Momentus XT, Momentus, Momentus Thin, MobileMax, and other notebook drives

[ajarmoniuk]

I think the lags people are experiencing might be caused by the drive copying the information to SSD (perhaps too aggressively?)

[KrypteX]

Yes, most likely. And because of that, it may be that the SLC deteriorates quickly and people get all kinds of strange behaviors...

[deroby]

Bit of history : over a year ago I  bought a Momentus 500 XT which came with SD23, replacing a 'stock' 80GB disk because I needed the extra space and found the hybrid concept 'promising'. To put it mildly : I was VERY happy about this upgrade. Compared to other people at the office using the same hardware (but the 'stock' disk) my laptop was suddenly running circles around them. (booting was faster, outlook launched much faster, Visual Studio was much faster etc...).

As time went by I upgraded the firmware to SD25 but never really noticed much differences.
Then, after months of no problems at all, my laptop started crashing every once in a while. It would BSOD on me for 'random' reasons. (IRQ errors, gfx driver issues, sound driver issues, ram issues, you name it...  )
In the end I had at least one BSOD per week. Internal IT was "kind enough" to provide me with another (identical) laptop, so we simply swapped the disk from one frame to the other but this did not fix the issue. It actually kind of proved the problem was with the disk. That said, I'm still not sure if this can be blamed on the disk itself, or on the OS. Fact is, I was having a problem. Anyway, since my laptop was bound to be replaced with a more modern one 'in a few months', I simply 'got used to it' and made sure to save often so the inevitable BSOD would not cost me too much work being lost.

Starting to 'doubt' my HDD by now, I ran CrystalDisk to read the SMART data, and what do you know : I had an unrecoverable sector ! I then tried SeaTools and it marked the drive for RMA. Off course I could not simply send the drive back to Seagate and wait for a replacement, so assuming this was a 'bad luck' hardware failure, I decided to buy a new disk (Momentus 500 XT again) and simply imaged the old disk to the new disk. The new one would take over from the old one, the old one would get replaced and afterwards I surely would find another use for it somewhere.

The new disk came with SD26 and I was *MAJORLY* disappointed about the performance of the new drive. Sure, it still was faster than the 80GB drive of the past, but it felt a lot slower than what the "old" Momentus once did. Around that time I got upgraded to a newer laptop which now came with a 320GB WD Black. That 320 Gb was fine (size & speed-wise) but Since I had the Momentus drive around anyway, I imaged the 320Gb to the 500Gb and although the the extra space was useful, it made zero effect on performance at all =( (**)

Last week I came to the forums again to see SD28 is out and all the discussion going on to which I'd like to add that :

• Yes, SD26 sucks and probably indeed has the SSD part simply disabled and I'm angry at Seagate to sell me a standard disk for a premium price. At least they could have put a BIG warning on it saying SD26 was put on there for safety reasons but would get a firmware upgrade soon that would (re)enable the flash part.
• Yes, my SD25 disk got corrupted (Win7 64bit, never had any other OS on it) but I can't prove it was because of the firmware or because of the OS or simply a  bad sector on a platter
• I've now been running SD28 on the new disk (the old one is still sitting on my desk, I should RMA it asap I guess) and performance has increased drastically; however I too am under the impression that it caches things a bit to eagerly.

So I did the following test (SD28, Win7 Enterprise 64bit)

• Reboot and log in to windows, wait for all disk-activity to cease (1)
• Reboot and log in to windows, wait for all disk-activity to cease (2)
• Copy a couple of data-files that I haven't touched in the past days to an external disk (9Gb), wait for all disk-activity to cease
• Reboot and log in to windows, wait for all disk-activity to cease (3)

if the cache would indeed be used to store the 9Gb of data files, boot time should increase drastically again for reboot nbr 3.

Boot times gathered from the Event Log (see : http://windows7themes.net/speed-up-windows-7-boot-time-2.html )

• (from the log : previous boot times : averaged around 135 seconds)
• 1: 96999ms
• 2: 94879ms
• 3: 96316ms

So the idea that SD28 swaps out the cache for something else 'right away' seems to be false, although I too have feeling that SD25 seemed to be 'smarter' at this when 'just working' with the laptop. Additionally,  yes, after the copy process of the 9Gb files to the external disk had finished, there still was A LOT of hard-disk activity I could not attribute to anything the OS did, strange.

Again, all of this is VERY subjective.

PS: Out of curiosity I tried to "No GUI" boot option in msconfig.exe and then got this : 80771ms, which is a remarkable difference again IMHO and probably can be attributed to both the Flash-cache and the option (it's the 5th time in the past hour I've rebooted so maybe the boot-files are being cached more & more after all)

Conclusion : so far I'm happy with SD28 but I still have a grudge about Seagate shipping the SD26 without proper notification. I too prefer stability (= no corruption) over speed, but selling me a drive that is only hybrid in theory is well, fraud.

[KrypteX]

Thanks a lot for your description, I see mostly the same behavior as you do.

I'd add something interesting.

When you have read 9 GB of files from the MXT... I wonder how much of that data was actually pulled in the SSD. I guess the drive makes a list of the files to be cached while reading the data and starts to pull them to the SSD only after the read procedure finished. You probably saw this. As soon as the read process finishes, the HDD LED stops, but one can hear the drive working at max afterwards, for a while. It caches a list of LBAs into the SSD.

Now, I wonder how big that list can be made. I mean, what if I read 300 GB of small files from the drive, which of those will actually be cached into the SSD ? How long can that "to-be-cached" list grow to ?

I'm asking this, since I've checked something. In HDTune, if one does a Quick Scan for the whole drive, only a part of those LBAs are cached. When a second Quick Scan is performed (after waiting let's say a minute, for the drive to be able to finish the caching), only a part of the reading will actually be faster... Try it out and see for yourself. It's, indeed, very subjective.

I wonder how did the SD25 handle this case. I won't go back to SD25 just to damage my files with some nice corruption/BSOD, but I was still wondering....

It's very interesting how the drive forgets the cached data with SD28. It needs to relearn every other day... that's why I said that it behaves almost as a normal mechanical drive... in terms of performance. SD25 was better....

[KrypteX]

OK guys.

I believe that the lags/stuttering issues (maybe also connected to the End-to-End errors in SMART, not entirely sure about this) are directly caused by the SD25 firmware. It seems that the problems occur ONLY on drives which, at a given point had the SD25 firmware (even if, afterward one updated further to SD28, for example).

SD25 messes up drives really badly. We've already shown that it corrupts files under all operating systems. But I strongly believe that it also messes with the SSD and/or with some very important cache-controlling firmware components.

I can tell you that I have never seen the lags/stutterings/freezes on the following 2 drives:

1. originally SD23, updated to SD24, then to SD28 directly. Perfect drive. No corruption

The second drive:

2. originally SD22, never updated to anything else. It works very well, no problems. No file corruption.

It seems that the file corruption was introduced in SD24 (I'm not entirely sure about this). Can anyone confirm ? However, SD25 clearly shows serious problems with corruption.

I've seen the lags/freezes only on drives which, at a given moment in time, had SD25 on it (especially for a longer amount of time). Those drives sometimes showed the End-to-End errors in SMART.

[Joeri]

Here is how I upgraded the firmware of my Momentus XT drive with a bootable USB device instead of using a bootable CD. This is particularly useful when the firmware update utility does not run within Windows, your netbook does not have a CD/DVD drive build in and you also don't have an external USB CD/DVD drive.

Configure your BIOS to be able to boot from USB.

Create a bootable USB device (with Rufus for example, format to FAT32 and include FreeDOS support).

Download the bootable CD ISO image for the SD28 firmware (here for example).

Extract the contents of the MomentusXT-ALL-SD28.iso file (with WinImage for example).

Extract the contents of the AN-SD28.ima file (also with WinImage for example).

Copy the contents extracted from the AN-SD28.ima file to the bootable USB drive except for the COMMAND.COM file.

Boot from the USB device and run the FDAUTO.BAT file.

Now follow normal procedures just like the bootable CD ISO image.

Cheers.