance.
But the firmware would of refreshed them slow blocks before it became a noticeable problem
the issue before was that it was allowed to get way to low voltage to the point where the ecc retry got high enough to be noticeable and did not trigger a page refresh
there first attempt at fixing still had the TLC NAND page refresh voltage set to low so it still had high ECC read retry was still happening on old data (about 2 months old from the looks of it, as to why this bug was one of them harder to test for) , the last 2 firmware updates has resolved it permanently and performes a refresh when it's at low voltage when drive is idle or high read retry ecc event happens (witch likey won't happen due to background idle scan)
So, I assume charge degradation would affect the drive in power off state too right? So if I for some reason left the drive off for 5 months (contrived example: pure archive drive), then I can't necessarily trust the data on the drive? That sounds slightly alarming.
As far as the current reports go, the drive would become dog-slow but the data would still be intact, according to the data retention spec (don't know what it is for this drive).
I have the exact same concern. I can drag out an old magnetic drive years after I've last used it and can read it fine. I recently booted up an old machine that was running Windows 98! I've never heard before that would be a concern with SSD, but it will apparently be a very real problem with the 840 EVO. I've already had one die exhibiting excessive ECC that sounds a lot like an extreme case of this same problem.
I'm very concerned that my Surface Pro 3 has a Samsung SSD in it (reportedly of the 840 EVO technology). If I stop using it for a year it's drive may be effectively erased? Better get the recovery partitions off onto a flash drive...
Like others have said, this may be my last Samsung SSD. Too bad I didn't know about one being in the Surface. Nothing I can do about that (except at least they are apparently preparing the same firmware update for the Surface drive - they never released the first version for it).
By mid 2014, problems were well documented. These are problems that made Samsung TLC drives definitely lower tier performance in the market and markedly different that the advertised and specified performance.
By late 2014, a first firmware 'fix' was found ineffective, but Samsung still promised to restore TLC performance to spec'd and advertised performance.
By Feb 2015, Samsung still promised to restore the drives to spec and their perceived market position in March, and evidence of slowdowns in the 840, 840evo and TLC based OEM drives were widely reported.
In mid April, (pardon the mixed metaphor), Samsung fed Anandtech the koolaid and said,, these are not the slowdowns you are looking for,, move along.......
So, to be quite clear, the slowdown disguise from Samsung ONLY works if you choose to install magician 4.x, and ONLY works if you have a consumer drive, and NOT if you have a Samsung OEM TLC based drive. (Oh and the extra bits written are also not of any concern, of course!)
So the 'fix' is to play musical chairs with the drive data every so often... ????
That's no fix,, restoring performance as advertised and spec'd is a fix!
did you fully read the article? There is a FW coming out as well, you won't need Magician if you update the FW since the FW will be doin everything for you automatically when you leave the system powered on and idle.
The middle chart for read speed clearly shows the FW update does not restore performance. It is not 'restored' until all the data is forcibly moved around. If the FW works, then why the need to include the full-drive data moving option at all? And so far, the indications are that the updates are for the 840evo only and not the other Samsung TLC drives.
Note that the drive wasn't given any idle time to fully restore the performance. With some idle time the performance should be fixed without the need to run the optimization tool. Its main purpose is for drives that have been used for archival purposes (although I wouldn't recommend an SSD, let alone a TLC drive for that) as the firmware's internal optimization process can only run if the drive is powered on.
The ssd was not given any time to idle after he did the firmware update (read the article) so it been able to restore the ssd even to that point was very good
more then likely just running the full ssd benchmark test after the firmware update most likely caused a force page refresh due to high ECC read retrys so the rest of the test was at expected speed (doing the adv performance optimisation should only need to be a one time event)
Want to hear the real kick in the nuts? I have multiple drives in my video editing machine. The OS drive is an EVO, which the performance "optimization" tool updates fine. But for the source file drive, the tool won't fix it. I called in to Samsung, and I have to do this whirly bird routine of taking the drive completely out of the machine, putting it in my laptop, booting from some boot disk that I have to make, and then running the tool with some command line garbage. I've already been through that routine once. Anyone here think I want to go through that again? Me neither. Unless these drives are replaced with something that works as advertised, I'll never touch another Samsung product of any sorts so long as I live. This is pure bull.
Well luckily there hasn't been any slowdown in the 850 evo drive. Since the cells are 40nm again I think the much larger cell size makes them a lot more robust, I believe the 840 evo used 20nm cells. Even so I still would not go for TLC nand. I have a 512GB 840 pro and the MLC it uses is flawless and it's speed is so good there is no point to upgrade until nvme and pci-e 3.0 x4 drives become commonplace. Tho I could use more space for more game installs as they have become so huge. May have to get a 1TB 850 pro to tide me over until 2TB drives are the same price as current 1TB drives and are 3.0 x4 interface and nvme.
My 850 EVO have had a notisable performance loss after being unused for a while now. A round with diskfresh dragged the performance up from ~200MB/s to the normal 500+ MB/s. I'm surpriced nobody else is experiencing this problem...? I have the 840 EVO as well, and i do agree the performance degration is not even on the same page, but still a loss of over 50% performance on the 850 should be reason for Samsung to take action?
I am at the end of my rope with my brand new Samsung evo 640 500gb ssd. I get MAX 100MB/s on Novabench no matter what settings in samsung magician I use. I am using AHCI mode, not ide, so this should be blazing fast but since the day I got it, no love.
Sabertooth r2.0 8gb ram FX-4100 GTX n760 hawk Newest stable bios on Sabertooth board and newest firmware on ssd.
My Adata 900 series 128gb was 3x as fast as this drive
Corporate systems might have Intel AMT available and with that you could do all the flashing remotely using the DOS tool for example. If you got consumer stuff just walk around starting the update. Pretty sure Magician doesn't allow automation / unattended firmware flashing. The server tool doesn't work with Evo's.
So how does this affect power consumption? 840 EVO was a good drive for lower power usage in a notebook (at least from what I remember), now with this new firmware, does it remain so?
Correct me if I'm wrong: All these tweaks to keep the aging cells of your drive alive seem to depend on that you actually use your drive regularly. But that doesn't help if you leave your drive with your precious data without power for a couple of months, does it?
That's actually my usage pattern, and I'm in the market for a new notebook and new SSD as well. I use my work notebook for private stuff, then boot my private PC very rarely (say 1-2 a year) to move private stuff from the work notebook to the private notebook.
It seems my comment got removed somehow. I was just asking if the firmware tweak helps if you are not using your drive for a couple of months. That's my (maybe not so common) usage pattern.
So: 1) How reliable are the EVO TLC drives when unpowered for a couple of months? 2) What is the expected difference (in say months not powered on) between the EVO and Pro SSDs?
Hi: Anyone experience 840 Pro firmware issue under raid controller? we have 6 x 840 Pro 512GB running under Dell PowerEdge 720 with H710 controller as raid1, scores around 400MB/s sequential 4K write constantly over past year until last month, we upgraded 2 x SSD firmware from DXM05B0Q to DXM06B0Q and the sequential 4K write speed dropped dramatically from 400MB/s to fluctuation of 20-200MB/s in the same hardware. We thought it might be our hardware, so we moved the two different firmware version SSD to another PowerEdge, (as 2 x raid0), as yield same result. We have been talking to Samsung helpdesk who has been nice but unhelpful, It seems to me this is a firmware bug. BTW, if we plug the old and new version of SSD back to a desktop with a SATA3 connector, re-do the 4K write test, it is constantly lower on new firmware but not much, something like 390MB/s vs 420MB/s
Here is the command we use to test with a 20GB test file, we even re-format the drive and tested as empty drive, still same result
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
78 Comments
Back to Article
Agent_007 - Tuesday, April 14, 2015 - link
Does anyone have any info about new firmware for vanilla/basic 840 drives? Only article I have seen is from PC Perspective, and that is from last October. It would be nice to know if Samsung is going to provide firmware updates for vanilla/basic 840 drives also.http://www.pcper.com/news/Storage/Samsung-Germany-...
Kristian Vättö - Tuesday, April 14, 2015 - link
For now all I know is that the fix is only for the 840 EVO. However, I'm trying to get a response from Samsung regarding the vanilla 840.kgh00007 - Tuesday, April 14, 2015 - link
What about the Samsung PM851 OEM drive as well, which is also TLC?Flunk - Wednesday, April 15, 2015 - link
Samsung has said it doesn't affect the 850 series.Kristian Vättö - Wednesday, April 15, 2015 - link
There will likely be a fix for OEM drives as well, though it will take a bit longer since it has to go through the PC OEM's validation first and will be distributed by the OEM.fade2blac - Tuesday, April 14, 2015 - link
I second this request. I have a laptop with a 500GB 840 (not EVO) and it suffers from extremely similar symptoms of significantly slower performance with older data. Many others have noticed this issue but Samsung seems to be willfully ignoring the possibility that the issue affects all TLC NAND drives. This is unfortunately turning out to be a bad customer experience, but in fairness I have yet to just try to RMA the drive directly.Per Hansson - Tuesday, April 14, 2015 - link
I third this request, it's really bad how Samsung just ignores the original 840, the issue with it is identical, here is a HD Tune graph of a 120GB 840 belonging to a friend of me who bought it on my recommendation:http://img.techpowerup.org/141020/27-september-201...
stickmansam - Tuesday, April 14, 2015 - link
I agree, I've seen the 840 with read issues as well. The 840 never got the first "fix" the 840 EVO got toomercutio - Tuesday, April 14, 2015 - link
for some reason i only get 10mb/sec on my samsung 840 non-evo with old data (over a year old; i don't use it anymore; it's pretty much unusable)Samus - Tuesday, April 14, 2015 - link
copy the data off (or image the drive to a spare hard disk) and secure erase it. only way to guarantee killing stale blocks.WorldWithoutMadness - Tuesday, April 14, 2015 - link
Nah, you just have to deal with it. That's how Samsung rolls.I've learnt my lesson with Samsung Galaxy S (their first money maker) and they still treated them like trash after a year.
You just have to wait until there is no major fault with the product (usually 6 months to a year), then you can buy with confident. Hey, another plus point of waiting is you get it cheaper :D
bigboxes - Tuesday, April 14, 2015 - link
Here's my HD Tune graph. It's of a 480GB SanDisk Extreme Pro. You can use it to compare. I have an 840 in the file server. I'll see what it shows in a bit.http://i256.photobucket.com/albums/hh175/bigboxes/...
bigboxes - Tuesday, April 14, 2015 - link
Well, it seems that your post is backed up with facts. Here's my HD Tune graph for the file server (with the Samsung 840 120GB).http://i256.photobucket.com/albums/hh175/bigboxes/...
Maltz - Tuesday, April 14, 2015 - link
While I certainly wouldn't call it a solution, per se, what about booting into a linux live CD and using dd to re-write every sector on the drive, and then (if you're running Windows 8) optimize the drive to re-TRIM it? You wouldn't want to do it often, for obvious write endurance reasons, but it would get you back to decent speed for some months.3DoubleD - Tuesday, April 14, 2015 - link
I've been using a utility called 'DiskFresh' which allows you to rewrite all of the data on your drive from within Windows. It is free and simple to use. Still, performance is not 100%. See before: http://1drv.ms/1H4JYL7 and after: http://1drv.ms/1CX3qnlI hope this update fixes this issue for good. Between myself and my family members, I have 5 of these drives to be disappointed about. The reviews and the prices were both great... sucks to get burned like this.
