02HA Poll.

[idata]

UPDATE 10-31-09 @3:20p

I=IDE 

A=AHCI R=RAID

BRAND NEW

III - GOOD=01 BAD=00 AAA - GOOD=02 BAD=00 AIA - GOOD=02 BAD=00

RIR - GOOD=00 BAD=01

INSTALLED (NO REFORMAT/ERASE)

III - GOOD=03 BAD=00 AAA - GOOD=06 BAD=02 AIA - GOOD=05 BAD=04 RAR - GOOD=02 BAD=00 RIR - GOOD=03 BAD=00

REFORMAT/ERASE

AAA - GOOD=01 BAD=00 AIA - GOOD=01 BAD=00 IIA - GOOD=01 BAD=00

Also reworded some of the questions. PLEASE POST ANSWERS ONLY. Thx

-----------------------------------------------------------------------------------------------------------

Hey, guys, I think we should take some info of those who updated their g2's. It's really hard to see what's going. I think people should list the stuff on their computer and what was on it before the firmware flash. It might be a lot easier to see a commonality. In my opinon, Intel should have made the flash update a destructive one. Wipe the drive off clean before updating the flash. That way a lot of variables for error is reduced. I have a feeling the trim command is getting some instructions all messed up with a filesystem already present. Yeah i know it's a lot of questions but...

Here' it goes...

1-Bios updated to Latest? (YES or NO)

2-What g2 size 80 or 160g? (80 or 160)

3-Was drive BRAND NEW / ERASED (hdd erase/reformat)/ INSTALLED (OS already installed) before update?

4-Is the drive partitioned? (YES or NO)

5-other harddrives installed? (YES or NO)

6-Was drive aligned BEFORE and AFTER update?

7-OS that you have installed or will be installing?

8-If using windows and already installed, what driver is being used before updating firmware? (Intel or MS, AHCI, IDE) if applicable

9-Bios setting before update? (IDE, AHCI, RAID, etc)

during update? (IDE, AHCI, RAID, etc)

after update? (IDE, AHCI, RAID, etc)

10-Did you run the flasher via USB or CD?

11-How many times did you run the flasher? (1x, 2x, 3x)

12-Was there an error when flashing? (YES or NO)

13-After it flashed and the computer rebooted, did you immediately change the settings back to what it was before booting/installing OS? (YES or NO)

14-When booted in Windows did it install a driver? (YES or NO) if applicable

15-What driver did it install?

16-If possible, what driver is being used in Windows after firmware? (Intel or MS, AHCI, IDE)

17-Is the G2 drive bad or working fine? (BAD or GOOD)

List your system specs too and if possible chipset (south bridge chip)

List you SSD manufacturing date.

Anything else you think I missed please say.

***FOR THOSE WITH BAD DRIVES AFTER UPDATE DID YOU TRY TO HDD ERASE AND REINSTALL? If so, did it work again?

***Don't do it if you're not comfortable with it.

Hopefully the information given can help INTEL resolve this issue quicker as it seems to be easier to get information quicker on the forums than individual emails.

[idata]

1-Bios updated to Latest? YES

2-What g2 size 80 or 160g? 80

3-Is the drive partitioned? NO

4-other harddrives installed? YES

5-What version of 7, x86, x64? Win 7 Pro x64

6-Was windows 7 Installed already or was it clean? INSTALLED

7-Was it aligned before and after update? YES (both)

8-Did you reformat or leave windows on before update? INSTALLED

9-Was Intel optimizer installed before or after update? AFTER

10-If optimizer was Installed BEFORE the update did you run it? n/a

11-Bios and windows setting before update? RAID

12-Setting during update? AHCI

13-Setting after update? RAID

14-Did you run the flasher twice before booting or installing Win7? 1x

15-Was there an error when flashing? NO

16-After it flashed and the computer started (Windows was installed before update), did you immediately change the settings back to what it was before booting into windows? YES

17-When booted in Windows 7 did it install a driver? NO

18-Is the G2 drive bricked or working fine? GOOD

System Spec:

Intel DX58SO

Core i7 950

6GB DDR3

NVIDIA 9800 GX2

Intel X25-M G2 80GB

WD VelociRaptor

2x WD Raptor X (RAID0)

ASUS Xonar DX

[idata]

1-Bios updated to Latest? (YES or NO) YES

2-What g2 size 80 or 160g? (80 or 160) 80

3-Is the drive partitioned? (YES or NO) NO

4-other harddrives installed? (YES or NO) YES

5-What version of 7, x86, x64? (using or will be using) X64

6-Was windows 7 Installed already or was it clean? (INSTALLED or CLEAN) INSTALLED

7-Was it aligned before and after update? BEFORE

8-Did you reformat or leave windows on before update?(INSTALLED or ERASED or BRAND NEW) INSTALLED

9-Was Intel optimizer installed before or after update? (BEFORE or AFTER) AFTER

10-If optimizer was Installed BEFORE the update did you run it? (YES or NO) N/A

11-Bios and windows setting before update? (IDE, AHCI, RAID, etc) -list both if applicable AHCI

12-Setting during update? (IDE, AHCI, RAID, etc) IDE

13-Setting after update? (IDE, AHCI, RAID, etc) AHCI

14-Did you run the flasher twice before booting or installing Win7? (1x, 2x, 3x) 2X

15-Was there an error when flashing? (YES or NO) NO

16-After it flashed and the computer started (Windows was installed before update), did you immediately change the settings back to what it was before booting into windows? (YES or NO) YES

17-When booted in Windows 7 did it install a driver? (YES or NO) YES

18-Is the G2 drive bricked or working fine? (BAD or GOOD) GOOD

WIN 7 ULTIMATE

CORE I7 920ASROCK X58 EXTREME3GB OCZ VARIOUS OTHER HD

[idata]

To check alignment, run this command from a command line (Start, type "cmd" into the search box):

wmic partition get BlockSize, StartingOffset, Name, Index

This is what an unaligned partition looks like. Mine is a leftover from XP days, upgraded to Vista and then again to 7:

BlockSize Index Name StartingOffset

512 0 Disk # 0, Partition # 0 32256

That's a 63 sector offset. What were they thinking.

An aligned partition created by Vista / 7 will have an offset of 2048 blocks, which will look like this:

BlockSize Index Name StartingOffset

512 0 Disk # 0, Partition # 0 1048576

Running an SSD unaligned is throwing performance away. Your SSD "thinks" in 4k flash "pages". If you're mis-aligned, filesystem clusters span across flash pages, which means a very real performance hit.

Quick aside: For any SAN, Exchange Server, SQL Server, anything with a RAID, even when using HDDs, alignment is also crucial. The amount of admins that don't know this and throw 20-30% performance away is maddening. Give this to your sysadmin and have her take it to heart: http://blogs.msdn.com/jimmymay/archive/2009/05/08/disk-partition-alignment-sector-alignment-make-the...

I think the picture that is emerging so far seems to be the best explanation:

- TRIM or "Optimize" is causing the issue

- Only when the boot partition is misaligned

It'd be great to know whether those people with failures have misaligned partitions. Obviously, you can't run wmic when your drive's bricked. Or, let's say, only with "great difficulty". Here's a way to estimate:

1) Did Vista or Windows 7 create the partition that ended up on this drive? This would be the case if the Vista / Windows 7 install was run on a completely unpartitioned, unformatted drive; or if you opted to delete the existing XP partition during install. That latter action is very much user-driven - you'd know if you'd done that. If yes: Your partition is aligned.

2) Did Vista or Windows 7 "inherit" an XP partition? This would be the case if you did in-place upgrades to Vista and then Windows 7, or if you ran a fresh install of Vista / Windows 7 but did not instruct the installer to delete your existing XP partition. By default, the installer will re-use the existing partition, not delete it; it may also have created a "windows.old" folder. If yes: Your partition is unaligned.

[idata]

Just curious, how did you come to conclusion that misalignment is the cause of the problem? All posters but 1 in this thread report no problems.

Thanks for the neat way to check alignment.

[idata]

It's not a conclusion, and it's not mine. It's a premise put forth by another poster in the "Windows 7 and bricked drive after upgrade" thread. It'd make sense in that:

- We now see the issue in Vista after a few days

- Which means it's likely related to the freeing of blocks

- Which happens automatically in Win7 through TRIM, so that's where it's been seen first

- TRIM lets the drive know which file system clusters can be reclaimed. On an aligned drive and assuming default 4k NTFS clusters, each cluster corresponds neatly to 1 flash page. On an unaligned drive, it doesn't: Each cluster spans part of two adjacent flash pages. Lots of potential to get the freeing of flash pages in that scenario wrong, as you can imagine.

It's the best premise I've heard so far. Now it's a matter of collecting data. If we get an overwhelming number of people with bad drives reporting that their partitions were unaligned, and an overwhelming number of people with good drives reporting that their partitions are aligned, the premise would be strengthened.