Hey everyone, I've been having some serious problems with my new motherboard and ram. I have the GA-P55m-UD4 with cmd4gx3m2a1600c8 ram (which is to say Corsair Dominator 4gb 1600 Mhz cas 8 dual channeled). Anyway my problem is that the motherboard only recognizes 1 of the sticks, however it knows the other one is there. It is very strange, the bios will say that it has two 2gig sticks and yet only use 1 of them. Does anyone else have this problem and have they fixed it?
I think i may have found out why the problem is happening, but i am not sure. On the manufacturer's, they have a list of compatible memory and my memory isn't on the list. I also had the same problem on a GA-p55m-ud2 board, so if anyone has similar problems on that board this thread might help them as well.
Thanks to anyone who figures this problem out or cares to even look at this.
So, I can load the System with either stick of memory, and when i set the timings and voltage it doesn't help the problem. I am running Windows 7 64 bit if that helps. i can use the XMP profile also which auto sets the ram speed and timings and suck but that doesn't help either. Also thanks for responding quickly hahah didn't expect anyone to see it so fast.
Message edited by darkpower45 on 09-21-2009 at 07:42:29 AM
ok so im guessing you want to know everything I have a 850 watt Coolmaster and a 4870x2 sapphire brand. The timings i set are 8-8-8-24 and voltage of 1.640 (which is the closest it can get without going over 1.65 which id rather not do) and the QPI set to 1.150. The timings and voltage are also set at those specs when set to Auto under XMP profile. If i dont enable the XMP the timings are 1333 Mhz, 9-9-9-24 with 1.5V. And yes i manually set the timings, unless stated otherwise. haha
So the memory sticks are "good" - at least they work independently.
I suppose you could try the relaxed timings the mobo set on auto, and kick the voltage up to 1.64.
I suppose you could also try relaxing the load on the psu by pulling one of the vid cards. If that works, the problem is likely the psu. If you have another psu, you could also try that with a single vid card if the psu is smaller.
well many other people have had this same problem with this motherboard, also i had all the parts running on my old Maximus Formula with a Q6600 before i switched to this new mobo and processor.
Had a similar problem with my Gigabyte P55-UD4P board and 8 GB or Corsair 1600 RAM... would boot fine, but after re-boot it would disable half the RAM. Tried replacing MB, no dice... new BIOS, not it... finally Replaced the Corsair RAM with OCZ Gold and presto, 8 GB all the time!
I've seen several people with this issue, with different MBs... almost all are using the same RAM or something similar from Corsair.
I am suspecting that Gigabyte will eventually release a bios update to fix this problem with the ram but not sure when that will come out, does anyone else know? Oh and thanks Mak-daddy, thats what i figured after looking at the supported memory sheet on the Gigabyte webpage. Oh one more thing that is strange, does anyone else have Turbo feature one where it raises each core by x1 instead of the 1st core by x4? i was curious on how to check if the turbo feature is working correctly? i have never seen the first core got to x24 multiplier even when running single thread tasks.
Message edited by darkpower45 on 09-21-2009 at 08:37:41 PM
I'm not sure it's an issue with the BIOS as I know of at least one person that replaced their Corsair RAM with another identical set from Corsair and that resolved the problem... I'm wondering about a bad batch of RAM getting out?
Use CPU-Z to monitor the speed and multiplier... Turbo worked much better for me after moving from the F3 BIOS to the F4p for what that's worth.
Watching CPU-Z while running video's in Premiere Elements I can see it jumping around quite a bit depending on the load.
I haven't written down all of the specifics of how the BIOS set it up, but I am having more or less the same problem with the same board, except I have 4x G.Skill 1333, a Corsair 450VX video card, and a Gigabyte GV-R487D5-1GD for video. Win7RTM says 8.00 GB (4.00 GB usable). CPU-Z sees all of the memory. Memtest+ reports SPD info for all the DIMMs. The memory values reported by CPU-Z appear to match the JEDEC #4 SPD profile.
Gigabyte Support is clearly aware of this issue. (They referred to it as a G.Skill compatibility issue but there are reports of other memory and other P55 boards having similar problems--there are several reports in amongst the 11 reviews on the GA-P55M-UD2.) Beyond that, all they'd say is engineering is evaluating and there is no date scheduled for a beta BIOS release that might fix it.
So now I'm in a quandry: return the memory for something else and eat a restocking charge but do it soon, or live with 4GB till there is a BIOS fix or, potentially, get stuck with the 4GB unusable if I doddle too long waiting for a BIOS fix that never shows. Say it's a P55 and/or rev. 1 board problem? Then only hope might be different memory that "gets lucky".
What to do? What to do?
(Other than that, the machine works great. I'm posting from it...)
I say return it for some memory that is listed on the compatible RAM chart, eat the re-stocking fee, and in the future shop somewhere with better customer care.
a lot of the memory on the Ram chart is actually triple channel memory or 2 gigabyte dual channel memory. The chart also has older DDR3 that uses 1.8V to use, so I dont know exactly gigabyte is testing this motherboard with those types of memory and not instead test the new memory that are dual channel and use less than 1.65V??
I say return it for some memory that is listed on the compatible RAM chart, eat the re-stocking fee, and in the future shop somewhere with better customer care.
I bought the whole pile of parts from newegg...
But following your thought process, I look at the qualified vendor list for memory for the GA-P55M-UD4 on the web as of this moment and find:
Size Module Supplier Module P/N. SS/DS Comp. Used Component P/N. Single-channel Dual-channel XMP
2GB G.SKILL F3-10666CL8D-4GBHK DS v v
That's EXACTLY what I'm using. The list never says that you can't use 4 of these. But if you do they won't all work. So if I return it to "get something on the list" then how do I trust anything on the list works? What good is the list if my case is failing?
Message edited by dick_w on 09-23-2009 at 01:01:41 AM
Alright I got new ram for this motherboard and they work fine so far. If anyone else wants to try them they should work too. I got Kingston 4gb 1600Mhz model number: KHX1600C8D3K2/4GX. These have lower timings that my Dominators, but i am not sure how high they can overclock yet. With my Dominators, i was able to reach 1900Mhz with 9-9-9 timings. we will see.
Thanks for your posts! Since I am planning to build my HTPC arround the GA-P55M-UD4, I was worried that the memory modules I choose would cause me troubles described here. I asked Gygabite techsupport to provide a feedback and here thier reply:
---------------------------------------------------------------------------------------------------
to resolve problem on forum, just adjust memory voltage match in bios.
Mother board bios memory voltage setting is base on 1.5v (default voltage), please check current model memory label for voltage spec information, if it's higher then 1.5v go in bios under M/B intelligent tweaker, on memory voltage control item adjust memory voltage to match. For example: current memory is 1.7v go in bios M/B intelligent tweaker on DDR3 voltage control item change from 1.5v to 1.7v, this will match the current model memory voltage spec, system will work stable.
Please log on these webs for memory list which has been validated on P55 chipset board :
Maybe it's that simple but I'm using 1333 memory straight off the Gigabyte qualified memory vendors list down to the exact part number. It lists "1.5 ~ 1.6V" on its label.
In "Fails Safe Defaults" Win7 reports 8.00GB (4.00 GB Usable). In "Optimized Defaults" Win7 reports 4.00GB. Gigabyte BIOS never reports more than 4096MB memory. CPUID sees all of the SPD ROMs.
My communications with Gigabyte TS suggest that there will likely be a BIOS fix for this at some point Real Soon Now. Let's hope.
This article appeared today and may shed some light on this problem. The overall article is a motherboard faceoff, but they reveal some "problems" their mobo testing to date has uncovered: