mm 500 FSB with a quad is already extreem, bundle that with an entry level board like the DS3P, I would opt for the DS4 rev 2.0 mate better PWM area, heatpipe etc... and surely will last longer when being heavily pushed then a DS3P
Printable View
mm 500 FSB with a quad is already extreem, bundle that with an entry level board like the DS3P, I would opt for the DS4 rev 2.0 mate better PWM area, heatpipe etc... and surely will last longer when being heavily pushed then a DS3P
kindda mixed up DS3P with DS3 :p my humble apologies, the heatsink works fab though on the DS4, in comparison to my P5K Dlx...
I can't recommend the DS3P for anyone wanting to OC a quadcore high and do anything 3D. The PCI-E x1 problem is a huge :banana::banana::banana::banana::banana:.
have you try it??? im goin to use dry ice with 1.7++ Vcore and of course would do 3d and some bench software....
Yeah, I used the board for a few weeks, and it kept reverting the PCI-E x16 slot to x1. I needed more than +.2 vMCH, which you probably will to, and that is what usually triggers the x1 problem.
That's what Gigabyte rep thinks... :p:
Indeed P35 chipset inherited a prob from 965 and that's PCI-E 1X prob. I've seen reports (very little though) about same prob happening to some P5K boards (Deluxe and Premium) but these are a minority. In gigabyte m/bs this is the trend.
I've used to own a P35 DS4 for 2 weeks and that problem was the main reason i returned it back. Grabbed a IP35 Pro and have been more than happy. No pci-e 1x prob and that brings us to the Gigabyte rep... ;)
There haven't been any reports with X38 boards reverting to 1X.
I just got my DS3R, and it's great except for the fact that it won't POST with my PC-8000 2.2v Crucial RAM.
It will, however, work with some old Corsair RAM which I think is rated at 1.9v.
I've tried both Crucial sticks individually, and I'm sure they aren't bad.
So, is this a known issue, and is there any way to resolve it other than switching out the RAM sticks after manually changing the voltage?
I never had the PCIex x1 issue.
For testing purposes I flashed my DQ6 to BIOS F7b.
Set my usual overclock (8.0x420, RAM divider 2.50, vdimm +0.25v)
And I raised the other voltages, including MCH. First to +0.2 then later to +0.325. I manually set PCIex frequency to 100.
PCIex x1 issue did not show up. I still have PCIex working at x16.
Can't see any useful changes from F6 to F7a/b so far. Now the CPU clock is displayed at the MIT menu. But the voltages are still shown "OK" rather than the actual readings.
So I don't know if the PCIex reverts to x1 in particular cases, or in combination of hardware used, or higher FSBs. Who knows...
Okay, this is weird...
I set the vdimm to 2.2 with my corsair sticks in, switched em out with my crucial sticks, and now when i boot, instead of posting, it loads up some bootblock bios thing that just says "scanning bios image in harddrive"??
Uh... help?
darcon, if you get that everytime you power up your PC it means that your BIOS is corrupt.
If its looking for the file it may be easy to make the BIOS recovery procedure.
Try copying the BIOS update file to a floppy disk or to a USB drive and power on the computer with either of them, maybe the computer will look for the BIOS update file and flash the BIOS.
I don't know what is the precise BIOS recovery procedure for these new BIOSes but it should be something like that...
Weird, did you try one DIMM at a time?
Well, the f7d bios didn't help, so I'm back to square one.
Something weird did happen, though. The last time I tried the crucial sticks, I heard a strange metallic squeak as I powered down the machine. It only did it once and I couldn't pinpoint the location, so I dont really know if it has anything to do with it at all.
Anyway, I guess I'm stuck with my slower ram till I figure out what's going on.
Did you try a Clear CMOS procedure?
The thing is, I really don't have any way to test them, and having them both go bad at the exact same time just seems so unlikely.
Would there be any hassle if I sent them back to crucial and they turned out to be fine?
PS-Ugh. I think my exact modules have been discontinued. I also found out that I payed over twice as much for them december last year than the pc2-8500 kits are now. That really pisses me off.
The droop on mine is just killing me. Wasnt too bad with dual core but with a quad I select 1.35v in bios and under load I get about 1.20v. Idle is about 1.32v. I need a board that has a simple pencil mod.
I need some help here with a DS3R rev1.0. I had an ASUS P5W DH Deluxe before and i had a 24/7 config:
E6400@3.2Ghz (400x8)
2x1 or 4x1GB Crucial Ballistix PC5300@800Mhz Cl4-4-4-15
.
.
.
Now i cannot make it work at the same speeds with the gigabyte (CPU and memory are the same), i think i need some help with the timings and some tricky advise too.
Any idea what to do? maybe is the vcore drop?
Yup did that as well. Mentioned that in my first post here. Thanks for the suggestion though.
Have you tried overclocking a quad on I35pro? This review wich was released just yesterday believes that the Vdrop/Vdroop is pretty bad when you try overclocking a quad core on that board. My curent board has its own Vdrop/Vdroop but it's not nearly as big as what is shown in that review. In fact I am pretty satisfied with the Vdrop/Vdroop of P35-DQ6 it's pretty slight and has not caused me any troubles wich again makes me think why such a perfect high quality board should be plagued with this annoying pcie issue .Seems like every board has its own achilles heel. :shakes: