Before, After, and Optimized Results
I tested multiple samples with the following procedure, but I will single out single units to show the various peformance differences. I'll start with a 120GB 840 EVO:
Before update:
Our test sample had slowed even further since the last time we tested it – nearly to the speeds we saw before performing Samsung's first attempted fix.
After update:
The above test was run immediately after the firmware update, giving the 840 EVO no time to refresh data. We see that just updating the firmware alone yields a large improvement to the reading of stale data. Apparently there is more to this update than what Samsung led on in their answers to our questions. It appears that they have tweaked their read algorithm to better adapt to cells that have drifted over time.
After 'Advanced Optimization':
The Advanced Optimization refreshes the data, likely doing so in a sequential fashion, meaning that files that had been randomly written (as was the case of our test file) are restored to an even better state than they were before.
Now we'll move onto another set of tests I performed on a 1TB 840 EVO:
The above chart says it all. Random IO performance was clearly suffering on this unpatched 840 EVO. Patching led to an immediate jump in performance (which did not change at all after the SSD was given additional time to perform any supposed background restoration). Triggering the Advanced Optimization resulted in further performance gains.
**Note** – the above and below Iometer data can not be compared directly with our other reviews, as this test was run in a different configuration and with different pre-conditioning than the typical test sequence run for our full performance reviews. The same conditions are impossible to achieve given that writing data to the 840 EVO would act to refresh data and nullify the accuracy of the next test performed.
Taking a look at latencies, we can see great improvements in the patched 840 EVO, and even further improvements once optimized.
So glad I made sure I stuck
So glad I made sure I stuck with MLC, I advised people on the 840 launch to buy 830s and avoid TLC.
I own 2 830s and an 850 pro.
The 850 EVO though… Have a
The 850 EVO though… Have a 1TB one in my laptop. Nice “bit of kit,” that (4+½ seconds full boot times in 8.1, etc). If I remember correctly, it even outperforms the PRO in some scenarios due to the SLC cache.
I do have a 1TB 840 EVO in my main home PC and a 500GB one in my wife’s laptop. At least now it looks like they’re going to continue to be solid performers. Thank goodness I’ve never had reliability issues with them, or I would have jumped ship a while ago.
Forgot to run the
Forgot to run the “SSDReadSpeedTester” before I updated to Magician 4.6 and new firmware, but ran it before and after running Advanced Performance Optimization. 250GB drive (used space 79.9GB of 232GB).
Before;
http://i.imgur.com/fczQuRU.png
After;
http://i.imgur.com/9THLien.png
The samsung firmware for my
The samsung firmware for my EVO 840 SSD EXT0DB6Q did NOT fix my issues, its even slower then before.
installed tool and firmware
rebooted twice
uninstalled antivirus
cleanup the drive so only windows 8.1 is there
Used the advanced optimisation option in the magician 4.6 tool.
Afther advanced optimisation:
Random Read in iops is 37000
Random Write in iops is 16000
in the beginning its was:
Random Read in iops is 93400
Random Write in iops is 85940
Hmmmm, just started using 840
Hmmmm, just started using 840 evo external hdd. Is there a way to apply the fix to external usb 3.0 enclosure? Don’t have a computer to swap out internal drives.
May 4th and Samsung Magician
May 4th and Samsung Magician still claims the installed version (4.5) is the latest version.
I just tried using the
I just tried using the bootable ISO update (EXT0DB6Q) on my mSATA 840EVO, and it said “no supported SSD’s detected for firmware update”.The Magician software, as has already been stated, says that EXT42B6Q is the latest firmware. Does this mean that the mSATA drives are not affected by this slowdown anymore?
+1
The same for me. Linux
+1
The same for me. Linux environment.
Added to that, had difficulties to prepare the bootable ISO (in an USB device), but finally could preparing an isolinux.
I have a mSATA 840 EVO too,
I have a mSATA 840 EVO too, and I can confirm that these drives have the same problem as 2.5″ EVOs . Samsung has not released any new FW for this drive so far. I am using diskfresh to periodically rewrite sectors in my drive.
I just updated my 750GB 840
I just updated my 750GB 840 EVO with firmware EXT0DB6Q on my Macbook pro. I’m not sure what happened with this firmware but my Write speed is at 55 MB/s and Read speed is 511 MB/s. How do I optimize my write speed? I’m using about 320 GB.
I was getting min 4.4 MB/s,
I was getting min 4.4 MB/s, max 306 MB/s, average of 77.9 MB/s.
I updated the firmware with Magician 4.6 as instructed by Samsung. I now can’t run HD Tune Pro. I get read error. I am also now losing files intermittently. So disappointed in Samsung.
