Official GIGABYTE Forum

GA-X58A-UD3R - Random failure to wake from sleep after BIOS Update to FG1

Rolo42

  • 260
  • 4
  • DON'T PANIC!
    • PC Parts List
Re: GA-X58A-UD3R - Random failure to wake from sleep after BIOS Update to FG1
« Reply #45 on: February 02, 2012, 08:24:41 pm »
I agree with pursuing the JMicron route; I can almost remember something in the mental cobwebs about those things not quite playing well with others.

Nealeb

  • 12
  • 2
Re: GA-X58A-UD3R - Random failure to wake from sleep after BIOS Update to FG1
« Reply #46 on: February 02, 2012, 09:44:38 pm »
I've just reconfigured the SATA connectors. System disk (non-RAID) is now on Intel controller slot 0 with the 4 RAIDed disks on 2-5. DVD writers now connected to the JMicron slots. Windows came back up and all the disks look present and correct. Testing begins - I have my PVR software configured to wake/hibernate the machine a few times a day. Let's see what happens.

Don't remember seeing any posts about JMicron controller issues, but you guys are making it sound like one of those "but I thought everyone knew that!" situations:) I am going to feel very silly if it all turns out to be something so trivial to fix...

Aussie Allan

  • 1964
  • 141
  • A yoyo uses all three fundermental laws of phisics
Re: GA-X58A-UD3R - Random failure to wake from sleep after BIOS Update to FG1
« Reply #47 on: February 02, 2012, 10:46:05 pm »


     Nealab

  "Don't remember seeing any posts about JMicron controller issues,"   ;D ;D ;D When I spotted your admission of using The JMicron ..... I had a real head slapping moment .... but the horse had already bolted with Lsdmeasap coming in with his usual impeccable diplomacy  :D

  Marvell 9128 was bigger than "WaterGate" there for a while and probably responsible for a spike in keyboards in monitor returns world wide ....

  But for your specific problem.... it's not over till the fat lady sings .... Although a big step forward ..... the 88SE9128 Marvell controller chip was the first mainstream chip that was touted to have 6GB capabilities .... so everyone (manufactures) used it ..... little did we all know ...

  Things have changed though .... I actually have and use the HighPoint RocketRaid 640 PCI-E card that has not one but 2 of these chips in them and it actually flies ..... this was my work-around to getting the X58 board up to full speed with more than 6 drives under the bonnet.

  Fingers crossed for you buddy  ;)

  Aussie Allan
i7-4790K @4.8GHz 24/7 water clock
MSI XPower AC
32GB corsair  2666Mhz
 GTX-1070Ti full cover
Lange DDC elite pump
G changer360 Rad x2
Phobya 450 balancer
W10 Pro-64
Zigor 2000 UPS
1x500GB for clone
6x2tb- raid5-Storage
C: Evo 970 Pro 512gb
Scratch:Evo 970 Plus 512gb

Rolo42

  • 260
  • 4
  • DON'T PANIC!
    • PC Parts List
Re: GA-X58A-UD3R - Random failure to wake from sleep after BIOS Update to FG1
« Reply #48 on: February 03, 2012, 02:12:27 am »
Don't remember seeing any posts about JMicron controller issues
Not in this thread, just from years ago (haven't used JMicron controllers in ages).

I'm an engineer; I need that guy who deals with the customers so I don't have to.


Lsdmeasap

  • 1166
  • 58
    • Gigabyte Support (TweakTown USA)
Re: GA-X58A-UD3R - Random failure to wake from sleep after BIOS Update to FG1
« Reply #49 on: February 03, 2012, 06:10:26 am »
I know for sure the J.Micron ports are slow, so they rarely get used, as for them causing issues directly I can't say with certainty because not enough users use them long enough for anyone to report issues.  Those I do see using them switch quickly to Intel ports, or sometimes they are used for additional storage ports when the Intel ones are all filled.   And of course that usually doesn't cause problems other than slow transfers, it's the use of the OS on these ports that concerned me, and since S1 writes to disk while windows tries to shut it all down quickly is what made me think this may be the cause of his issues.

Hopefully now that he's moved the drives around we can find out for sure!

I don't know if anyone has addressed this or not, but have you also tested your hard drive for errors with the manufacturers tool?   Just thinking of additional things that could be causing these issues, no need to hurry and test now unless the drive move doesn't help anything.

