Review Index:
Feedback

Samsung 840 EVO Performance Restoring Firmware Only Partially Effective

Subject: Storage
Manufacturer: Samsung

Introduction

Well here we are again with this Samsung 840 EVO slow down issue cropping up here, there, and everywhere. The story for this one is so long and convoluted that I’m just going to kick this piece off with a walk through of what was happening with this particular SSD, and what was attempted so far to fix it:

View Full Size

The Samsung 840 EVO is a consumer-focused TLC SSD. Normally TLC SSDs suffer from reduced write speeds when compared to their MLC counterparts, as writing operations take longer for TLC than for MLC (SLC is even faster). Samsung introduced a novel way of speeding things up with their TurboWrite caching method, which adds a fast SLC buffer alongside the slower flash. This buffer is several GB in size, and helps the 840 EVO maintain fast write speeds in most typical usage scenarios, but the issue with the 840 EVO is not its write speed – the problem is read speed. Initial reviews did not catch this issue as it only impacted data that had been stagnant for a period of roughly 6-8 weeks. As files aged their read speeds were reduced, starting from the speedy (and expected) 500 MB/sec and ultimately reaching a worst case speed of 50-100 MB/sec:

View Full Size

There were other variables that impacted the end result, which further complicated the flurry of reports coming in from seemingly everywhere. The slow speeds turned out to be the result of the SSD controller working extra hard to apply error correction to the data coming in from flash that was (reportedly) miscalibrated at the factory. This miscalibration caused the EVO to incorrectly adapt to cell voltage drifts over time (an effect that occurs in all flash-based storage – TLC being the most sensitive). Ambient temperature could even impact the slower read speeds as the controller was working outside of its expected load envelope and thermally throttled itself when faced with bulk amounts of error correction.

View Full Size

An example of file read speed slowing relative to age, thanks to a tool developed by Techie007.

Once the community reached sufficient critical mass to get Samsung’s attention, they issued a few statements and ultimately pushed out a combination firmware and tool to fix EVO’s that were seeing this issue. The 840 EVO Performance Restoration Tool was released just under two months after the original thread on the Overclock.net forums was started. Despite a quick update a few weeks later, that was not a bad turnaround considering Intel took three months to correct a firmware issue of one of their own early SSDs. While the Intel patch restored full performance to their X25-M, the Samsung update does not appear to be faring so well now that users have logged a few additional months after applying their fix.

Continue reading our look at the continued problems with the Samsung 840 EVO SSD!


January 28, 2015 | 10:04 PM - Posted by Anonymous (not verified)

So basically it looks like drives in cold storage a cell drifting causing a reduction in read speeds, but hot drives are refreshing their data in the background and thus retaining their speed, unless...

My guess is that the reason some of those hot drives are slowing down is probably due to cell damage that is either caused by the refreshing process itself or was caused by excessive writes prior to the performance restoration update.

Allyn, it might be worth asking some of those forum members for their SMART attributes, particularly relating to the 'total bytes written' metric. My guess is that you'll find a correlation between a high number of program/erase cycles and the performance degradation they're seeing even after the performance restoration update.

As far as the speculative internal refreshing process, if true, it probably means that EVO NAND is being degraded at a rapid pace on hot drives. And if true, Samsung must know this and they did it to shut everyone up with the hope that warranties would expire before the NAND does. If true, this is some shady bullshit. I've got a 1TB EVO and now i'm waiting for it to die.

Good work sir! Keep it up please.

January 29, 2015 | 10:14 AM - Posted by SSBrain (not verified)

On my non-EVO Samsung 840 250GB I found a correlation between temperature and speed with old data. Basically if I increase it without making it reach the throttling threshold, performance improves. I've made a gif animation showing this:

http://i.imgur.com/5NjyAHS.gif

It would be useful if Samsung 840 EVO owners could verify whether the same behavior occurs on their SSD.

January 28, 2015 | 11:18 PM - Posted by P0ci (not verified)

Maybe a new firmware will fix it?

