![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
Asus P8P67 Pro SSD issues
- Thread starter ochadd
- Start date
I didn't do a Win7 reinstall when I switched from X58 to P67, and my Intel SSD performance is still fine. I did go into device manager and clean up all the old X58 stuff that I could find. Google devmgr_show_nonpresent_devices to see how to find the old stuff...careful though. Maybe that will help get some improvement without a complete reinstall, or maybe I was just lucky.
Ok, I'll take your word for it, random internet guy, when every other resource and guide I've ever seen says not (sometimes even NOT in all caps, just for emphasis) to run Defrag on one.
Or, you know, maybe I won't. At all.
A quote from another post here on [H]:
Or, you know, maybe I won't. At all.
A quote from another post here on [H]:
blue.chord2
n00b
- Joined
- Jan 7, 2011
- Messages
- 2
Vipera66 flash drives do not store memory the same way as magnetic hard drives do.
In a magnetic hard drive windows will sometimes store 1 file in several "fragments" - they are not stored contiguously. This increases hard drive seek time and thus decreases performance. Defrag aims to put these pieces closer to one another so the hard drive does not have to move as much to access one file.
Flash memory has a different architecture. They do not have moving parts- yes there is still "seek time" but in this case it is delay from hardware signal relay, and its something almost impossible to notice (>.1 ms). Flash also stores memory in individual memory cells, and these cells have a finite amount of program-erase cycles before they burn out.
Thus there is no reason to defragment a ssd. The only thing defragging your ssd will do is burning out the flash cells more quickly. HOWEVER- there are 2 methods (that i know of) for restoring your SSD to its original "out of box" speed- one of these is secure erasing the drive and the other is re-imaging it. I would recommend against doing either of these as frequent SE/re-imaging of a SSD could burn it out within months - with that said you can do your research and see if its worth it.
In a magnetic hard drive windows will sometimes store 1 file in several "fragments" - they are not stored contiguously. This increases hard drive seek time and thus decreases performance. Defrag aims to put these pieces closer to one another so the hard drive does not have to move as much to access one file.
Flash memory has a different architecture. They do not have moving parts- yes there is still "seek time" but in this case it is delay from hardware signal relay, and its something almost impossible to notice (>.1 ms). Flash also stores memory in individual memory cells, and these cells have a finite amount of program-erase cycles before they burn out.
Thus there is no reason to defragment a ssd. The only thing defragging your ssd will do is burning out the flash cells more quickly. HOWEVER- there are 2 methods (that i know of) for restoring your SSD to its original "out of box" speed- one of these is secure erasing the drive and the other is re-imaging it. I would recommend against doing either of these as frequent SE/re-imaging of a SSD could burn it out within months - with that said you can do your research and see if its worth it.
Ok, so I *did* have this same issue, but my problem was solved entirely by either that Intel Rapid Storage software install or some windows updates related to solid state storage larger than 32gb. I'm not sure which, as both got installed before the restart that brought the fixed performance.
Last edited:
This is pretty much what the official SSD Guide from Intel states. In fact Win7 disables auto defrag on SSD's by default. If you check the Defrag's schedule dialog you'll see that hard drives are scheduled for defrag early in the AM by default, this is not the case for SSD's.
Last edited:
We have finished testing every SSD currently on the market. JJ and I will have some additional information up shortly but here are the quick results.
1. As suggested in the tip and tricks and guide section, if you are having subpar performance from your SSD, please uninstall Intel RST 10.1.1088 until the new firmware and driver set is released. There have been some problems with it depending on drive configurations and loads. Also, the inbox Win7 driver (AHCI) is generally a tad faster than the Intel RST 10.0 driver set on P67 at this point.
2. If items in number 1 have been completed and the drive is still not up to task, please ensure you completed a clean install of the OS as we have noticed with several users that tried to bring over their AM3 or S775 OS install that it created numerous problems, not just drive speeds but other items also. In addition, even moving from X58/P55 can cause similar problems. I know this is touchy but a lot of support calls were solved by having the user not install a hacked OS.
Properly clear/clean the SSD before the reinstall to also improve performance.
3. Disabling C-States on this platform can actually decrease drive performance in several situations. A complete opposite from previous platforms. Also ensure the latest Intel INF and MEI driver are loaded.