And EuP/ErP, yes please disable this  ;)

Aussie Allan

  • 1964
  • 141
  • A yoyo uses all three fundermental laws of phisics
Re: GA-X58A-UD3R - Random failure to wake from sleep after BIOS Update to FG1
« Reply #50 on: February 03, 2012, 08:06:13 am »

    I can't say with certainty because not enough users use them long enough for anyone to report issues.  :D :D :D :D

  That would be one of the most diplomatic statement I've had the pleasure to read in some time Lsdmeasap   .... My attempts usually end up sounding like a "Hand grenade in a chicken coop" .... to be fair to the Marvell logo, .... and head off a slapping around the head for a retraction..........Marvells 9128 chip is only one of many ..... there market penetration is also huge .... if you turn over your hard drive ... odds on you'll see a Marvell logo.

  The 9128 fiasco was limited to just that chip and only under certain platforms .... as I stated earlier I have a raid card that using this same chip(S) with no problems .... lets just hope the advice given to Nealab helps him to a solution.

  Aussie Allan

 
i7-4790K @4.8GHz 24/7 water clock
MSI XPower AC
32GB corsair  2666Mhz
 GTX-1070Ti full cover
Lange DDC elite pump
G changer360 Rad x2
Phobya 450 balancer
W10 Pro-64
Zigor 2000 UPS
1x500GB for clone
6x2tb- raid5-Storage
C: Evo 970 Pro 512gb
Scratch:Evo 970 Plus 512gb

Dark Mantis

  • *
  • 18405
  • 414
  • 10typesofpeopleoneswhoknow binaryandoneswhodont
    • Dark Mantis
Re: GA-X58A-UD3R - Random failure to wake from sleep after BIOS Update to FG1
« Reply #51 on: February 03, 2012, 09:24:14 am »
I can't say that I remember seeing any issues directly connected to the JMicron chips either other than the slow speed problem. It still surprises me how many people are unaware of the Marvell 9128 chip troubles. I would have thought that every amnm and his dog has seen all the write ups about it by now.
Gigabyte X58A-UD7
i7 920
Dominators 1600 x6 12GB
6970 2GB
HX850
256GB SSD, Sam 1TB, WDB320GB
Blu-Ray
HAF 932

Gigabyte Z68X-UD5-B3
i7 3770K
Vengeance 1600 16GB
6950 2GB
HCP1200W
Revo Drive x2, 1.5TB WDB RAID0
16x DLRW
StrikeX S7
Full water cooling
3 x 27" Iiy

Nealeb

  • 12
  • 2
Re: GA-X58A-UD3R - Random failure to wake from sleep after BIOS Update to FG1
« Reply #52 on: February 03, 2012, 09:43:18 am »
Just for clarity (and it's not directly relevant to what I'm testing at the moment) can I ask which controller you guys mean when you say "JMicron"?

According to the motherboard manual (decoded from the original Japanese by a blind Urdu speaker who once learnt French at school and who speaks neither Japanese nor English) I think that there are four disk controllers on the UD3R board.(*)
  • Intel RAID controller. 6 ports, SATA2, can support mix of one RAID set and a bunch of other single drives.
  • Marvell 9128 controller. 2 ports, SATA3. I have this one disabled at BIOS level as it seemed to be associated with hangs earlier in the boot sequence than I'm now seeing - although that might have been the USB3 controller. Anyway, I don't have any drives fast enough to justify it
  • Gigabyte controller. 2 ports, SATA2, which I'm now using for a couple of DVD drives so speed is not an issue.
  • JMicron controller, 2 ports but eSATA and hard-wired to a couple of USB/eSATA ports on the back panel.

The ICH10R drives the Intel SATA ports directly, but the other three controllers hang off PCIe bus(es) driven by the ICH10R, so it's reasonable to assume that the Intel ports would be faster overall. So, whether or not the controller choice is the root of my problem, from a performance point of view it wasn't the best idea in the world to put the system disk on the "Gigabyte" controller. Ho hum - we live and learn! In any case, the ICH10R is an integral part of the chipset while the Gigabyte (and Marvell) controllers are a level of driver away - although presumably the BIOS code should be able to handle drives connected to these for reading, for example, hiberfil.sys. But maybe that's the point where it gets a bit flaky.

