The Tech Museum - Inside Microchips - 01

PROJECT: Inside Microchips
PRODUCED BY: Created entirely by Gary Backus.
PRODUCED FOR: The Tech Virtual Museum Workshop
VIDEO NOTES: A bit more technical and formal than the 2nd one.
PROJECT PAGE: http://thetechvirtual.org/blog/projects/microchip-clips/videos/inside-microchips

DESCRIPTION: The result of this project is a presentation-style video about the inside of a microchip, then it showcases the inside of a CPU while mentioning some details of what a CPU is and how it works. I've decided to create two videos with identical animations but somewhat different narration approaches; this one is a bit more technical, the other much less so. The goal of this presentation is to have the viewer gain a sense of the small size yet complex workings of a microchip and CPU (as an example of an advanced microchip), without getting too technical, but I do throw some terms out there, and explain what I'm talking about. I didn't show any layers of semiconducting materials inside the microchip (2nd scene of the video) in order to represent the whole in a more interesting and fluid way. Also, the CPU seen in the video was not modeled to any exact architecture or common design for copyright and accuracy reasons; it's a generic model for demonstration purposes. Thank you for taking the time to view this video. Source animation clips created for this video, along with the video itself, are available for use in any way by The Tech Virtual Museum Workshop. This video may be featured in a playlist of other short videos about microchips at "The Tech Museum" in San Jose, California. The exhibit is called "The Tech Silicon Valley Innovation Gallery".

VIDEO ATTRIBUTES:
YouTube: MKV container, 1280x720 at 29.970 fps, MPEG4-AVC Main at 3.8 Mbps, 192k 2-channel audio.
Source: MPG container, 1280x720 at 29.970 fps, MPEG2 High at 19.8 Mbps, 320k 2-channel audio.

SOFTWARE USED:
Autodesk Maya: For all 3D and 2D content creation.
Autodesk Toxik: For compositing rendered sequences.
Sony Vegas: For editing the video, and rendering to MPEG2.
x264, ffmpeg, mkvmerge: CLI tools used to convert MPG to MKV.

RANDOM TECHNICAL DETAILS: The scenes were modeled using the brute force technique, and the CPU model itself contains a little over 3.1 million triangles; geometry instancing, however, was used to lower memory consumption. Mental Ray for Maya was used to render the scene, and no indirect lighting algorithms were used for lighting; I set up a basic manual lighting rig, comprised of around 16 directional lights and shadow-only spot lights for depth-map shadows. Ambient occlusion, z-depth, and motion vector passes were also rendered for use in applying indirect lighting, depth of field, and motion blur in post production. If you're wondering, in Maya, shadow-only lights are achieved by setting the light's color to black and the shadow color to a negative value, leaving the light intensity a positive value. There's another way too; set a light's intensity to a positive value, then duplicate it and set the intensity of the new light to a negative value exactly the opposite of the positive one (negate the intensity). Now set shadows to cast on the positive intensity light. If you need a light that sucks light out of the area it shines upon, simply set the light color to white (for uniform subtraction) and the intensity to a negative value.

Category:

Education

Tags:

• inside
• the
• chip
• microchips
• microchip
• tech
• virtual
• museum
• workshop
• integrated
• circuit
• ic
• microprocessor
• CPU
• photolithography
• ion
• implantation
• semiconductor
• silicon
• wafer
• doping
• dopant
• atoms

[TacTiCOrc]

Can someone explain to me how the transistors work?

I mean who changes their position from 0 to 1 ?

I dont understand how the Data Input is being transformed into Data Output. Would be thankful for explanation.

[Asephei]

@TacTiCOrc Again on the most basic level what makes a computer useful are the logic gates, an electronic calculator is a very simple computer. I can't answer your question because the answer would require too much space. There's an amazing amount of information online about any topic so the answer is there. You might also want to buy an electronics kit and build your own logic gates. I'm not an electrical engineer or anything, just someone interested in science and technology as a side-hobby.

[marcoamartins2]

Isso não é de Deus não. É complicado demais!!!

[Asephei]

@Asephei Correction, not 255, 256 I meant. In 8-bit unsigned integer color, you have, well, 8 bits, and 2 to the power of 8 is 256. In steps it goes from 0 to 255, which is a total of 256 levels of color value change in each red, green, and blue sub-pixel. The 8 bit thing is completely inadequate in representing anything close to the brightness values encountered in the real world, that's why 8-bit imagery must be gamma-corrected to get the most out of such a small amount of color values.

[axelasdf]

@TacTiCOrc

Usually it's a voltage. 5V(or some other arbitrary voltage) is a 1 and 0V is a zero. Transistors work by amplifying current (a small amount of electric flow from one side opens the door for a LOT more current to flow.) Hope that helped. Reply if more explanation is wanted.

[GoCubsGo2012]

youtube.com/watch?v=BL2DPFISUg­M

[truckrally]

@TacTiCOrc

Isnt that at leasst a 3 year long engineering degree? I know, but would never take the time to explain

[Asephei]

@TacTiCOrc I know I digressed from your original question and threw in an example about digital data but the raw details about how information is worked on in a processor are best found in scientific and engineering literature, but there's plenty of information online as well. I wouldn't be able to give you the answer satisfactorily because whatever I know there are countless things I don't know and never will know, but keep knowledge seeking.

[TacTiCOrc]

@Asephei You are writing so much stuff i already know, but that wasnt my question. It seems like you dont understand what im talking about. But since this seems to be a bit too hard to explain i will just try to find answers on google. Thank you anyway for your efforts.

[Asephei]

@TacTiCOrc Sophisticated maths such as the "Discrete Fourier Transform" are used to develop compression schemes to reduce file sizes of such image and video content, and the same goes for audio. Keeping it general, visual and audio is usually compressed in a "lossy" scheme (though some formats like FLAC audio are lossless), while a ZIP archive compresses data "losslessly" (it has to). The simplest form of lossless compression is "Run-Length Encoding" and anyone can understand it.

[Asephei]

@TacTiCOrc Also consider digital video. If you have a 1920x1080 image (frame), that is 2,073,600 pixels. Each pixel is 3 bytes of information (in 8-bit integer color), resulting in 6,220,800 bytes (6.22 MB) per frame. Multiply this by, for example, 2 hours at 24 frames per second: 6.22 MB * 7,200 seconds * 24 fps = 1.075 Terabytes (1,075 GB) of data for a typical 2 hour movie, uncompressed. That would be more than 20 dual-layer 50 GB BluRays.