Alexvrb - Tuesday, April 14, 2015 - link
Looks to me like the new firmware will have these drives running great for their useful lifespan, which is years. By that point they'll be well overdue for replacement anyway. I'm glad Samsung is fixing the issue rather than just telling you to buy an 850 Evo. 840 I'm not too sure about, some don't seem to have major issues others do - I would bet running DiskFresh every once in a blue moon would help those guys even more so than it did you.Maltz - Wednesday, April 15, 2015 - link
Does DiskFresh run on the boot drive? Because if it's running on a drive with any mounted volumes, it can't be rewriting ALL sectors.leexgx - Friday, April 17, 2015 - link
at least access time is unaffected (it does not stall under load)MrSpadge - Tuesday, April 14, 2015 - link
+1I recommended the vanilla 840 for about a year, when it was the price champ. The 840 EVO was never price competitive with the Crucials, which had arrived in the mean time.
dyermaker - Tuesday, April 14, 2015 - link
A 26% reduction in endurance is a huge issue.Kristian Vättö - Tuesday, April 14, 2015 - link
It's hypothetical worst-case scenario assuming a full drive refresh (in reality only some cells are refreshed). It's unlikely that the cells would be refreshed on a weekly basis because it generally took at least a month for the read performance degradation to occur. In any case, the data refresh won't have any impact on typical client workload endurance.dyermaker - Wednesday, April 15, 2015 - link
Until Samsung says what the real update rate will be, I'll just assume 26%. Seriously, if you bought any other product in the world, and then were told that it was going to die 26% sooner than you were expecting, I'll bet you would say there was an impact. I completely understand that most people are unlikely to see any effect within the warranty period, but don't most of us expect products to last far longer ? In order to make me whole, it seems fair to refund 26% of my purchase price. I'll bet those Samsung guys are really happy about your 26% hypothetical scenario. Maybe it will force them to release some actual info, so that we can see it's not that bad. Until then - I want my 26% back !leexgx - Friday, April 17, 2015 - link
so it die in 5-10 years instead of 10-20 years (you most like not have it by then)Romberry - Tuesday, April 14, 2015 - link
Is really an issue when that 26 percent reduction takes you from "Far, far, far, far more write endurance than any consumer class user will use over the course of the hardware's service life" to only "Far, far, far more write endurance than any consumer class user will use over the course of the hardware's service life"?cygnus1 - Tuesday, April 14, 2015 - link
So, does anyone want to join me in contemplating a real class action lawsuit and recall of the entire product line?I'm honestly fairly serious about this. If the drive has to constantly refresh (re-write) cells and the cells have a specific endurance, this drive can't possibly end up lasting as long as it's rated for.
Romberry - Tuesday, April 14, 2015 - link
Your B ("this drive can't possibly end up lasting as long as it's rated for") does not necessarily follow A (" If the drive has to constantly refresh (re-write) cells and the cells have a specific endurance")See the SSD endurance test over at Storage Review that ran for a year or two and only ended a month or two back to see why I say this. SSD endurance is very, very conservatively rated. Very, very.
menting - Tuesday, April 14, 2015 - link
Storage Review has only a few samples. Hardly what I'd call representative of the population. Excursions do happen, and you only need a few of those that barely make the cut to get out in the field to snowball into a PR nightmare But if Samsung guardbands their qual process enough (which i'm pretty sure they do), a reduction in NAND cycles will not be felt by the user before the warranty runs out.leexgx - Friday, April 17, 2015 - link
and it taken them a year and a half of constant writing to make them fail (not something you normally do ) the Samsung 840 Pro was very impressive it never skipped a beat (until it did and died very silently) but the lack of any windows warning about pending failure in windows or the Samsung SSD toolbox is not very good but 2.4PB of data thats insane(but the 840 mostly went bad only after 300TB but did contune on to 700TB before doing other issues)
jann5s - Wednesday, April 15, 2015 - link
perhaps you mean this one at Tech Report:http://techreport.com/review/27909/the-ssd-enduran...
cygnus1 - Tuesday, April 14, 2015 - link
"We know that Samsung's 19nm TLC NAND is rated at 1,000 P/E cycles, so if the drive was to refresh all cells once a week, even that would only consume 52 cycles in a year. In five years time the total would be 260 cycles, which leaves you with 740 cycles for user data writes (for the record, that's 52GB of NAND writes per day for five years with the 120GB 840 EVO). "This is a bullshit response and justification. Did Samsung feed you that? If it's rated for 1000 p/e cycles and it dies after I send it 740, then it didn't survive what it's rated and warrantied for, plain and simple. Samsung needs to strongly consider preemptively cutting checks to owners for at least 25% of the MSRP of the drive or just announce a straight up recall/replacement program and replace them with non-TLC drives.
LeftSide - Tuesday, April 14, 2015 - link
The old anandtech would have much set a much harsher tone for a manufacturer that pulled something like this.semo - Wednesday, April 15, 2015 - link
They were just as soft on OCZ with their Vertex 2 25nm "upgrade" and with Kingston's V300 bait and switch.For that and few other such examples I no longer frequent this site.
DanNeely - Tuesday, April 14, 2015 - link
Real world there's always write amplification; so some of the cycles the flash is rated for on paper will be consumed by the firmware not by user writes.Also real world is that the number of GB of writes a consumer SSD is rated for is a small fraction of the total number of rewrite cycles. That's partly to make sure that even worst case amplification doesn't cause a drive to fail while still under warranty; and so they can sell drives with much higher rated write amounts at enterprise markups.
Kristian Vättö - Tuesday, April 14, 2015 - link
Samsung doesn't have any official endurance spec for the 840 EVO -- the 1,000 figure is based on our own tests. It's fully warrantied for three years, meaning that if it dies during that time frame you are eligible for a new, working drive, even if the failure was due to exhausted NAND endurance.Maltz - Tuesday, April 14, 2015 - link
The cycle endurance rating NEVER refers to the user's activity - it measures the actual writes to the NAND. Whether you "send" it 740 cycles worth of data or 1000 is irrelevant. There is always going to be some write amplification just from wear leveling, if nothing else.Anyway, it's highly unlikely that a user that had heavy enough usage for this to be a real problem would go anywhere near a consumer SSD, much less a TLC drive.
Romberry - Tuesday, April 14, 2015 - link
You're in a tizzy over theoretical numbers. Why? Why does it matter whether or not it actually hits 1000 P/E cycles when the fact is that unless you plan to use the drive for a decade and just absolutely hammer it with writes the entire time, you'll never get there anyway? (And ya know, if you are a person that really plans to hammer a drive with writes -- maybe in a database scenario -- the odds are very good that you won't be using a last generation TLC device anyway.)leexgx - Tuesday, March 6, 2018 - link
Bit old but these drives have expected 1000 p/e, but reality it's more like 10x more then that (you can write over 1PB of data before it becomes an issue so at the time people are splitting hairs over it)None samsung drives can fail around 300-600gb of written data
JatkarP - Tuesday, April 14, 2015 - link
Samsung made trade off to bear those extra 260 P/E cycles to achieve performance by refreshing mechanism. This is not a perfect solution I would say, if some extensive user finishes his 760 cycles within 3 years then soon after 850 cycles product would fail (156 cycles have already completed by refresh). What samsung would say in this case ? Technically it has completed 1000 cycles so they might say this is not under warranty. But why would user want to trade off when he has paid full money. User wants both the things 1000 cycle endurance for HIS USAGE and performance mentioned in spec. Issue has been just patched but not resolved completely.MrSpadge - Tuesday, April 14, 2015 - link
Do any of you complaining read the previous answers in this thread? Everything you're asking here has been asked before and has been properly answered.leexgx - Friday, April 17, 2015 - link
properly notall drives perform a refresh to make sure data is still accessible, but i guess the trigger for refresh was set way way to low on the 840 and 840 evo forcing the drive to do muti reads and ECC
Hulk - Tuesday, April 14, 2015 - link
I don't own an 840 Evo but feel compelled to comment. It certainly is starting to look like Samsung sold a bill of good with the 840 Evo's that they can't deliver on.Also, Kristian I realize that you are trying to be impartial but it seems like you are defending Samsung. While the "letter of the warranty" doesn't say Samsung much do anything as long as their drive lasts 3 years that does not bode well for customer service.
I for one can tell you I won't touch a Samsung TLC drive with a 10 foot pole. In fact I'm going to stay away from all Samsung products in solidarity for the 840 Evo's that in my opinion are getting screwed here.
