During the last couple of weeks, numerous reports of Samsung SSD 840 and 840 EVO having low read performance have surfaced around the Internet. The most extensive one is probably a forum thread over at Overclock.net, which was started about month ago and currently has over 600 replies. For those who are not aware of the issue, there is a bug in the 840 EVO that causes the read performance of old blocks of data to drop dramatically like the HD Tach graph below illustrates. The odd part is that the bug only seems to affect LBAs that have old data (>1 month) associated with them because freshly written data will read at full speed, which also explains why the issue was not discovered until now. 

Source: @p_combe

I just got off the phone with Samsung and the good news is that they are aware of the problem and have presumably found the source of it. The engineers are now working on an updated firmware to fix the bug and as soon as the fix has been validated, the new firmware will be distributed to end-users. Unfortunately there is no ETA for the fix, but obviously it is in Samsung's best interest to provide it as soon as possible.

Update 9/27: Samsung just shed some light on the timeline and the fixed firmware is scheduled to be released to the public on October 15th.

I do not have any further details about the nature of the bug at this point, but we will be getting more details early next week, so stay tuned. It is a good sign that Samsung acknowledges the bug and that a fix is in the works, but for now I would advise against buying the 840 EVO until there is a resolution for the issue.

POST A COMMENT

116 Comments

View All Comments

  • FunBunny2 - Saturday, September 20, 2014 - link

    Refresh my memory? I recall that controllers *routinely* moved all data on the NAND as part of wear leveling, including unmodified blocks. When did controllers (or, only Samsung's) stop doing this? Reply
  • cooldadd - Saturday, September 20, 2014 - link

    Interesting...I wonder if a full backup and restore of the drive would (temporarily) restore its full speed. (Not a practical solution, just wondering...) Reply
  • Musafir_86 - Saturday, September 20, 2014 - link

    -Excuse me, but aren't ALL SSD (and NAND flash storage) have wear-leveling algorithm that supposed to move around (shuffle) the data internally? So those old data never get shuffelled at all?

    -What about data retention? What'll happen to the data in a drive that has been unplugged for months/years?

    -And how about the earlier 840 (non-EVO, non-Pro) which use 21nm TLC NAND (versus 19nm TLC NAND in 840 EVO)? Does it has the same issue too? I helped a relative upgraded his ProBook 6555b to a 120GB 840. :(

    Regards.
    Reply
  • hojnikb - Sunday, September 21, 2014 - link

    840basic is also reported to have this issues. Reply
  • GTVic - Monday, September 22, 2014 - link

    OLD data wouldn't get shuffled, that would make the drive wear out faster. Leveling means that when it is time to write NEW data that it is placed in an area that has the fewest previous writes. Reply
  • Solid State Brain - Wednesday, September 24, 2014 - link

    Yes, it looks as if old data never gets shuffled, which means it's likely to be a issue in the firmware's programming logic rather than hardware problems. Wear leveling is indeed supposed to shuffle all the data, both old (static) and new, so there isn't supposed to be "old data" inside the SSD in need of refreshing (reprogramming) in the first place with normal usage patterns.

    The data retention time is supposed to be at least one year for cells at the end of their rated life (1000 program/erase cycles for Samsung TLC NAND).

    Earlier 840 seem to be affected as well (mine is, at least) by this issue.
    I wonder if this actually started showing with one of the latest firmwares. It seems strange to me that people started experiencing it only relatively recently.
    Reply
  • kgh00007 - Saturday, September 20, 2014 - link

    I just checked my 120GB 840 Evo in my laptop and the seq read is below 50MB/s across half of the drive and dropping to single figures. This thing has performed terribly since a few weeks after I installed the OS, my laptop sometimes takes longer to start up then it did with the HD. I know it's only a 120GB, but I hope a firmware update can sort this!

    I actually tried to return it after about a month, but the return shipping and restocking fees made it not worth while and they would not accept that it was underperforming for an RMA.
    Reply
  • poohbear - Monday, September 22, 2014 - link

    Oh my, i just bought one these as my sys drive last week!!! Thank goodness a new firmware is coming out, i just purchased it based on all the good reviews....didnt realize there was a problem past the one month mark!! Good thing Samsung is releasing a fix soon... Reply
  • chas_martel - Monday, September 22, 2014 - link

    When did they release info about an updated firmware? I've seen/heard nothing. Reply
  • Gothmoth - Wednesday, September 24, 2014 - link

    it´s written in this anandtech article:

    I just got off the phone with Samsung and the good news is that they are aware of the problem and have presumably found the source of it. The engineers are now working on an updated firmware to fix the bug and as soon as the fix has been validated, the new firmware will be distributed to end-users. Unfortunately there is no ETA for the fix, but obviously it is in Samsung's best interest to provide it as soon as possible.
    Reply

Log in

Don't have an account? Sign up now