info
Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
SSD freezes from time to time
81 views
Skip to first unread message
Yousuf Khan
Apr 3, 2012, 12:48:15 AM4/3/12
to
Okay, so I've installed my SSD as my boot disk. It's working
fantastically for the most part, but every now and again, I see the
C-drive just become 100% busy, and which in turns prevents other things
from functioning on the computer. The 100% busy periods don't last for
more than 10 seconds, and then it drops off immediately to 0% again.
I've tuned the SSD as well as possible according to the optimization
guides. I've turned off the disk indexing, I've moved the swapfile to
another disk, disk defragging is disabled for that drive, and I've
turned off prefetching, and system restore points. The only things I
haven't moved off of it are some of the high disk activity programs like
Thunderbird or Firefox, and that's because I'd actually like to have
those apps speeded up by the SSD, so I don't want to move them.
I have Resource Monitor loaded and I see that the Disk Activity goes
upto 100%, but the actual disk queue remains around 0.0, so it doesn't
seem like any particular application is actually creating the disk
activity, so I assume it must be the OS itself doing something.
Any ideas what else might need to be turned off?
Yousuf Khan
fantastically for the most part, but every now and again, I see the
C-drive just become 100% busy, and which in turns prevents other things
from functioning on the computer. The 100% busy periods don't last for
more than 10 seconds, and then it drops off immediately to 0% again.
I've tuned the SSD as well as possible according to the optimization
guides. I've turned off the disk indexing, I've moved the swapfile to
another disk, disk defragging is disabled for that drive, and I've
turned off prefetching, and system restore points. The only things I
haven't moved off of it are some of the high disk activity programs like
Thunderbird or Firefox, and that's because I'd actually like to have
those apps speeded up by the SSD, so I don't want to move them.
I have Resource Monitor loaded and I see that the Disk Activity goes
upto 100%, but the actual disk queue remains around 0.0, so it doesn't
seem like any particular application is actually creating the disk
activity, so I assume it must be the OS itself doing something.
Any ideas what else might need to be turned off?
Yousuf Khan
Rob
Apr 3, 2012, 7:46:52 AM4/3/12
to
Otherwise, this is going to be difficult to track down and you
might want to look at Process Explorer (now owned by ms) or
similar tools.
HTH
--
Rob
Yousuf Khan
Apr 3, 2012, 8:12:22 AM4/3/12
to
On 03/04/2012 7:46 AM, Rob wrote:
> Check that your SSD doesn't need a firmware update.
> Otherwise, this is going to be difficult to track down and you
> might want to look at Process Explorer (now owned by ms) or
> similar tools.
> HTH
Yeah, I checked that at the beginning before installing the os, it was
> Check that your SSD doesn't need a firmware update.
> Otherwise, this is going to be difficult to track down and you
> might want to look at Process Explorer (now owned by ms) or
> similar tools.
> HTH
already at the newest rev.
Yousuf Khan
Allen Drake
Apr 3, 2012, 3:53:25 PM4/3/12
to
Amazon and a new one arrived in a few days no charge. They even paid
for the shipping to RTM the original. Have you consulted the SSD maker
Forum?
Yousuf Khan
Apr 3, 2012, 6:47:39 PM4/3/12
to
On 03/04/2012 3:53 PM, Allen Drake wrote:
> I'd simply return it. Out of the 10 I purchased one was returned to
> Amazon and a new one arrived in a few days no charge. They even paid
> for the shipping to RTM the original. Have you consulted the SSD maker
> Forum?
No, that's an idea though.
> I'd simply return it. Out of the 10 I purchased one was returned to
> Amazon and a new one arrived in a few days no charge. They even paid
> for the shipping to RTM the original. Have you consulted the SSD maker
> Forum?
Yousuf Khan
Yousuf Khan
Apr 5, 2012, 11:35:33 PM4/5/12
to
On 03/04/2012 3:53 PM, Allen Drake wrote:
> I'd simply return it. Out of the 10 I purchased one was returned to
> Amazon and a new one arrived in a few days no charge. They even paid
> for the shipping to RTM the original. Have you consulted the SSD maker
> Forum?
Well, I did go to the forums, and then eventually I did exchange it. The
> Amazon and a new one arrived in a few days no charge. They even paid
> for the shipping to RTM the original. Have you consulted the SSD maker
> Forum?
Corsair forums are full of threads about this freezing problem, lots of
theories about how to fix it (upgrade the firmware, turn off services,
etc.). First of all, exchanging the drive didn't help at all. The second
drive had exactly the same problem.
Eventually, I discovered that the problem lay with the AHCI vs. IDE
drivers. Specifically, the AHCI drivers were the problem. I switched
back over to IDE mode in the BIOS, and the problems went away. One
reason I had it on the AHCI drivers was because I thought TRIM was only
supported through AHCI, but it turned out that TRIM works in IDE too. So
there's no reason to stick with AHCI at this point.
Yousuf Khan
Paul
Apr 6, 2012, 12:35:12 AM4/6/12
to
being sent. There is also an fsutil variable, indicating whether
it's enabled or not. I'd really be surprised if IDE supported it,
as the IDE drivers include crusty old drivers from 2002 or so that
don't know a thing about TRIM. I'd want some evidence it is really
being sent. They suggest emptying the trash bin as a potential trigger event.
http://www.ocztechnologyforum.com/forum/showthread.php?66696-New-FW-Flashing
Paul
YKhan
Apr 6, 2012, 3:27:39 AM4/6/12
to
On Apr 6, 12:35 am, Paul <nos...@needed.com> wrote:
Windows 7 didn't even come out till 2009, so it has much newer IDE
drivers. TRIM support didn't even appear until Win 7, so I'd be
inclined to agree that the 2002 XP drivers wouldn't know anything
about TRIM, but this is a new set of drivers, which are aware of SSD's
and TRIM.
Anyways, I'm not using Procmon, I'm using a utility called Hard Disk
Sentinel, which can monitor this sort of thing, and it confirms that
TRIM support is enabled. And I'm also using the fsutil command-line
utility that also confirms it. You can see how to use the "fsutil"
commands here: http://is.gd/rfYxxO.
> http://www.ocztechnologyforum.com/forum/showthread.php?66696-New-FW-F...
I don't really see what you're referring to in this website.
Yousuf Khan
> Yousuf Khan wrote:
> > Eventually, I discovered that the problem lay with the AHCI vs. IDE
> > drivers. Specifically, the AHCI drivers were the problem. I switched
> > back over to IDE mode in the BIOS, and the problems went away. One
> > reason I had it on the AHCI drivers was because I thought TRIM was only
> > supported through AHCI, but it turned out that TRIM works in IDE too. So
> > there's no reason to stick with AHCI at this point.
>
> > Eventually, I discovered that the problem lay with the AHCI vs. IDE
> > drivers. Specifically, the AHCI drivers were the problem. I switched
> > back over to IDE mode in the BIOS, and the problems went away. One
> > reason I had it on the AHCI drivers was because I thought TRIM was only
> > supported through AHCI, but it turned out that TRIM works in IDE too. So
> > there's no reason to stick with AHCI at this point.
>
> There is a claim here, that you can use ProcMon to verify TRIM is
> being sent. There is also an fsutil variable, indicating whether
> it's enabled or not. I'd really be surprised if IDE supported it,
> as the IDE drivers include crusty old drivers from 2002 or so that
> don't know a thing about TRIM. I'd want some evidence it is really
> being sent. They suggest emptying the trash bin as a potential trigger event.
The IDE drivers in XP might be the crusty old ones from 2002, but
> being sent. There is also an fsutil variable, indicating whether
> it's enabled or not. I'd really be surprised if IDE supported it,
> as the IDE drivers include crusty old drivers from 2002 or so that
> don't know a thing about TRIM. I'd want some evidence it is really
> being sent. They suggest emptying the trash bin as a potential trigger event.
Windows 7 didn't even come out till 2009, so it has much newer IDE
drivers. TRIM support didn't even appear until Win 7, so I'd be
inclined to agree that the 2002 XP drivers wouldn't know anything
about TRIM, but this is a new set of drivers, which are aware of SSD's
and TRIM.
Anyways, I'm not using Procmon, I'm using a utility called Hard Disk
Sentinel, which can monitor this sort of thing, and it confirms that
TRIM support is enabled. And I'm also using the fsutil command-line
utility that also confirms it. You can see how to use the "fsutil"
commands here: http://is.gd/rfYxxO.
> http://www.ocztechnologyforum.com/forum/showthread.php?66696-New-FW-F...
I don't really see what you're referring to in this website.
Yousuf Khan
Paul
Apr 6, 2012, 3:46:41 AM4/6/12
to
YKhan wrote:
>
>> http://www.ocztechnologyforum.com/forum/showthread.php?66696-New-FW-F...
>
> I don't really see what you're referring to in this website.
>
> Yousuf Khan
"Part 3 – In Depth Monitoring of SSD Metrics
>
>> http://www.ocztechnologyforum.com/forum/showthread.php?66696-New-FW-F...
>
> I don't really see what you're referring to in this website.
>
> Yousuf Khan
TRIM
Tracking ATA TRIM in Win 7 using Procmon:"
The idea is to see if a TRIM command of some sort,
is actually being sent when you empty the trash.
The advice in the past has always been, that TRIM
support existed in AHCI (like the MSAHCI driver)
but that it didn't exist in the IDE driver. That's
why I'm skeptical, that your discovery goes against
past advice. It would be reassuring to see something
like actual trace data from ProcMon, as a double check.
HTH,
Paul
YKhan
Apr 6, 2012, 9:19:41 AM4/6/12
to
out of Procmon. The fact of the matter is that the fsutil command
provided by Microsoft lets you change it on both the AHCI driver and
the IDE driver. If the flag wasn't supported on the IDE driver, then
it wouldn't allow you to change it, it would always send back
"disabled" when you query it.
And it's confirmed to be enabled and in use by a 3rd party util, Hard
Disk Sentinel. I've been using HDS for years to monitor the SMART
health of my hard drives, and when I first installed the SSD, it
immediately figured out that it was an SSD because it had special
SMART parameters that aren't available in HHD's.
http://www.hdsentinel.com/
Paul
Apr 6, 2012, 2:40:38 PM4/6/12
to
For Windows 7, and the built-in (MS) IDE driver.
http://www.anandtech.com/show/2829/15
(found the ref on this page, to that Anandtech link)
http://hardforum.com/showthread.php?t=1475177
Paul
Loren Pechtel
Apr 7, 2012, 12:04:07 AM4/7/12
to
same problem. I need the AHCI for RAID on the spinning storage,
though!
Paul
Apr 7, 2012, 2:05:43 AM4/7/12
to
*******
The problem with shopping for a separate controller, is finding a chip that
doesn't have issues. There is an example here, of one based on a Marvell 9128.
I could also find some cards with Asmedia chips on them, but I don't know right
off hand, whether they have all the drivers needed to do everything you need.
http://www.newegg.com/Product/Product.aspx?Item=N82E16815158207
http://ca.startech.com/Cards-Adapters/HDD-Controllers/SATA-Cards/2-Port-SATA-6-Gbps-PCI-Express-SATA-Controller-Card~PEXSAT32
Some of the cards, will have AHCI drivers as the default choice. And then
you need to verify in advance, you can get the driver flavor you need.