Romberry - Tuesday, April 14, 2015 - link
The 840 EVO is gonna in the end turn out to be a mistake I think. We had a fix, and now there's a fix that fixes what the first fix didn't fix. People in the know just aren't gonna use this particular series of drives. Maybe it took 3-D V-nand for TLC to really be ready for prime time?abhaxus - Tuesday, April 14, 2015 - link
I know that I personally will not be buying any more Samsung drives until the issue is conclusively fixed on not just the 840 EVO (of which I own a 500gb mSATA version) and the 850 EVO. I wouldn't mind getting a Pro series ifbit weren't for this debacle. I have also made sure friends in the market for drives are qware of how Samsung is mishandling this situation.kgh00007 - Tuesday, April 14, 2015 - link
The problem is that from the minute the performance optimization is completed, the performance is degrading until it is run again! Just like it is now after runinng disk fresh or whatever to restore performance.I just can't trust another Samsung drive right now. This issue has been on going for a very long time now, way too long to inspire any confidence.
The fact that the 840 is being ignored just makes this worse. And what about the Samsung PM851 OEM drive that come in laptops? That's TLC, will that get the fix?
MrSpadge - Tuesday, April 14, 2015 - link
Did you actually read the article? The new firmware is designed to detect which blocks need a refresh and to then refresh them during idle time, without user intervention.kgh00007 - Tuesday, April 14, 2015 - link
I did thanks! But regardless if the refresh was started manually or without user intervention, as soon as the refresh is complete the data on your drive begins to degrade. And quite quickly at that, within a month there is a noticeable drop in performance.I don't like that idea at all!
FlushedBubblyJock - Friday, April 17, 2015 - link
Yes, that's the question not answered, the real topic avoided in the article.There's no data on time/degradation.
So thank you kgh00007.
leexgx - Tuesday, March 6, 2018 - link
But the firmware would of refreshed them slow blocks before it became a noticeable problemthe issue before was that it was allowed to get way to low voltage to the point where the ecc retry got high enough to be noticeable and did not trigger a page refresh
there first attempt at fixing still had the TLC NAND page refresh voltage set to low so it still had high ECC read retry was still happening on old data (about 2 months old from the looks of it, as to why this bug was one of them harder to test for) , the last 2 firmware updates has resolved it permanently and performes a refresh when it's at low voltage when drive is idle or high read retry ecc event happens (witch likey won't happen due to background idle scan)
hammer256 - Tuesday, April 14, 2015 - link
So, I assume charge degradation would affect the drive in power off state too right? So if I for some reason left the drive off for 5 months (contrived example: pure archive drive), then I can't necessarily trust the data on the drive? That sounds slightly alarming.MrSpadge - Tuesday, April 14, 2015 - link
As far as the current reports go, the drive would become dog-slow but the data would still be intact, according to the data retention spec (don't know what it is for this drive).Solrax - Tuesday, April 14, 2015 - link
I have the exact same concern. I can drag out an old magnetic drive years after I've last used it and can read it fine. I recently booted up an old machine that was running Windows 98! I've never heard before that would be a concern with SSD, but it will apparently be a very real problem with the 840 EVO. I've already had one die exhibiting excessive ECC that sounds a lot like an extreme case of this same problem.I'm very concerned that my Surface Pro 3 has a Samsung SSD in it (reportedly of the 840 EVO technology). If I stop using it for a year it's drive may be effectively erased? Better get the recovery partitions off onto a flash drive...
Like others have said, this may be my last Samsung SSD. Too bad I didn't know about one being in the Surface. Nothing I can do about that (except at least they are apparently preparing the same firmware update for the Surface drive - they never released the first version for it).
Achaios - Tuesday, April 14, 2015 - link
Thank you for the update, Kristian. You are the best.KenPC - Tuesday, April 14, 2015 - link
I am amazed, dazed, and shocked by this article.Samsung launched TLC in a major way 2013..
By mid 2014, problems were well documented.
These are problems that made Samsung TLC drives definitely lower tier performance in the market and markedly different that the advertised and specified performance.
By late 2014, a first firmware 'fix' was found ineffective, but Samsung still promised to restore TLC performance to spec'd and advertised performance.
