OCZ Vertex 3 SATA III + WDC WD1003FBYZ + WDC WD1002FAEX

  • Release Candidate 6
    Guest:
    We are at a “proposed final” true release candidate with nothing known remaining to be changed or fixed. For the full story, please see this page in the "Pre-Release Announcements & Feedback" forum.
    /Steve.
  • Be sure to checkout “Tips & Tricks”
    Dear Guest Visitor → Once you register and log-in:

    This forum does not automatically send notices of new content. So if, for example, you would like to be notified by mail when Steve posts an update to his blog (or of any other specific activity anywhere else), you need to tell the system what to “Watch” for you. Please checkout the “Tips & Tricks” page for details about that... and other tips!

    /Steve.

iefbr14

Member
Oct 12, 2020
9
1
Code:
Driv Size  Drive Identity     Location:    0      25%     50%     75%     100
---- ----- ---------------------------- ------- ------- ------- ------- -------
 81  120GB OCZ-VERTEX3                   309.0   255.6   388.0   482.2   526.7 
 82  1.0TB WDC WD1003FBYZ-010FB0         129.6   119.9   105.2    90.5    62.0 
 83  1.0TB WDC WD1002FAEX-00Z3A0         121.3   116.9   101.6    86.2    60.2 

                  Benchmarked: Saturday, 2020-12-26 at 21:01
 
Last edited:
Thanks for posting your results!

Interesting pattern across the SSD there! Weird that the 25% measurement is the lowest!

How full is the 120GB drive? My Windows 10 install is about 70GB. No, wait ... just ran Disk Cleanup and found 5GB of "Windows Update Cleanup" to remove! Call it 65GB!

That would make sense with the first half of your drive having (presumed) wear right across?
 
Thanks for posting your results!

Interesting pattern across the SSD there! Weird that the 25% measurement is the lowest!

How full is the 120GB drive? My Windows 10 install is about 70GB. No, wait ... just ran Disk Cleanup and found 5GB of "Windows Update Cleanup" to remove! Call it 65GB!

That would make sense with the first half of your drive having (presumed) wear right across?
Hi there,

My OCZ SSD was purchased new and dates back to 2010. It has over 18,000 hours behind it and has always worked flawlessly. Firmware version 2.22 is installed and I never bothered to install the latest 2.25 version. It still runs on a fully patched Windows 7 Pro configuration taking about 80GB of space. Depending on usage conditions, it keeps between 20 and 30GB of free storage which is getting a bit tight for my taste.

I agree with your assessment that the drive's first half portion might be wearing down although I don't know much about flash-based memory wear patterns. My take is that, apart from space getting low, the drive is starting to get pretty old and I just purchased a new 500GB Samsung 860 EVO to replace it. Should highly improve performance and long term reliability for this box on which I will eventually replace Windows by Linux.
 

Attachments

  • smartctl report.txt
    2.3 KB · Views: 321
Thanks for the detailed information, that's really helpful! I have a 90GB system partition, which I now similarly regret as being a bit on the tight side! I use the rest of the (512GB) SSD for VMs, and anything else I want to be fast, so I wanted to keep plenty of space for that when I was deciding partition sizes. I understimated how much my Windows 10 install would grow over time! It hasn't got to the stage where I need to repartition the drive just yet though. I have about 20GB free, according to Windows, which is just about okay, as you say. :)

Yes, I feel I was perhaps inaccurate using the term "wear", so apologies for my looseness of teminology:

What I meant to refer to was the typical slowdowns in speed that the ReadSpeed benchmark is revealing on these forums. I don't think it is "wear" in the sense of wearing out. It seems that speeds can be improved somewhat by running Spinrite at level 2 (read only), or even more by running level 3 (which reads and writes back to the drive, so will increase the long term wear on the drive, but generally not considered significant enough to worry about). If you don't have access to Spinrite, then there are discussions around the forums about other ways of improving drives performance.

Having said that, congratuations on your new Samsung SSD! :) They seem to be of very good quality (disclaimer: I do own one myself so am biased!), and your old OCZ is still doing really well for 10 years of service! Once you've got the new drive installed, you can maybe explore the ways of improving the OCZ drive's speeds. And if so, do let us know how you get on? :)
 
That would make sense with the first half of your drive having (presumed) wear right across?
Or, how about this?

By default Windows runs it's defragger once every week. The unused part of the drive is therefor trimmed every week, assuming in this case it is scheduled to run once a week. As RE.exe is probing these areas blocks that are known to be empty aren't read as where LBA blocks towards the start of the drive (well not really of course) are actually read. IOW, SSD 'knows' LBA blocks at 100% are empty so it immediately can return zeros without even ever having to read the flash memory.

