Need Latest 02M3 Firmware for X-25-M 80GB

[idata]

I get an error from the SSD Optimizer in the Intel SSD Toolbox to update my drive's firmware; the utility will not run. It says I have the old firmware and informs me I need to update (the drive's firmware).

Intel Product Change Notification number 110343-00, dated December 3, 2010, tells me that the current firmware (for this drive) is version 02M3, not the version 02HD as is in the ISO update on the Intel site and is noted at the latest version of the firmware (110902HD88208850.iso).

The Intel SSD Toolbox "View Drive Information" confirms the firmware is version: 2CV102HD.

The disk created from the ISO file also confirms the firmware is version 02HD and will not update the SSD drive.

I need to update the firmware to version 02M3 to allow the SSD Optimizer to run.

I am running XP media Center Edition with all the latest patches and upgrades. No other problems with the drive or OS.

Please help.

Thanks!

[idata]

mistermokkori is correct, the Toolbox will not work with Intel G1/50nm SSDs.

infocus is correct that the Toolbox should not be needed when an SSD is used with an OS that sends TRIM commands, and the Toolbox is useful for OS's that do not. The fine point of being in AHCI mode and having the optimal driver (not a TRIM driver) which is iaStor for TRIM to work, I am not sure about. I'm not saying you are wrong, I'm just not sure if TRIM will not work in IDE mode, and with non-AHCI drivers, I always forget that point.

I too use the Toolbox although I use Win 7. I like to run a Windows Diskcleanup, and also remove all but the latest restore point from the second tab's options, and then run the Optimizer. I ought to turn Restore off actually, but I've decided I'm not going to "baby" the SSD, beyond the usual performance and basic tweaks for SSDs, just to see how it works over the long term. So far so good.

One general issue regarding SSDs that I think is rather ridiculous, and I mean ALL SSDs made by ANY manufacture, and moreso the manufactures of the NAND chips used in SSDs and the controller chips as well, is this:

Given the TRIM issue and that the need for it is due to the way NAND chips and the controllers operate, in conjunction with the manner OS's implement file system management, and the problems caused by that with early SSDs, which is still seen today with some products, it seems ridiculous to me that all of this was not understood from the beginning. That is, it seems SSDs were sold by manufactures with no knowledge that these issues existed.

How could that be possible? Didn't the engineers designing these products know this, or notice it when they were testing them? How could the NAND chip manufactures not know that the operational differences of the chips would need new and different file management? Did they simply decide it was "not my job man..."? Or was it simply a high level executive decision to "get those SSDs on the shelves now" with full knowledge of the issues, but with the hope that the public would not notice?

IMO, Intel is one of if not the only manufacture to address the unique needs of SSDs, with their controllers and the Toolbox. To be fair, the Intel G1 SSDs certainly were "not ready for prime time", and belong in what I call the "...Uh oh..." club of SSDs.

There may be a simple and reasonable explanation for all this, but it still seems strange to me that it seems as if... no one knew.

[mmokk]

parsec wrote:

mistermokkori is correct, the Toolbox will not work with Intel G1/50nm SSDs.

that isn't what i said. i said only the optimizer function does not work with the g1's. the diagnostic scans and smart attribute reading work fine.

The fine point of being in AHCI mode and having the optimal driver (not a TRIM driver) which is iaStor for TRIM to work, I am not sure about. I'm not saying you are wrong, I'm just not sure if TRIM will not work in IDE mode, and with non-AHCI drivers, I always forget that point.

neither ahci nor iastor are needed for trim to work. windows 7 trims just fine with its default ide drivers.

To be fair, the Intel G1 SSDs certainly were "not ready for prime time", and belong in what I call the "...Uh oh..." club of SSDs.

g1's were and still are very very good drives. with the latest firmware, they are quite resilient under desktop workloads. i've got several with terabytes of data written on them and they benchmark close to how they did new. subjectively, they 'feel' as fast as they did out of the box. it is the ssd's before the g1 that fall into that 'uh oh' club.

[idata]

mistermokkori wrote:

"that isn't what i said. i said only the optimizer function does not work with the g1's. the diagnostic scans and smart attribute reading work fine."

That is true of course, sorry but I was thinking of what I consider to be truly useful about the Toolbox, the Optimizer, since I don't run the diagnostic often, and SMART attributes can be seen with other programs. To me, the Optimizer is 95% of the Toolbox.

"neither ahci nor iastor are needed for trim to work. windows 7 trims just fine with its default ide drivers."

While I would tend to believe you, I have read elsewhere exactly the opposite. Some have even claimed the Toolbox won't work if AHCI mode is not on. Regrettably, bad information can be found all over forums, and at times I don't know what to believe anymore. I am not saying you are wrong, on the contrary, I believe you. Are you also saying that msahci is not needed as well?

"g1's were and still are very very good drives. with the latest firmware, they are quite resilient under desktop workloads. i've got several with terabytes of data written on them and they benchmark close to how they did new. subjectively, they 'feel' as fast as they did out of the box. it is the ssd's before the g1 that fall into that 'uh oh' club."

Ok, that's great, I don't own a G1, and the implication of the Toolbox is that G1's don't support TRIM, which I imagine you'll correct me on as well.

Your comment is the first positive one I've ever heard regarding G1 SSDs, thanks for the info.

[idata]

Hi parsec,

SSD's are extremely complicated and the controllers between different brands of SSD's work in slightly different ways. I am only a layman when it comes to SSD, but I try to explain below, as best as I can, the issues behind TRIM and SSD development.

First off it is worth considering that operating systems and programs have been developed with consideration to how a HDD performs. A SSD performs very differently and has different weaknesses and strengths.

Win 7 is the first OS that took steps to becoming SSD friendly. Intel G1 drives however pre-dated Win 7 and TRIM.

The provision of TRIM is one of the key Win 7 developments for SSD.

With HDD (or SSD for that matter) when you delete a file the "delete" only occurs in the file system. The data is not deleted on the HDD (or SDD); it is only marked as available to be used. That is why it is possible to get data back after it has been "deleted" because it has only been marked as available, it has not been erased.

With HDD data that has been marked as available can be overwritten without a performance penalty. That is unfortunately not the case with SSD. When data is overwritten on the SSD it has to organise a block to relocate any valid pages and then erase the block before data can be written. That process incurs a write penalty and is why some of the early SSD's stuttered. This was a problem on some of the SSD's that pre-dated Intel G1 drives, which had significantly less ability to overcome the write penalty.

The "stutter" was a result of the write penalty, which became apparent when lots of really small random files were written in quick succession. When this occurs the SSD has to organise and write significantly more physical data compared to what actually needs to be written from the file system perspective. Hence apps that generated lots of small writes could cause stuttering. The types of app's that generated small writes were not app's that would normally be considered to generate a significant load on the storage system. It was apps like MS Messenger, web browsing and Windows itself.

The fact that the write penalty resulted in unacceptable performance did indeed seem to catch out early vendors (like OCZ) and product reviewers alike.

The G1 drives from Intel did not suffer from stuttering, however reviewers, who were caught out by the stuttering on other brand SSD's, subsequently hammered the G1 drive with writes to make sure that the drive did not stutter. The amount of writes that reviewers hammered the drives with were well in excess of what would typically occur in normal use. Whilst they could not get Intel drives to stutter they did manage to get performance to degrade. It's easy to understand the reviewers paranoia after getting caught out with the stutter problem on previous drives, but it could be argued that they then went OTT. In essence they were creating a fairly unrealistic scenario and then making a big deal out of it when they found it caused a performance drop.

Regardless however Intel issued a firmware update for G1 drives that went a long way to reducing performance degradation when the drives were subjected to excessive writes over a short duration.

After the firmware update the G1 drives were highly resilient to degradation and they remain a great drive to this day. The fact that Intel did not issue a firmware update for G1 drives to enable TRIM is however something that upset a lot of customers. Not a great way to reward earlier adopters. Apart from that gripe however Intel drives have performed admirably and they are the most reliable SSD out there. Even today the G1 access times outperform the latest gen drives from other vendors and they have managed to do that without DRAM.

The emergence of TRIM helps prevent a write penalty from occurring. It is issued by Win 7 when a delete occurs on the file system. This gives the SSD the opportunity to erase blocks in advance of a fresh write occurring, thereby limiting the occurrence of an overwrite penalty when a new write is performed. The timing between the OS TRIM command being issued and actually being executed is something the SSD decides, however from my understanding once the TRIM command is issued a subsequent read command (in-between the TRIM command actually being executed) would be cleared to zero by the device.

So, whilst TRIM helps to provide consistent write performance the down side is that once a TRIM command has been issued the chances of recovering "deleted" data is effectively zero. I believe a format command also generates a TRIM command, so if you accidently delete data or accidently format the drive the chances of getting data back is zero.

The manual TRIM command from the toolbox generates temporary files that completely fill the SSD. As far as I know all data marked as available is then cleaned up ready for a fresh write to occur without a write penalty occurring first. I don't know, but would assume that the chances of getting "deleted" data back afterwards would also be zero.

Sorry for the long post, but I thought it was worth explaining some of the issues behind TRIM that arose during the evolution of SSD.

[mmokk]

parsec wrote:

While I would tend to believe you, I have read elsewhere exactly the opposite. Some have even claimed the Toolbox won't work if AHCI mode is not on. Regrettably, bad information can be found all over forums, and at times I don't know what to believe anymore. I am not saying you are wrong, on the contrary, I believe you. Are you also saying that msahci is not needed as well?

the toolbox and trim both work fine without ahci. i say that because i've used them. if people report they don't work, then it is due to some configuration issue specific to their case.

when you hear someone say 'msahci', they are referring to microsoft's default ahci driver that comes with windows 7.