(*)Actually, that's rather unfair and it's better than some I've seen, but you always feel that it nearly but not quite tells you the things that you really want to know - especially around BIOS settings which are never well-explained.

Lsdmeasap

  • 1166
  • 58
    • Gigabyte Support (TweakTown USA)
Re: GA-X58A-UD3R - Random failure to wake from sleep after BIOS Update to FG1
« Reply #53 on: February 03, 2012, 10:11:02 am »

   I can't say with certainty because not enough users use them long enough for anyone to report issues.  :D :D :D :D

  That would be one of the most diplomatic statement I've had the pleasure to read in some time Lsdmeasap   .... My attempts usually end up sounding like a "Hand grenade in a chicken coop" .... to be fair to the Marvell logo, .... and head off a slapping around the head for a retraction..........Marvells 9128 chip is only one of many ..... there market penetration is also huge .... if you turn over your hard drive ... odds on you'll see a Marvell logo.

  The 9128 fiasco was limited to just that chip and only under certain platforms .... as I stated earlier I have a raid card that using this same chip(S) with no problems .... lets just hope the advice given to Nealab helps him to a solution.

  Aussie Allan

  

Are you being mean or sarcastic, or is this a compliment?   Sorry I'm not sure, and this is the second time you referred to me as diplomatic, so I wanted to ask to be sure what you meant.   Hopefully it's a nice comment, I don't try to make anyone mad, rather the opposite as I try to avoid making anyone mad or stepping on anyone's toes.  If it was meant as a compliment, thank you!   If you were being bad, shame on you and Grrrrrr!  >:(  ;D J/K

Yes, I have a RR 640 here too, and it's much better than a single 9128, due to dual chips and an extra PCIE lane, plus the bridge chip.  But it can easily be overloaded too, with two SSD's or four.   Found in RAID will slow it down just like the onboard single chips, and 2 SSD's can do that too if you don't connect them correctly.   When using 2 SSDs on the 640 you have to make sure you only put one SSD on each controller, so they split the load.   But even when you don't, using 2 SSD's, it's still not as unstable or slow as the onboard solutions.  

I used it in my awesome C300 review, if you have a few hours you can read it here if you're interested.
http://www.xtremesystems.org/forums/showthread.php?t=261789

Ans yes, I use Marvell based drives, C300's and soon to be M4's.   I also like their new onboard solution, it's a much needed improvement to the 9128, the 9182 and 9172 are pretty good controllers, but since they are now used on P67/Z68 boards that's a little bit too late since Intel finally stepped up!  

Now, wheres our Native 4-6 6Gb/s ports Intel, they were supposed to arrive with X79???  ;D




Just for clarity (and it's not directly relevant to what I'm testing at the moment) can I ask which controller you guys mean when you say "JMicron"?

According to the motherboard manual (decoded from the original Japanese by a blind Urdu speaker who once learnt French at school and who speaks neither Japanese nor English) I think that there are four disk controllers on the UD3R board.(*)
  • Intel RAID controller. 6 ports, SATA2, can support mix of one RAID set and a bunch of other single drives.
  • Marvell 9128 controller. 2 ports, SATA3. I have this one disabled at BIOS level as it seemed to be associated with hangs earlier in the boot sequence than I'm now seeing - although that might have been the USB3 controller. Anyway, I don't have any drives fast enough to justify it
  • Gigabyte controller. 2 ports, SATA2, which I'm now using for a couple of DVD drives so speed is not an issue.
  • JMicron controller, 2 ports but eSATA and hard-wired to a couple of USB/eSATA ports on the back panel.

The ICH10R drives the Intel SATA ports directly, but the other three controllers hang off PCIe bus(es) driven by the ICH10R, so it's reasonable to assume that the Intel ports would be faster overall. So, whether or not the controller choice is the root of my problem, from a performance point of view it wasn't the best idea in the world to put the system disk on the "Gigabyte" controller. Ho hum - we live and learn! In any case, the ICH10R is an integral part of the chipset while the Gigabyte (and Marvell) controllers are a level of driver away - although presumably the BIOS code should be able to handle drives connected to these for reading, for example, hiberfil.sys. But maybe that's the point where it gets a bit flaky.

(*)Actually, that's rather unfair and it's better than some I've seen, but you always feel that it nearly but not quite tells you the things that you really want to know - especially around BIOS settings which are never well-explained.


