Which is the best testing method?

bigben2k · 02-22-2006, 02:56 PM

I'm opening up the question to get as many people's perspectives:

Which testing method do you believe is best:

1- testing with a simulator (heat die)
or
2- testing using a real CPU

billbartuska · 02-22-2006, 05:14 PM

Quote:

Originally Posted by bigben2k

1- testing with a simulator (heat die)
or
2- testing using a real CPU

Nice idea, but what sinulator and which CPU?

The way you ask it's kinda like "Which is bigger, red or blue?" (I love that question!)

Hasn't this all been hashed out ad nausium here?

I'm waiting to see what MaxxxRacer comes up with.

bigben2k · 02-22-2006, 05:41 PM

I was going to add more options as they came up, but for the sake of arguments, let's say that:
1) the simulator is home made (it's not the Intel TTV).
and
2) the CPU is an Intel with the groove machined into the IHS, for fitting a thermocouple.

snowwie · 02-22-2006, 05:56 PM

I think tests are far more dependent on the tester. a good tester will determine their methodology depending on a variety of things: their budget, their audience/purpose, what exactly they're testing (blocks or systems), etc.

jaydee · 02-22-2006, 08:43 PM

Either way your results will be spacific to that test bench temp wise. I am in favor of the CPU method simply to prove the cooler functions properly on a actual computer being we don't use die sims to surf the net, play games and what not. Your temp results will be pretty much useless either way to end users.

snowwie is pretty much correct. What is the goal? Product development, finding the product thermal performance on that test bench, over all computer performance or?....

9mmCensor · 02-23-2006, 12:38 AM

Basically make basic grasp is that...

If you do it one way, its not real world, and if you do it the other way, you can get consistant results. So basically both aren't perfect.

If one is more right that the other, it is so marginally.

But if the writer does a good job on either, and states how it was done, it should be good enough for a consumer to decern what is roughly accurate.

View Poll Results: Which testing method is best?
Use a simulator (heat die)	2	22.22%
Use a real CPU	7	77.78%
Voters: 9. You may not vote on this poll

02-22-2006, 02:56 PM	#1
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	Which is the best testing method? I'm opening up the question to get as many people's perspectives: Which testing method do you believe is best: 1- testing with a simulator (heat die) or 2- testing using a real CPU __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

02-22-2006, 05:41 PM	#3
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	Re: Which is the best testing method? I was going to add more options as they came up, but for the sake of arguments, let's say that: 1) the simulator is home made (it's not the Intel TTV). and 2) the CPU is an Intel with the groove machined into the IHS, for fitting a thermocouple. __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

02-22-2006, 05:56 PM	#4
snowwie Cooling Savant Join Date: Mar 2003 Location: New Jersey Posts: 154	Re: Which is the best testing method? I think tests are far more dependent on the tester. a good tester will determine their methodology depending on a variety of things: their budget, their audience/purpose, what exactly they're testing (blocks or systems), etc.

02-22-2006, 08:43 PM	#5
jaydee Put up or Shut Up Join Date: Dec 2001 Location: Spokane WA Posts: 6,506	Re: Which is the best testing method? Either way your results will be spacific to that test bench temp wise. I am in favor of the CPU method simply to prove the cooler functions properly on a actual computer being we don't use die sims to surf the net, play games and what not. Your temp results will be pretty much useless either way to end users. snowwie is pretty much correct. What is the goal? Product development, finding the product thermal performance on that test bench, over all computer performance or?....

02-23-2006, 12:38 AM	#6
9mmCensor Cooling Savant Join Date: Apr 2004 Location: Canada Posts: 219	Re: Which is the best testing method? Basically make basic grasp is that... If you do it one way, its not real world, and if you do it the other way, you can get consistant results. So basically both aren't perfect. If one is more right that the other, it is so marginally. But if the writer does a good job on either, and states how it was done, it should be good enough for a consumer to decern what is roughly accurate.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)