April 14, 2015 | 05:58 PM - Posted by Anonymous (not verified)

no firmware can fix hw issue with design flaw

January 28, 2015 | 11:55 PM - Posted by Rick (not verified)

I have two 250gb evos in raid0 in my pc. I booted to bios and disabled the Intel raid and ran the utility on both drives from a bootable ISO. Once this was done I was able to turn the raid back on and everything came up normal. I did this about a week after the fix came out and I use my system almost everyday.

When I saw this story I tested my system and everything came back normal.

For the drives that are affected, can they run the fix again to restore performance?

January 29, 2015 | 12:24 AM - Posted by Allyn Malventano

The tool will only run in conjunction with the firmware update, and won't run on drives that have already been updated, so no.

January 29, 2015 | 12:28 AM - Posted by MRFS (not verified)

Many thanks, Allyn.

You are THE BEST!

MRFS

January 29, 2015 | 02:21 AM - Posted by Topinio

Interesting, looks to me there might be no way to have a firmware rev that fixes EVOs so that they can safely be put into in storage.

FWIW, my solitary data point is that the drive's still good: 250GB 840 EVO, 75 GB used, POH 1771, 1.4 TB written (which seems far too high).

January 29, 2015 | 02:31 AM - Posted by pdjblum

Stay away from cheap-ass nand: TLC, that is.

January 29, 2015 | 03:30 AM - Posted by Anonymous (not verified)

Hi

I was going to buy it (the 120gb) next week,
but after I saw this issue I'm not sure.

so now which one should I get ?

Samsung 840 EVO 120GB TLC $105

Intel 335 Series 80GB MLC $88

thanks

January 29, 2015 | 07:52 AM - Posted by Jabbadap (not verified)

Have to be between those?

Crucial bx100 or mx100, they are both mlc and cost about $70 for 120GB/128GB(newegg).

January 29, 2015 | 11:58 AM - Posted by Anonymous (not verified)

Thanks
but there is only mx100 and it cost $122 in my country.

January 29, 2015 | 12:04 PM - Posted by Anonymous (not verified)

and no I can get another ssd,
but I thought samsung 840 evo was the best performance/value ssd in my country.

anyway here is a website from my country with a list of ssd:

http://www.egprices.com/en/category/computers/hard-disks/ssd-hard-disks

thanks again

January 29, 2015 | 05:36 AM - Posted by JohnGR

UltimateDefrag was a nice program for defragmentation when HDDs where still the only option. The program comes with an option to move all data to the outer region of a HDD or to the inner region. Well SSDs are not like HDDs to have outer and inner regions, but I bet in the case of SSDs with less that 50% usage, that translates to move all the data to different cells. That could help with performance EVO owners. Probably Samsung created the first SSD that needs defragmentation.

EDIT
After using UltimageDefrag in my own 128GB 840 EVO, HDTach gives me 240MB/sec and SSDReadSpeedTester 380MB/sec(AMD AM3+ platform). HDTach scores where less than half before running defrag program. Haven't run SSDReadSpeedTester before doing the defrag i am afraid.
I have NOT updated to the newer firmware and I have NOT used the Restoration Tool. The disk was formated 6-8 months ago.

January 30, 2015 | 12:56 AM - Posted by Techie007 (not verified)

MyDefrag is a much better defragmenter from a functional standpoint. For the purpose of refreshing the whole drive, run its System Disk Monthly script (for C:\ drive) or Data Disk Monthly script (for storage drives) once. MyDefrag's "monthly" scripts are very thorough defragmentation scripts that should reshuffle pretty much all the files on the disk any time you run them, thus rewriting everything.
You can download MyDefrag from here: http://mydefrag.com/

January 29, 2015 | 06:39 AM - Posted by Anonymous (not verified)

I have looked around a bit to find the ssdreadspeedtester, but only found deal links.

anyone have a good link?

January 29, 2015 | 11:47 AM - Posted by Anonymous (not verified)

thank you

January 29, 2015 | 06:45 AM - Posted by Anonymous (not verified)

My 250gb Evo is used daily, and sometimes left on for an hour or two while I'm away from my PC.

Since the "fix", I`ve secure erased and restored at least twice. Each time, performance stays good for about 4 or 5 weeks, then starts to deteriorate.

January 29, 2015 | 08:04 AM - Posted by Laststop (not verified)

Man I am so glad I went with a 840 pro over an 840 evo. I was so close to just saving the money but i had slight doubts about TLC nand. Always trust your gut fellas.

You would have to be crazy to get an 850 evo with TLC v nand. 2 new technologies mashed together ur only asking for trouble there. I wish samsung would make a 256GB SLC v nand drive. It should be the same price as a 512GB MLC drive and thats only 330 dollars I wouldn't mind paying 330 for a 256GB SLC V Nand drive. But it would have to be sata express based so you can actually take advantage of the increased speed of SLC

January 29, 2015 | 09:03 AM - Posted by BBMan (not verified)

I'm glad I resisted the cheap-side as well- though I opted for the 850 Pro. But I do have to wonder if this is uniquely a Samsung problem. Aren't there others who use this process?

January 29, 2015 | 11:50 AM - Posted by Patrik (not verified)

Apple used TLC in there new Iphone 5. They suffered the same problem, slowdowns until the phone was so slow to use it rebooted spontaneously. Apple abandoned TLC after that and went MLC only but apparently the users with TLC based Iphones where left with some "patch" probably similar to what samsung tried to do.

Considering the fact that the 840, 840 EVO, 840 EVO mSATA and all samsung OEM TLC SSD's are effected, enterprise versions not confirmed yet and Apple having the same problem that only leaves SanDIsk Ultra 2 and the new 850 EVO as unknowns. Other then that the TLC track record is not very good, pretty bad I would say.

TLC might work but it seems problematic on small nodes with out more testing and research.

TLC is more and more common in SD cards and USB sticks so theres a risk of slow read speed from them to if engineers have missed something with TLC.

January 29, 2015 | 10:43 AM - Posted by razor512

When will people learn to avoid TLC flash? Less reliable, and more expensive than MLC when you take the price and write endurance into account.

January 31, 2015 | 10:06 PM - Posted by Anonymous (not verified)

We saw it was Samsung,so...

January 29, 2015 | 11:06 AM - Posted by trparky

If this is indeed a hardware design flaw that can't be fixed then Samsung should be forced to recall every one of these SSDs and replace them with unaffected hardware.

January 29, 2015 | 11:16 AM - Posted by SSBrain (not verified)

My thinking is that if this could be easily fixed it with a firmware update, Samsung would have already done so. It could be something as bad as their 21/19nm TLC NAND requiring higher operating temperatures than what most (but not all) users are used to, for being reliably programmed. This would imply, were a fix be implemented through a firmware update, either slower write speeds, higher average power consumption or both.

January 29, 2015 | 11:34 AM - Posted by Allyn Malventano

I'm pretty sure this has nothing to do with temps, but it may be an issue of the original patch not working with some early production runs. Nothing solid there yet though, just a theory.

January 29, 2015 | 12:03 PM - Posted by SSBrain (not verified)

What I have observed on my drive (an unfixed, 1st gen TLC Samsung 840 250GB) is that the higher the temperature gets, the faster read speed on the data affected by the issue gets. This is a repeatable result and makes me think that there is a firmware problem with the temperature-based voltage drift correction.

I've made a gallery of screenshots from SSD Read Speed Tester showing the correlation: http://imgur.com/a/rUGoI

I also made additional test with a program called HDDScan, which performs a sequential read along the entire LBA range. I've made this with a 8192 Kb block size. Again, performance overall appears to improve with drive temperature:

~32°C:
http://i.imgur.com/7zCp1jo.png

42 (start) - 49 (end) °C:
http://i.imgur.com/7zCp1jo.png

The question is: why does this happen? My hypothesis is that as the drive is clearly having problems applying the correct temperature adjustments when reading data, the same could be happening when it's writing data.

The cooler a NAND memory cell, the longer it takes to program.

So, what if due to the same bug affecting read speeds the drive won't take correctly into account drive temperature when writing? Then, it could be that over time such data will develop read errors earlier. Could it be that this happens more when new data is written at low temperatures? Could this explain why some users don't seem to be experiencing it? Perhaps they are running their SSDs hotter in general.

January 29, 2015 | 10:05 PM - Posted by davepk (not verified)

I wonder if people who leave their PCs on 24/7 and thus achieve some thermal equilibrium for the SSDs exhibit this problem to a lesser degree than those who power their computers only periodically.

January 29, 2015 | 10:31 PM - Posted by SSBrain (not verified)

The PC with the Samsung 840 SSD I have that is showing this problem is left on 24/7, so it might not be correlated with that.

February 4, 2015 | 09:34 PM - Posted by Anonymous (not verified)

I purchased 2 new 500G drives a week before this was first reported. Both drives are suffering slowdowns after the patch was installed, so I'm not sold on the idea of early batches being the issue.

January 29, 2015 | 02:13 PM - Posted by mangel (not verified)

Has anyone had similar problems with the Samsung XP941?
I have not benchmarked mine but it sure seems to slow after using programs with large databases like Forte Agent.

January 29, 2015 | 04:51 PM - Posted by Jon (not verified)

Can anyone speculate as to whether this will effect the 850 Evo? I just purchased one. Now I'm thinking I should have just gone up to the Pro to not have any possibility of this same problem cropping up. It uses TLC as well, right?

January 30, 2015 | 01:01 AM - Posted by Techie007 (not verified)

I'd love to find out, and you could help with this process. Every month or so, you could run my SSD Read Speed Tester, and see if the graph starts to show declining read speeds with increasing file age. If it does, you could share your findings at the link below.

You can download SSD Read Speed Tester from here: http://www.overclock.net/t/1512915/

January 29, 2015 | 11:13 PM - Posted by Ramon (not verified)

So in the end, the only manufacturers that did not have a horror story has been Crucial and Intel. I'm so glad I bought 2 Crucial M4! Still rocking at full speed in RAID0 for the last 3 years now! :) Not a SINLGE issue!

January 30, 2015 | 04:13 AM - Posted by Anonymous (not verified)

Actually... I used to buy lots of the Crucial M4 drives and yes they did have a big issue. A few firmware versions in someone writing their firmware didn't think to account for SMART data properly. Once the drive reached right over 5000 power on hours they would not become bootable into Windows and had a host of other issues making them appear to be dead. A firmware update fixed the issue and no data was lost. A customer of mine had this issue.

Still I'm currently buying MX100 drives instead of the 840 EVO due to price and until this 840 EVO deal blows over.

January 30, 2015 | 11:51 AM - Posted by Dissapointed=( (not verified)

I got 2x250GB in raid0 and HDTach shows 1.1GB/s mostly, but also 9 short sudden drops to 700-800MB/s. Updated the drives couple weeks ago.

January 31, 2015 | 05:29 AM - Posted by Engell (not verified)

yeah mine still have the problem aswell
http://i.imgur.com/V6K8UxF.png

Main and only drive in my laptop , used constantly and patched with samsungs "fix" tool.

January 31, 2015 | 09:30 AM - Posted by jtiger102

I have a 250GB 840 EVO that was purchased in Feb 2014. I updated the firmware and used the Restoration Tool.

I checked this morning to see if my drive was affected.
It was running slower similar to what users are now seeing.

I ran DiskFresh on the drive and performance is now back up to what it should be.

http://imgur.com/a/VsRbj

February 1, 2015 | 01:50 PM - Posted by Smash (not verified)

Allyn, you have EVOs that are both affected and unaffected by this problem. Can you identify their manufacture dates/batches etc to see if maybe somewhere there is a correlation? Thanks.

February 1, 2015 | 02:15 PM - Posted by Smash (not verified)

I also noticed that the ppl with older 840 evos seem to read above 500MB/s, while the newer ones seem to be reading about 100MB/s avg slower. Again goes to manufacturing methinks...

February 1, 2015 | 02:38 PM - Posted by hechacker1 (not verified)

Has there been any accounting for over provisioning and space used? I imagine it could affect any background or trim operations. The samsung magician makes it easy to OP, and I think some users would take advantage of that. I certainly do because I don't keep a lot on my OS drive.

February 1, 2015 | 02:49 PM - Posted by ciddotech

I saw this post and decided to reuse the SSD speed tester you guys posted on last post on performance degradation. My speeds are a little slower than at the beginning. I got an average of about 375 MB/s read speed on my 2.5" 250GB 840 Evo. However, I wasn't able to use my computer as much as I'd have liked during that time because life happens.

Maybe the new firmware that Samsung created just learns and adjusts to the voltage drifts and stuff. If everyone actually used their drive during the period of the firmware release, the firmware has been learning and adapting while those who haven't been using their drives haven't given the firmware a chance to monitor the shift in voltages.

So perhaps the more you use your drive in between voltage drifts = the less degradation you'll face since the firmware will be able to adapt more frequently? Just my assumption.

February 2, 2015 | 01:23 PM - Posted by trparky

Nope, I'm actively using my SSD as my system drive and I still have the slowdowns.

February 2, 2015 | 07:33 PM - Posted by Anonymous (not verified)

I put one of these in my laptop and have had significantly noticeable IO pauses when accessing disk. I migrated my Linux install to a different SSD and the pauses have disappeared.

I did boot into Windows 8 to do this firmware patch when it was announced.

hdparm -i /dev/sda

/dev/sda:

Model=Samsung SSD 840 EVO 500GB, FwRev=EXT0CB6Q, SerialNo=S1DHNSAF620781D
Config={ Fixed }
RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=0
BuffType=unknown, BuffSize=unknown, MaxMultSect=1, MultSect=1
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=976773168
IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 udma2 udma3 udma4 udma5 *udma6
AdvancedPM=no WriteCache=enabled
Drive conforms to: unknown: ATA/ATAPI-2,3,4,5,6,7

February 3, 2015 | 08:42 AM - Posted by Anonymous (not verified)

The new firmware is not effective.

http://imgur.com/b1qg2xX,XHGgwtU#1 shows massive degradation for a Samsung 840 EVO mSATA 1TB.

The SSD has the data as shown at http://imgur.com/b1qg2xX,XHGgwtU#0 - confirming that I did have the latest firmware installed

February 3, 2015 | 02:14 PM - Posted by Allyn Malventano

Thanks for the data point. One of our 'good' results was also a 1TB 840 EVO MSATA, so it appears those are hit or miss on good results, just like the other models.

February 4, 2015 | 03:48 AM - Posted by Anonymous (not verified)

I am glad to support wherever I can.

The SSD was purchased in Europe in November 2014 and almost immediately upgraded to the new firmware.

Yesterday I ran "DiskFresh" on the raw disk. This took ages to complete (1 TB) and appeared to stall every once in a while. Presumably, read times were just through the roof.

I still need to measure the performance with "SSD Read Speed Tester" to investigate the validity of employing DiskFresh as a work-around. Will do eventually.

Personally (and assuming that it works), I am hesitant to use DiskFresh on a schedule:
a) it really takes a lot of time to execute
b) at worst, each run will add 1 TB to "Total Bytes Written" of the SSD and thus consume lifetime

[I am aware of the fact that Samsung have an unconditional three year warranty on that SSD, with no upper limit on the total bytes written, but I'd rather make use of that investment for longer than three years...]

February 6, 2015 | 01:54 AM - Posted by Anonymous (not verified)

... and http://imgur.com/E5vfJcy shows the results of SSD Read Speed Tester after having run DiskFresh (http://www.puransoftware.com/DiskFresh.html) on the raw disk.

Conclusions:
a) my Samsung 840 EVO mSATA 1 TB has a problem
b) running DiskFresh helps to mitigate that problem

These performance numbers were not gathered following a rigorous scientific approach (laziness in collection, working in the background while speed testing etc), but, structurally, the difference appears so utterly significant that I think I can be forgiven for the lack in rigor.

February 4, 2015 | 02:56 PM - Posted by Alex1002222222222222222222222 (not verified)

I own over 25+ Samsung SSDS. I dont have time to play with firmware, etc.I am very dissapointed with SAmmyyy. In the future I will buy another brand.

February 27, 2015 | 01:19 PM - Posted by Armien Cassiem (not verified)

I agree.
My one keeps crashing, BSOD, even after I've connected it to another PC.

