Official GIGABYTE Forum
Questions about GIGABYTE products => Motherboards with Intel processors => Topic started by: RobD on July 09, 2012, 05:02:22 pm
-
I've tried this with 2 x 500Gb Samsung drives and 2 x 2Tb WD [full 4Gb capacity uefi 64 bit] on intel ports 0 and 1 and the result is the same.
When resuming from sleep, it either hangs at the welcome screen or hangs on a blank screen after the welcome screen or bsod's
Install is barebones with only chip and vga drivers.
The full 4Tb with 7x64 setup was even worse, it did nothing but bsod!
Bios is F10, will try F11
-
Ok, last 3 times system has locked up on me resuming from sleep :-[ Yet I can prime95 for hours & can't even resume from sleep ??? ??? ???
Almost had enough of this board. For real.
-
Almost had enough of this board. For real.
Noooo...don't kill it! I need you to compare notes with! :)
F11 didn't make any difference. I think this is a driver issue, might try vga from nvidia and will try installing xp to see what happens.
-
Tried gtx 680 ver. 301.42 from Nvidia - epic fail!
-
Tried win7 x86 sp1 with 2 x 500Gb drives on the marvell raid and it works fine!! :)
So...something to do with the intel raid driver?
-
Almost had enough of this board. For real.
I was nearly with you, ready to wave the white flag!! ["Can't....install.....windows...any....more...."]
But after reading a bit more around the net, I found others with the same issue and it seemed the culprit was that the intel raid driver would fail to re-initialise after resume from sleep.
So the OS had a hdd when it went to sleep, then wakes up to find the hdd has gone awol!!
So I devised a scheme to get hard evidence.
I set both controllers to ahci, connected a single drive to port 7 [marvell controller], installed win 7 x86 to it with the marvell ahci driver, then chip and vga drivers then tested sleep and all worked fine.
I then went back to bios and changed the intel contoller to raid mode [no drives connected to it] and rebooted.
At the desktop I get the obligatory message that windows couldn't find a driver, went to device manager and pointed it to the driver which it installed - no problems.
Did a restart anyway to check all was still ok, then put it to sleep and when I woke it up, viola, we arrive at the desktop ok but with an error message as per below pic and device manager showing the intel raid controller as now having a problem - 'This device cannot start code 10'
Now, besides gigabyte/intel sorting it out, is there anything we can do about it - I'm open to suggestions.
Bear in mind, the problem only occurs when resuming from sleep, if I leave everything as-is and simply restart the computer, everything is happy again - until the next sleep/resume.
(http://members.iinet.net.au/~dream/images/gigabyte/raid%20issue.jpg)
-
I've nearly nailed it!! :)
Look more closely at the pics - what do you see......
Mine now resumes without issue, though via a method with unexpected results. Need to do a bit more homework. Back soon...
There is hope yet!
-
I've nearly nailed it!! :)
Look more closely at the pics - what do you see......
Mine now resumes without issue, though via a method with unexpected results. Need to do a bit more homework. Back soon...
There is hope yet!
Power PC processor?? LoL... One thing I can clearly see now is that this board is epicly lame... Was really happy with my previous gen G1 boards, but the 2011 G1 is way to much fkin around. Shouldn't be this hard.
-
Yeah the powerpc processor had me wondering - "I can install mac OS9??" ??? :D
No. the lower part of the pic shows it trying to load the older ICH8R raid driver which is why it falls over.
Still trying to work out a way to stop it doing that.
This is not a gigabyte issue, it's a MS/Intel issue.
I can fix it when the OS is not running on that chip, but when it is on that chip....so far a nightmare! >:(
-
Well I'm about to raise the white flag here, everything I try ends in tears :'(
It's trying to load iastorV.sys on resume which is seemingly for the older ich8r etc. chipset and of course, that won't work.
Why only on resume from sleep, I'm stuffed if I know, I'm not that smart.
The workaround to get the intel raid to cooperate when the OS is on the marvell chip. is to get a copy of iastorA.sys from the F6 drivers, rename it to iastorV.sys and overwrite the existing iastorV.sys in c\windows\system32\drivers with it.
On resume everything is fine and the raid array will still mount in my computer and can be navigated.
In device manager, even though it shows as a ich8r 9r etc. raid, it's working fine.
If you do a restart, it will then show correctly as a C600 raid device.
But using the same trick when the OS is booting over the intel raid is still a fail!
In fact, you can't even boot safe mode over the intel in raid mode with or without mods, no doubt for the same reason.
Seriously, Intel/MS/Gigabyte [and others] should offer a bounty to anyone who provides evidence that directly leads to the arrest of a bug!! :D
I've spent a fortune in blood, sweat and tears trying to overcome the problems with this board and like C-BuZz, I'm about to bin it and go elsewhere!
-
Oh dear, another wierd observation.
I was having another go, yup, yet another install to the intel raid. :-\
When I got to the point where setup asks for a raid driver, I navigated to the flash drive and while it was 'seemingly' loading the driver, I also happened to notice that the win 7 dvd was also being accessed.
So I restarted and at the same point, I ejected the win 7 dvd - totally removed it, then clicked next to begin the driver loading from the flash.
Lo and behold, it failed to find any new devices.
So to try and prove the point further, I put the gigabyte driver cd in the drive, clicked browse again and to my surprise, I had D: drive. On past attempts I could not get a D drive and so could not get to the cd drivers and I think I know why...more on that later.
So I changed tack and clicked my way into the driver cd to the raid drivers and clicked next and yet again it failed.
So for some reason, when loading them from a flash with win 7 setup dvd still in the drive, these raid drivers will not load and win setup is cheating it's way to some other driver or something on the win setup dvd and win setup is able to continue.
This also helps explain why it seems to take forever for the driver to load. On past raids it was almost instant - done - quick, click next!
As for the dvd drive suddenly turning up as a browsable drive, I think that happened because my flash drive is actually a multi card reader [4 port] and it was consuming all available drive positions in bios.
On this lastest attempt with another theory in mind, I went into bios and disabled all floppy flash drives except one and I think that is why my dvd drive suddenly showed up in the browse window - but yet to be proven.
-
Nope, can't get anything to work. The appearance of the dvd drive when browsing for raid drivers is totally upredicatble. When it is there, they don't load anyway.
The number of floppy drives that appear in bios is variable [mutli port card reader]. If I diasbale them, bios re-enables them. Bios drops in and out trying to post. Bios sometimes doesn't initialise the video card. Bios goes nuts with OC button in, del no longer works, brings up tab menu instead etc. etc. etc.
Very skittish, I quit!!
-
Have replied to Gigabyte with all this info + pics for them to consider. Will update here with their response.
Here are the pics I sent them showing the fundamental problem.
This one shows the correct state of the c600 raid after a restart with drivers iastorA,sys and iastorF.sys loaded
(http://members.iinet.net.au/~dream/images/gigabyte/correct_driver_after_restart.jpg)
This one shows the error state of the raid after resuming from sleep.
It now shows as a IHC8R chip and trying to load iastorV.sys
(http://members.iinet.net.au/~dream/images/gigabyte/incorrect_driver_after_resume.jpg)
I have tried deleting every reference to iastorV.sys [the entire folder and contents - about 4 of them] in the registry, but it made no difference.
-
Oh dear, another wierd observation.
I was having another go, yup, yet another install to the intel raid. :-\
When I got to the point where setup asks for a raid driver, I navigated to the flash drive and while it was 'seemingly' loading the driver, I also happened to notice that the win 7 dvd was also being accessed.
So I restarted and at the same point, I ejected the win 7 dvd - totally removed it, then clicked next to begin the driver loading from the flash.
Lo and behold, it failed to find any new devices.
So to try and prove the point further, I put the gigabyte driver cd in the drive, clicked browse again and to my surprise, I had D: drive. On past attempts I could not get a D drive and so could not get to the cd drivers and I think I know why...more on that later.
So I changed tack and clicked my way into the driver cd to the raid drivers and clicked next and yet again it failed.
So for some reason, when loading them from a flash with win 7 setup dvd still in the drive, these raid drivers will not load and win setup is cheating it's way to some other driver or something on the win setup dvd and win setup is able to continue.
This also helps explain why it seems to take forever for the driver to load. On past raids it was almost instant - done - quick, click next!
As for the dvd drive suddenly turning up as a browsable drive, I think that happened because my flash drive is actually a multi card reader [4 port] and it was consuming all available drive positions in bios.
On this lastest attempt with another theory in mind, I went into bios and disabled all floppy flash drives except one and I think that is why my dvd drive suddenly showed up in the browse window - but yet to be proven.
Update:
Whilst the appearance or non-appearance of the optical drive is totally random, I do withdraw my issue with the laoding of the drivers from either flash or cd [when it appears] without the win setup dvd in the drive.
I have never tried loading drivers from a cd before and have suddenly realised that the behavior may be totally normal.
This is because even though you get a message that no new devices could be found, as soon as you re-insert the win setup dvd, everything continues to run as normal.
If this is normal, then the appearance of the message is a little misleading - should be 'Please re-insert the windows setup cd-dvd'
Please let me know if I've got this right.
-
Gigabyte asked me to try a slightly different setup - marvell in ide and intel in raid, hdd connected to marvell.
My response back:
The results are different - mixed - unpredictable - worse.
I have set up the board as per your request with the following steps.
Disconnect all power and connect a single hdd [wd 2tb] to a marvell port.
Press clear cmos and power on the system.
At the load defaults screen, choose load optimised defaults and reboot.
Press del to enter setup and change marvell controller to ide and intel controller to raid. Save and exit.
Install win 7 sp1 x86 with F6 Intel raid driver.
Install chipset and vga drivers.
A check of device manager showed the intel raid was correct with c600 drivers.
A restart worked fine with device manager still showing correct c600 raid.
I then put the system to sleep and then woke it up. I clicked the account icon and the welcome message was displayed.
The system remained on the welcome message for a long time - more than 1 minute.
It eventually went to the desktop and I was able to access device manager and the raid was still correctly showing c600.
I then put it to sleep again, but this time on resume the welcome screen quickly disappeared [normal] but the desktop was a black screen with an empty taskbar at the bottom.
The mouse was still working - could move it - but nothing more happened.
I then momentarily pressed power and the system momentarily showed the desktop but then went into shutdown mode.
Wether the system was about to show the desktop or wether my pressing the power button caused it to show the desktop, I can't say.
When the system rebooted, I checked device manager and all was still ok,
I tried sleep again, but this time it refused to fully sleep, it only turned the screen off but after a further minute or so, power was still on the board [various led's still illuminated]
So, I thought I would try and duplicate the results.
I reinstalled windows [delete existing partitions], but now on resuming I would either have a hang at the welcome screen followed by a bsod, or arrive at the desktop followed by a bsod, pic of which is attached.
So, it never did the same thing twice in a row, very unpredictable.
(http://members.iinet.net.au/~dream/images/gigabyte/bsod.jpg)
-
Jesus christ Rob, you still battling with this POS mobo? I picked up a Ramp IV Formula X79 yesterday & was fully up & running within 30min. Not a single issue. I'll just repeat that... NOT A SINGLE FKING ISSUE!!!!!!!!!! My god what a breath of fresh air after that nightmare G1.Asshole2 conundrum.
Everything on this mobo just works (Ramp IV Formula X79). No bull$hit, no headaches.
Here my my 24/7, good for 2hrs of BF3 straight with these settings ;D I seriously suggest you do the same..
(http://s17.postimage.org/3yzncbcfz/image.gif)
-
Hey C-BuZz, good to hear from you. I'm right with you mate but still on a waiting list. Should have another board by the end of the week, so for now I'm doing more tests for gigabyte to help them track down this problem, but assassin's days are numbered!
The Rampage looks good, especially the creative audio chip that now also supports front jack re-tasking!
What vga are you running? Get 'Furmark' and give me some numbers!! :D
http://www.geeks3d.com/20120510/furmark-1-10-1-gpu-burn-in-stress-test-videocard-geforce-radeon/
Scroll down a bit to get to the dl button on the left side. There's some other vga testers there as well [right hand side] with some pretty cool 'fluid' graphics to bench with.
My trusty old geforce 7950gt agp [ p4 extreme 3.4 ghz ] did the Furmark 1/4 mile at 4 fps @ 1280 x 720 full screen and hit a temp of 83c :o :D
-
Hey C-BuZz, good to hear from you. I'm right with you mate but still on a waiting list. Should have another board by the end of the week, so for now I'm doing more tests for gigabyte to help them track down this problem, but assassin's days are numbered!
The Rampage looks good, especially the creative audio chip that now also supports front jack re-tasking!
What vga are you running? Get 'Furmark' and give me some numbers!! :D
http://www.geeks3d.com/20120510/furmark-1-10-1-gpu-burn-in-stress-test-videocard-geforce-radeon/
Scroll down a bit to get to the dl button on the left side. There's some other vga testers there as well [right hand side] with some pretty cool 'fluid' graphics to bench with.
My trusty old geforce 7950gt agp [ p4 extreme 3.4 ghz ] did the Furmark 1/4 mile at 4 fps @ 1280 x 720 full screen and hit a temp of 83c :o :D
Good to hear your moving on. Seriously, don't waste any more time on this mobo it's a waste of your life & a dead end. As for GPu's, I've got about 30 GPU's here to play with. But at the moment, I'm rocking 3 x 8800Ultra's in Tri-Sli ;D Love my retro Gpu's which is why I selected the Rampage, so I can run Tri/Quad NV or Amd whenever I like ;D I also do a lot of benching 3D06 & 3D11.
-
Gigabyte have sent me a link to beta bios 12a to address this issue. :) Will try it myself tonight.
http://ggts.gigabyte.eu/FileList/1242694/g1assas2.12a
-
This has fixed the problem with only one small caveat.
The only config that still fails is if the marvell chip is set to ide with the boot drive connected to it which is probalbly not worth worrying about.
All other configs of drive connections,single dual, raid non riad, intel/marvell, x86/x64 now resume from sleep perfectly, though all my tests were with the intel chip in raid, but I expect other configs of the intel chip to work still.
So that's one major flaw fixed! :)
-
So here's a wrap-up of the oustanding issues with this motherboard - mostly ok now.
The below is a copy n paste of what I just sent to Gigabyte.
Hi, This version works correctly! :) There is only one minor issue. If the marvell controller is in ide mode [intel in raid] and boot is over the marvell controller, it will still fail, but this is maybe not worth chasing as most users will use ahci mode.
Maybe update the user manual with an addendum - 'no ide on marvell with intel in raid'
There are some other issues, though not all critical as there are workarounds with most of them.
1. With bios version F7, it was possible to install win 7 x64 to a raid volume on the marvell controller greater than 2TB.
At version F11 [or maybe earlier] and version F12a this ability has been broken.
There is a caveat however with version F7.
Normally, when you have a fresh drive/s and you start the computer with a bootable windows dvd in the drive, after the bios has finished post it will automatically boot the dvd drive. You will NOT see the message "Press any key to boot cd/dvd", because bios can see that there are no other bootable devices connected.
However, when the drives are connected to the marvell chip and are configured to raid, for some reason bios thinks there is an alternative boot drive besides the dvd and you do see the message "Press any key to boot cd/dvd".
If you ignore that messsage, there is a momentary flash on the screen with text on the left top quoting the bios date and version and on the right top 'press del or f2 to enter setup"
If you allow this to happen, you will find that you cannot partition more than 2tb of the raid volume.
If you DO press a key when you see the boot message and not allow the flash on the screen to happen, you can partition the whole raid volume as a single partition [with the usual small system partitons of course]
However, this is only possible with bios F7, it is not possible to partiton >2tb on the marvell chip with F12a.
No matter what I do, with F12a I always get a message that windows cannot install to this volume, when you click details it says "The computers hardware may not support booting to this disk, make sure the device is enabled in bios".
At which bios version this was broken, I don't know. All I know is that F7 works [with attention to boot message] and F12a doesn't work.
2. When installing windows to a raid volume on the intel c600 controller, at the point where your asked to load f6 drivers, only on rare occassions is the dvd drive visible, therefore impossible to install drivers from driver cd.
Workaround - put drivers on flash drive first.
3. Silverstone FP34 usb flash drive bay is not recognised by q-flash. Works fine for F6 driver install.
This never worked on any version of bios. In fact, version F12a now sees my dvd drive as a place to save a bios!! :)
Workaround - Ignore dvd drive and use other flash drive to save/load bios.
4. The F6 x64 drivers for the G1.Assassin 2 Marvell controller on Gigabytes website are incorrect and do not work.
The drivers on the driver cd in the floppy64 folder are correct and do work.
I have not tested the x86 driver.
Workaround - use the F6 drivers on the cd
So, the board is quite usable now, but bug No. 1 really needs to be addressed
-
I am facing some similar kind of problem. Hope you will give me solution for the problem. Thanks. :)
-
Hi, did you try Gigabytes beta bios version F12a? It works quite well and fixes the hang/bsod when intel c600 is in raid mode.
http://ggts.gigabyte.eu/FileList/1242694/g1assas2.12a
I notice that it's still not available in their downloads section and probably won't be until it is elevated from beta to rtm status.