Gigabyte EP45-UD3P fails to resume from sleep

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

howzz1854

Junior Member
Apr 16, 2009
4
0
0
Originally posted by: zagood
I finally got S3 to resume on my UD3P. Followed the advice in this thread:
http://www.vistax64.com/tutori...eep-mode-problems.html

in BIOS (F6):
1. Enabled legacy USB for kb/mouse
2. Enabled PME event

in Vista:
1. Restored Hibernation file (link on how to do this in the thread above)
2. Used "Recommended Settings for Sleep"

I did all of these at the same time so I don't know which one did it. In addition, this is the first time I've tried in a few months, so it may have also been new nV or Creative drivers, who knows. I'm just stoked it's working.

FYI - Enabling the Hibernation file will eat up a chunk of HDD space the same amount as your ram: i.e. 4gb ram/4gb file on your hard drive. So I'll probably test to see if that was the culprit, and if not, delete it for good.

-z


but what happens when you raise the FSB. the problem is when FSB increases to certain level, S3 cease to function. and what makes it even more problematic, the S3 FSB wall is different for everyone. some say 400 and below work fine, others 420, for me, it's 370fsb.

have you tried raising your FSB>?

 

BonzaiDuck

Lifer
Jun 30, 2004
16,358
1,894
126
I feel like a thread-vampire, but I got a UD3R. Had too much trouble with an older processor; got a E8400 to work with my 680i motherboard -- testing configurations for a server. If I had problems with sleep/wake on the configuration, I was going to switch back to the UD3R mobo.

So this is "interestin'." Anybody remember whether the sleep features are any good an a 680i motherboard?

we live in a energy demanding world where the prices of energy has gone up 10 fold in the last decade. this makes every price conscious consumer care deeply about their energy consumption.

You said it, howzz1854!! You said it! . . . . :frown:
 

zagood

Diamond Member
Mar 28, 2005
4,102
0
71
Originally posted by: howzz1854
but what happens when you raise the FSB. the problem is when FSB increases to certain level, S3 cease to function. and what makes it even more problematic, the S3 FSB wall is different for everyone. some say 400 and below work fine, others 420, for me, it's 370fsb.

have you tried raising your FSB>?

I'm on an e5200, FSB craps out around 370, using 333mhz right now. So no, can't test that for you.

I'll just say that I couldn't get it to resume from S3 before making the steps I referenced earlier. You may want to try it.
 

zagood

Diamond Member
Mar 28, 2005
4,102
0
71
Looking into the issue of FSB increase killing the S3 functionality, I'm thinking that maybe it has something to do with the amount of voltage that the ram is supplied with - since the state of your computer before sleep is stored in memory.

Ram isn't like non-volatile memory, it needs juice to perform correctly - so with ram at higher speeds, needing more juice, there may be a problem with motherboards keeping volatile memory stable in sleep mode.

My 2c.

Try upping your ram voltage or lowering ram frequency. Worth a shot just for testing.

-z
 

howzz1854

Junior Member
Apr 16, 2009
4
0
0
I updated to F9, it's iffy at best. it not only doesn't fix the problem, it seems might have made it worse. i used to be able to do 370fsb with 333 NB strap and S3 would still work. now it wouldn't... i almost flashed it back to F6 untill i finally got 370fsb to do S3 again but with 266 NB strap.

so there you have it, Gigabyte has made no progress. and i give up trying to figure it out, there're too many other things in my life that i have to deal with than killing my own brain cell trying to figure out why this motherboard wouldn't do S3 properly.

i do have to say, i think it all has something to do with gigabyte's NB strap VS memory setting. maybe someone who has enough time can pick up from here and figure out for the rest of us what the cause of the problem is.

should've stuck with Asus.
 

howzz1854

Junior Member
Apr 16, 2009
4
0
0
Originally posted by: zagood
Looking into the issue of FSB increase killing the S3 functionality, I'm thinking that maybe it has something to do with the amount of voltage that the ram is supplied with - since the state of your computer before sleep is stored in memory.

Ram isn't like non-volatile memory, it needs juice to perform correctly - so with ram at higher speeds, needing more juice, there may be a problem with motherboards keeping volatile memory stable in sleep mode.

My 2c.

Try upping your ram voltage or lowering ram frequency. Worth a shot just for testing.

-z

unfortunately, i wish it was that simple. i've tried pumping Dram voltage before, or lowering the memory frequency, still no effect. the problem seems to be certain FSB frequency that causes S3 to go irratic, and since FSB is so closely related to NB strap it just makes the matter even more ambiguous.

btw, i tried your previous suggestion, it didn't work for me. i think in "your" case, it was more of an OS issue, you were probably having issues already with Vista's sleep function, which evidently you need to disable "allow this device to wake the computer" and other Hibernation related setting under the vista device manager keyboard/mouse advanced setting.

in my case, and some other's cases, the S3 already worked in stock FSB and CPU frequency (which eliminates the possibility of the OS issue). only when you increase the FSB, the S3 starts to behave erratically.
 

jrichrds

Platinum Member
Oct 9, 1999
2,537
3
81
Originally posted by: howzz1854in my case, and some other's cases, the S3 already worked in stock FSB and CPU frequency (which eliminates the possibility of the OS issue). only when you increase the FSB, the S3 starts to behave erratically.
Yup, that's my case. S3 works fine until ~420FSB with my E7200 and UD3P. If I go higher, I get an endless start-stop loop when resuming from S3 and have to shut off the computer with the PSU switch in the back.

I had some S3 issues on overclocked E7200/ECS boards that were memory voltage related (memory passed all torture tests at 1.8v, but I needed to feed it 2.0v to properly resume from S3). Feeding my 2x2GB Ballistix plenty more voltage than needed didn't fix the S3 issues on my UD3P though.