Aerocool Deep Impact
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
-
- Posts: 135
- Joined: Tue Jun 01, 2004 10:09 am
- Location: Fort Worth, Texas
Aerocool Deep Impact
I am currently running a Aerocool Deep Impact Limited Edition. It runs at
around 22db, but my load cpu temp is around 64 celcius. I have a Pentium 4
3.06 with HT turned on. i wanted to replace the two stock Deep Impact
fans with the Silenx 14db fans. Is this ok. i think my temps are a little off
because i am using the one built onto the motherboard and i know how they
can be inacurate. Whats your advice and how can i get my temps down.
around 22db, but my load cpu temp is around 64 celcius. I have a Pentium 4
3.06 with HT turned on. i wanted to replace the two stock Deep Impact
fans with the Silenx 14db fans. Is this ok. i think my temps are a little off
because i am using the one built onto the motherboard and i know how they
can be inacurate. Whats your advice and how can i get my temps down.
How do you like the Deep Impact?
What are your case temps? Does the case have a free air path from in to out?
What is the room temp?
Sometimes software messes up the temp readings, sometimes the temp reading hardware/BIOS settings are badly inaccurate. Not much can be done about the latter. For the former, many recommend Motherboard Monitor 5 (MBM 5).
If the CPU temp read is the "on-die" temp, then 64C is not too alarming tho obviously not too great either.
Anyhow if your case temps are well above room temp (> 10 or 15C diff), and your CPU temp is not much above case temp, then you'd want to tackle case temps. Contrariwise, if case temps are near room temp but your CPU temp is way above case temp (>25 or 30C diff), then you'd want to tackle CPU HSF efficiency.
The Deep Impact should be good for ducting to the exhaust and ducting from an intake, since the fans are standing up and can face the case front/back. That would be a fairly easy way to get temps down. There is no "reversal penalty" for ducting to exhaust with the DP.
Guesstimate about your proposed fan replacement:
A 14 CFM fan replacing a 25 CFM fan: 0.12 - 0.067 = 0.053 C/W difference ... times 80W ... = 4C rise at CPU.
If I were you, I'd want to get temps down in some way before swapping CPU HS fans for anything substantially less powerful.
My guess is that like most cases, your case is not designed for good ventilation ...
the wesson
What are your case temps? Does the case have a free air path from in to out?
What is the room temp?
Sometimes software messes up the temp readings, sometimes the temp reading hardware/BIOS settings are badly inaccurate. Not much can be done about the latter. For the former, many recommend Motherboard Monitor 5 (MBM 5).
If the CPU temp read is the "on-die" temp, then 64C is not too alarming tho obviously not too great either.
Anyhow if your case temps are well above room temp (> 10 or 15C diff), and your CPU temp is not much above case temp, then you'd want to tackle case temps. Contrariwise, if case temps are near room temp but your CPU temp is way above case temp (>25 or 30C diff), then you'd want to tackle CPU HSF efficiency.
The Deep Impact should be good for ducting to the exhaust and ducting from an intake, since the fans are standing up and can face the case front/back. That would be a fairly easy way to get temps down. There is no "reversal penalty" for ducting to exhaust with the DP.
Guesstimate about your proposed fan replacement:
A 14 CFM fan replacing a 25 CFM fan: 0.12 - 0.067 = 0.053 C/W difference ... times 80W ... = 4C rise at CPU.
If I were you, I'd want to get temps down in some way before swapping CPU HS fans for anything substantially less powerful.
My guess is that like most cases, your case is not designed for good ventilation ...
the wesson
-
- Posts: 135
- Joined: Tue Jun 01, 2004 10:09 am
- Location: Fort Worth, Texas
Deep Impact
i really like the deep impact. my case temps range from 30 celcius to 32 celcius. The room temp usually stays around 74 farenhiet but in texas it can become a little hotter during the day. I am currently using MBM5 and the temp reading ias on die. I was also was wondering what i could use to dust the deep impact fans. i currently have a Thermaltake Xaser III which has ok airflow. I am hoping to get a new silverstone sst-tj03 which has 120mm fans. I just got the new deep impact after replacing the coolermaster jet 4 which is about the most annoying thing ever. i am open to any other recommendations for how to cool this thing down.
