Question What would be the ideal CPU testing PC?

[ReveurGAM]

I realize this is a tricky question.

Please list all of the components you'd recommend for two computers - one for AMD and one for Intel.

My thought is that you don't need a PC - you just need a device that can simulate various CPUs by mimicking the heat output (including the size). Since LonGwin and any other companies that make such devices charge an arm and a leg for a complete device, and it's liable to be massive and require more limbs to pay the shipping, that doesn't seem viable. I don't have the technical skills to engineer and build such a device, either.

So, I'm stuck with this problem.

I know, I know, it can't just be one CPU from each to truly be able to test all coolers. The problem is, short of being majorly sponsored and finding a team of experts to design, build and run the tests, I don't think ANYONE has another option.

Unless you can design and build it for me.

 

[MrN1ce9uy]

If you can buy multiple coolers but only one CPU, then just work with what you have. Why pretend to test something that isn't what you are testing?

 

[hotaru.hino]

You can make one yourself, but the problem is how accurate and repeatable it's going to be. But generally speaking, if you're doing repeatable thermal loads, you usually attach a power resistor to a lab bench power supply so you can precisely control how much power is being dissipated across the resistor. From there, you attach the resistor to heat spreader that you can then attach to the CPU cooler.

That'll get you at least a basic setup, you'll have to find a way to attach sensors and such to it.

Take a look at how AnandTech did their CPU cooler test rig: https://www.anandtech.com/show/18985/the-be-quiet-pure-rock-2-fx-cpu-cooler-review/2

 

[ReveurGAM]

You can make one yourself, but the problem is how accurate and repeatable it's going to be. But generally speaking, if you're doing repeatable thermal loads, you usually attach a power resistor to a lab bench power supply so you can precisely control how much power is being dissipated across the resistor. From there, you attach the resistor to heat spreader that you can then attach to the CPU cooler.

That'll get you at least a basic setup, you'll have to find a way to attach sensors and such to it.

Take a look at how AnandTech did their CPU cooler test rig: https://www.anandtech.com/show/18985/the-be-quiet-pure-rock-2-fx-cpu-cooler-review/2

I understand the basics like that because another reviewer explained it to me. I'll look at that link asap. Thanks!

 

[ReveurGAM]

You can make one yourself, but the problem is how accurate and repeatable it's going to be. But generally speaking, if you're doing repeatable thermal loads, you usually attach a power resistor to a lab bench power supply so you can precisely control how much power is being dissipated across the resistor. From there, you attach the resistor to heat spreader that you can then attach to the CPU cooler.

That'll get you at least a basic setup, you'll have to find a way to attach sensors and such to it.

Take a look at how AnandTech did their CPU cooler test rig: https://www.anandtech.com/show/18985/the-be-quiet-pure-rock-2-fx-cpu-cooler-review/2

What they wrote on that page makes a lot of sense. If only I knew how to build something like that (but more complete)!