So it may have little to do with actual wear.
 
Last edited:
The defragger SHOULD be automatically turned off by Windows when running on an SSD. The moving of blocks around to be at the front of the drive is harmful to the drive's lifespan and is unnecessary on an SSD. If the defragger is active on an SSD, the user should turn it off manually, AFAIK.

Regarding the drive knowing blocks are empty and faking its performance, what if you fill all the empty space with random gibberish files, which I do when I burn in a drive anyway, and then delete the files. Regardless of what the file allocation table (or whatever it's called these days) says, the blocks themselves would not be empty. Would those blocks be cleared again when TRIM runs? Maybe they would.

May your bits be stable and your interfaces be fast. :cool: Ron
 
The defragger SHOULD be automatically turned off by Windows when running on an SSD. The moving of blocks around to be at the front of the drive is harmful to the drive's lifespan and is unnecessary on an SSD. If the defragger is active on an SSD, the user should turn it off manually, AFAIK.
Nonsense. Defrag will refrain from doing actual defragging for the most part for SSD. Instead it will send the TRIM command for all unused clusters which is quite useful.
Regarding the drive knowing blocks are empty and faking its performance, what if you fill all the empty space with random gibberish files, which I do when I burn in a drive anyway, and then delete the files.
If you delete files they'd be trimmed again.
 
Or, how about this?

By default Windows runs it's defragger once every week. The unused part of the drive is therefor trimmed every week, assuming in this case it is scheduled to run once a week. As RE.exe is probing these areas blocks that are known to be empty aren't read as where LBA blocks towards the start of the drive (well not really of course) are actually read. IOW, SSD 'knows' LBA blocks at 100% are empty so it immediately can return zeros without even ever having to read the flash memory.

So it may have little to do with actual wear.
Defraging an SSD drive isn't recommended and may reduce its lifespan so it's disabled for the SSD drive on my Windows configuration.
 
Last edited:
Thanks for the detailed information, that's really helpful! I have a 90GB system partition, which I now similarly regret as being a bit on the tight side! I use the rest of the (512GB) SSD for VMs, and anything else I want to be fast, so I wanted to keep plenty of space for that when I was deciding partition sizes. I understimated how much my Windows 10 install would grow over time! It hasn't got to the stage where I need to repartition the drive just yet though. I have about 20GB free, according to Windows, which is just about okay, as you say. :)

Yes, I feel I was perhaps inaccurate using the term "wear", so apologies for my looseness of teminology:

What I meant to refer to was the typical slowdowns in speed that the ReadSpeed benchmark is revealing on these forums. I don't think it is "wear" in the sense of wearing out. It seems that speeds can be improved somewhat by running Spinrite at level 2 (read only), or even more by running level 3 (which reads and writes back to the drive, so will increase the long term wear on the drive, but generally not considered significant enough to worry about). If you don't have access to Spinrite, then there are discussions around the forums about other ways of improving drives performance.

Having said that, congratuations on your new Samsung SSD! :) They seem to be of very good quality (disclaimer: I do own one myself so am biased!), and your old OCZ is still doing really well for 10 years of service! Once you've got the new drive installed, you can maybe explore the ways of improving the OCZ drive's speeds. And if so, do let us know how you get on? :)
Thanks for your elaborate input and for your suggestions! Once I have my new Samsung 860 EVO installed I'll indeed try to tune my OCS Vertex 3. I think it would make an excellent choice for hosting the Windows page file among other things. However, not being a registered Spinrite user I'll have to try the alternative ways you mentions to try improving its performance. I will post my results once I've done that. Thanks again for your suggestions!

Regards.
 
This is incorrect. Windows defrag does not actually do much defragging on a SSD, but it does TRIM unused space.
Okay I'll byte.

If defrag only TRIMs unused space on an SSD drive, then what's the advantage of defraging it if TRIM is already activated on the drive? Is it that unused space isn't TRIMed except if the drive is defraged?
 
Okay I'll byte.

If defrag only TRIMs unused space on an SSD drive, then what's the advantage of defraging it if TRIM is already activated on the drive? Is it that unused space isn't TRIMed except if the drive is defraged?
Under circumstances not every TRIM command may actually reach the drive, perhaps some API calls don't trigger it when deleting data. Anyway since the extra TRIM will not harm the drive I see no reason to disable defrag/optimize.

We call the tool defrag because that's what it has been called for years. Maybe we should just drop the term and refer to it as optimize.
 
Under circumstances not every TRIM command may actually reach the drive, perhaps some API calls don't trigger it when deleting data. Anyway since the extra TRIM will not harm the drive I see no reason to disable defrag/optimize.

We call the tool defrag because that's what it has been called for years. Maybe we should just drop the term and refer to it as optimize.
All right then. Thanks for your help!

Regards.
 
AFAIK, Windows 7 does not have optimize, only defrag. As of this moment I have been unable to determine if Windows 7 defrag executes trim. But, Windows 7 users should NOT use defrag on an SSD unless it can be proven that it does not, in fact, defrag and that it does execute trim. 100 million people still use Windows 7.


May your bits be stable and your interfaces be fast. :cool: Ron
 
AFAIK, Windows 7 does not have optimize, only defrag. As of this moment I have been unable to determine if Windows 7 defrag executes trim. But, Windows 7 users should NOT use defrag on an SSD unless it can be proven that it does not, in fact, defrag and that it does execute trim. 100 million people still use Windows 7.


May your bits be stable and your interfaces be fast. :cool: Ron

 
Thanks for that link.I've been googling around and I did see that one. I have skimmed it but not read every word. I have a vested interest in finding the correct answers. I do know that if the program on the screen says "Optimize", then it's Windows 8.1 or 10, not Windows 7. The relevant program in Windows 7 says "Disk Defragmenter". Best info I have at this moment is that if you install Windows 7 on an SSD, it's supposed to disable automatic defrag and turn on TRIM. I found out years ago that if you clone Windows 7 to an SSD, you may have to tell it there's an SSD and turn on TRIM (which I did). I read something somewhere that said if AHCI is off, Windows 7 may not recognize the SSD. And, I'm pretty sure that Disk Defragmenter does not have any discretion regarding SSD's. So scheduled defrag is off on my PC's. I'm currently doing research to vet this information and determine what is actually the truth, and will be sharing links here once I compile them into something coherent. It's also worth noting that the disk drive manufacturers own utilities, like Samsung Magician, sometimes have a "Performance Optimize" function or something like it that you can activate. I think all that does is run trim, but it would be interesting to know.

May your bits be stable and your interfaces be fast. :cool: Ron
 
OK, I've spent much of the day researching Windows 7 vs Defrag vs TRIM on SSD's. I just googled windows 7 trim defrag. It took longer than I meant for it to. But, I want to share the results here. However, I cannot spend another day writing a report about all this. So, I've decided the next best thing is just to give you all the links. I'll leave it to you to read and digest everything if you're inclined. I also leave it up to you to determine if you think these are credible. But, I tried to cite what appeared to be credible sources. So, see below. Enjoy.

Bottom Line (I think):
* auto defrag should NOT be on in Windows 7. (may or may not be disabled automatically)
* TRIM should be on in Windows 7. (may or may not be enabled automatically)
* TRIM should work even in IDE mode with a compatible drive. (maybe) (controversy about that)
* There's also a bunch of tweaks that some people recommend that users (including me) may or may not have done.
* Things are different depending on whether you install Windows 7 to an SSD or clone it to it.

May your bits be stable and your interfaces be fast. :cool: Ron

Links. Not all articles are talking about Windows 7 and not all are exclusively talking about Windows 7.



















 
AFAIK, Windows 7 does not have optimize, only defrag. As of this moment I have been unable to determine if Windows 7 defrag executes trim. But, Windows 7 users should NOT use defrag on an SSD unless it can be proven that it does not, in fact, defrag and that it does execute trim. 100 million people still use Windows 7.


May your bits be stable and your interfaces be fast. :cool: Ron
Thanks Ron. Following my discussion with Disktuna I chose to leave my defrag setup as it has been since I installed Windows 7, that is with defrag disabled for my SSD drive.

I will use Windows 7 until the end of extended support in 2023. I just don't like Windows 10 enough to go to it so after 2023, I will already be on a Mac and will renew with the pleasure of running Linux on my current Windows box which will be erased and converted to Linux. Much fun in perspective!

Thanks again!

Regards.
 
  • Like
Reactions: rfrazier
OK, I've spent much of the day researching Windows 7 vs Defrag vs TRIM on SSD's. I just googled windows 7 trim defrag. It took longer than I meant for it to. But, I want to share the results here. However, I cannot spend another day writing a report about all this. So, I've decided the next best thing is just to give you all the links. I'll leave it to you to read and digest everything if you're inclined. I also leave it up to you to determine if you think these are credible. But, I tried to cite what appeared to be credible sources. So, see below. Enjoy.

[snip]
Hey Ron, thank you very much for your extensive research. I didn't expect so much. It's very generous of you.

I will read it all and keep the best of these articles in PDF reference files once I've completely reviewed my SSD configuration and implemented the best setup for reliability and performance.

Thanks again. I do appreciate.

Regards,

Martin