right now I have a DD m6 in a cpu only loop with a rx360, rx120 and a laing dd5/ ek bay rez. There is no air in the system, the processor is at 4.6 @ 1.38 while under load of blend tests in prime 95 (I run prime95 for 24 hours as my end all stability test). The temps on the hottest core hit 75 briefly during one test. the ambient temp is around 20c to 22c in the house. I know what your first question is " why is your voltage so high for that clock", I usually set the voltage as high as im comfortable with then I see what the maximum clock speed I can get stable for that voltage. Personally I think that im running a little hot and I've read that the DD m6 is not as good a performer as I thought it was and I'm planning on replacing it with either a koolance cpu-380I or a ek supremacy. I'm wondering if replacing the block would result in lower overall temps and if so what do you guys think the temp difference would be?
-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
SOLVED danger den m6 to koolance cpu-380I
- Thread starter ehhwhat
- Start date
- Joined
- Dec 13, 2002
- Location
- Southern California
Martinsliquidlab has the m6 about tied with a raystorm or koolance 370, a roundup on xtremesystems has the m6 2 degrees behind the raystorm/370 and 5 behind the 380. So you are looking at a best case drop of 5 degrees, according to those reviews.
- Thread Starter
- #3
Thanks for the info, I was hoping it would me more but I don't think that a 5c drop in temp is worth another 80usd. I'll have to start looking at those websites see what they think about the rx series radiators.
- Joined
- Dec 13, 2002
- Location
- Southern California
RX rads are great, and you have more than enough for just a sandy cpu, unless you run your fans very slowly.
- Thread Starter
- #5
That's what I figured I just thought that I should be able to hit 5ghz on this processor @ 1.4-1.42 but it just seems to run hot. I got some new fans coming though and I'm gonna drain the system and refill with new coolant, so I'll see how that goes. Thanks again
- Joined
- Dec 13, 2002
- Location
- Southern California
Mine will do 5ghz with hyperthreading on at around 1.44 volts, but it will hit 90+ degrees in prime, on custom water.
- Joined
- Jan 12, 2012
I agree with Supertrucker, I don't think you'll see substantial drops in temps going with a new Cpu block. You have plenty of raddage for just the cpu. When was the last time the loop was pulled apart and thoroughly cleaned, including pulling apart the Cpu block and cleaning that? Sometimes debris can get caught in the nooks and crannies and cause the temps to rise.
The other thing I noticed is you set X Cpu V and then try to find the highest stable OC for that voltage, or did I read that wrong? If so have you tried lowering the voltage and testing for stability? Reason being, if I'm correct with how you do your overclocking. Just because you couldn't do 4.7 at 1.38 Cpu V doesn't mean you can't do 4.6 at a lower Cpu V 1.38, am I making sense?
The other thing I noticed is you set X Cpu V and then try to find the highest stable OC for that voltage, or did I read that wrong? If so have you tried lowering the voltage and testing for stability? Reason being, if I'm correct with how you do your overclocking. Just because you couldn't do 4.7 at 1.38 Cpu V doesn't mean you can't do 4.6 at a lower Cpu V 1.38, am I making sense?
- Thread Starter
- #8
you read that right, I set my voltage at what I think is comfortable and raise the clock till the voltage won't support the overclock. I'm actually in the middle of testing a 4.8 clock im just waiting on some fans and coolant to come. I plan on tearing the loop apart and cleaning the whole thing then shooting for that 4.8 again. For now I'm just leaving everything be, because I know that this voltage/clock speed is 24 hours prime95 stable. Currently the temps are not Ideal and I know I could probably hit 4.6 @ 1.36, I just think theres some more headroom in this chip
- Joined
- Jan 12, 2012
Understood, only reason I said this
is because you mentioned getting lower temps.
On the shooting for 4.8 OC part, have you found a stable 4.7 OC. I usually find when, really starting to push the clocks making smaller jumps is less frustrating. It is easier to lose our way sometimes when really trying to push and making large jumps in mhz. I know on my 2500k I was able to get it stable @ 4.7 with 1.35 Cpu V but 4.8 took a whole heap of volts to get to.
- Thread Starter
- #10
I see what your saying, for some reason I just decided to test 4.8 first and it was 12 hours prime 95 stable but it was just getting really hot. So I'm just gonna chill until the fans and coolant and other stuff gets here, then I'll post again. To be honest I'm hoping that the temp flux is due in part to build up on the block, I think im going to lap the processor as well. Thanks for all the advice guys I really appreciate all the help.
- Joined
- Jan 12, 2012
I know everyone has their own definition of stability, but do you really need 12 hrs to determine it? I know that if I can pass 2 hours Prime Blend that my rig does everything I need it to without a hitch.
- Joined
- May 22, 2011
- Location
- USA
What coolant are you purchasing? Color? You're not using distilled water + kill coil? Depending which coolant you have, I wouldn't be surprised if there's build up in that block.
How long have you had the system and whens the last time you took it apart for maintenance?
If you find anything, pic pls.
How long have you had the system and whens the last time you took it apart for maintenance?
If you find anything, pic pls.
- Thread Starter
- #13
In my testing I've had overclocks fail at 18 hours 20 hours, I think I had one fail at 22 hours but 24 and after the computer seems to run rock solid forever. I work a lot with UDK and maya/zbrush so I can't afford to have a crash while working or baking out lights etc... in my opinion there really is no varying degree of stable. It is either stable or some degree of unstable and testing for 24 hours just assures me that the overclock is 100%.
The coolant is in my sig and I ran a little of an experiment this time around with EK Blood Red. Before the torches and pitchforks come out let me explain myself. This coolant has been changed from its original formula and it was the original formula that people had all the horror stories from. The second gen of the formula was not cloudy like the first and was also supposed to be fixed according to EK. I talked to a local pc builder Geekbox and they said they had been using the new red for some time. Geekbox told me the last system they tore apart with the formula in it was fine except for a little cloudiness in the tubing. So I took a chance with this new formula of EK Red, when I tear apart the system I plan on making a video. Hopefully the formula didn't do any damage, but if it did I'll document it and it will serve as a warning to others the might want to experiment with it. I am switching coolants this time regardless to Fesser 1. the system has been not been cleaned in like 6-8 months I can't really remember the exact date. The new fans and coolant should get here on Tues so I'll have a video up around thurs or friday.
- Joined
- Jan 12, 2012
Understood, about not being able to afford a crash. Nothing wrong with it and if it works for you, then thumbs up to that. I also do work on my Rig's, though non of it requires the Cpu to be maxed out at 100% for hours upon end. Though I have been able to run Folding at Home for the Chimp Challenge,which is 10 days running 24/7 @ 100% if you're familiar with that program. Never had a crash only using 2 hours but again if it works for you then that is all that matters.
- Joined
- Jan 14, 2011
There are so many variables in wc the only way to find out is to do research and try it out. My 2500k @ 4.9 folding 24/7 for a month topped out at about 60c granted I had an ek supreme hf and a 480 rad. On a 240 rad and 4ghz im hitting about 55 depending on ambient.
- Thread Starter
- #16
24 hours, I've never had a problem. I have experienced problems with 18 hour failures and I think that I had one fail at 21 hours. These were older processors from the athlon 64 days. I don't know if these newer processors are just less prone to bsod or corrupt a boot.ini file, but I have not had a problem. It's just an old habit that has never failed me, the 24 hour prime95 testing. It really is not how strenuous the work is on the cpu and gpu is, its the loss of time. I had a map in UDK build lighting for 36 hours straight and had the build crash at 98%, I almost failed the class because of it. This all happened on a school workstation. I told myself that my home computer would never let me down like that, but I still wanted to overclock so I choose to test for 24 hours. By the way the school workstations were in horrible disrepair, unknown to me at the time, but the school tech support was as awful as could be. Nevertheless good habits die hard or don't die at all.
- Thread Starter
- #17
EK blood red coolant video
As I promised a quick video of the after effects of 6-7 months of Ek blood red coolant (formula 2). Now before everyone starts saying " I told you so", I knew there was a possibility of buildup. To be honest I thought that it would be much worse. I knew it was going to be experimental, and I thought that the aesthetic benefits would out weigh the possible performance negatives. That being said, I was only able to find 2 downsides to the EK coolant. The first downside being, it stained the tubes green. I know that all dyes are going to stain so this wasn't a huge surprise, although I have never seen fesser stain like this. The staining was pretty extensive and like I said, was a dark/forest green. The second downside is the small deposits of dye on cpu block fin array, to be fair the dye did not stick and simply washed off with some luke warm distilled water. This leads me to believe that it might not have had a impact on the temps being it was loose and free to flow in the loop. I also cleaned out the rads and gpu block and everything came clean really fast. Now the dye might be a little more stable then the first formula, but it is still not stable. Note that the color did stick to its original hue and tone for quite some time, although not UV reactive, it did eventually fade. So the question now is " would I ever recommend this stuff?" and the answer is no, not even if the person just simply didn't care about buildup or color loss. There is too many other coolants on the market that leave a minimal footprint in favor of aesthetics, that cost the same and leave very little behind.
As I promised a quick video of the after effects of 6-7 months of Ek blood red coolant (formula 2). Now before everyone starts saying " I told you so", I knew there was a possibility of buildup. To be honest I thought that it would be much worse. I knew it was going to be experimental, and I thought that the aesthetic benefits would out weigh the possible performance negatives. That being said, I was only able to find 2 downsides to the EK coolant. The first downside being, it stained the tubes green. I know that all dyes are going to stain so this wasn't a huge surprise, although I have never seen fesser stain like this. The staining was pretty extensive and like I said, was a dark/forest green. The second downside is the small deposits of dye on cpu block fin array, to be fair the dye did not stick and simply washed off with some luke warm distilled water. This leads me to believe that it might not have had a impact on the temps being it was loose and free to flow in the loop. I also cleaned out the rads and gpu block and everything came clean really fast. Now the dye might be a little more stable then the first formula, but it is still not stable. Note that the color did stick to its original hue and tone for quite some time, although not UV reactive, it did eventually fade. So the question now is " would I ever recommend this stuff?" and the answer is no, not even if the person just simply didn't care about buildup or color loss. There is too many other coolants on the market that leave a minimal footprint in favor of aesthetics, that cost the same and leave very little behind.
- Joined
- May 22, 2011
- Location
- USA
Don't take it personal and thank you for the video but the "I told you so's" have been around and so have the experiments and tests proven already for years now.
Hard to reach places like the radiator can still have stains and gunk in it. You said you had loose debris in the loop. Check and make sure the pumps ok. Those can damage the pump.
We have threads with stories and pictures of less, same and way worse than what you have there. Distilled water carries a better heat transfer characteristic than having stuff in the liquid like colored fluid. It will be a negative impact on your temps and only get worse over time. With colored you need to do tear downs every 2-3 months if you want to almost be stained and gunk free.
As for the fesser tubing, its no good. I used it in the past and clouded with the use of distilled within 2 months. Get plasticizer free tubing like Primochill Advanced LRT. You want color? Primochill PrimoFlex Advanced Bloodshed Red Tubing.
We can keep telling you but hey its your stuff and money. Its all up to you if you want to take our advise.
Hard to reach places like the radiator can still have stains and gunk in it. You said you had loose debris in the loop. Check and make sure the pumps ok. Those can damage the pump.
We have threads with stories and pictures of less, same and way worse than what you have there. Distilled water carries a better heat transfer characteristic than having stuff in the liquid like colored fluid. It will be a negative impact on your temps and only get worse over time. With colored you need to do tear downs every 2-3 months if you want to almost be stained and gunk free.
As for the fesser tubing, its no good. I used it in the past and clouded with the use of distilled within 2 months. Get plasticizer free tubing like Primochill Advanced LRT. You want color? Primochill PrimoFlex Advanced Bloodshed Red Tubing.
We can keep telling you but hey its your stuff and money. Its all up to you if you want to take our advise.
Last edited:
- Thread Starter
- #19
I don't use fesser tubing, I have used fesser coolant and after a year of use of the orange uv coolant I had very little staining. I use tygon e-3606 and it seems to work fine, I haven't found any plasticizing from previous builds. I do believe that some dyes are better then others but you do have to tear down the build once a year even with a good dye. I have used distilled water in the past and maybe some day I might go back to it. For now though fesser seems to work fine and I don't mind the additional maintenance for the aesthetics. I'll probably never use another EK coolant though. When its all said and done it really isn't that much work or money to test coolants, sure would have been nice though to have that red coolant work out. It really is good looking stuff, it literally looked like blood running through the system.
