View Single Post
Unread 02-24-2006, 12:18 PM   #37
jaydee
Put up or Shut Up
 
jaydee's Avatar
 
Join Date: Dec 2001
Location: Spokane WA
Posts: 6,506
Default Re: How would you re-structure ProCooling?

Quote:
Originally Posted by gmat
The problem is about the same for car performance parts. Let's say i want a header for my engine. There are reviews, dynos and so on. But this header will give huge gains with high lift cams, and almost no gains (or losses) with stock cams.
It all comes back to what we've always said: think about the whole system, not individual parts. Still, testing each individual part to see how it performs under controlled circumstances, is very important, to engineer the whole system in a coherent manner. Back to my header. If i take it, it means i'm upgrading the cams, putting a decent intake, and sizing the exhaust properly. I wont focus on the individual gains of the header, since i know that will ALL parts sized properly the whole setup will do what i want (like, +25 whp for example in my case).
The same goes for computer "aftermarket" parts: each must be tested in all circumstances to see how they fit into a whole system. And which complete system produces "good" results. By "good" in our case it will mean higher OC, and less noise (maybe define a noise vs OC ratio ? in dBA per % of OC, no real unit i know ?)
I am not seeing how this is any different then what I am suggesting. Only difference is I am taking the main focus off temps and placing it on performance.

There are only 3 main parts to a water cooling system for a PC. Pump, rad and water block (or multiples of each). It has been pretty well proven replacing any one of those parts and swapping in another is not going to drastically change things and if it does my tests will prove it anyway.

Your head example doesn't fit the bill in this day and age. You can gain huge amounts of performance from changing parts on a head. Changing from a Apogee to Strom Rev. 2 will get you hardly anything. Changing from a BI to a BIX will hardly gain you anything. Changing from a Ehime 1260 to a MD30 with hardly gain you anything..... Changing all 3 will hardly gain you anything. Temp wise that is. So why focus on temps instead of over all performance, noise, looks and over all ease of use?

You can interchange all the modern water cooling parts you want in any order you want and you will be hard pressed to notice any significant changes in computer performance.

Also how is testing each part on a test bench not designed for what the part was made for telling us how that part will work on the thing it was made for? Also how is testing these parts on a Intel 775 CPU that no one uses going to tell us any different?

We can test how ever we want but the end results are not going to be consistent for every type of comp out there. The temp difference from one cooling system to another is hardly relevant over noise and ease of use these days.

From en engineering stand point I can certainly understand the higher level of testing and that is the standpoint I have had the last few years. Now I don't really care about that and looking at it from and end users perspective. I don't really care about all the tech stuff, just want to know if the cooling system is reliable, quiet and cost effective. Not concerned about 1-6C temp difference. Just doesn't matter in reality. Games still play the same, internet is just as fast and my OC will hardly matter with that small temp swing.

Back in the day it was a lot more important. When you could take a Duron 600 and OC it to 1100mhz. That was a useful overclock! Today you can add 1000mhz to a 2000mhz CPU to get 3000mhz and it is hardly noticeable. And you can do it with air cooling. Water cooling might add a 100mhz or so but what does it matter? Only to the uber geeks that have nothing better to have a dick size contest about.


Just seems some of us take testing way to seriously. Certainly has not done anyone any good here lately.
jaydee is offline   Reply With Quote