Mythbusters Demo GPU versus CPU

The Mythbusters, Adam Savage and Jamie Hyneman demonstrate the power of GPU computing.


balamadras says:

Amazing blast !! The machine you guys built for this demo incredible !!!

MadeBySaints says:

Great resolution on that thing 🙂 I bet it would have been better, but there was a malfunction 1:22

Joseph Benson says:

In then the idea of minining came and along with it the most expensive graphics card generation ever.

Janusz Biela says:

If you think about no interpolations system then of course CUDA wins…but CPU is millions times faster with interpolation system. For interpolation system you need genius code engine and only that can handle CPU. CUDA can only operate brute force system with very small code without reverse channel and is easy to write down.

Kfftfuftur says:

That’s why computer only gives one frame every ten seconds. Because it has to count down from ten before rendering the frame

powervac1 says:

взрослые дядьки а занимаются херней

drew6017 says:

That was a BLAST

shaun themanc says:

i can do that without a computer ffs

OptimisticPessimist says:

The way it looks to me as that each processor type is gradually evolving, while adding features of other types. Some companies have built experimental neural processors. Those will probably take over most work that quantum computers are not suitable for. GPUs will probably go back to being hard wired specifically for fixed purpose pixel programming or each core having only all functions needed for graphics rendering.

Murad Beybalaev says:

Such a suckass letdown!

Axel White says:


Bassem Alhalabi says:

Good job, that is what parallel processing is, doing thousands of actions all at once. But what people often forget is the extensive time and cost of the overhead to prepare for that one fascinating shot! Nevertheless parallel processing is still a powerful concept but the software community still have miles of development to make a meaningful use of true parallelism.

FlySpyGuy says:

This implies that the GPU is superior to the CPU, but in reality, the CPU would be handling much larger tasks, one at a time. They both have their uses. I wanna see the AMD CPU/GPU hybrid.

Anuraag Vaidya says:

At 1:22, one overclocked GPU core broke a tube 😛

Marcus Lai Kai Cong says:

how long does it take to reload

Enoch Rogahn says:

This idea blew my mind.

1990idris says:

Is GPU used for gaming?
My CPU’s temperatur stands 73°C on the app in my laptop but my GPU ‘s temperatur was UMA. What does UMA stand for?
Was my GPU not turn on?

Justin Felder says:

This actually is a an interesting way to explain how a GPU works as opposed to a CPU

joessmax says:

This is only for graphics processing?

james fordin says:

Why was this only just recommended to me?

nooneisback says:

A CPU connected to the 2.0 would do just as fast. It is true that GPUs can simultaneously compute multiple functions, but the speed of an individual computation is so fast that nobody would notice it, even on a scale of 1000×1000.

Radwan HAMDI says:


Meter reading says:

Linus torvalds: “Nvidia, Love you!”

Alaa T says:

While I agree that GPU computing has massive potential, this video at best is a stunt for high-schoolers to all scream WOW while the camera captures their reaction.

Even commodity CPUs now have 4 hyper-threaded cores, the video emulates an 8-bit single core CPU with a clock rate of 1 Hz at best.

It would have been great if Mr 2.0 was able to pick up the right ink colors out of a pool load them into the parallel pipes (emulating CUDA cores) in order to match a target pattern and fire them on time.

Victor Konovalov says:

америкосы хуесосы

ParaSwarm says:

It’s… just a bunch of tubes pre-loaded with a specific colored paintball and fired at the same time. I don’t get how people find this amazing.

ampwizzard2012 says:


Akhil Katpally says:

Very intuitive demo

AntiZippo says:

like novideo cards didnt explode enough. they created a nuke?

Supahfly Adrian says:

It’s like Homer Simpsons makeup shotgun 😀

V!K says:

this is absolutely not a good way of portraying the comparison. Why don’t they show where a CPU shines? my 1080 ti is supposedly super fast, but it still lags every few seconds. i7-7700k setup

Muhammad Naufal Alauddin says:

I miss mythbusters.. i love their episodes..

senseTUBE says:

wheres the science ?
this was just engineering..

Victor Konovalov says:

american cocksuckers

CGShareNet says:

Thank you for sharing in
Visit for digital art

Marcos Espinoza says:

Se habla de CPU, y GPU, si pero me trae a la mente el símil de SQL, y BIGDATA, parecerá loca mi comparación, pero es así, en SQL, se va armando el conjunto de datos, usando join, unio, y demás , en BIGDATA, es simplemente muéstrame el todo.
Loco pero NO..

angel Calderón says:

I’m confused??

Tim S says:

Anyone got video of the full thing?

VengefulSage says:

yah…that was useless and not at ALL what happens. stupid

DJ Programer says:

2009?!!! Wow I was really living in a cave back then.

Eyevenear says:

1:16 Creampied.

Lantz Greer says:

wait so this seems backwards the first one was way for impressive the second one was just gas going through tubes or i just don’t understand

World of AI - Games says:

Not a fair comparison. The GPU machine is fast, but it is also massive in size. The puny CPU machine is only about 1/20 the size. A true CPU machine would also be able to switch colors with its one nozzle and paint more than just blue.

subfeel says:

will it blend?

신우섭 says:

1. 지금. 다시보니. 물체가 맞고. 파편’이 튄다… 그것도 보인다는 말이 된다.2. CPU나 GPU 사이에 가속 판넬을 연결할 수 있다.

 Write a comment


Do you like our videos?
Do you want to see more like that?

Please click below to support us on Facebook!