By Feb 2015, Samsung still promised to restore the drives to spec and their perceived market position in March, and evidence of slowdowns in the 840, 840evo and TLC based OEM drives were widely reported.
In mid April, (pardon the mixed metaphor), Samsung fed Anandtech the koolaid and said,, these are not the slowdowns you are looking for,, move along.......
So, to be quite clear, the slowdown disguise from Samsung ONLY works if you choose to install magician 4.x, and ONLY works if you have a consumer drive, and NOT if you have a Samsung OEM TLC based drive. (Oh and the extra bits written are also not of any concern, of course!)
So the 'fix' is to play musical chairs with the drive data every so often... ????
That's no fix,, restoring performance as advertised and spec'd is a fix!
gohanrocs12 - Wednesday, April 15, 2015 - link
did you fully read the article? There is a FW coming out as well, you won't need Magician if you update the FW since the FW will be doin everything for you automatically when you leave the system powered on and idle.KenPC - Thursday, April 16, 2015 - link
The middle chart for read speed clearly shows the FW update does not restore performance. It is not 'restored' until all the data is forcibly moved around.If the FW works, then why the need to include the full-drive data moving option at all?
And so far, the indications are that the updates are for the 840evo only and not the other Samsung TLC drives.
Kristian Vättö - Thursday, April 16, 2015 - link
Note that the drive wasn't given any idle time to fully restore the performance. With some idle time the performance should be fixed without the need to run the optimization tool. Its main purpose is for drives that have been used for archival purposes (although I wouldn't recommend an SSD, let alone a TLC drive for that) as the firmware's internal optimization process can only run if the drive is powered on.leexgx - Tuesday, March 6, 2018 - link
The ssd was not given any time to idle after he did the firmware update (read the article) so it been able to restore the ssd even to that point was very goodmore then likely just running the full ssd benchmark test after the firmware update most likely caused a force page refresh due to high ECC read retrys so the rest of the test was at expected speed (doing the adv performance optimisation should only need to be a one time event)
aggiechase37 - Tuesday, April 14, 2015 - link
Want to hear the real kick in the nuts? I have multiple drives in my video editing machine. The OS drive is an EVO, which the performance "optimization" tool updates fine. But for the source file drive, the tool won't fix it. I called in to Samsung, and I have to do this whirly bird routine of taking the drive completely out of the machine, putting it in my laptop, booting from some boot disk that I have to make, and then running the tool with some command line garbage. I've already been through that routine once. Anyone here think I want to go through that again? Me neither. Unless these drives are replaced with something that works as advertised, I'll never touch another Samsung product of any sorts so long as I live. This is pure bull.Laststop311 - Wednesday, April 15, 2015 - link
Well luckily there hasn't been any slowdown in the 850 evo drive. Since the cells are 40nm again I think the much larger cell size makes them a lot more robust, I believe the 840 evo used 20nm cells. Even so I still would not go for TLC nand. I have a 512GB 840 pro and the MLC it uses is flawless and it's speed is so good there is no point to upgrade until nvme and pci-e 3.0 x4 drives become commonplace. Tho I could use more space for more game installs as they have become so huge. May have to get a 1TB 850 pro to tide me over until 2TB drives are the same price as current 1TB drives and are 3.0 x4 interface and nvme.Jan-Erik - Saturday, June 18, 2016 - link
My 850 EVO have had a notisable performance loss after being unused for a while now. A round with diskfresh dragged the performance up from ~200MB/s to the normal 500+ MB/s. I'm surpriced nobody else is experiencing this problem...?I have the 840 EVO as well, and i do agree the performance degration is not even on the same page, but still a loss of over 50% performance on the 850 should be reason for Samsung to take action?
rexbass24 - Wednesday, April 15, 2015 - link
I am at the end of my rope with my brand new Samsung evo 640 500gb ssd. I get MAX 100MB/s on Novabench no matter what settings in samsung magician I use. I am using AHCI mode, not ide, so this should be blazing fast but since the day I got it, no love.Sabertooth r2.0
8gb ram
FX-4100
GTX n760 hawk
Newest stable bios on Sabertooth board and newest firmware on ssd.
My Adata 900 series 128gb was 3x as fast as this drive
rexbass24 - Wednesday, April 15, 2015 - link
oops, I meant evo 840 hahaBlobTheCop - Wednesday, April 15, 2015 - link
Now seriously, I have about 300 of these drives deployed in my users desktops.How do I deploy this fix? Where is the command line tool?
Coup27 - Wednesday, April 15, 2015 - link
Pretty sure there isn't one, after all these are consumer level drives.leexgx - Friday, April 17, 2015 - link
walk to each system and press a button reboot job done :) (just take you a week to do it)Penti - Friday, April 17, 2015 - link
Corporate systems might have Intel AMT available and with that you could do all the flashing remotely using the DOS tool for example. If you got consumer stuff just walk around starting the update. Pretty sure Magician doesn't allow automation / unattended firmware flashing. The server tool doesn't work with Evo's.Pajos - Thursday, April 16, 2015 - link
please, how days must no function disk, for file degraded?I thing buy 840 EVO and i dont now is it good choice, price 40 USD
chucky2 - Sunday, April 19, 2015 - link
So how does this affect power consumption? 840 EVO was a good drive for lower power usage in a notebook (at least from what I remember), now with this new firmware, does it remain so?BenjiSt - Sunday, April 19, 2015 - link
Correct me if I'm wrong: All these tweaks to keep the aging cells of your drive alive seem to depend on that you actually use your drive regularly. But that doesn't help if you leave your drive with your precious data without power for a couple of months, does it?That's actually my usage pattern, and I'm in the market for a new notebook and new SSD as well. I use my work notebook for private stuff, then boot my private PC very rarely (say 1-2 a year) to move private stuff from the work notebook to the private notebook.
So, should I fear data loss if I use a TLC drive?
andrewbaggins - Sunday, April 19, 2015 - link
maybe go with intel, sandisk, ore crucial .....leexgx - Tuesday, March 6, 2018 - link
All ssds have this issue just samsung setting for refresh was set to lowiceblitzed - Sunday, April 19, 2015 - link
Does this bug affect random read or just sequential read (or both)? Im guessing both.BenjiSt - Sunday, April 19, 2015 - link
It seems my comment got removed somehow. I was just asking if the firmware tweak helps if you are not using your drive for a couple of months. That's my (maybe not so common) usage pattern.So:
1) How reliable are the EVO TLC drives when unpowered for a couple of months?
2) What is the expected difference (in say months not powered on) between the EVO and Pro SSDs?
Lothsahn - Monday, April 27, 2015 - link
apparently this new firmware doesn't work properly with fstrim. See here:http://www.overclock.net/t/1507897/samsung-840-evo...
You should really update this to warn your readers. I'd also like a statement from Samsung as to whether they plan to address this.
fiveandhalf - Wednesday, April 29, 2015 - link
Hi:Anyone experience 840 Pro firmware issue under raid controller?
we have 6 x 840 Pro 512GB running under Dell PowerEdge 720 with H710 controller as raid1, scores around 400MB/s sequential 4K write constantly over past year until last month, we upgraded 2 x SSD firmware from DXM05B0Q to DXM06B0Q and the sequential 4K write speed dropped dramatically from 400MB/s to fluctuation of 20-200MB/s in the same hardware.
We thought it might be our hardware, so we moved the two different firmware version SSD to another PowerEdge, (as 2 x raid0), as yield same result. We have been talking to Samsung helpdesk who has been nice but unhelpful, It seems to me this is a firmware bug.
BTW, if we plug the old and new version of SSD back to a desktop with a SATA3 connector, re-do the 4K write test, it is constantly lower on new firmware but not much, something like 390MB/s vs 420MB/s
Here is the command we use to test with a 20GB test file, we even re-format the drive and tested as empty drive, still same result
Sqlio.exe -kW -t4 -s30 -dS -o128 -fsequential -b4 -BH -LS testfile.dat
Hope anandtech.com lab can reproduce this issue, and if it is a bug, we can get Samsung to do something about it.