Show us your SSD performance

[essenbe]

Hey, I've been wondering now just how good this SSD actually is. Glad I found the tests to run here

OCZ Vertex2 3.5" 120GB SSD

The transfer rate should be above 200MB/sec and the access time should be 0.1ms. But I had the same problem with HD Tune - the Vertex2 results were all wrong whilst the Intel results were OK. Atto is a better measuring tool.

OCZ tech support only accept ATTO results. Of course, they give the best results for Vertex2.

 

[whs]

The thing is that I get very similar HD Tune and Atto results on my Intels. But for the OCZs, HD Tune can be anything. And it is not consistant from run to run.

I just made a quick run on my 3 year old Vertex in this system (which is a Dell). It came out relatively well - given that this is a very old Vertex. But in the next run it may be different.

 

[essenbe]

I know WHS, I have the same issue. OCZ forums had a thread that said it is how the tests are done. I really am not technical enough to keep up eith what they were saying. but it had to do with compressible testing vs I/O reads and non comressible tests.

 

[essenbe]

Whs, also on another issue you may find interesting. I was reading on secure erase, and it was said that a secure erase on Vertex SSDs did not actually erase anything. The vertex encodes and compresses the data you write to the SSD. A secure erase simply changes the encoding key thus making the previously written info unreadable.

 

[Cr00zng]

Whs, also on another issue you may find interesting. I was reading on secure erase, and it was said that a secure erase on Vertex SSDs did not actually erase anything. The vertex encodes and compresses the data you write to the SSD. A secure erase simply changes the encoding key thus making the previously written info unreadable.

Depending on the type of encoding, that would be a OK just to delete the key; it's pretty similar to what the self-encrypting drives do. The probable difference is that the hardware based FDE utilizes AES encryption while the Vertex SSD encoding is unknown. At least it is unknown to me...

 

[whs]

Whs, also on another issue you may find interesting. I was reading on secure erase, and it was said that a secure erase on Vertex SSDs did not actually erase anything. The vertex encodes and compresses the data you write to the SSD. A secure erase simply changes the encoding key thus making the previously written info unreadable.

Yeah, I have seen that too. But unless you know the key, you can still not get at the data.

 

[Hopalong X]

Sounds like a job for Diskpart "clean all" or software to write 0's to actually clear the drive.

Good to know.

 

[tw33k]

We all now know that the TRIM command is not passed on if using the Marvell controller. One way I found to clear the free space on my C300 was to use Ccleaner's wipe free space function. Under Options and Settings make sure "Wipe MFT Free Space" is checked (I use the Normal mode). Then click "Cleaner" and right down the bottom click "Wipe Free Space" It took about 12 minutes for 80GB on mine. It's best to let your PC idle for a few minutes after it's finished

 

[Miranova23]

I just used HD Tune cuz of the 2 that the op, Nate42nd, suggested, only 1 was in English. Here's Atto. & then as recommended, I won't be testing my SSD again for a while unless need be.

 

[Cr00zng]

Well, evidently the OCZ Vertex SSDs utilize AES 128 and/or 256 bit encryption, as indicated at this site, quote:

DuraWrite is the name SandForce gives this black box of compression mojo, and the company has been reluctant to give us a closer look inside. We suspect there's an element of encryption involved, since SandForce controllers offer on-the-fly AES encryption that can't be disabled. The latest SF-2000 series support 256-bit AES encryption, which is a nice upgrade over the 128 bits of protection offered by the SF-1000 family.

The drive is also Opal SSC compliant, which means that one could control access to the drive via software that can manage self encrypting drives. That's neat and takes no time to provision access to and/or decommission the drive...

Here's a link for what DuraWrite does that has more information that you'd need...

 

[Fumz]

We all now know that the TRIM command is not passed on if using the Marvell controller. One way I found to clear the free space on my C300 was to use Ccleaner's wipe free space function. Under Options and Settings make sure "Wipe MFT Free Space" is checked (I use the Normal mode). Then click "Cleaner" and right down the bottom click "Wipe Free Space" It took about 12 minutes for 80GB on mine. It's best to let your PC idle for a few minutes after it's finished

I am in the process of trying to get Asus to answer the question about Trim and the Marvell controller. Here's our last exchange:

Hi krisenda,
Thanks for the reply. It was my understanding that the Trim signal is sent to the Marvell
controller, which supports Trim, but that the Marvell drivers do not pass along the
signal; therefore, the drive will never Trim. Do you know if the latest Marvell drivers also
support Trim... and that they do now indeed pass along the signal?

Hello Sir/Madam

Yes this board and the drivers support the TRIM feature on SSD's.

If you have any further questions please feel free to contact us.

Krisenda
ASUS Tech Support

I'm not sure whether to believe this or not... or whether to believe she knows what she's talking about because if Trim were now finally supported, I think we'd have heard about it... right? Do you have any new news?

Until I get here it from other sources, I'm sticking with the Intel controller.

 

[tw33k]

I don't believe it simply because everywhere else I've read says it doesn't support TRIM. If it did, we'd all know by now.

 

[essenbe]

I don't believe it simply because everywhere else I've read says it doesn't support TRIM. If it did, we'd all know by now.

I'm sure if it was true, the Crucial forums would know about it.

 

[Fumz]

Yeah... that's what I was thinking too. The degree to which Asus has avoided giving a direct answer to a direct question was really peculiar. It took 5 email exchanges and 2 different reps just to get to this point... which instills zero confidence.

 

[Hopalong X]

SSD Write Endurance 25nm Vs 34nm - Page 16

Post#390-391
Kingston SSDNow- Intel 25 rebrand
69.88 TB
MWI-61%

Intel 320
92.2TB writes.
MWI- 53%

Last check both had 5 RaS.

 

[essenbe]

SSD Write Endurance 25nm Vs 34nm - Page 16

Post#390-391
Kingston SSDNow- Intel 25 rebrand
69.88 TB
MWI-61%

Intel 320
92.2TB writes.
MWI- 53%

Last check both had 5 RaS.

Seems the vertex is out of the race.

 

[GeneO]

Well I haven't seen much performance benches posted on this thread lately!

Well just got a 128 GB Crucial M4 for my Asus P7P55D-E Pro system. Running it on the not so great Marvell 912x 6 Gb/s controller - which I don't believe supports TRIM - will see how that goes. MS AHCI driver. WEI score is 7.9. Here are the AS-SSD and CDM. Pretty happy with it so far. Have my OS, browser and Photoshop caches on it.

 

[essenbe]

Good scores GeneO. Looks like it is so fast it broke AS SSD.

 

[A Guy]

Forgot I had Crystal Disk Mark



A Guy

 

[essenbe]

Good scores A Guy. Those write speeds look above spec. and read speeds are right on. Looks real good.