• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

OCZ Agility 3 90GB

A

AQ_OC

2[H]4U
Joined
Jan 23, 2011
Messages
2,131
I got one of these on sale at Amazon.com.

I have it on the sata 3 port on my Asus P8P67-Pro.

The packaging claimed Reads up to 525MB/s and writes up to 500MB/s. 4kb Random write aligned up to 50000 IOPS.

Well, I'm getting nothing like those numbers:

-----------------------------------------------------------------------
CrystalDiskMark 3.0.1 x64 (C) 2007-2010 hiyohiyo
Crystal Dew World : http://crystalmark.info/
-----------------------------------------------------------------------
* MB/s = 1,000,000 byte/s [SATA/300 = 300,000,000 byte/s]

Sequential Read : 214.784 MB/s
Sequential Write : 117.712 MB/s
Random Read 512KB : 204.888 MB/s
Random Write 512KB : 117.818 MB/s
Random Read 4KB (QD=1) : 25.114 MB/s [ 6131.4 IOPS]
Random Write 4KB (QD=1) : 63.074 MB/s [ 15398.9 IOPS]
Random Read 4KB (QD=32) : 114.365 MB/s [ 27921.1 IOPS]
Random Write 4KB (QD=32) : 115.719 MB/s [ 28251.8 IOPS]

Test : 500 MB [E: 0.2% (0.1/83.8 GB)] (x5)
Date : 2012/02/14 21:15:36
OS : Windows 7 Ultimate Edition SP1 [6.1 Build 7601] (x64)

What gives? Am I doing something wrong here?
 
BTW, the mobo is set to AHCI mode. I have an Intel X25 (160GB) as the boot drive in this PC.
 
Are you using random data or all zeros for the sequential reads and writes?

The quoted numbers are for compressible data involving large files with reasonably large buffers.

I can't speak to the IOPS.
 
I used the default random test data. Should I be getting something different?
 
Ok...I'll give that a try once I get home.

I didn't know that test results were based on the type of data you write. Is this some form of false advertizing or is there some reason why using 0 fill is representative of real-world usage? The reason I ask is because when I test my Samsung 830 256GB ssd I get the kind of results I'd expect based on reviews I have seen.

Thanks.
 
The SF controller uses compression to limit the amount of data written to the flash. Since disk I/O is the long pole in the tent, reducing the amount of data written and read can speed up access.

While a lot of files are highly compressible -- especially text files -- there is a ton of stuff that is already compressed, like music, video, photos, etc. So not every file access is going to experience the "up to" numbers that are advertised.

Also, this drive uses asynchronous flash which is cheaper but doesn't perform as well as synchronous flash. The question is whether or not you feel the drive performs poorly, or if it's just a matter of looking at the benchmarks. Anand has some benches of the Agility 3 if you want to see if your setup is achieving what's expected: http://www.anandtech.com/bench/Product/371

He doesn't have a 90 GB bench though so you might have to extrapolate a bit between the 120 GB and 60 GB benches.
 
Thanks for that. I can see that evaluating SSD performance is a very complex task. I guess in the end, I just want it to be fast compared to HDD access. Anything more seems to be too hard to quantity (at least for me). I'm going to put this SSD into one of my HTPCs as the sole drive.
 
You must log in or register to reply here.
Back
Top