The option to use 3.6+ can be a good thing with BH-5, they still don't give as much heat as TCCD on 3.1 v when they are on 4 V.
And the perform better for benching.
Printable View
The option to use 3.6+ can be a good thing with BH-5, they still don't give as much heat as TCCD on 3.1 v when they are on 4 V.
And the perform better for benching.
yeah, having the increased Vdimm is good for benching. But 3.3v maybe enough for a 24/7 computer. But still a nice option to have.
jjcom
Such a simple solution... HOW COME I DIDNT THINK OF IT??? :brick:
Hopefully they will use this on their NF3 939 board as well
Quote:
Originally Posted by DaGoof
I doubt they will have 939 board. :p:
OPB
... you sure? I've heard a lot recently that they were finally going to put it out
Yes, Vivian told the same thing @ AMD Fall Tech tour.Quote:
Originally Posted by DaGoof
so this board would be definitely the best choice for OC players? And DIMM area still ought to be actively cooled?
It should be, from the looks and from what DFI has done in the past. The DIMM area should be activly cooled when using the Vdimm from the 5v rail but otherwise, no more than on anyother board it would seem.
jjcom
3.96V max ;)
That's cool, can you run some benches with the vdd set to 3.8v-4.0v and feel if the mofsets are getting warm?
Or are you afraid to blow your mem?, I can understand that offcourse.
Aaargh :stick: Are you sure of that? i can't live with the neo2Quote:
Originally Posted by Onepagebook
3.8V tested with 512*2 Mushkin BH-5 is fine for benching, with a fan blowing on MOSFET... :)Quote:
Originally Posted by alexio
Wonderful! :banana:Quote:
Originally Posted by TinTin
So NF4 ultra isn't 939?Quote:
Originally Posted by Onepagebook
you know what he meant...
how about booster on the dfi nf4 on the 3.3V line?
how about booster on the dfi nf4 on the 5V line?
I know there's no point switching to 5V line AND using booster but just for safety cuz I thought that with all boards + booster that draw Vdimm from 5V eather the booster burns or the board burns out but it never works.
Anyone know wich voltage regulator they used with 5V line on memory ?
Anyone ??Quote:
Originally Posted by Cyrus The Virus
http://www.silverstonetek.com/products-tj06.htm
This case looks like it could help with cooling CPU/MEM/MOSFET area!
:D O.K. guys - maybe I got somethin' wrong but DFI manual says if jumper is set to 3,3 V rail max Vdimm is 2,85 and OPB and Oscar told us to switch the jumper to the 5 V rail only when using more than 3,3 Vdimm.
But what to do if something between 2,85 and 3,3 is needed ???
Does anyone know ???
:toast:
That manual is more than likely incorrect. Using the 3.3V rail, it should at least get you 3.1V-3.2V on your VDimm.Quote:
Originally Posted by HARDCORECLOCKER
yeah, since it came from Oskar chances are the manual is wrong...maybe an early beta version? Yeah, if its like the other boards...about .2-.15v below the 3.3v rail is the Vdimm. So should be nice :D
jjcom
:D Good news here - my G.Skill LA like to get 2,9 - 2,95 at higher speeds so it will be working fine just from the 3,3 V rail.
Thanks all !!! :toast:
Then, why did the DFI Infinity and LanParty nForce2 motherboards draw VDIMM off of the +5VDC rail for 24/7 use? My old Infinity NF2 didn't even heat up very much around the VDIMM Mosfets.Quote:
It's in OPB's post, using the 5V causes too much of a heatload on the mosfets in the RAM voltage regulator. That's why they prefere to use it only when it's really necessary, like when you need more than 3.2
Add some nice copper heatsinks, and you're all set so long as you have good airflow. I don't think that a fan specifically for the Mosfets will be too easy to implement well (look at all the power cables around there), and hopefully not even neccesary for stable operation.
As far as I knew they didn't...Quote:
Originally Posted by felinusz
jjcom