February 7, 2015 | 07:37 PM - Posted by Yeltnerb (not verified)

This is precisely why I purchased an 850 EVO.

February 13, 2015 | 08:56 AM - Posted by Anonymous (not verified)

That also uses some "clever" NAND memory.

I'd avoid Samsung EVO SSDs until it's clear that there is no problem. It's early days for the 850's.

February 8, 2015 | 03:08 PM - Posted by nathan118 (not verified)

Any info about the 850 EVO? Does it suffer same problems?

February 17, 2015 | 08:20 PM - Posted by Anonymous (not verified)

Same issue here, ran the recovery tool on my 120GB 840 evo on 15th October 2014 and read speeds improved to 350MB/s across the whole drive from below 50MB/s in places.

I ran HD Tach again the other day and reads have dropped below 100MB/s and in some cases below 50MB/s across 50% of the drive....

The 120GB is especially poor in performance I reckon, laptop feels like it has a HDD with how long it takes to boot from shutdown. Booting from hibernate is a lot quicker, probably due to the hibernate file having been recently written upon hibernate!

February 18, 2015 | 01:56 AM - Posted by Matt Lathrum (not verified)

I got my 840 EVO with the new firmware from the factory back in early December. It's my C drive. I just ran HD Tach and I'm seeing some worrying results:

http://imgur.com/a/ao9Qq

Still no news on a response from Samsung?

February 18, 2015 | 04:39 PM - Posted by Matt Lathrum (not verified)

So, I ran DiskFresh and recovered my performance:
http://i.imgur.com/LBkLAUZ.png

So there's that at least, until Samsung has a response.

February 20, 2015 | 10:30 AM - Posted by Anonymous (not verified)

Just had this reply from Samsung (after being told previously.... "According to your screenshots, the performance of the SSD seems quite normal.")

"Please send us a screen shot from Magician tool with the performance history.

Only like this we can check the real performance of the SSD.

Thank you."

Looks like they're going into denial mode.

February 27, 2015 | 01:08 PM - Posted by Armien Cassiem (not verified)

Yep, they're in denial.
I should have bought an Intel.

February 23, 2015 | 10:29 AM - Posted by Michael Goldshteyn (not verified)

Not sure if anyone suggested this, but simply defragmenting your drive may restore its performance back to the peak. I know this is not advised for SSDs, but in this case it may be a good interim solution that will rewrite many blocks restoring them to peak performance. This will work until Samsung issues a fixed firmware that hopefully resolves this issue once and for all.

February 27, 2015 | 01:06 PM - Posted by Armien Cassiem (not verified)

Bought this drive after much hype, 1 month after purchase the drive just started giving BSOD and auto restart, subsequently chkdsk reported drive needed to be checked for "consistency".
Chkdsk kept on finding corrupt data on the drive, this would happen all the time to the point where my Windows 7 would be corrupt after this happened for a few weeks. I was left with having to restore my Windows via an uncorrupt image.
I eventually got fedup, returned the drive to the store and eventually got a replacement drive.
Lets fast forward 1 month, same problem, PC would just freeze, followed by a BSOD, restart and chkdsk reports the drive needs to be checked for consistency.
Samsung ZA seems oblivious that the drive is problematic.
All they're offering is I return it once again to the store and once again go through the whole process.
I'm pretty much fedup. Seems Samsung simply overhypes their products. I've learnt my lesson.
Samsung, never again.
So glad I have an iPhone and iPad, and not Samsung equivalents.

March 25, 2015 | 08:10 AM - Posted by Wrend (not verified)

I have 2 840 EVOs, a 500GB one in my wife's laptop and a 1TB one in my main home PC. I've had them in both computers for over a year and I never get BSOD other than from when I have occasionally done overclock limit testing. They're both the main OS drives using Windows 7 in these computers.

My wife's laptop is used daily and is shutdown, suspended, and restarted on a regular basis. My home PC is up and running crunching for scientific and humanitarian research and hosting three Minecraft servers running in a Debian virtual machine on a RAM drive within Windows nearly 24/7/365.242 and is only restarted once every few weeks or so for software updates. I also simultaneously use this computer as my main home entertainment system, game on it, and do everything else most people use computers for.

Based on my experience with these drives, I can assure you that they're very stable and have no problem holding onto bits of data.

I think there is likely something wrong with your drive image, drivers, software you're using, or similar. The issues you're having also have nothing to do with the issues reported in this article.

Based on your mobile device preferences, it would also seem that you have an irrational bias against all things Samsung. I also don't use their mobile devices, because I prefer rooted vanilla Android Nexus devices, but that likewise has very little to do with the SSDs I use.

March 13, 2015 | 10:01 PM - Posted by Anonymous (not verified)

Bought an 840 EVO last summer, and just turned the computer on for the first time in several months (traveling). Performance is around 1MB/s; much worse than earlier reported. Running the performance restoration tool now hoping the issues are really fixed.

March 18, 2015 | 03:25 PM - Posted by 840 devo (not verified)

I own 4 840 EVOs - 3x1TB and 1x500GB - 2x1TB in a Mac Mini Server, 1x1TB in a Macbook Pro (Core 2 Duo), and 1x500GB in a W7 PC. (I also have a 250GB non-EVO 840 in the PC).

Total outlay = $1575.08 (from Amazon whenever they had a deal)

I couldn't get the PRT to work on the Mac, so I did the following to each drive:

copied all the data off of each drive to a spinner
reformatted each drive to NTFS on Windows
ran the Windows PRT v1.0 on each drive
reformatted each drive to Mac OS
copied the original data back onto the drive.

But now I'm reading that the first PRT does not guarantee success - and furthermore, since I already used v1.0, v1.1 will not work.

Samsung's page says - "you don't need it" and "it will not work" - those are different things.

Shouldn't Samsung treat this like a defective part, and issue equivalent replacements to those affected? (ie - 840 EVO owners should be allowed to trade in their drives for 850s or 850 EVOs). So I guess I will just wait like everyone else at this point until the class action suit happens.

March 22, 2015 | 06:33 PM - Posted by Anonymous (not verified)

Samsung is the most dishonest company in the world. So don't count on them being honest in this case.

http://www.vanityfair.com/news/business/2014/06/apple-samsung-smartphone...

April 6, 2015 | 04:08 AM - Posted by Anonymous (not verified)

Who would quote a tech analysis by vanity fair? Actually there were no analysis in the piece, resembled more of a short novel.

March 30, 2015 | 02:59 AM - Posted by Anonymous (not verified)

I have an EVO SSD since december 2014. Yesterday I run a test tool called SSDReadSpeedTester, and the read speed of files on my C: drive was between 200 an 350 MBps, much lower than specs. Then I run a disk refresh tool called DiskFresh, and now the read speed of files is between 480 and 550 MBps, as it should be.
Waiting for a new (second) firmware update by Samsung, you can test and restore the original speed of EVO SSD's with the two tools above mentioned.
My suggestion is to do that at 3-4 months intervals.
Regards

April 5, 2015 | 09:50 PM - Posted by Anonymous (not verified)

Hello! Although this Performance fix utility is only meant for Read, I tried it anway. My problem with my 840 Evo 250GB was that write speed was below 300MB/s (usually about 220MB/s or so), however read speeds were quite fine (averaga of about 540MB/s). I ran the tool and after restarts my write speed was up again to 540MB/s. Nice!

May 1, 2015 | 01:14 AM - Posted by Carlo (not verified)

Any experience with the new firmware EXT0DB6Q and Samsung Magician v.4.6 ?
Thanks

June 29, 2015 | 04:11 PM - Posted by Rodrigo (not verified)

I have a 840 EVO with serious degradation. The Performance Restorer from Samsung did not work. It keeps telling me to update a driver. But following their manual, the procedures were done accordingly.

Shitty drive.

June 1, 2016 | 04:59 AM - Posted by Vanex (not verified)

840 evo EXT0DB6Q after 8.39 TB written, 5196 power-on hours, 999 power-on count

http://imgur.com/NZ72heG

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Lines and paragraphs break automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd> <blockquote><p><br>
  • Web page addresses and e-mail addresses turn into links automatically.

More information about formatting options

By submitting this form, you accept the Mollom privacy policy.