Official GIGABYTE Forum
Questions about GIGABYTE products => Motherboards with AMD processors => Topic started by: stuart on September 18, 2009, 11:26:31 am
-
Will the UD5P auto-detect DIMM voltage requirements and lower the voltage supplied if the requirement is below 1.6v?
Looking at the BIOS options, DDR3 voltage increases are possible, with the smallest increase (0.05v) leading to a total of 1.65v being supplied. However, I have 1.5v RAM. That extra 0.1v may not be doing any damage (I hope) - but at the best it's needlessly raising temperatures, and at worst it could limit to life of the component.
There should, at the very least, be BIOS options to reduce as well as to raise these voltage levels.
-
You should be able to manually set the voltage to 1.5v in the bios RAM setttings.
-
You should be able to manually set the voltage to 1.5v in the bios RAM setttings.
Nope: the only options are to increase the base voltage, with the lowest increase being 0.05v, and with this selected the actual DDR3 voltage is shown as 1.65v - which seems to confirm that the default level is 1.6v.
This is pushing too much voltage through a 1.5v part (which is what I have) and, at the very least, will be generating more heat and being less efficient than is necessary.
Are there any hidden (or otherwise) BIOS options to lower the voltage to 1.5v - nothing additional seems to appear after pressing Ctrl+F1?
-
Oh right strange! Does it let you type the voltage in? Have you tried updating to the latest BIOS?
-
I've had to downgrade back to F5 because F6 and F7a appear to be completely broken for use under Linux - but none of them appear to have any option to reduce DDR3 voltage below 1.6v. Voltage selection is only via a drop-down menu - there's no way I can see to enter a voltage directly.
If this is correct, it strikes me as a fairly fundamental design-flaw...
Hopefully Gigabyte is aware of this issue (I've raised a technical support query on the topic) and will address the issue with a future update.
-
Stuart
do you experience any BSOD's because of this higher voltage value?
-
No - if that was the case I'd have said so.
However, I'm hoping that you're not angling for the "it's not hurting anyone" approach, because needlessly over-volting a component because of a lack of a configuration option will not only generate more heat and consume more power than is necessary, but it is also running the component out-of-spec (by almost 7%) and could potentially reduce its operating lifetime.
Just because the system is stable doesn't mean that it's not an oversight that should be fixed...
-
ok, i see your point.
i'm curious what gigabyte will tell you
-
Highlight the required voltage setting, using the arrow keys, make certain you have selected manual voltage control 1st, then use the + and - keys to raise or lower.
Works on mine using F6 BIOS, BTW what and where is F7A ?
-
Having selected manual voltage control, the only options are (from memory) "Default", "+0,05", "+0,1", etc.
"Default" (or "Auto") sets the voltage to 1.60v, and all of the other options are incremental - none reduce the voltage.
I haven't tried using "+" and "-" directly - possibly these will allow unlisted options?
F7a was a test-release of the F7 BIOS, which is probably now obsolete. I now have a new F7 release which I'll be testing this evening.
Thanks for the suggestion - I'll keep you posted of how I get on.
-
Any chance of a link to that F7 BIOS ?
-
I saw sent the BIOS by Gigabyte technical support, but I'll have to wait until this evening to get my hands on it myself!
-
Highlight the required voltage setting, using the arrow keys, make certain you have selected manual voltage control 1st, then use the + and - keys to raise or lower. ....
i haven't heard about such way of changing voltage, i have different mobo but will try this. thanks for info m8 !
btw you use linux + firefox, as i can see, right?
-
I use many operating systems, as a PC hardware tech I come across lots of different op systems in use by customers, so I have about 5 installed at the moment.
Also I've been a Beta tester for Microsoft since Win 98SE.
When web browsing and doing day to day stuff Ubuntu AMD64 is my system of choice.
-
Interesting... your may be able to help with my other problem:
http://forum.giga-byte.co.uk/index.php/topic,587.0.html
In short, after brief use of my first 955BE + UD5P combination, I booted into Linux and found that it took minutes to boot, pausing for 30 seconds for each processor core and then reporting "Failed to initialize APIC on processor #{x}" (or something very similar. Thinking this to be a processor or motherboard problem, I returned both and received replacements. The new combination worked - until I upgraded to BIOS F6 - at which point the same thing happened. I downgraded back to BIOS F5, and the problem went away. Upgraded to BIOS F7a, and the problem came back. I'm now back on F5, and seeing warnings about not being able to calculate loops-per-jiffie (which I don't believe were there before) but otherwise appears to work.
This problem has manifested on my system with self-build 2.6.30 and 2.6.31 kernels on Gentoo, and Ubuntu's stock AMD64 2.6.28-15 kernel. I made no changes at all except for a BIOS upgrade/downgrade between the working and non-working situations.
I've reported this to Gigabyte technical support, who report that they are unable to reproduce the problem in the same circumstances :(
(This is two different processor and motherboard combinations that I've experienced this problem with, and two different Linux distributions with at least three different kernels - I can't be that unlucky that it's just me!)
Over the weekend I'll re-install Ubuntu yet again running the latest BIOS (F7?) in the hope that it'll work...
Do you the latest 9.04 64bit release of Ubuntu? Have you noticed any delays in the boot process (the progress bar should move smoothly from left to right within about 5 seconds or so)? Could you please post the full 'dmesg' output so I can compare it with mine?
Cheers,
Stuart
-
I use many operating systems, as a PC hardware tech I come across lots of different op systems in use by customers, so I have about 5 installed at the moment.
Also I've been a Beta tester for Microsoft since Win 98SE.
When web browsing and doing day to day stuff Ubuntu AMD64 is my system of choice.
5 systems (what are these?) in one PC :o, how can you manage this? i look up to you. i tried myself Win XP + Ubuntu but it was not success, maybe i did not have enough patience and time to fight it.
-
dmesg output attached. 8)
-
Many thanks for that - I now seem to have solved the problem: I upgraded from DOS, and all of the problems (Linux booting issues, NIC problems, etc.) went away. To test this, I then downgraded and upgraded again using my original method - using the Gigabyte @BIOS tool under 64bit Windows 7 (which I have legitimately from a Microsoft Technet subscription). Lo and behold, all of the problem returned. On performing the upgrade from DOS again, the problems again went away.
My conclusion? @BIOS should be considered dangerous, and should not be used.
The great news is that everything now seems to be working as intended - I just need a way to run my RAM at 1.5v!
-
No problem happy to be of assistance.
Beekeeper, the Operating systems I have currently installed are;
On the RAID array, I have Ubuntu AMD64 & Windows XP Pro 32 bit.
On one of my Seagate drives I have for partitions with 3 operating systems, these are all 64 bit;
Vista
XP
Windows 7 Enterprise.
I also have a small (40gig) IDE drive running OSX Panther.
The RAID array is using Grub to select operating system.
The Seagate is under control of the Windows 7 boot loader, I then use the F12 key to get BIOS based Boot Manager and select the Disc to boot from.
No way would I consider trying to use a boot manager for all the lot. ;)
BTW anyone have a link fot the F7 BIOS I've searched everywhere and can find lots of people mentioning it but no link to access it. ???
-
G'ah! It's still happening, but no longer all the time now. After loading Windows 7, a reboot into Linux triggers the APIC problem (although I also powered down an eSATA drive), but powering the machine off completely and starting Linux works correctly.
This is an improvement, since before I upgraded the BIOS from DOS this bad behaviour would occur on every boot.
It makes be think that the BIOS isn't completely resetting the state of the machine on reboot - what else could trigger these issues?
-
........My conclusion? @BIOS should be considered dangerous, and should not be used.......
fully agree m8!
-
No problem happy to be of assistance.
Beekeeper, the Operating systems I have currently installed are;
On the RAID array, I have Ubuntu AMD64 & Windows XP Pro 32 bit.
On one of my Seagate drives I have for partitions with 3 operating systems, these are all 64 bit;
Vista
XP
Windows 7 Enterprise.
I also have a small (40gig) IDE drive running OSX Panther.
The RAID array is using Grub to select operating system.
The Seagate is under control of the Windows 7 boot loader, I then use the F12 key to get BIOS based Boot Manager and select the Disc to boot from.
No way would I consider trying to use a boot manager for all the lot. ;)
BTW anyone have a link fot the F7 BIOS I've searched everywhere and can find lots of people mentioning it but no link to access it. ???
nice set of OS'es.
F7 bios is normally available at : http://www.giga-byte.co.uk/Support/Motherboard/BIOS_Model.aspx?ProductID=3005
while previous beta version (F7a) was mentioned in this post: http://forum.giga-byte.co.uk/index.php/topic,449.msg2499.html#msg2499