Just Putzen Round

RGone said:

Plus the mobo stays a good bit drier.
RGone...

Click to expand...

Except when you have a leak of course.

I've knocked off some LN2 runs with H2O myself in the past and have to agree, it's a nice feeling to do that. Also proves there's more to OC'ing than just getting it cold.

Ohhhhh I am liking this case and the Chipset / VRM block. Just started stepping on this puppy. Thought I would just jump to the chase and start at 4.6GHZ. Gotta admit I am liking what I see so far. In general the view over in the water cooling section is that cooling the motherboard and ram is not needed except in rare occasions. After roughly 30 min of prime @ 4.6 I see just a meager 3*C temp difference on the CPU Socket temp compared to the CPU CORE Temps. This might be something to look at for those looking to push hard. The unfortunate part I have only ever seen Northbridge / VRM blocks for the CHV and CHVz.



@4.7GHZ



@4.8GHZ

Looking good bassnut. I think you're right about the blocks. You'd have to use generic ones. I don't think I ever found one for the Sabertooth. Not that I was going to get one I just like to look at shiny computer stuff now and then.

It's not too often on these Fx's that you see the Socket and Cores as close as that Bassnut. I know for myself with a 80mm fan on the Vrm and a 120mm on the backside of the motherboard, I'm usually in the 9-10c difference range.

Edit out of curiosity Bass what do you have the Cpu NB voltage set at?

Here's my rig at 4.7 Cpu V 1.4625 Cpu Nb 1.25 LLC High on both Cpu and Nb, 20 minutes blend. With all the power saving features on for 24/7

Take a look at these two pics. It might explain a bit of the temp difference.


Here is the Northbridge / Vrm cooler from my Motherboard. You notice the back plate for it.



Here is the back side of the Motherboard where the cooler is attached. I have placed the backplate roughly where they go. With each backplate there is some thermal tape when you look close there is 10 chips that these backplate cover that I assume need cooling as by the presence of the thermal tape. My waterblock does not need these backplates these chips are now open to any airflow I can give them which in this case is allot.

Johan45 said:

Looking good bassnut. I think you're right about the blocks. You'd have to use generic ones. I don't think I ever found one for the Sabertooth. Not that I was going to get one I just like to look at shiny computer stuff now and then.

Click to expand...

Check this out.
http://www.overclock.net/t/1468642/how-to-watercool-the-asus-990fx-sabertooth-rev2-0-vrms-and-nb

Thanks Doc will check that out

Dr. McCoy said:

Check this out.
http://www.overclock.net/t/1468642/how-to-watercool-the-asus-990fx-sabertooth-rev2-0-vrms-and-nb

Click to expand...

That's pretty much what I came up with when I looked at it. Just looks like too much work an too many tubes for the benefit it gives. Thanks for the link though doc.

I've been doing dome experiments today with bios revisions and 2400 ram. We've had some posters lately with the high speed ram it's it's been nothing but a headache so that is what got me on to this. We've talked about this on and off for a few months now about how the newer releases have been messy especially with some brands of ram . All I had was G.Skill to test but I still found dome things that may shed light on this phenomenon.
I crashed my bios last night so I figured what the hey as good a time as any.
So the first candidate was the latest and greatest which I think was 2002. First off I noticed that the SPD settings that are available in the tolls section were all messed up. This is on 2400 and 2133 ram.

Here's some shots
2400 2002 bios


2133 2002 bios




As you can see XMP#1 is just wrong on both these shots.
One thing I noticed though was that the busses seemed to be whacky as well
Here you'll see what I mean This is default setting in bios and all I set was the ram speed and timings



Now doesn't that just seem backwards 2200 HT and 2400 NB that's default. I found a correlation for this and will explain in next post.

Here's the same with bios 508 beta but the SPDs look like they should I still got the strange bus speeds though and I did have some trouble with the 2400 ram on this version it would drop down to 800 to start at default. Set speed and timings It would boot up to windows but when I restarted it wouldn't boot or it would it was just not right. The ram was at stock so it was a bit baffling TBH Oh ya and I noticed the 1.4+ NB voltage on this bios and the last. With the same unexpected bus speeds

2400



2133



Next I tried 1403 which was the first "official" bios for the 9370 When I started out I had the 2133 sniper in the board the SPD looked good busses were at the expected speeds of 2600HT and 2200NB and NB voltage wa at 1.1ish as usual. Then I put the 2400 in the board it booted right up without issue at 1333 then I set the speed and timings. That's when I noticed the bus speeds changed. When the 2400 ram is in the board it sets 2400 as the minimum NB speed. When I went to change it anything below that was unsupported. Still not sure why it changes the HT as well but it does again to 2200 as well as the CPU_NB volts back up to 1.4+

So in conclusion I did learn a couple of thing the new bios does get timings a bit messed up and 2400 ram is going to be a pain no matter what IMO since it requires a higher NB speed.
Now I see why 2 different sets of Kingston 2400 2x8 kits are causing some problems with some of our members

I knew there was a good reason not to buy that DDR3-2400 ram for AMD since it never did much good performance wise, but this with the bios is even more of a pain. I hope the DDR3-2400 stuff can be tightened up and run at a more reasonable DDR3-2000 to DDR3-2200-ish where most of the power is anyway.

Thanks for sharing this "johan". I know it takes time to sort out the odd stuff that we seem to know in our gut, but most ask why.
RGone...

i THINK THE HEAT KILLER MOTHERBOARD BLOCK GOES ON ONE OF THE SABERTOOTH BOARDS BUT THERE IS A TRICK WITH A PIECE OF TAPE TO MAKE IT WORK.

This might make it official, I have truly lost my marbles. In the quest to keep winter in my house so I can cold bench. This is my new quest, filling a bucket with DICE and Acetone and running the big loop through copper coils, which are submerged in the DICE/Acetone mixture. So far the experiment hasn't gone well due to my theory of more is always better. My first attempt was putting roughly 10lbs of DICE in the bucket and getting the mixture to -60c +, my thermometer only goes as low as -60. I figured I had enough antifreeze in the loop and enough flow from the 700gph pump to keep it flowing, not!!!! It froze in 2-3 minutes. The second attempt went a bit better because I added the DICE slowly but the temps in the mixture were only down to -1c and the water/antifreeze +1-4c. I wanted it colder so again going back to the more is better theory I dumped 2+ pounds of dice in it to get it colder and froze the loop again. Not giving up yet, so I'm going to add more antifreeze to the loop so I have a 70% mixture and then add the DICE slower and see what happens. Oh and that's the Phenom 9650 C/D and RGone passed on to me.

View attachment 142112
View attachment 142113
View attachment 142114

Well I said a little back that opening the can of worms is often not good unless you want to use all the worms because it is hard to get the top back on the container.

I don't think in your current mental state that you have let the worms out...think it is more like snakes or maybe dragons. U have flipped for sure, but that is what happens to a lot of Rodeo Clown members.

One thing is for sure though...the copper is transferring lost heat to the water in your loop. Hehehe.

Luck man.
RBobertGone...skisterstignatorzzzzzzzzzzzz

Mandrake4565 said:

This might make it official, I have truly lost my marbles. In the quest to keep winter in my house so I can cold bench. This is my new quest, filling a bucket with DICE and Acetone and running the big loop through copper coils, which are submerged in the DICE/Acetone mixture. So far the experiment hasn't gone well due to my theory of more is always better. My first attempt was putting roughly 10lbs of DICE in the bucket and getting the mixture to -60c +, my thermometer only goes as low as -60. I figured I had enough antifreeze in the loop and enough flow from the 700gph pump to keep it flowing, not!!!! It froze in 2-3 minutes. The second attempt went a bit better because I added the DICE slowly but the temps in the mixture were only down to -1c and the water/antifreeze +1-4c. I wanted it colder so again going back to the more is better theory I dumped 2+ pounds of dice in it to get it colder and froze the loop again. Not giving up yet, so I'm going to add more antifreeze to the loop so I have a 70% mixture and then add the DICE slower and see what happens. Oh and that's the Phenom 9650 C/D and RGone passed on to me.

Click to expand...

So you wasted DICE in a bucket?

Run acetone through the water loop....

Or get a dice pot? I got a couple. Are you in the US? perhaps I could send you one?

Would rather seen Dice run properly......

LOL shrimp just experimenting.

Way to go MrClean, I was wondering if you were gonna share. I can't help it everytime I think about it I start laughing harder than the last time. I can just picture the point where the big grin on your face changes to that ruh roh puzzeled look replaced quickly by the OHHH NOOOOOOO! #*&&*#%#^&#*#*#

Johan45 said:

Way to go MrClean, I was wondering if you were gonna share. I can't help it everytime I think about it I start laughing harder than the last time. I can just picture the point where the big grin on your face changes to that ruh roh puzzeled look replaced quickly by the OHHH NOOOOOOO! #*&&*#%#^&#*#*#

Click to expand...

Even if nothing comes out of this little experiment at least we got a good laugh out of it. At least I know the sound the pumps makes when the loop freezes so I am smart enough to shut the Pc down before it overheats.

Johan45 said:

your face changes to that ruh roh puzzeled look replaced quickly by the OHHH NOOOOOOO! #*&&*#%#^&#*#*#

Click to expand...

Now that is what I really wanted a picture of. Hehehe. I can imagine it to be quite similar to the one I had when I forgot to reconnect one of my hoses on water block with a tightening of my clamp on the hose. Imagine the look on my face when a half inch stream of water started bouncing off the celing. Ruh Roh was real. Hehehe.
RBobertGone...skisterzzzzzzzzzzzz <<

Johan45 said:

Way to go MrClean, I was wondering if you were gonna share. I can't help it everytime I think about it I start laughing harder than the last time. I can just picture the point where the big grin on your face changes to that ruh roh puzzeled look replaced quickly by the OHHH NOOOOOOO! #*&&*#%#^&#*#*#

Click to expand...

LMAO

Mandrake4565 said:

Even if nothing comes out of this little experiment at least we got a good laugh out of it. At least I know the sound the pumps makes when the loop freezes so I am smart enough to shut the Pc down before it overheats.

Click to expand...

Ya I know that rattly sound quite well Manny! I've settled down a bit now, thanks for that bud!

RGone said:

Now that is what I really wanted a picture of. Hehehe. I can imagine it to be quite similar to the one I had when I forgot to reconnect one of my hoses on water block with a tightening of my clamp on the hose. Imagine the look on my face when a half inch stream of water started bouncing off the celing. Ruh Roh was real. Hehehe.
RBobertGone...skisterzzzzzzzzzzzz <<

Click to expand...

I'm sure we've all had those moments I know I did when i thought running that big azz pump was a good idea to flush the line. It works way too goooder.