That particular card has a PCI Express x1 Rev.2 interface, so cannot transfer
data faster than 500MB/sec (minus whatever overhead is typical for such things).
It has the same kind of problem a SIL3132 would have, in terms of the match
between what is on the SATA side, versus what is on the PCI Express side.
The issue would be, if the thing supported something like RAID 0, and you
tried to pass 2 x SATA III rates through a 500MB/sec bottleneck. For
single standalone drives connected to the SATA ports, it's not going
to be nearly as much of a practical issue. (The PCI Express bus has
separate TX and RX bus, and copying from drive to drive for example,
uses separate buses at the instant the transfer happens. It's just RAID 0
configurations you have to watch, because it all goes through the RX or
all goes through the TX path.)
Some of those peripheral chips, have had a checkered history, in terms of
the grief they caused with early drivers. Which is why shopping for one
of those little cards, is not a trivial exercise.
In the past, the problem with early SATA cards, was the need to flash the
firmware, to change how the card behaves. It's possible there is still
a need to change the BIOS add-on module stored in the flash chip onboard,
to get the driver you want to work with them. What is stored in the flash
chip, is an Extended INT 0x13 routine, to enable the motherboard BIOS
to read boot info during POST and boot. But in some cases, the chip
configuration may be programmed with that code too (like, a bit that
says it is a RAID chip or is an IDE chip etc).
Paul
Yousuf Khan
Apr 7, 2012, 10:42:04 AM4/7/12
to
On 07/04/2012 12:04 AM, Loren Pechtel wrote:
> Argh! It's sounding more and more like my longstanding mystery is the
> same problem. I need the AHCI for RAID on the spinning storage,
> though!
Let me guess, you're using the Intel drivers, right?
> Argh! It's sounding more and more like my longstanding mystery is the
> same problem. I need the AHCI for RAID on the spinning storage,
> though!
Yousuf Khan
Loren Pechtel
Apr 7, 2012, 5:58:19 PM4/7/12
to
I tried flashing the drive, the ver2 to ver6 flash worked but the ver6
to ver7 is saying there's nothing for it to do.
YKhan
Apr 8, 2012, 3:00:37 PM4/8/12
to
of the Intel one? That way you can go back to standard Microsoft
drivers. The Microsoft drivers are standardized and they can even work
within Linux if you like. The Intel ones, are hit and miss.
Yousuf Khan
Loren Pechtel
Apr 8, 2012, 3:40:06 PM4/8/12
to
On Sun, 8 Apr 2012 12:00:37 -0700 (PDT), YKhan <yjk...@gmail.com>
wrote:
>Well, actually there is, why not use Microsoft's software RAID instead
>of the Intel one? That way you can go back to standard Microsoft
>drivers. The Microsoft drivers are standardized and they can even work
>within Linux if you like. The Intel ones, are hit and miss.
>
> Yousuf Khan
What driver are you talking about? How do I get RAID 1 out of some
spinning drives on the Intel controller that's on my motherboard?
wrote:
>Well, actually there is, why not use Microsoft's software RAID instead
>of the Intel one? That way you can go back to standard Microsoft
>drivers. The Microsoft drivers are standardized and they can even work
>within Linux if you like. The Intel ones, are hit and miss.
>
> Yousuf Khan
spinning drives on the Intel controller that's on my motherboard?
Yousuf Khan
Apr 9, 2012, 2:34:35 PM4/9/12
to
functionality. Microsoft's Disk Management has everything from simple
concatenation to RAID5 available, if you're using Windows 7 Ultimate or
Professional.
Also you should know that Intel's RAID is not something comes from its
controller, it is really a software RAID too. Microsoft's software RAID
is more compatible with more things.
Yousuf Khan
Loren Pechtel
Apr 10, 2012, 2:36:00 PM4/10/12
to
Yousuf Khan
Apr 10, 2012, 3:46:02 PM4/10/12
to
On 10/04/2012 2:36 PM, Loren Pechtel wrote:
> On Mon, 09 Apr 2012 14:34:35 -0400, Yousuf Khan
> On Mon, 09 Apr 2012 14:34:35 -0400, Yousuf Khan
>> Sorry, I shouldn't have said driver when talking about the RAID
>> functionality. Microsoft's Disk Management has everything from simple
>> concatenation to RAID5 available, if you're using Windows 7 Ultimate or
>> Professional.
>>
>> Also you should know that Intel's RAID is not something comes from its
>> controller, it is really a software RAID too. Microsoft's software RAID
>> is more compatible with more things.
>
> Any way to migrate?
Not without backing up and restoring, that I can think of.
>> functionality. Microsoft's Disk Management has everything from simple
>> concatenation to RAID5 available, if you're using Windows 7 Ultimate or
>> Professional.
>>
>> Also you should know that Intel's RAID is not something comes from its
>> controller, it is really a software RAID too. Microsoft's software RAID
>> is more compatible with more things.
>
> Any way to migrate?
Yousuf Khan
Yousuf Khan
Apr 12, 2012, 1:49:29 AM4/12/12
to
On 06/04/2012 2:40 PM, Paul wrote:
> OK, I can see a mention of the IDE driver having TRIM here.
> For Windows 7, and the built-in (MS) IDE driver.
>
> http://www.anandtech.com/show/2829/15
>
> (found the ref on this page, to that Anandtech link)
>
> http://hardforum.com/showthread.php?t=1475177
>
> Paul
Hold the presses! The actual reason for the problem has been pinpointed
> OK, I can see a mention of the IDE driver having TRIM here.
> For Windows 7, and the built-in (MS) IDE driver.
>
> http://www.anandtech.com/show/2829/15
>
> (found the ref on this page, to that Anandtech link)
>
> http://hardforum.com/showthread.php?t=1475177
>
> Paul
inside the AHCI drivers now. The problem lay in an obscure power
management feature in the AHCI drivers that don't exist in the IDE
drivers, called the HIPM/DIPM attributes. HIPM means Host-Initiated
Power Management, while DIPM is Device-Initiated ... etc. Since the IDE
driver doesn't touch these features, it doesn't have any problem with
them. To avoid the problem you need to disable these completely on the
AHCI drivers.
You need to set a key in the Registry to allow this feature to be
accessed by the Power Management advanced options. You need to go to the
following registry key:
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Power\PowerSettings\0012ee47-9041-4b5d-9b77-535fba8b1442\0b2d69d7-a2a1-449c-9680-f91c70521c60
And set the "Attributes" field to a DWORD of 0x2.
You can then follow this article on what to do to disable it:
http://www.sevenforums.com/tutorials/177819-ahci-link-power-management-enable-hipm-dipm.html
I followed the article, and switched back to the AHCI drivers, and no
longer had any freeze-up problems.
Yousuf Khan
Paul
Apr 12, 2012, 2:14:21 AM4/12/12
to
http://www.ocztechnologyforum.com/forum/showthread.php?88340-Vertex-3-Windows-install-in-AHCI-mode-on-Intel-6-series-Chipset-platforms
I suppose the thinking goes, "where would the fun be if
you didn't have to tweak something". :-(
You'd think the SSD manufacturers, could come up with an "SSD ready"
software, to pre-check these things for you. Instead of this week long
"search the web for hints" thing the customers have to do now. That
way, they could encapsulate all they've learned about their product
versus Windows or Linux.
Paul
Yousuf Khan
Apr 12, 2012, 10:48:08 AM4/12/12
to
On 12/04/2012 2:14 AM, Paul wrote:
> That reminds me of "LPM", which was an issue at one time.
>
> http://www.ocztechnologyforum.com/forum/showthread.php?88340-Vertex-3-Windows-install-in-AHCI-mode-on-Intel-6-series-Chipset-platforms
>
> I suppose the thinking goes, "where would the fun be if
> you didn't have to tweak something". :-(
>
> You'd think the SSD manufacturers, could come up with an "SSD ready"
> software, to pre-check these things for you. Instead of this week long
> "search the web for hints" thing the customers have to do now. That
> way, they could encapsulate all they've learned about their product
> versus Windows or Linux.
This does sound like a very similar issue. I noticed in the LPM issue
> That reminds me of "LPM", which was an issue at one time.
>
> http://www.ocztechnologyforum.com/forum/showthread.php?88340-Vertex-3-Windows-install-in-AHCI-mode-on-Intel-6-series-Chipset-platforms
>
> I suppose the thinking goes, "where would the fun be if
> you didn't have to tweak something". :-(
>
> You'd think the SSD manufacturers, could come up with an "SSD ready"
> software, to pre-check these things for you. Instead of this week long
> "search the web for hints" thing the customers have to do now. That
> way, they could encapsulate all they've learned about their product
> versus Windows or Linux.
one of the registry entries was for DIPM.
I suppose I could test to see which setting in particular caused the
issue, the HIPM or the DIPM, but having them both disabled fixed it, and
I don't think I'm that concerned about energy savings on a desktop, like
I would be on a laptop.
Yousuf Khan
YKhan
Apr 12, 2012, 3:47:42 PM4/12/12
to
On Apr 12, 2:14 am, Paul <nos...@needed.com> wrote:
> You'd think the SSD manufacturers, could come up with an "SSD ready"
> software, to pre-check these things for you. Instead of this week long
> "search the web for hints" thing the customers have to do now. That
> way, they could encapsulate all they've learned about their product
> versus Windows or Linux.
Also, I don't think the power issue is necessarily a problem with the
> You'd think the SSD manufacturers, could come up with an "SSD ready"
> software, to pre-check these things for you. Instead of this week long
> "search the web for hints" thing the customers have to do now. That
> way, they could encapsulate all they've learned about their product
> versus Windows or Linux.
SSD, it could be an issue with the motherboard chipset, or even
cabling between drive and motherboard for all we know. So it's not an
issue that affects all installations, just a large number of them.
Yousuf Khan
Rick
Apr 14, 2012, 10:57:15 AM4/14/12
to
Intel just released an update for their 320 SSD if that is the one you
have you might check it out
Yousuf Khan
Apr 14, 2012, 2:19:32 PM4/14/12
to
On 14/04/2012 10:57 AM, Rick wrote:
> Intel just released an update for their 320 SSD if that is the one you have you might check it out
Nope, that's not my drive, but the problem has been solved. It turns out
> Intel just released an update for their 320 SSD if that is the one you have you might check it out
that it was an obscure power management setting in the AHCI drivers
called HIPM-DIPM. Once those settings were turned off, the freeze-ups
went away.
Yousuf Khan
Loren Pechtel
May 3, 2012, 11:07:55 PM5/3/12
to
On Thu, 12 Apr 2012 01:49:29 -0400, Yousuf Khan
<bbb...@spammenot.yahoo.com> wrote:
>You need to set a key in the Registry to allow this feature to be
>accessed by the Power Management advanced options. You need to go to the
>following registry key:
>
>HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Power\PowerSettings\0012ee47-9041-4b5d-9b77-535fba8b1442\0b2d69d7-a2a1-449c-9680-f91c70521c60
>
>And set the "Attributes" field to a DWORD of 0x2.
>
>You can then follow this article on what to do to disable it:
>
>http://www.sevenforums.com/tutorials/177819-ahci-link-power-management-enable-hipm-dipm.html
>
>I followed the article, and switched back to the AHCI drivers, and no
>longer had any freeze-up problems.
>
> Yousuf Khan
I wish that was the answer for me.
<bbb...@spammenot.yahoo.com> wrote:
>You need to set a key in the Registry to allow this feature to be
>accessed by the Power Management advanced options. You need to go to the
>following registry key:
>
>HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Power\PowerSettings\0012ee47-9041-4b5d-9b77-535fba8b1442\0b2d69d7-a2a1-449c-9680-f91c70521c60
>
>And set the "Attributes" field to a DWORD of 0x2.
>
>You can then follow this article on what to do to disable it:
>
>http://www.sevenforums.com/tutorials/177819-ahci-link-power-management-enable-hipm-dipm.html
>
>I followed the article, and switched back to the AHCI drivers, and no
>longer had any freeze-up problems.
>
> Yousuf Khan
The 0b2d... section of the key doesn't exist on my box.
The link shows an option that's not on my system, either.
Yousuf Khan
May 4, 2012, 4:10:33 AM5/4/12
to
"0b2d...1c60" key did exist, the "Attributes" field underneath it didn't
exist. So I created that one. Now, everything is fine. You'll simply
have to create the key and then the field too.
It seems that not all registry entries are created by default on all
machines, and are only created on an as needed basis, otherwise default
values are taken.
Yousuf Khan
Yousuf Khan
May 4, 2012, 4:15:12 AM5/4/12
to
On 04/05/2012 4:10 AM, Yousuf Khan wrote:
>> I wish that was the answer for me.
>>
>> The 0b2d... section of the key doesn't exist on my box.
>>
>> The link shows an option that's not on my system, either.
>
> If it doesn't exist, then create it. In my case, though that
> "0b2d...1c60" key did exist, the "Attributes" field underneath it didn't
> exist. So I created that one. Now, everything is fine. You'll simply
> have to create the key and then the field too.
>
> It seems that not all registry entries are created by default on all
> machines, and are only created on an as needed basis, otherwise default
> values are taken.
>
> Yousuf Khan
BTW, this is what my registry looks like in that section (you can copy
>> I wish that was the answer for me.
>>
>> The 0b2d... section of the key doesn't exist on my box.
>>
>> The link shows an option that's not on my system, either.
>
> If it doesn't exist, then create it. In my case, though that
> "0b2d...1c60" key did exist, the "Attributes" field underneath it didn't
> exist. So I created that one. Now, everything is fine. You'll simply
> have to create the key and then the field too.
>
> It seems that not all registry entries are created by default on all
> machines, and are only created on an as needed basis, otherwise default
> values are taken.
>
> Yousuf Khan
and paste):
[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Power\PowerSettings\0012ee47-9041-4b5d-9b77-535fba8b1442\0b2d69d7-a2a1-449c-9680-f91c70521c60]
"FriendlyName"=hex(2):41,00,48,00,43,00,49,00,20,00,4c,00,69,00,6e,00,6b,00,20,\
00,50,00,6f,00,77,00,65,00,72,00,20,00,4d,00,61,00,6e,00,61,00,67,00,65,00,\
6d,00,65,00,6e,00,74,00,20,00,2d,00,20,00,48,00,49,00,50,00,4d,00,2f,00,44,\
00,49,00,50,00,4d,00,00,00
"Description"=hex(2):43,00,6f,00,6e,00,66,00,69,00,67,00,75,00,72,00,65,00,73,\
00,20,00,74,00,68,00,65,00,20,00,4c,00,50,00,4d,00,20,00,73,00,74,00,61,00,\
74,00,65,00,2e,00,00,00
"IconResource"=hex(2):00,00
"Attributes"=dword:00000000
Paul
May 4, 2012, 10:30:50 AM5/4/12
to
A person seeing the above, might conclude it was "complicated and scary", when
in fact that could be a device name, a file path name, and so on.
For fun, let's translate this much. There are ten non-zero items, and
those correspond to ten ASCII letters/numbers.
41,00,48,00,43,00,49,00,20,00,4c,00,69,00,6e,00,6b,00,20 = "AHCI Link "
You can get an ASCII table from a UNIX man page, to finish the translation
process, and understand what is stored there. Unicode encoding is a nuisance,
but it's here to stay. The 41 is 0x41 hex, so you use the hex column when
translating to ASCII.
http://unixhelp.ed.ac.uk/CGI/man-cgi?ascii+7
Paul
Loren Pechtel
May 4, 2012, 11:55:29 AM5/4/12
to
rigs--such stuff won't be so scary.
Not to mention that I have written code professionally for more than
20 years. It's not one bit scary.
Loren Pechtel
May 4, 2012, 11:55:29 AM5/4/12
to
On Fri, 04 May 2012 04:10:33 -0400, Yousuf Khan
>It seems that not all registry entries are created by default on all
>machines, and are only created on an as needed basis, otherwise default
>values are taken.
The thing is the AHCI settings in the URL don't show up on my machine
either.
<bbb...@spammenot.yahoo.com> wrote:
>On 03/05/2012 11:07 PM, Loren Pechtel wrote:
>> On Thu, 12 Apr 2012 01:49:29 -0400, Yousuf Khan
>> <bbb...@spammenot.yahoo.com> wrote:
>>
>>> You need to set a key in the Registry to allow this feature to be
>>> accessed by the Power Management advanced options. You need to go to the
>>> following registry key:
>>>
>>> HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Power\PowerSettings\0012ee47-9041-4b5d-9b77-535fba8b1442\0b2d69d7-a2a1-449c-9680-f91c70521c60
>>>
>>> And set the "Attributes" field to a DWORD of 0x2.
>>>
>>> You can then follow this article on what to do to disable it:
>>>
>>> http://www.sevenforums.com/tutorials/177819-ahci-link-power-management-enable-hipm-dipm.html
>>>
>>> I followed the article, and switched back to the AHCI drivers, and no
>>> longer had any freeze-up problems.
>>>
>>> Yousuf Khan
>>
>> I wish that was the answer for me.
>>
>> The 0b2d... section of the key doesn't exist on my box.
>>
>> The link shows an option that's not on my system, either.
>
>If it doesn't exist, then create it. In my case, though that
>"0b2d...1c60" key did exist, the "Attributes" field underneath it didn't
>exist. So I created that one. Now, everything is fine. You'll simply
>have to create the key and then the field too.
Created, I'll see what happens. I *HATE* these blue-screens!
>On 03/05/2012 11:07 PM, Loren Pechtel wrote:
>> On Thu, 12 Apr 2012 01:49:29 -0400, Yousuf Khan
>> <bbb...@spammenot.yahoo.com> wrote:
>>
>>> You need to set a key in the Registry to allow this feature to be
>>> accessed by the Power Management advanced options. You need to go to the
>>> following registry key:
>>>
>>> HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Power\PowerSettings\0012ee47-9041-4b5d-9b77-535fba8b1442\0b2d69d7-a2a1-449c-9680-f91c70521c60
>>>
>>> And set the "Attributes" field to a DWORD of 0x2.
>>>
>>> You can then follow this article on what to do to disable it:
>>>
>>> http://www.sevenforums.com/tutorials/177819-ahci-link-power-management-enable-hipm-dipm.html
>>>
>>> I followed the article, and switched back to the AHCI drivers, and no
>>> longer had any freeze-up problems.
>>>
>>> Yousuf Khan
>>
>> I wish that was the answer for me.
>>
>> The 0b2d... section of the key doesn't exist on my box.
>>
>> The link shows an option that's not on my system, either.
>
>If it doesn't exist, then create it. In my case, though that
>"0b2d...1c60" key did exist, the "Attributes" field underneath it didn't
>exist. So I created that one. Now, everything is fine. You'll simply
>have to create the key and then the field too.
>It seems that not all registry entries are created by default on all
>machines, and are only created on an as needed basis, otherwise default
>values are taken.
either.
Paul
May 4, 2012, 12:24:04 PM5/4/12
to
Are you and Yousef using the same chipset ?
Can you search in Regedit, for a Unicode string, like "ACPI Link" etc ?
That's what I don't like about Unicode, is there aren't necessarily good ways
of doing searches. In one case, I had to write my own program, to do
such a search, when debugging a problem.
Paul
Yousuf Khan
May 4, 2012, 3:25:13 PM5/4/12
to
On 04/05/2012 10:30 AM, Paul wrote:
exported it from the regedit as a *.reg file. But when I was viewing it
in the regedit itself it was simply text.
Yousuf Khan
> Yousuf Khan wrote:
>> BTW, this is what my registry looks like in that section (you can copy
>> and paste):
>>
>> [HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Power\PowerSettings\0012ee47-9041-4b5d-9b77-535fba8b1442\0b2d69d7-a2a1-449c-9680-f91c70521c60]
>>
>> "FriendlyName"=hex(2):41,00,48,00,43,00,49,00,20,00,4c,00,69,00,6e,00,6b,00,20,\
>>
>>
>> 00,50,00,6f,00,77,00,65,00,72,00,20,00,4d,00,61,00,6e,00,61,00,67,00,65,00,\
>>
>>
>> 6d,00,65,00,6e,00,74,00,20,00,2d,00,20,00,48,00,49,00,50,00,4d,00,2f,00,44,\
>>
>> 00,49,00,50,00,4d,00,00,00
>> "Description"=hex(2):43,00,6f,00,6e,00,66,00,69,00,67,00,75,00,72,00,65,00,73,\
>>
>>
>> 00,20,00,74,00,68,00,65,00,20,00,4c,00,50,00,4d,00,20,00,73,00,74,00,61,00,\
>>
>> 74,00,65,00,2e,00,00,00
>> "IconResource"=hex(2):00,00
>> "Attributes"=dword:00000000
>
> An alternating sequence of 00 and non 00 hex value, is a Unicode text string.
> A person seeing the above, might conclude it was "complicated and scary", when
> in fact that could be a device name, a file path name, and so on.
>
> For fun, let's translate this much. There are ten non-zero items, and
> those correspond to ten ASCII letters/numbers.
>
> 41,00,48,00,43,00,49,00,20,00,4c,00,69,00,6e,00,6b,00,20 = "AHCI Link "
Yeah, I know I noticed that too, but that's how it got saved when I
>> BTW, this is what my registry looks like in that section (you can copy
>> and paste):
>>
>> [HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Power\PowerSettings\0012ee47-9041-4b5d-9b77-535fba8b1442\0b2d69d7-a2a1-449c-9680-f91c70521c60]
>>
>> "FriendlyName"=hex(2):41,00,48,00,43,00,49,00,20,00,4c,00,69,00,6e,00,6b,00,20,\
>>
>>
>> 00,50,00,6f,00,77,00,65,00,72,00,20,00,4d,00,61,00,6e,00,61,00,67,00,65,00,\
>>
>>
>> 6d,00,65,00,6e,00,74,00,20,00,2d,00,20,00,48,00,49,00,50,00,4d,00,2f,00,44,\
>>
>> 00,49,00,50,00,4d,00,00,00
>> "Description"=hex(2):43,00,6f,00,6e,00,66,00,69,00,67,00,75,00,72,00,65,00,73,\
>>
>>
>> 00,20,00,74,00,68,00,65,00,20,00,4c,00,50,00,4d,00,20,00,73,00,74,00,61,00,\
>>
>> 74,00,65,00,2e,00,00,00
>> "IconResource"=hex(2):00,00
>> "Attributes"=dword:00000000
>
> An alternating sequence of 00 and non 00 hex value, is a Unicode text string.
> A person seeing the above, might conclude it was "complicated and scary", when
> in fact that could be a device name, a file path name, and so on.
>
> For fun, let's translate this much. There are ten non-zero items, and
> those correspond to ten ASCII letters/numbers.
>
> 41,00,48,00,43,00,49,00,20,00,4c,00,69,00,6e,00,6b,00,20 = "AHCI Link "
exported it from the regedit as a *.reg file. But when I was viewing it
in the regedit itself it was simply text.
Yousuf Khan
Yousuf Khan
May 4, 2012, 3:36:25 PM5/4/12
to
On 04/05/2012 12:24 PM, Paul wrote:
using the IDE drivers, these power management fields don't show up,
because that sort of power management didn't exist in that generation of
the ATA specifications, i.e. version 7, so the registry entries for
power management are simply ignored. But when you go upto the newer
Microsoft AHCI drivers, which is ATA version 8, these specs include this
power management, so if the registry entries exist, and there is a value
set for them, then it'll take effect.
On the other hand, if he's using something other than the standard
Microsoft drivers, such as the Intel drivers, then these entries may or
may not have any effect. The original webpage I linked to mentioned that
these settings don't have any effect except with Intel ICH9M drivers and
later.
Yousuf Khan
> Loren Pechtel wrote:
>> The thing is the AHCI settings in the URL don't show up on my machine
>> either.
>
> Any chance the GUIDs are specific to brands of chipsets ?
> Are you and Yousef using the same chipset ?
>
> Can you search in Regedit, for a Unicode string, like "ACPI Link" etc ?
> That's what I don't like about Unicode, is there aren't necessarily good ways
> of doing searches. In one case, I had to write my own program, to do
> such a search, when debugging a problem.
>
> Paul
I'm simply using the standard Microsoft AHCI and/or IDE drivers. When
>> The thing is the AHCI settings in the URL don't show up on my machine
>> either.
>
> Any chance the GUIDs are specific to brands of chipsets ?
> Are you and Yousef using the same chipset ?
>
> Can you search in Regedit, for a Unicode string, like "ACPI Link" etc ?
> That's what I don't like about Unicode, is there aren't necessarily good ways
> of doing searches. In one case, I had to write my own program, to do
> such a search, when debugging a problem.
>
> Paul
using the IDE drivers, these power management fields don't show up,
because that sort of power management didn't exist in that generation of
the ATA specifications, i.e. version 7, so the registry entries for
power management are simply ignored. But when you go upto the newer
Microsoft AHCI drivers, which is ATA version 8, these specs include this
power management, so if the registry entries exist, and there is a value
set for them, then it'll take effect.
On the other hand, if he's using something other than the standard
Microsoft drivers, such as the Intel drivers, then these entries may or
may not have any effect. The original webpage I linked to mentioned that
these settings don't have any effect except with Intel ICH9M drivers and
later.
Yousuf Khan
Yousuf Khan
May 4, 2012, 3:46:18 PM5/4/12
to
On 04/05/2012 11:55 AM, Loren Pechtel wrote:
> The thing is the AHCI settings in the URL don't show up on my machine
> either.
If you've taken my registry settings, then chances are that you won't
> The thing is the AHCI settings in the URL don't show up on my machine
> either.
see the link power management options show up, because I've set the
"Attributes" field to a value of zero. This setting should disable the
link power management feature for all power schemes, by default. So you
*won't* see those settings in the power management settings.
However, if you do want to see those settings, then you'll have to set
the Attributes field to a value of 2, and then all of those settings
will show up, provided you're using the Microsoft AHCI drivers for your
disks. But then you'll have to edit and disable the power management in
each of your power schemes individually. That's because after you set
the Attributes field to 2, the link power management then becomes a
local setting for each power scheme, rather than a global setting for
all of them.
Yousuf Khan
Loren Pechtel
May 4, 2012, 8:28:37 PM5/4/12
to
some screenshots. Those settings don't show up here.
While I do agree it's likely chipset related I don't think it's just a
matter of different keys.
Loren Pechtel
May 4, 2012, 8:28:37 PM5/4/12
to
On Fri, 04 May 2012 15:36:25 -0400, Yousuf Khan
<bbb...@spammenot.yahoo.com> wrote:
>On the other hand, if he's using something other than the standard
>Microsoft drivers, such as the Intel drivers, then these entries may or
>may not have any effect. The original webpage I linked to mentioned that
>these settings don't have any effect except with Intel ICH9M drivers and
>later.
I'm using the Intel driver--it's what came with the motherboard, I
<bbb...@spammenot.yahoo.com> wrote:
>On the other hand, if he's using something other than the standard
>Microsoft drivers, such as the Intel drivers, then these entries may or
>may not have any effect. The original webpage I linked to mentioned that
>these settings don't have any effect except with Intel ICH9M drivers and
>later.
didn't even know there was a useable Microsoft option.
Loren Pechtel
May 4, 2012, 8:28:37 PM5/4/12
to
Bob H
May 5, 2012, 10:55:10 AM5/5/12
to
was a doddle to install and setup.
After I connected it to the main board, I switched on my pc, went into
the BIOS to set the boot order, then restarted with the windows 7 dvd in
the dvd rom drive.
After a brand new installation of windows 7, with AHCI already set and
used previously, windows now boots up to the desktop in under 30 seconds.
I had hardly anything to set and change apart from disabling a couple of
things which samsung ssd magician software recommended.
After that it was just a matter of following a couple of other
recommedations I found on the net, and all is honky dory now.
Yousuf Khan
May 5, 2012, 1:00:03 PM5/5/12
to
RAID or anything like that.
Yousuf Khan
Yousuf Khan
May 5, 2012, 1:02:44 PM5/5/12
to
On 04/05/2012 8:28 PM, Loren Pechtel wrote:
> They were missing before I made the registry changes.
That doesn't matter.
Yousuf Khan
Loren Pechtel
May 5, 2012, 1:28:47 PM5/5/12
to
On Sat, 05 May 2012 13:00:03 -0400, Yousuf Khan
<bbb...@spammenot.yahoo.com> wrote:
>On 04/05/2012 8:28 PM, Loren Pechtel wrote:
>> On Fri, 04 May 2012 15:36:25 -0400, Yousuf Khan
>> <bbb...@spammenot.yahoo.com> wrote:
>>
>>> On the other hand, if he's using something other than the standard
>>> Microsoft drivers, such as the Intel drivers, then these entries may or
>>> may not have any effect. The original webpage I linked to mentioned that
>>> these settings don't have any effect except with Intel ICH9M drivers and
>>> later.
>>
>> I'm using the Intel driver--it's what came with the motherboard, I
>> didn't even know there was a useable Microsoft option.
>
>Try switching over to the Microsoft drivers, if you're not using Intel
>RAID or anything like that.
I am using the RAID. Sys is a SSD, data is RAID 1.
>On 04/05/2012 8:28 PM, Loren Pechtel wrote:
>> On Fri, 04 May 2012 15:36:25 -0400, Yousuf Khan
>> <bbb...@spammenot.yahoo.com> wrote:
>>
>>> On the other hand, if he's using something other than the standard
>>> Microsoft drivers, such as the Intel drivers, then these entries may or
>>> may not have any effect. The original webpage I linked to mentioned that
>>> these settings don't have any effect except with Intel ICH9M drivers and
>>> later.
>>
>> I'm using the Intel driver--it's what came with the motherboard, I
>> didn't even know there was a useable Microsoft option.
>
>Try switching over to the Microsoft drivers, if you're not using Intel
>RAID or anything like that.
Yousuf Khan
May 5, 2012, 7:12:00 PM5/5/12
to
On 05/05/2012 1:28 PM, Loren Pechtel wrote:
>> Try switching over to the Microsoft drivers, if you're not using Intel
>> RAID or anything like that.
>
> I am using the RAID. Sys is a SSD, data is RAID 1.
Then you have no choice, but to keep using it. You could of course,
>> Try switching over to the Microsoft drivers, if you're not using Intel
>> RAID or anything like that.
>
> I am using the RAID. Sys is a SSD, data is RAID 1.
break the mirror and mirror it instead through the Microsoft dynamic
disk system.
Yousuf Khan
Loren Pechtel
May 15, 2012, 7:36:21 PM5/15/12
to
work.
Loren Pechtel
Jun 23, 2012, 12:07:51 AM6/23/12
to
dynamic instead of using the Intel mirroring software. Any
suggestions on how to install the Microsoft drivers without having to
redo my box?
Yousuf Khan
Jun 23, 2012, 1:49:22 AM6/23/12
to
use them in RAID mode. The Microsoft dynamic disks feature works at a
level just above the device drivers, so you can use Microsoft dynamic
disks alongside Intel drivers. But as I said, you'll have to first break
the Intel mirroring first.
Yousuf Khan
Loren Pechtel
Jun 23, 2012, 1:36:30 PM6/23/12
to
On Sat, 23 Jun 2012 11:49:22 +0600, Yousuf Khan
<bbb...@spammenot.yahoo.com> wrote:
>I think you can continue to use the Intel drivers, just don't have to
>use them in RAID mode. The Microsoft dynamic disks feature works at a
>level just above the device drivers, so you can use Microsoft dynamic
>disks alongside Intel drivers. But as I said, you'll have to first break
>the Intel mirroring first.
>
> Yousuf Khan
The Intel mirroring is broken, it doesn't help with the blue screens.
<bbb...@spammenot.yahoo.com> wrote:
>I think you can continue to use the Intel drivers, just don't have to
>use them in RAID mode. The Microsoft dynamic disks feature works at a
>level just above the device drivers, so you can use Microsoft dynamic
>disks alongside Intel drivers. But as I said, you'll have to first break
>the Intel mirroring first.
>
> Yousuf Khan
Throwing that much data around really pushed it, I had multiple blue
screens copying it back.
YKhan
Jun 24, 2012, 6:41:48 PM6/24/12
to
switching to them.
Loren Pechtel
Jun 25, 2012, 4:20:26 PM6/25/12
to
On Sun, 24 Jun 2012 15:41:48 -0700 (PDT), YKhan <yjk...@gmail.com>
wrote:
What driver is that, though?
wrote:
Yousuf Khan
Jun 26, 2012, 2:38:04 PM6/26/12
to
called pciide.sys. Both located in Windows system folders.
Yousuf Khan
0 new messages