J.Micron ports are the ones labeled GSATA2_8/9, the last set of ports at the bottom of the SATA stack.

Yes, there is four hard drive controllers, and they are as you mentioned, ICH10R (SATA2_0-6), Marvell 9128 (GSATA3_6/7), J.Micron JMB362 (2 eSATA on Rear I/O panel), and Gigabyte/ Also J.Micron (GSATA_8/9 +  IDE x 2 ports).  If you want a good view on what's what, generally the motherboard block diagram shows you everything fairly well towards the front of the manual (Page 8 for this board).  A better description of the controllers used on each board is best described on the motherboards homepage, in specifications, in case you want to look up other boards or future board details.   Once you look at the specs on those pages, and then the block diagram, it's much easier to see what's really under the hood.

Any results yet on the move to the Intel ports?   Another thing I thought to mention to you earlier today is this, sometimes with any board if you put it to sleep S1 or S3 all the time, but never actually shut it down or reset it, this can end up leading to issues eventually.   So from time to time be sure you do an actual shut down or reset when time permits.
« Last Edit: February 03, 2012, 10:27:07 am by Lsdmeasap »

Nealeb

  • 12
  • 2
Re: GA-X58A-UD3R - Random failure to wake from sleep after BIOS Update to FG1
« Reply #54 on: February 03, 2012, 12:46:32 pm »
I didn't see any reference to JMicron against the "Gigabyte" ports, as they are described in the MB manual, but I'm quite prepared to believe that that's who made the controller. I guess that the manual author just wanted to differentiate from the eSATA controller. I had a look at the block diagram earlier, and I have to admit that the text describing what plugs in where was more than adequate when I first put the system together. What frustrates me, though, is the lack of indication of many of the board-specific BIOS settings mean and it's forums like this one that help the user to understand. For example, the "S5" power save state setting (now disabled on my system). If you read most descriptions of the power states, S5 is supposed to be fully powered-off with no saved state. Now, that's clearly not what Gigabyte mean!

However, it's done 3-4 hibernate/resume cycles since I swapped cables last night and it's all been fine so far, but unfortunately that's no guarantee of anything. Sometimes it can go for a week or so and a dozen or two cycles without problem, and then it fails when it's supposed to be recording something important...

I note the point about a reboot from time to time; I've been working with Windows for many years now and that's always seemed like a good idea. In practice, though, there seems to be enough patches and updates to make sure that you reboot on a regular basis anyway!

Aussie Allan

  • 1964
  • 141
  • A yoyo uses all three fundermental laws of phisics
Re: GA-X58A-UD3R - Random failure to wake from sleep after BIOS Update to FG1
« Reply #55 on: February 03, 2012, 02:07:23 pm »

  Firstly the most important Lsdmeasap

     " If you were being bad, shame on you and Grrrrrr!  Angry  Grin J/K"

    Your "Grrrrrrrr!" sent me scurrying under the table :D .... Yes Mate it was a complement (did you not notice the karma I pinged you) ..... Please don't read it any other way!

   Being raised in OZ ..... we tend to not hide behind witticism and shaded remarks ..... but calmly walk up to people with a sledge hammer over ones shoulder .... If I have beef with anyone .... I'm pretty up front and most will know the difference. ;) ....  the emphases was directed to those who read that have been less tactful of past .... and yours was a good example of how to tackle a subject still rather sensitive to many, including Giga and the manufacture .... And thanks for the review link ...... might be applicable reading as I'm shortly going to need ideas on raiding 4 Virtex3s when they dip under £100 .... got a little project on the hop ( X79 Monster build) that will be steeling my Revox2.... and the editing machine gets the Vertex upgrade ....need a good solution to max out 4 of them under Raid0


  The 640! .... Yup! worked out some time ago that it performs optimally with a single SSD per controller .... as a X58 "fixya" it was a excellent cost effective solution that I've recommended several times and still do now .... for the money, is there something more current that performs as well for X58 owners ?