-
- Posts: 135
- Joined: Tue Jun 01, 2004 10:09 am
- Location: Fort Worth, Texas
undervolting
How much of a temperature change would it make to undervolt my Pentium 4 3.06 GHZ 533FSB. anyone have recommendations of how much i should undervolt this thing.
-
- Posts: 135
- Joined: Tue Jun 01, 2004 10:09 am
- Location: Fort Worth, Texas
Fans
The silenx fans i want are 11db with 21cfm and the stock deep impact fans are 24.6 dBA with 31.5 cfm.
Re: Fans
Yep, and Christian Andersen was another one telling fairy tales.TheDarkHacker wrote:The silenx fans i want are 11db with 21cfm and the stock deep impact fans are 24.6 dBA with 31.5 cfm.
Sorry man, I have nothing against SilenX-fans, but 21 CFM @ 11 dBA is highly unlikely. Belongs in the fairy-tale department.
Re: Fans
That's not true. It's entirely possible. You just need to take your measurements from the other room down the hall.SpyderCat wrote:Sorry man, I have nothing against SilenX-fans, but 21 CFM @ 11 dBA is highly unlikely. Belongs in the fairy-tale department.
Re: Fans
Ok, you got me therelenny wrote:That's not true. It's entirely possible. You just need to take your measurements from the other room down the hall.SpyderCat wrote:Sorry man, I have nothing against SilenX-fans, but 21 CFM @ 11 dBA is highly unlikely. Belongs in the fairy-tale department.
-
- Posts: 135
- Joined: Tue Jun 01, 2004 10:09 am
- Location: Fort Worth, Texas
New Temps
My new temps are 35 celcius for the processor when idle after underclocking it from 3.06GHz to 2.53Ghz. My case temps are the same and i will get back to you about my load temps. anyone have any recommendations about software that can max out your processor.
-
- Posts: 1283
- Joined: Wed Sep 03, 2003 1:35 am
- Location: Sweden, Linkoping
Prime95 is the most common choise. It produces slightly less heat than cpuburn, but on the other hand it will notice if your CPU gets the wrong result from calculations etc. This means that Prime95 finds problems in setups that CPUburn does not, thus making prime95 a better program for testing a setup.anyone have any recommendations about software that can max out your processor.
If you have a CPU with hyperthreading you should run 2 instances of whatever test program you choose to run.
-
- Posts: 135
- Joined: Tue Jun 01, 2004 10:09 am
- Location: Fort Worth, Texas
load temps
After using every porgram to load my cpu my load cpu temp is is now 48 celius. i have also found that the futurmark pcmark is the best program to get an processor at 100% even if it is hyperthreaded.
-
- Posts: 1283
- Joined: Wed Sep 03, 2003 1:35 am
- Location: Sweden, Linkoping
Now I get a bit uncertain of what you actually want to achieve with the tests.i have also found that the futurmark pcmark is the best program to get an processor at 100% even if it is hyperthreaded.
Are you talking about getting the OS to report 100% CPU activity?
Do you want software to make the CPU produce most heat?
Do you want software to make the whole machine produce most heat?
Do you want software to ensure your CPU will run stable regardless of what software you run?
Re: load temps
Incorrect, sad to say - least, by my experience.TheDarkHacker wrote:After using every porgram to load my cpu my load cpu temp is is now 48 celius. i have also found that the futurmark pcmark is the best program to get an processor at 100% even if it is hyperthreaded.
The best way to do it (having done it a few hours ago meself) is to run 2 instances of CPU Burn - that pumped my temps a lot more than any other benchmark I could find.
CPU-burn is a nifty way to find what the lowest "safe" voltage is for your fans, as no real-world app will stress the CPU as much as 2x CPU-burn (or 1x, for non-HT CPU's).
Hope this helps .