• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Need help decifering data graphs

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

my_republic

New Member
Joined
Jul 1, 2004
Sorry if this is a repeat question, but here it goes. Last summer i was tempted to build a water-cooled PC. I ended up not doing anything. Now i have upgraded to a Pentium D and want wc because it is so hot here in Denver. I have a dillema...

I want to purchase the switech mcw6002, the new black and chrome :p

I just finished talking myself out of the Swifty storm block due to my small budget. Then i noticed the performance graphs...

It seems that the data published for the MCW6000 has C/W at best at about 0.19... not bad...

Then i noticed that on the Storm data graphs, the Mcw 6002 has a performance (at best) of 0.05 C/W just 0.01 C/W worse than the storm?

This is why i am going with the economy block, but is this true? Can there be a 0.15 C/w difference between the MCW 6000 and Mcw 6002. Are there any differences between them other than barb size?

Please help. I want a 3/8 id system, but if the data is true, for the money, i would be a fool to go with the 6000 if there is a real difference of 0.15C/W.

The links are here:

http://www.swiftnets.com/products/Storm.asp#
http://www.swiftnets.com/products/mcw6000.asp#

???????? what to do?!?!
 
http://www.procooling.com/html/pro_testing.php <- the MCW6002 is just slightly better than the MCW6000 and the Swiftech Storm is the same as the Storm G4. As you can see at higher flow rates the Storm does clearly win out against the MCW6002. However you need a stronger pump to keep up with the Storm because it has 2x the pressure drop as a MCW6002. At flowrates past .5gpm the Storm comes out the winner so any pump stronger than the eheim 1048 will win on the Storm. Also 1-2C is what you are looking at here so its a close race (That 1-2C happens at a 71W heat load according to the sim used at procooling).

Edited because I forgot the pressure drop, thanks Voodoo_Rufus :)
 
Last edited:
Ok so that graph is a little more what expected, but it does not tell about the heat input so as to determine the c/w. If the heat input in that graph is less than 100 (nice round number), then does that mean that the actual c/w is higher than what is published over at swift? This is where i get a little confused. What swift has published and what procooling has published seems drastically different. Tell me if ia m wrong with my assumption below:

6002 according to swift: .05c/w (or 5 deg C over ambient at 100watts???)

6002 according to procool: .09c/w (or 9 deg C over ambient if tested at exactly 100watts, worse if tested at less, right??)

Thanks again...
 
The testing methods may be different for each site.
http://www.swiftnets.com/products/mcw6000.asp# - Swiftech references ProC for comparison against other companies' blocks, and only use their test bench for comparison against their own blocks.

The TIM interface and mounting pressure can have an impact on results, and Swiftech does not state their methods. Without knowing that, it's impossible to compare results on different test benches.
 
Ah... so ProC is like the real world test and the in-house test at swift is a simulator. That helps a great deal to understanding the discrepancy... One last question: why then use a simulator if proc's test setup is closer to real life results? Why publish what seems to good to be true (on swift's site)?

That bothers me a little, but not enough not to buy one...

Thanks again for a quik reply.
 
Voodoo Rufus said:
The TIM interface and mounting pressure can have an impact on results, and Swiftech does not state their methods. Without knowing that, it's impossible to compare results on different test benches.

there is a link to the test method (for the storm) on the page...
testconditions.gif


not sure why the C/Ws are so much lower ..... possibly they have stripped out the c/w of the TIM?

Bill?
 
Back