• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Bad TRIM behaviour OCZ Vector 512 GB

JouMxyzptlk

Junior Member
The drive is used for a lot 4K Ultra HD video work, so I actually need the read/write performance of that drive, and it delivers.
But one thing bugs me a lot: The bad TRIM behavior of the drive.

The setup: OCZ Vector 512 GB, firmware 2.0, the drive is my OS drive, OS uses 35 GB. Rest is demoted to video recording.
The effect(s):
Deleting one 10 GB file if video data (or random data, no compressible) causes an OS stutter of one second within the next ten seconds. Some services don't like the slow response and stop working properly from time to time.
Deleting one 100 GB file causes a bad OS response for about 10 seconds.
Deleting 350 GB of video data make my mouse freeze several times for a few seconds, the whole time needed to recover until normal response is 39 seconds.

The ONLY fix up to now: Disable TRIM via registry.

Up to now it is confirmed by OCZ and by some other users. Trouble tickets (yes, plural) get a first response, but time out since the second level doesn't have a fix yet.

Can anyone confirm this behavior? If yes, create a support ticket to make them MOVE.

To Anandtech: You should add TRIM behavior testing, a time graph on how fast the drive responds to various sizes of deleting big file(s) filled with random data. Suggested sizes: 32 GB, 128 GB, 256 GB and "90% of drive capacity" GB.
 
To Anandtech: You should add TRIM behavior testing, a time graph on how fast the drive responds to various sizes of deleting big file(s) filled with random data. Suggested sizes: 32 GB, 128 GB, 256 GB and "90% of drive capacity" GB.
If they can come up with a way to measure it accurately, I think that's a great idea.
 
OCZ has added a TRIM command to it's utility program you can run in windows.

If you have issues with TRIM in daily use, you could theoretically set up a macro now to TRIM in the middle of the night each night when away from the PC.

I honestly do not know if TRIM is enabled properly or not in my new install of win7 on my main box with an ocz agility 3 60gb (old install it was enabled), but it performs great and the one time I ran TRIM with the ocz utility it said it was successful along with taking around 25 seconds so *shrug*.
 
That is really disappointing to hear because so far I haven't had an SSD do that (my Intel X25-M, OCZ Vertex 2e and now my Crucial M4). Its certainly not a universal problem with trim and SSDs, its largely a problem with a particular model and having a very aggressive trim that presumably is impacting performance of the drive in a remarkable way. Its likely worth sending Anand an email from the site to tell him about it, he would love to do such a comparison of the drives and break such a story I suspect.
 
TRIM behavior testing

I agree, it would be interesting. For instance you can see from this post there is quite a difference in trim time between different models. Trimming 4kB time are Vertex4 0.6ms, Agility3 11ms and Vertex3 14ms. Also trim can return successful when it is not, the driver telling fibs just to keep the system happy perhaps, in which case the completion is very quick, just a few microseconds.

With the Windows trim on delete I have found it is quite usual for there to be a delay of up to 20 or 30 seconds before the trim command is sent after delete so this is probably why you are experiencing the slowdowns a few seconds later.

There are some trim testing features in the Windows HCK but I've not used them so do not know if they are suitable or not.
 
Back
Top