  Nealeb .... Glad you have been able to get in touch with people that seem to have helped you move forward ..... please keep in contact so the fix goes up for prosperity when you finally get there .... now you have got this far.....if it does wobble ..... get the dump codes ext and what time local the event happened....... I still think power may be in the equation

  Aussie Allan
i7-4790K @4.8GHz 24/7 water clock
MSI XPower AC
32GB corsair  2666Mhz
 GTX-1070Ti full cover
Lange DDC elite pump
G changer360 Rad x2
Phobya 450 balancer
W10 Pro-64
Zigor 2000 UPS
1x500GB for clone
6x2tb- raid5-Storage
C: Evo 970 Pro 512gb
Scratch:Evo 970 Plus 512gb

Lsdmeasap

  • 1166
  • 58
    • Gigabyte Support (TweakTown USA)
Re: GA-X58A-UD3R - Random failure to wake from sleep after BIOS Update to FG1
« Reply #56 on: February 04, 2012, 09:24:21 am »
I didn't see any reference to JMicron against the "Gigabyte" ports, as they are described in the MB manual, but I'm quite prepared to believe that that's who made the controller. I guess that the manual author just wanted to differentiate from the eSATA controller. I had a look at the block diagram earlier, and I have to admit that the text describing what plugs in where was more than adequate when I first put the system together. What frustrates me, though, is the lack of indication of many of the board-specific BIOS settings mean and it's forums like this one that help the user to understand. For example, the "S5" power save state setting (now disabled on my system). If you read most descriptions of the power states, S5 is supposed to be fully powered-off with no saved state. Now, that's clearly not what Gigabyte mean!

However, it's done 3-4 hibernate/resume cycles since I swapped cables last night and it's all been fine so far, but unfortunately that's no guarantee of anything. Sometimes it can go for a week or so and a dozen or two cycles without problem, and then it fails when it's supposed to be recording something important...

I note the point about a reboot from time to time; I've been working with Windows for many years now and that's always seemed like a good idea. In practice, though, there seems to be enough patches and updates to make sure that you reboot on a regular basis anyway!

Well they don't reference it anywhere, but I know for the past several years when there was only J.Micron controller as secondary they would label those ports Gigabyte/J.Micron in some places, so I assume the same controller is used here and they just label it Gigabyte, and yes likely to differentiate between those ports and the eSATA ones as you mentioned.

I agree, they could benefit (Or end users could anyway) from adding more info in the manuals for the BIOS settings.  They have improved over the years though, at least now the manual actually includes all of the BIOS settings, in the past many were missing from the manual altogether.

Here's some real information for you on the various sleep / Power Saving / C-States
http://www.hardwaresecrets.com/article/611
http://software.intel.com/en-us/blogs/2008/03/27/update-c-states-c-states-and-even-more-c-states/

Awesome to hear sleep and resume have now  been working for you so far, hopefully this will continue and you can call this solved!!



  Firstly the most important Lsdmeasap

     " If you were being bad, shame on you and Grrrrrr!  Angry  Grin J/K"

    Your "Grrrrrrrr!" sent me scurrying under the table :D .... Yes Mate it was a complement (did you not notice the karma I pinged you) ..... Please don't read it any other way!

   Being raised in OZ ..... we tend to not hide behind witticism and shaded remarks ..... but calmly walk up to people with a sledge hammer over ones shoulder .... If I have beef with anyone .... I'm pretty up front and most will know the difference. ;) ....  the emphases was directed to those who read that have been less tactful of past .... and yours was a good example of how to tackle a subject still rather sensitive to many, including Giga and the manufacture .... And thanks for the review link ...... might be applicable reading as I'm shortly going to need ideas on raiding 4 Virtex3s when they dip under £100 .... got a little project on the hop ( X79 Monster build) that will be steeling my Revox2.... and the editing machine gets the Vertex upgrade ....need a good solution to max out 4 of them under Raid0


  The 640! .... Yup! worked out some time ago that it performs optimally with a single SSD per controller .... as a X58 "fixya" it was a excellent cost effective solution that I've recommended several times and still do now .... for the money, is there something more current that performs as well for X58 owners ?

  Nealeb .... Glad you have been able to get in touch with people that seem to have helped you move forward ..... please keep in contact so the fix goes up for prosperity when you finally get there .... now you have got this far.....if it does wobble ..... get the dump codes ext and what time local the event happened....... I still think power may be in the equation

  Aussie Allan

Haha, sorry, didn't mean to scare you  ;D   Haha, that's too funny, I know your playing though, as was I!

Great to hear it was a compliment!   I just wasn't sure because I know sometimes language or regional differences end up making many words or comments mean different things to different people around the world, so I thought I better ask to be sure in case I was offending you somehow!

