View Single Post
Old 10-05-2007, 11:19 AM   #22
tw
Read? I only know how to write.
 
Join Date: Jan 2001
Posts: 11,933
Quote:
Originally Posted by PointsOfLight View Post
I had a question for tw though on fans...are you saying that all I need to cool my case is the power supply fan and that's it? Cause with a overclocked Core Duo and a pretty big graphics card it just doesn't seem like enough.
How many watts are you dissipating inside that case? That number and 35 CFM for a standard 80 mm fan is sufficient to calculate reality.

Appreciate how widespread are computer myths. Most who assemble (not build) a computer and consider themselves expert don't even know how electricity works. They cannot even perform that heat calculation and would get angry if pushed to do so.

One fan for the chassis - not to be confused with another fan on the heatsink that has a completely different task.

Use lessons from junior high science. First form a hypothesis. That means doing the theoretical calculations given watts and fan CFM. Then do experiments. Build the system and measure those results. Without both, then you would know nothing except speculation. These concepts from science get lost on many computer assemblers who just know only because they assembled (not built) a system. Myths will be widespread because observation alone somehow becomes fact. This ignorance is why Asian exporters make massive profits selling inferior power supplies to clone computer assemblers. Sell a supply that is missing parts for less - and have a higher profit margin? Yes because so many computer assemblers recommend power supplies only on dollars and watts - and don't even know what the watts are really measuring.

You don't need all those chassis fans. When done, put your computer in a 100 degree room and operate it all day. Yes, all computers must work just fine at temperatures above 100 degrees - most all with only one 80mm chassis fan.

Many hype more cooling for reliability because they have a trend but don't have a grasp of the numbers. Same applies to another question about disk drive life expectancy that no one answered with numbers. The naive will hype 'leave it on constantly' to extend hardware life. They have not a clue - only speculations justified by observation. Yes, higher temperatures will age a component faster. And then we put numbers to that CPU. We have lowered its life expectancy from 200 years to 100 years. So what? Notice how trends hyped by the naive become irrelevant once we apply what they avoid - the numbers.

You will hear hype about Arctic Silver. Myths to have you spend more. Most heatsink compounds selling for far less money are same as Arctic Silver. But ask one who recommends Arctic Silver to provide its thermal resistance numbers. He cannot. He just 'feels' Arctic Silver is better because it is hyped like Listerine.

When assembling a computer, also challenge claims by asking for numbers. Another number for that CPU heatsink is 'degrees C per watt'. They cannot provide that spec in writing? Then suspect a product promoted only on hype.

Again, it is the 'Tool Man Taylor' joke. The naive just know something is better ... "More Power".

Last edited by tw; 10-05-2007 at 11:26 AM.
tw is offline   Reply With Quote