My SSD Rocks!

[crazyfool]

Oh... I must be going nuts then lol

[Paul Adams]

1. controller supports the TRIM command (i.e. forwards it to the drive)
2. drive supports TRIM
3. drive reports (via controller) TRIM is available
...
1-3 are effectively indicated by crystaldisk's TRIM highlighting - hence it being such a useful tool - MS could of made this a darn sight easier to understand IMHO.

Actually (2) isn't exposed by CrystalDisk, I found today... and because of the nature of the cause, I don't think any software can guarantee this.

I have an OCZ 120GB Vertex SSD that has been running firmware 1.3 for ages as it was supposed to support TRIM (and the reason I got it in the first place, when getting ready to install Win7 RTM).
CrystalDisk reported TRIM as active/supported on this drive.

Apparently the drive firmware reports that it supports it, but it just ignores all the TRIM commands - you have to go up to at least 1.4 w/TRIM support to get it properly implemented.
(This is straight from the OCZ staff on their forums.)

I just did 2 scary firmware updates after booting from a USB stick (1.3 > 1.4 > 1.5) and held my breath as Windows booted correctly after restoring the drive to AHCI.
As TRIM has been ignored until now, the bitmap of the cells is incorrect and I ought to write to the whole disk to fix this, but I'm lazy and will wait until SP1 is out, then flatten & reinstall.

[dangel]

Actually (2) isn't exposed by CrystalDisk, I found today... and because of the nature of the cause, I don't think any software can guarantee this.

To be fair, in your case, the drive was erroneously reporting it being supported (which is a bug) rather than this being the fault of crystaldisk itself. I can see that being a PITA but it doesn't really change what was said. As I said, the whole thing is an unnecessary mess and should of been handled much better by all concerned.

[Paul Adams]

To be fair, in your case, the drive was erroneously reporting it being supported (which is a bug) rather than this being the fault of crystaldisk itself. I can see that being a PITA but it doesn't really change what was said. As I said, the whole thing is an unnecessary mess and should of been handled much better by all concerned.

Agree entirely, it was just to point out that you can't necessarily trust even what the device reports to the software as there's no way to ensure the commands are being honoured.
One could write a device driver which presents itself as a disk and calls its IoComplete routine immediately, ignoring the actual I/O contents - it would appear to be the worlds fastest disk with zero latency or faults

[dangel]

Agree entirely, it was just to point out that you can't necessarily trust even what the device reports to the software as there's no way to ensure the commands are being honoured.
One could write a device driver which presents itself as a disk and calls its IoComplete routine immediately, ignoring the actual I/O contents - it would appear to be the worlds fastest disk with zero latency or faults

Well for that matter, a bug in Windows, the device driver etc etc could all do the same - point is, it's not (as you well know) all that likely and people are constantly asking how they figure out (to the best of their ability) whether it's working or not.. The issue is muddy as is!

[TooNice]

TRIM aside, would you install this drive as you would any normal HD? Doing a new build tonight with that drive, so I just want to make sure that it is just a case of plugging, formatting and installing Windows.

[dangel]

TRIM aside, would you install this drive as you would any normal HD? Doing a new build tonight with that drive, so I just want to make sure that it is just a case of plugging, formatting and installing Windows.

Yup, W7 will adjust accordingly (assuming you're using that) anyway.

[Terbinator]

Having ACHI enabled improves drive performance so you may want to flick that on.

[kalniel]

Having ACHI enabled improves drive performance

In what way?

[Terbinator]

In what way?

I don't know I just know that seems to be the general consensus - with the Intel drives at least.

[rsmarsha]

Not sure what exactly it does but this is from the intel SSD pages.

Is the Intel SSD a drop-in replacement for SATA Hard Disk drives?
Yes. The Intel SSD a drop-in replacement provides rugged, reliable performance at lower power. Intel SSDs support the ATA-7 command sets and the SATA II command extensions. AHCI must be supported and enabled by both the system BIOS and OS (May require F6 installation of compatible windows storage driver like IMSM. Note: Vista AHCI support is native)