I like that!  >> Calmly walk up to people with a sledgehammer!  ;D  ;)  :D

Yes, the 640 was a great solution for many with X58, that wanted multiple 6Gb/s SSD's!   I was sent one for review, so I didn't have to buy it, not sure if I would have one otherwise because I was happy personally with RAID on the Intel ports anyway.

I haven't been keeping up with RAID controllers recently, but I do know there are many new 6Gb/s capable controllers out now, and many that don't use 9128.   And Marvell has improved too with onboard controllers, the 9172 and 9182 work pretty damn good actually.   Well I know they bench much better than 9182, I have not used them for testing with an OS so I can't vouch for their stability or lack of.


Aussie Allan

  • 1964
  • 141
  • A yoyo uses all three fundermental laws of phisics
Re: GA-X58A-UD3R - Random failure to wake from sleep after BIOS Update to FG1
« Reply #57 on: February 04, 2012, 10:05:23 am »

   Lsdmeasap

  sometimes language or regional differences end up making many words or comments mean different things to different people around the world,

   Tell me about it! ..... I remember one Russian chappie on another forum praising me for helping him and commenting on my in-depth  knowledge on the subject......His words of praise of ....."You seem to know F*%^K all on this subject" eventually made me wet myself with laughter.

  This thread has been Valuable ..... I've picked up a couple of goodies for the skull drive .... and Nealab may be back in calm waters.

  One rather specific question though ..... if you had 4 Vertex3 in your hand..... and an X58 in the other....... what would you add as an interpreter to make them truly fly (Raid0 for OS) ? ..... All opinions are welcome ;)

  Aussie Allan
i7-4790K @4.8GHz 24/7 water clock
MSI XPower AC
32GB corsair  2666Mhz
 GTX-1070Ti full cover
Lange DDC elite pump
G changer360 Rad x2
Phobya 450 balancer
W10 Pro-64
Zigor 2000 UPS
1x500GB for clone
6x2tb- raid5-Storage
C: Evo 970 Pro 512gb
Scratch:Evo 970 Plus 512gb

Lsdmeasap

  • 1166
  • 58
    • Gigabyte Support (TweakTown USA)
Re: GA-X58A-UD3R - Random failure to wake from sleep after BIOS Update to FG1
« Reply #58 on: February 04, 2012, 11:58:57 am »
Hahaha, ya that is funny!!   I'd be laughing too, and actually I am right now, so he's still funny all this time later!

Yes, this is a good thread with lots of helpful information for others!   And hopefully Nealab's issues are finally sorted out for him, I know it sucks to have issues drag out for long periods with no one able to help, that's terrible.   So hopefully he's good to go now!

I can't help on the SSD question, I haven't read into any new controllers lately, but if you are not already a member or have not visited this site you should check it out.  They are always doing tons of SSD and RAID controller testing, aside from all the overclocking in the extreme sections.

Here's the storage area, you'll find some nice controller results here that should help you pick out a nice one!
http://www.xtremesystems.org/forums/forumdisplay.php?62-Storage

Nealeb

  • 12
  • 2
Re: GA-X58A-UD3R - Random failure to wake from sleep after BIOS Update to FG1
« Reply #59 on: February 04, 2012, 08:54:42 pm »
I've had around 10-12 hibernate/resume cycles over the last couple of days with no problems seen. That's no guarantee that it's fixed but it looks promising!

I can't say that I haven't seen people trying to help in other forums. Trouble is that it's difficult to take them seriously when it's blindingly obvious that they know even less about it than I do. It's been good to have a sensible discussion with people who both know a lot and are prepared to accept that there are still more things to learn. I've been playing with computers for about 45 years now and still playing :) but PCs aren't really part of the day job and it's difficult to keep up with the changes in hardware.

Thanks again and I'll drop in from time to time to update. A definite conclusion might help other people who pass this way via Google in the future.

But to finish, a small admission... Some few months ago I joined the forum under a different name (Oyster16) via a thread about USB3-related hangs. I had completely forgotten that I was already a member when I came across this thread via Google - didn't even realise that I had been to this forum before. That particular problem does not seem to have quite been satisfactorily resolved but maybe I'll go back to it if I can feel confident that the current problem has been fixed. I'll also do some research on disk controllers if I ever need SATA 3 speeds and forget the onboard 9128...