RISC processors kept things simple, but when do you need to make your hardware more complicated and when can you leave it to the software? We asked ARM’s Jem Davies.
Note: At 4min 12secs the graphic says ‘nearly 1GB’ when it should say ‘nearly 10GB’
Brain Scanner: https://youtu.be/TQ0sL1ZGnQ4
Finite State Automata: https://youtu.be/vhiiia1_hC4
CPU vs GPU: https://youtu.be/_cyVDoyI6NE
3D Rock Art Scanner: https://youtu.be/Ahdnkj77rAY
Steve Furber on ARM: https://youtu.be/1jOJl8gRPyQ
http://www.facebook.com/computerphile
https://twitter.com/computer_phile
This video was filmed and edited by Sean Riley.
Computer Science at the University of Nottingham: http://bit.ly/nottscomputer
Computerphile is a sister project to Brady Haran’s Numberphile. More at http://www.bradyharan.com
Uncompressed video is used… In broadcast for one. 1080i is 1.5 Gbps for example
Bring back fractal encoding !!!!!
Whos bright idea was it to put the link at the bottom of the screen beneath the youtube interface so it cant be clicked? :/
this made me look for some color noise videos on youtube, but sadly I couldn't find any :(
So what does the most computationally expensive image to compress look like?
I find standard definition on TV really bad when the image gets a bit darker, the motion prediction doesn't track properly and you see a face move and the eyes stay still for a bit before catching up.
Funny fact: I watched this video with a 5xx series NVidia GPU, that doesn't integrate hardware VP9 compression. So, the reading was made by the CPU.
And it was f*** slow :-p
I think he forgot what the question was.
Great explanation!
Now, how does the most complex frame you can reasonably expect to get look like?
:D
Anyone else noticed how short his middle finger is?
Lovely, thank you!
This man is amazing. Would love to see more interviews with him.
Great vid. Get a portable tripod though guys :D
So why are we still using software for 10-bit h.264?
Don't use HEVC! It's a patent bomb!
Nice video.
Luckily I shoot my films in RAW (CinemaDNG) ~24MB per frame at 24fps so my camera doesn't waste energy on compressing 😉 it wastes it on buffer space and SSD writing.
You only realize how hard things are compressed when you do color grading or VFX work (especially keying) a shot in RAW keys far easier and cleaner near the edges than a shot in a compressed format.
Plus compositing is easier too because instead of only 10bits of colourspace you have 24bits more space to push and pull the elements to match more.
And no BANDING! If only YouTube had better compression, youtube kills videos.
that's what i call getting it from the source.