New Samsung Magician software
New Samsung Magician software :
http://www.samsung.com/global/business/semiconductor/minisite/SSD/global/html/support/downloads.html
“Magician version 4.7 features a number of improvements over version 4.6:
– New model support: 850 PRO 2TB, 850 EVO 2TB
– Advanced Performance Optimization support for 840 EVO mSATA
– New firmware is required to use Advanced Performance Optimization
– Minor bug fixes”
I don’t see the “new firmware” on their site for the mSATA EVO, but the new 4.7 Magician may find it for the mSATA users.
I’ve just read and re read
I’ve just read and re read the info on the pages and all the comments and it was informative and interesting so thanks
I am new to SSD’s and have a EVO840
in windows 7 it still gives the option to defrag, in windows 10 (2x boot setup) it will not let you defrag
so the new firmware refreshing the data on the SSD
could we just do the same thing by doing a defrag of the SSD ?
or even doing a true image backup and restore writing the files back to the SSD from a HDD backup?
Also what about the over
Also what about the over provisioning space? is this refreshed or is it best to turn of OP while the refresh goes on
also OP on or off does it work (by saving about 20% and keeping the SSD fast)
During this so called refresh
During this so called refresh that the firmware does transparently(which happens at least 1 times per day, lasting hours) and cannot be stopped by anything other than a Windows shutdown (ie. power off, because reboot won’t do it): got 2 meg per second random write speed (eg. CDM’s 4KiB Q1T1)
screenshots:
Crystal Disk Mark: https://cdn.minds.com/fs/v1/thumbnail/870241363528581120/xlarge
Magician (4300iops, takes 5min36sec to run the test): https://cdn.minds.com/fs/v1/thumbnail/870241424523509760/xlarge
Normal, no refresh happening: 26 meg per second –
screenshots
CDM: https://cdn.minds.com/fs/v1/thumbnail/870243723101405184/xlarge
Magician (27241iops, takes 1min6sec to run the test – aka normal speeds): https://cdn.minds.com/fs/v1/thumbnail/870243759886811136/xlarge
There doesn’t seem to be any way to tell the SSD to not run the refresh. I don’t understand why Samsung thought this was a good idea: running the refresh at least once a day (assuming I shutdown my PC every day, ofc) seems like such a terrible idea, not only for writes performance while refresh is going on, but more importantly for the SSD lifetime!! I expect this SSD will last no more than 3 years due to this EXT0DB6Q firmware doing such refreshes, and there’s no way to downgrade to a previous firmware version!
Needless to say I’m avoiding all Samsung stuffs forever from now on!
let me correct that for
let me correct that for you:
Q: The new firmware appears to restore read performance without the need for Magician. How was this accomplished?
A: Samsung revised the firmware algorithm to maintain consistency in performance for old data under exceptional circumstances. Therefore, read performance was restored without the need for Magician. This algorithm is based on a periodic refresh feature that can maintain the read performance of this older data. The algorithm does affect normal user scenarios (i.e. occasional PC performance degradation due to background work of SSD) and the lifespan of the SSD, but can otherwise maintain its performance without the help of Magician. However, this algorithm does not operate when the power is off.
In the comment above “July
In the comment above “July 29, 2018 | 06:49 AM – Posted by xftroxgpx (not verified)” I’m sorry to say that I WAS WRONG about the 2MB/sec low speed being caused by Samsung SSD. It was instead a quirk in my Lenovo Ideapad Z575 laptop whereby after a Windows 7 Restart, while the SSD is connected on the optical disk drive bay (via a caddy), it would start exhibiting that 2MB/sec low random write speed! I don’t know why still! But I’m sure now that it’s because of the laptop/BIOS, and NOT because of Samsung SSD or its new firmware.
When I place a Kingston SSD on the ESATA port, keep the Samsung SSD on the ODD bay, then it’s ONLY the Kingston SSD that experienced the 2MB/sec low random write speed after the first Linux reboot. Samsung SSD’s write speed remains as high as usual (like 36MB/sec) regardless of how many restarts happen.
This is how I know it’s not Samsung SSD’s fault, it’s Lenovo Ideapad Z575’s fault or something.
Therefore there’s no way to know when the refreshing happens and I’ve not seen any performance slowdown, even though the lifespan of the cells may still decrease with each refresh(I don’t know that, but I’m assuming that there’s no way to refresh the cells without decreasing their lifetime – I may be wrong).
So ignore my previous comment too, this one: “July 29, 2018 | 07:52 AM – Posted by xftroxgpx (not verified)”