Personal computing discussed

Moderators: renee, Flying Fox, Thresher

 
Airmantharp
Emperor Gerbilius I
Topic Author
Posts: 6192
Joined: Fri Oct 15, 2004 10:41 pm

Memory and USB header woes...

Sat Mar 12, 2011 12:43 am

I recently moved, and recently had to move my computer around until I assembled a proper desk for it (Staples desks rock, btw, Office Depot not so much). Upon the last move, the system didn't turn on- crazy stuff. I figured it was the board, an EP45-DS3L from Gigabyte that had been a Fry's return, which couldn't run my G.Skill memory at it's full 1066MHz speed, with all four DIMMs populating the four slots. I did some testing, reset the CMOS, but couldn't get it to appear to breathe again, so I picked up one of the last EP43-DS3Ls that Microcenter had in stock (they have since disappeared completely). This board is a newer revision, and can run the memory at full speed.

Three weeks later, messing around with the USB stuff in the back, as the board was losing USB input upon waking from sleep, the new board appeared to kick out too. At this point, I'm extremely frustrated, looking for P43/P45 boards online that support the 1600MHz FSB and 4 DDR2 DIMM slots I need, and staring suspiciously at my aging Corsair 620HX. I found exactly two boards online that met my needs, and one was refurbished, though nicer than the DS3Ls I had been using.

Then, when I decided to try and do some last minute troubleshooting just for the heck of it, I was able to get some life out of my first P45 board. After playing around with it for a while, I found that I could get the system to boot with some of the DIMMS installed and not others- and zeroed on the newer kit. It was then that I realized that my DS3L wasn't running the DIMMs at the appropriate voltage- and the two newer G.Skill DIMMs would not even post without proper voltage! Thankfully, I was able to boot in with a single DIMM from the older pair, change the voltage (now at 2.2v), and get all four DIMMs running. This got me a little excited, but I realized quickly that this older board still couldn't run the DIMMs at their full speed.

Moving everything back to the newer P43 board from Microcenter, I slowly as in 50DKP Minus slowly put the pieces back together. I get board posting, the memory all checking in and at full speed, the overclock set, the wireless card (my single expansion card) installed and working, and I'm pretty happy. When I get almost everything back in, suddenly poof- it stops posting. Back apart, everything except the bare minimum necessary to post, and it's back alive again! Real Whiskey Tango Foxtrot.

With some additional testing, bringing it down to literally one connection change per test, I narrow it down- the USB headers were killing it! With those unplugged, the system boots perfectly the first time, stable, CPU overclocked per my signature and all RAM accounted for running at full speed, using their XMP profiles.

And this leaves me with a hanging question- I would like to have the USB ports working, but I don't know if I trust my old board, and I don't know if I want to go through the process of attempting to return a defective board to Microcenter. I know they're going to try and replace it per their return policy, and I'd accept a fully working replacement of this board if they had one in stock, but they do not and probably will not.

Further, in haste, I ordered the only new board I could find off of Newegg, and it's shipped already- it's the bare base P43 from Gigabyte, even lower end than these aging DS3Ls (the UD3's were the upgrade). While it will probably work, possibly as good as the P43 DS3L from Microcenter, I don't think it's worth the extra cost- so I'm about to figure out how to ship brand new things back to Newegg. Hmmp!

Who is online

Users browsing this forum: No registered users and 1 guest
GZIP: On