1. As suggested in the tip and tricks and guide section, if you are having subpar performance from your SSD, please uninstall Intel RST 10.1.1088 until the new firmware and driver set is released. There have been some problems with it depending on drive configurations and loads. Also, the inbox Win7 driver (AHCI) is generally a tad faster than the Intel RST 10.0 driver set on P67 at this point.
2. If items in number 1 have been completed and the drive is still not up to task, please ensure you completed a clean install of the OS as we have noticed with several users that tried to bring over their AM3 or S775 OS install that it created numerous problems, not just drive speeds but other items also. In addition, even moving from X58/P55 can cause similar problems. I know this is touchy but a lot of support calls were solved by having the user not install a hacked OS.
Properly clear/clean the SSD before the reinstall to also improve performance. 3. Disabling C-States on this platform can actually decrease drive performance in several situations. A complete opposite from previous platforms. Also ensure the latest Intel INF and MEI driver are loaded.
Old Hippie
Supreme [H]ardness
- Joined
- Oct 31, 2005
- Messages
- 6,013
LOL!
ochadd
[H]ard|Gawd
- Joined
- May 9, 2008
- Messages
- 1,317
Plot thickens as of last night. I'll preface saying that I never managed to get the drive back to the state it was before the change in motherboards.
Decided to pick up another identical X25-m g2 and move to RAID 0. The SATA settings were already set to RAID and the original SSD and my secondary hard drive were detected in the Intel RAID manager screen. The original SSD was plugged into Port 0 of the grey SATA 6GBps port, second hard drive and DVD drive plugged into the blue SATA 3GBps ports
* Imaged the old SSD to an external hard drive using Clonezilla.
* Shutdown.
* Plugged the new SSD into port 1 using the other grey SATA port and the SATA 6GBps cable bundled with the mobo.
* Both SSDs detected in BIOS on the correct ports.
* Saved settings and exited BIOS
* Only the original SSD and secondary hard drive detected in the Intel RAID manager.
* Booted to Windows 7.
* Both disks show up in Disk Manager. The second SSD showed a "Drive Signature Conflict". Initialized the new disk, created a volume and made sure I could write files to it.
* Verified it was still listed in BIOS and again it was not detected in pre-boot Intel RAID manager.
* Shutdown and moved both SSDs over to blue SATA 3GBps and still only 1 showed up in RAID manager.
* Genius that I am I came to the conclusion that the Intel RAID manager was only listing one because they were the exact same part, revision, etc.
* Created a RAID volume and it didn't give me a choice in disks. Proposed a 149GB RAID 1 set and I accepted. Nuked my friggen data drive...
* 16 hours of DR later my critical data is back and there is still only a single disk in the preboot Intel RAID manager.
How the heck do I get it to show up? BIOS sees it, Windows sees it, Linux distro I used to recover files can read and write to both of them.
Decided to pick up another identical X25-m g2 and move to RAID 0. The SATA settings were already set to RAID and the original SSD and my secondary hard drive were detected in the Intel RAID manager screen. The original SSD was plugged into Port 0 of the grey SATA 6GBps port, second hard drive and DVD drive plugged into the blue SATA 3GBps ports
* Imaged the old SSD to an external hard drive using Clonezilla.
* Shutdown.
* Plugged the new SSD into port 1 using the other grey SATA port and the SATA 6GBps cable bundled with the mobo.
* Both SSDs detected in BIOS on the correct ports.
* Saved settings and exited BIOS
* Only the original SSD and secondary hard drive detected in the Intel RAID manager.
* Booted to Windows 7.
* Both disks show up in Disk Manager. The second SSD showed a "Drive Signature Conflict". Initialized the new disk, created a volume and made sure I could write files to it.
* Verified it was still listed in BIOS and again it was not detected in pre-boot Intel RAID manager.
* Shutdown and moved both SSDs over to blue SATA 3GBps and still only 1 showed up in RAID manager.
* Genius that I am I came to the conclusion that the Intel RAID manager was only listing one because they were the exact same part, revision, etc.
* Created a RAID volume and it didn't give me a choice in disks. Proposed a 149GB RAID 1 set and I accepted. Nuked my friggen data drive...
* 16 hours of DR later my critical data is back and there is still only a single disk in the preboot Intel RAID manager.
How the heck do I get it to show up? BIOS sees it, Windows sees it, Linux distro I used to recover files can read and write to both of them.
ochadd
[H]ard|Gawd
- Joined
- May 9, 2008
- Messages
- 1,317
Finally got the RAID 0 to work. Needed to unplug, fully shutdown, reboot, shutdown, plug in, startup and then they were detected in the SATA 3GBps ports. Also updated to the 1204 bios.
