TeckQuickyesque Videoscript feedback

[ChalkChalkson]

Hey guys,

I recently wrote a script for a techquicky like video with a harder focus on a physics approach to computer hardware, just because I felt like it. I am not sure yet if I am going to make a video out of it (thus there is no actual data in here).

I hope you have some things to say

If I ever made that video, I'd probably use CGP-Gey like powerpoint animations, moving around the different symbols to represent the "==> " lines and excel charts to present the data

 

Also: would you be interested in a video like this, and or the possible video hinted at in here?

 

The script:

Spoiler

There are tonnes of different PC cooling solutions, blower-style, Tower, AIOs and custom water-cooling loops.

Most of which advertise their performance in highly relative terms, or if they give you concrete numbers, most of the time they are just plain wrong.

To see why this is let's look at two different often advertised metrics: absolute temperature and a watt rating.

Image a 65° cooler on a 30W CPU and the same cooler on a 180W CPU,

since heat is work and watts are work/time this cooler would suddenly have to displace 6 times as much energy in the same time.

On the other hand, a cooler that always displaced 100W put on to a chip that already ran at room temperature would cool it below ambient straight of violating the second law of thermodynamics

So, what is the proper metric?

Let's do an experiment!

My setup here is an i7 6700k running on the be quiet! pure rock with the fan maxed out.

The variable we are going to change is the power used by the CPU.

Since we want to measure a relationship rather than an absolute number the unit of measurement is irrelevant.

For simplicity sake, I am going to use watts as my unit of power and measure Temperature of the CPU and ambient in degree Celsius.

And here are the results:

On the X-axis, you see the power, on the Y axis temperature, the blue dots are ambient, the orange dots CPU temperatures

What you can see here is that the gap between the dots gets bigger as the power increases.

So now it is time to form a first hypothesis: The temperature difference between a CPU with a given cooler depends and ambient temperature is a constant times the power the CPU draws.

To check this, I am repeating this experiment at a total of 3 different ambient temperatures, plotting the difference between CPU and ambient temperatures against the power draw.

The new data clearly confirms our hypothesis, so let’s now start to think about what this tells us about the cooling solution.

When the temperature is dependent on ambient and a constant the only thing the cooler can change is the constant the unit of which is °C/W, from now on I will refer to 1 divided by this constant though, for no other reason than that I like bigger to be better.

This new measure with the unit W/°C gives you the amount of power the cooler can move per degree of temperature difference.

 

This could be the end of this video, having found a proper measure for cooler performance, but what fun is a measure without using it?

For starters, I am going to compare 3 representative points of the spectrum: The Intel stock cooler, the be quiet pure rock and my total overkill custom water-cooling loop.

*****

So, what would you do with these numbers? Well, first look up the OC corrected TDP of your CPU eg 100W, get a temperature reading during summer in your room 30°C, pick a maximum temperature you’re comfortable with your CPU running at eg 80°C and

do the math

(CPU temp – ambient)*C=TDP

ð  C=TDP/(CPU temp-ambient)

ð  C=100W/(80°C-30°C)

ð  C=100W/50°C

ð  C=2 W/°C

With this knowledge, you can know check for a cooler that performs that well, neither overspending nor getting an unpleasant surprise, when your new cooler arrives.

I hope you enjoyed this video, if you want to know how to adjust the TDP for overclocking, check out the video linked in the video description, also feel free to leave any feedback and or subscribe.

 

Thanks for reading,

Chalky

[Mug]

I'm a science man, I have a degree in Computational Chemistry and am studying to do a masters. I will probably go on to become a teacher.

 

I don't like TechQuickie. I think the videos they do are actual garbage, they don't usually provide you with useful information, they often confuse more than they explain (due to the writing style) and they go into needless detail for simple concepts that doesn't need to be in there.

 

I'll have to watch another TechQuickie video to give you an example but I really don't want to do that.

 

I think for you, the best way to write scripts for these videos is to bullet point key points that you want in there. What makes CGP Grey so good is that he isn't the video-churning-out machine that TechQuickie is; his scripts are better than mediocre.

[ChalkChalkson]

3 hours ago, Mug said:

I think for you, the best way to write scripts for these videos is to bullet point key points that you want in there. What makes CGP Grey so good is that he isn't the video-churning-out machine that TechQuickie is; his scripts are better than mediocre.

I know, and I don't want to be neither of them, I am going to write a script as a full text and in sever iterations (this is the first draft, just to get an idea whether people are actually interested in this approach based on the scientific method 

[ChalkChalkson]

3 hours ago, Mug said:

I'm a science man, I have a degree in Computational Chemistry and am studying to do a masters. I will probably go on to become a teacher.

That is pretty cool, I am doing physics, I guess my folks are torturing you with a lot of Schrödinger