Hi, I've recently installed Win7 x64 on my new DP55KG and I'm expecting some quite boring issues.
First of all, most unsuitable, I've on every boot a message from event log that says :
"Le chargement de \SystemRoot\SysWow64\Drivers\EBIOS32.SYS a été bloqué en raison d’une incompatibilité avec ce système. Contactez l’éditeur de votre logiciel pour obtenir une version compatible du pilote."
The SMI driver is not compatible on x64 OS ? So I can't use the .exe flash utility to update my bios ("This program is unable to continue. Internal system error: There is a programming or internal problem"), worst I can't use the Intel DCC (XTU service can't start) or other software that uses SMI to get infos from the motherboard such as cpuz etc... It's incredible for a mobo that is supposed for enthusiast users !!
Up : It might be cause of UEFI boot enabled, so DON'T ENABLE UEFI boot before windows installation if you don't have a partition larger than 2TB you'll avoid problems !
Then the other boring issue is since bioses after the first relased 3206, my computer have to boot twice to load windows. When I push the power button, it starts, I can see some post codes scrolling, it makes a short pause on 66 and when it switches to 69 it restarts and boots normally. I've tested all of bioses from 3456 to the new 4507 and only the first can boot properly. I've tired to unplug all sata devices or increase the sata drive delay but nothing changed.
Does someone else is expecting this ???
An other issue happens when I switch my computer into sleep mode, I've a message in the event log that says :
"Le microprogramme de la plateforme a endommagé la mémoire lors de la précédente transition d’alimentation du système. Recherchez un microprogramme plus récent à utiliser pour votre système."
It happens with every bioses I've tested, all versions, but no problem when the computer gets out of sleeping.
An other issue with the sleep S3 mode : I've overclocked my processor with 160mhz blck and 1.35Vcore dynamic and it's stable on windows with all turbo modes. But when the computer gets out of sleeping, after few seconds, the system hangs (no blue screen) or no display anymore. I could see from Everest that it could be because of the Vcore got reinitialized, above 0.86V in EIST mode.
New issue I've forgotten to mention, sometimes the fullscreen boot logo isn't display in center of the screen, often a little strip can be seen at the right side of screen...
Some other problems like low speed fans can't start from cold boot because of not enough power (http://communities.intel.com/message/78438), wrong sata port detection (http://communities.intel.com/thread/9432) or bluetooth disabled (http://communities.intel.com/message/81492) but they have been already posted.
If I found new issues, I'll post them here.
Message was edited by: Ultraboss
The EBU (Express BIOS Update) method uses the same technical path as an S3 sleep state cycle (suspend to RAM). If S3 fails, your BIOS update will fail. In your case, you said you are running overclocked... well in your overclocked state your memory might be unstable enough that it does not survive an S3 transition and therefore your BIOS updates are also failing. In general... here's a very solid rule of thumb...
If you are going to flash on a new BIOS, don't be overclocked during the process.
Bluetooth enable/disable which accidentally broke on BIOS 4507 is already fixed and will be available in a new BIOS update soon (probably within 1-2 weeks from now)
About the restarts on your system, this might also be due to your overclocked state. Getting to POST 66 means you made it through Memory Init, but perhaps your memory is getting stressed later. I suggest first trying to go back to BIOS defaults and see if that problem goes away. If yes... then it means you need to keep working on your overclocking profile because you made your system unstable. Don't forget... when you increase the Base Clock, to a decent value (like up at 160 or more) you need to increase CPU voltage, UnCore Voltage, and Memory Voltage. Try 1.3v 1.3v and 1.65v with that base clock and see if that stables it out at 160 BClock.
About the fan thing, we have a kick start algorithm in place, but some fans are just too old, too low quality, or too big for that kick to get them going. You can disable the fan control to run it at full speed and that might work, but replacing with a different fan might be best.
About the SATA drive issue, we are looking into it now. I have thankfully not yet heard that it causes a failure... just reports wrong.
Thanks a lot for taking time to answer to all of us.
As it's written in the EBU topic, it's a driver bug with UEFI enabled, but we can't disable it after windows is installed, beause the disk is formatted into a special partition and the mobo won't be able to boot on it if UEFI is disabled...
My system is rock stable with 160blk, all voltages at stock except vcore set to 1.35v dynamic and vram to 1.4v (1.35v by default). I haven't restarting problem with the original bios, with or without oc, only with the others.
And for the fans issue, why don't you increase a little the kick start, I think about the lastest Noctua fan 14cm...
I just wanted to add a few things to this discussion for Edwin:
1. I am a firm believer in running stock clocks on everything. I was running 8GB (4*2GB DIMMs) DDR3 1333MHz dual channel RAM at JEDEC standard timings of 9-9-9-24 @ 1.5V and the Intel i7-860 CPU. Windows 7 Ultimate x64 full retail was BSODing like crazy until I upped the RAM voltage to 1.6V. Could there possibly be a bug with the voltage being supplied to the RAM on the DP55KG?
2. The Noctua 140mm fans are considered by some to be the best fans money can buy for high-end performance / quiet computing. Would you please considering upping the initial boot-up voltage on the fan headers so these fans will spin-up properly at boot? That way we can still use the BIOS controlled fan option and set it to SLOW for quiet fan operation.
Ultraboss: If you increase the Core voltage, you should really increase the UnCore voltage as well. For me, if I put one of them at an increased value, I match it with the other one. In your setup, I would try putting the UnCore voltage at 1.35v as well. I might even push up the RAM voltage a bit more too, you might get a stable system from it. It's important to note that there are MANY RAM modules out there that just flat out don't run well at the posted specs or what settings are contained in their own SPD chip. Some RAM companies literally make a very bad product and customers often end up blaming board vendors when that bad RAM stick does not run.
Zorlac: There is no issue I know of with RAM voltage, in fact I use a DMM to measure the voltage of the RAM and it's pretty much right on. But see my comments above about some RAM vendors.
Both of you: Yeah, we can look into increasing the "kick-start" for the fan control to a bigger kick. Perhaps even put in a BIOS option for big kick, or small kick?
Regarding RAM, I am using Corsair XMS3 which I suppose is a midrange series from them, but again I have no desire to overclock. Oh well, 1.6V is working great for me.
Regarding fan voltage, that would be awesome to have a big-kick/small-kick option! I know the X58 folks have been complaining of the same issue, so you might offer this setting for that board as well.
One more thing, any noticeable bugs fixed with the latest P55 stepping offered on the latest AA revision 405 for the DP55KG? Basically is it worth it to track this AA revision down and replace a revision 403?
Perhaps a little vdroop on vmem ?
Yes an option to choose big or small kick will be useful !
I have to report another thing about fan management, I have 2 Arctic F12 PWM and if I set them to 3pins, they run lower than 4pins...
Another option could be useful is the possibility to modify the cpu fan management, like silent normal and turbo.
Zorlac: No, both the 403 and the 405 boards are both production quality.
Ultraboss: I doubt if there's any V-Droop on the memory voltage rail, it's significant. It's FAR more likely that the memory you bought just flat out requires more voltage.
I assume the Arctic F12 PWM is a 4 wire fan, because it's called "PWM" in it's name and 4 wire fans are driven from PWM signal on the 4th pin (which does not exist on 3 wire fans). So you should not be plugging it into a 3 wire port, which is what Kingsberg ports will be if you configure them as 3-wire in BIOS setup. Remember my answer above, if you do that, the fan is being under-voltaged on a + rail it expects to be full. Now... perhaps the fan was meant to work with both types and they added control circuitry for both? I don't know, never used that fan. But if it has 4 pins on the connector, I would set BIOS to 4-wire.
Unfortunately I can't comment on future features that we do or don't have in development. Helping you guys with bugs, or giving you some idea when a bug will be fixed, or collecting current product feedback is all good, but I'll put myself in a hotseat if I say anything about future development.
From that though, if there's something you would like to see... tell us about it.
I've just manage to flash my bios to 4507 via recovery mode, it was quite laborious, it has hung during my first flash try.
When the flash was done, I've checked for my "boot twice" and still there altough a fresh flash with all default settings
So I've loaded my custom settings and when I start, the computer restarts at post 10 looping. My previous oc settings doesn't work anymore with this new bios !
But the only positive point is that I still have my bluetooth !
And what about this option to disable C2 state report? Was it enabled by def in previous bioses? What is the utility of C2 state report?
I've just tested the new 4752 bios and it made me sad.
I've read in the release notes "Fixed issue where system will perform two POST cycles before boot" but my mobo still boots twice
I've read too that an XTU processor compatibility was fixed but XTU still doesn't work, I guess it's a software bug, I've mailed them about it but no answer yet.
And like the previous 4507 my oc settings don't work anymore