This is a fresh install of Windows 7. Top image is without write back cache enabled, forgot. Bottom is with it enabled. These look ok?
This is a fresh install of Windows 7. Top image is without write back cache enabled, forgot. Bottom is with it enabled. These look ok?
Last edited:
Been following this thread for a while now. Man this board is just plain awful from all the horror stories I have been reading on the net about it with memory problems, drive detection issues, cold boot problems, etc, etc, etc. While I do appreciate Asus' support, I have to wonder what they were thinking releasing their new boards in this early beta state? These things are just awful! I hope you can work out your issues chadd. For me, I will probably be going with the Gigabyte or Biostar P67 offering. Asus: get your shit together!
All the P67 boards have problems. If you want guaranteed stability don't build one for a few months after all the issues have been worked out. There may even be new board revisions out then if the problems are hardware related.
I never buy new technology right after they come out, mostly I am talking about computer' part, now days the competition between each manufacture is so big that they just release product that are not fully check or finish, a lot of time if you wait a while they will come up with better BIOS with revisions.
dandragonrage
[H]F Junkie
- Joined
- Jun 5, 2004
- Messages
- 8,298
I have had zero problems with my setup.
I should have said all the manufacturers of P67 boards have had problems. There are likely a majority of people having no problems with any boards. I doubt Asus has significantly greater problems than any other, and the fast bios updates are addressing a lot of what is happening. I hope mine comes together with no problems like yours. Just waiting on the memory and a new HD. I like to install on a new HD and set aside the old one in case I have serious problems.
This may be your issue. Looks like there is a silicon/hardware issue with the chipsets:
http://www.intc.com/releasedetail.cfm?ReleaseID=546454&ReleasesType=Financial%20News
"SANTA CLARA, Calif.--(BUSINESS WIRE)-- As part of ongoing quality assurance, Intel Corporation has discovered a design issue in a recently released support chip, the Intel® 6 Series, code-named Cougar Point, and has implemented a silicon fix. In some cases, the Serial-ATA (SATA) ports within the chipsets may degrade over time, potentially impacting the performance or functionality of SATA-linked devices such as hard disk drives and DVD-drives."
http://www.intc.com/releasedetail.cfm?ReleaseID=546454&ReleasesType=Financial%20News
"SANTA CLARA, Calif.--(BUSINESS WIRE)-- As part of ongoing quality assurance, Intel Corporation has discovered a design issue in a recently released support chip, the Intel® 6 Series, code-named Cougar Point, and has implemented a silicon fix. In some cases, the Serial-ATA (SATA) ports within the chipsets may degrade over time, potentially impacting the performance or functionality of SATA-linked devices such as hard disk drives and DVD-drives."
ochadd
[H]ard|Gawd
- Joined
- May 9, 2008
- Messages
- 1,317
It looks like I could have wrenched on it forever without getting it back where it was. Wonder how Asus and/or Newegg is going to handle this.
dandragonrage
[H]F Junkie
- Joined
- Jun 5, 2004
- Messages
- 8,298
Same... I am expecting a new motherboard out of this. If they require an actual provable failure then I will not be buying Asus ever again, because that would be like not recalling a brand new car that only runs on half its cylinders on the basis that it still runs.
First note: The bug only applies to the SATA 2 3gb/s ports. The SATA 3 6gb/s ports are fine and unaffected.
Also, Intel expects the total number of boards effected by this to run about 5%, so you may not ever have a problem.
Anandtech has all the details in what is without a doubt the most even handed and least reactionary of all the articles I've read about this so far.
It alleviated my fears significantly.
That said, I will definitely be expecting something out of Asus on this, even if that something doesn't manage to happen until April when Intel expects to catch back up to demand for the fixed chipset.
Also, Intel expects the total number of boards effected by this to run about 5%, so you may not ever have a problem.
Anandtech has all the details in what is without a doubt the most even handed and least reactionary of all the articles I've read about this so far.
It alleviated my fears significantly.
That said, I will definitely be expecting something out of Asus on this, even if that something doesn't manage to happen until April when Intel expects to catch back up to demand for the fixed chipset.
TonyToronto
n00b
- Joined
- Jan 23, 2011
- Messages
- 17
and so have i! i'm not buying the "only the sata2" ports are affected. Twice now i've had the SSD not boot and mangle the drive. Could not see data on another PC. also, the 6gb sata port wouold not detect the SSD, the old p55 board had no problem seeing it, just couldn't read shit from it.
I think i just "mangled" my vertex 2. I turned off the machine to move it to another room, went to boot it up and had the ram led come on, turned it back off, turned it on again and my ssd is gone. Not detected at all, gonna see if another computer can read it, but i'm not getting my hopes up based on this.
VenomousDesigns
Limp Gawd
- Joined
- Jan 4, 2009
- Messages
- 173
Oh jesus. I'm not sure if people are just panicking cause they saw the 'Sky is falling' or if these recently discovered claims about SATA II now run into SATA III - I have my boxed system still sitting next me with a OCZ Vertex 2 E.
VenomousDesigns
Limp Gawd
- Joined
- Jan 4, 2009
- Messages
- 173
Was this in a SATA II or SATA III port mate?
Sata III the grey port #1. It was there, reboot, not there. Tested the drive in an external box, as well as internal on my x58 setup, nothing, no ghost of a drive, just nothing.
Now this could be an isolated thing to just me so I don't want to start a panic over the 6gbit ports but I didn't do anything but turn the system off, move it to another room and turn it on. Computer wasn't open, nothing was rearranged. Sure SSD's have been known to die randomly, but this system was bought and put together on the 20th of january.
Now this could be an isolated thing to just me so I don't want to start a panic over the 6gbit ports but I didn't do anything but turn the system off, move it to another room and turn it on. Computer wasn't open, nothing was rearranged. Sure SSD's have been known to die randomly, but this system was bought and put together on the 20th of january.
I had to do a windows reinstall today because of the corruption. I'm hoping it was a fluke, but just to be safe I'm running bare minimum on the boot drive and making sure all my documents and things get pushed to my backup drives
I'm having the same issue with an Intel 80G, with roughly the same free space. Used to get around 7.4-7.6 in WEI, but now get 5.9. It might be the free space.
As for the extra space reserved...Intel drives do not. Trim frees up that space to be used. Drives that do not support TRIM, like Sandforce and Crucial drives, do have extra space put aside to move stuff around.
edit: this is with a fresh install.
Last edited:
I do think there is something fishy going on with SSDs on this board, but I'm dubious as to whether it's the SATA ports. I think it's likely a EFI / driver combination bug.
For one thing, WEI is unreliable. I've seen it give me 5.9 when the other testers (AS SSD and CystralDiskMark) were registering numbers that I was getting on my p35 mobo. Right now I'm getting 7.7 on WEI, but I don't see any differences in the other two programs' results and I certainly don't notice any difference in performance when it was giving 5.9. There should be a huge difference between 5.9 and 7.7.
And isn't it curious that WEI will bounce between 5.9 and 7.x (7.7 for me)? Why 5.9? If the SSD is under performing, why it is jumping down to HD speeds? Why not 6.2 or 6.7, or even 4.3? The fact that it always gives 5.9 is telling, I think. WEI gets easily screwed up. That's why lots of folks don't consider it of much value for benching.
I don't recall any of the Intel SSD users in this thread mentioning that they installed the latest SSD firmware which Intel released in January 2011. It might be wise to do that.
For one thing, WEI is unreliable. I've seen it give me 5.9 when the other testers (AS SSD and CystralDiskMark) were registering numbers that I was getting on my p35 mobo. Right now I'm getting 7.7 on WEI, but I don't see any differences in the other two programs' results and I certainly don't notice any difference in performance when it was giving 5.9. There should be a huge difference between 5.9 and 7.7.
And isn't it curious that WEI will bounce between 5.9 and 7.x (7.7 for me)? Why 5.9? If the SSD is under performing, why it is jumping down to HD speeds? Why not 6.2 or 6.7, or even 4.3? The fact that it always gives 5.9 is telling, I think. WEI gets easily screwed up. That's why lots of folks don't consider it of much value for benching.
I don't recall any of the Intel SSD users in this thread mentioning that they installed the latest SSD firmware which Intel released in January 2011. It might be wise to do that.
ochadd
[H]ard|Gawd
- Joined
- May 9, 2008
- Messages
- 1,317
I did update a few nights ago to the rev released 1/26/2011. Numbers are worse. This is on the SATA 6GBps ports. I ran the benchmark today and it's actually lower than the below screenshot. 428 was the score...
Edit: Moved pics to first post.
Last edited: