How Computers Work

Could it be someone is testing an automated system for analyzing and responding to discussion? Maybe they want to develop and sell some kind of AI for a search engine or other automatic knowledge processor.

This post has been edited by light in the tunnel on Apr 20 2010, 07:36 PM

If a processor "crashes" you're not going to see a "frozen" screen. You would probably see a blue screen or no screen at all. "Freezing" is usually the processor getting bogged down with instructions and becoming unable to recognize user input.

RAM is what your document was in when your computer crashed. Luckily, modern word processors will periodically save your document to the hard drive in case of such a failure.

Yes.

Typically your display is handled by a graphics processor and it has it's own memory (cheaper systems will use a hunk of main memory), but regardless of how it's done, what is to be displayed is transfered to this dedicated piece of memory and the graphics processor turns the data in those memory locations into a video signal that the particular display can use (XGA, VGA etc). Typically the display is refreshed at 60 frames per second.

If the computer freezes, which is typically the CPU executing a tight system loop such that Ctl/Alt/Delete doesn't do anything and it won't detect any external interupts, like moving the mouse, the graphics processor is unaffected and goes on it's merry way displaying what is in the graphics memory.

Arthur

I suspect that they are trying to get a bot that is able to post enough of these filler messages such that they can automate getting past the "have to post X messages" before you can post a link restriction.

But that's just a guess.

Arthur

This must also be why DVDs and digital TV signals respond to interference with series of "hiccups" in which a single frame freezes until a later one re-activates the display to go on changing. Am I overextending this?

Achieving the post-quota to be able to post links is a more practical explanation for the bot posts than my AI-experiments hypothesis. The question is are they going to be banned for plagiarizing?

Not really.

The process is similar.

Digital TV signals are processed using a graphics engine somewhat like the graphics processor in your computer, and so when they have problems with the input, like a smudge on a DVD, or a garbled signal, the last good frame in memory will typically be shown while the device trys to reacquire a good signal.

Different graphics processors handle input problems differently.

I had a older DVD player that would essentially freeze if it got a bad enough spot on a DVD, while my newer player generally manages to work through it, advancing the read head and figuring out how to resume playing, even if it might have to skip a few seconds of the movie. It sorta stutters along, until it gets a good signal.

Arthur

But isn't some kind of RAM required where the digital 1s and 0s of the DVD code is translated into a signal the TV can display as an image? I would guess that that RAM is encoded by the head that is reading the disc and recodes with each new frame, but that the encoding process itself is very similar to the display-RAM on the computer receiving frames at 60/sec from the processor. How else could the disk code be converted to image-signal without RAM?

Yes,

When I said "the last good frame in memory", I was speaking of RAM.

A frame of 1080p has about 2 million pixels, and then you have to specify for each pixel it's color and intensity, then you need to hold the frame being displayed and a frame or two being built, but even so, one still only needs a small amount of memory to do this processing.

Arthur

Arthur is correct in that the data can be processed with a small memory footprint.The first layer of encoding and decoding is done by an ASIC on the DVD player's logic board, and the resulting stream is handled by the graphics adapter. The processor doesn't have much to do at that point.

So back to the original topic: once the code has been read from the disk, it is held in a pattern of "on/off switches" in the ram until the following frame has been built in a separate section of RAM, which is then switched to for the following frame while a new image/frame is then built in the first one?

And as long as there is current going through the circuit/chip of the RAM, the frame/image "on/off switches" remain in their current configuration, if no signal is given to move on to a new frame.

In other words, each frame pattern/code is held in the RAM until something reprograms that pattern with a new one. It doesn't just automatically fade after being played, which is why a scratched disk can get hung up and freeze on a given frame - i.e. because the RAM is a group of switches that hold their positions as long as electricity is present?

I.e. like when the computer crashes and the desktop image freeze-frames?

Yes

See if these helps to understand how your DVD player works:

http://focus.ti.com/docs/solution/folders/print/530.html

Note ASSP is Application Specific Standard Product

http://focus.ti.com/dsp/docs/dspsupporto.t...Id=4&tabId=1443


Arthur

Neat flowchart, but the textual explanations in both sites was too macro to describe exactly how the RAM works. The OP seemed to want to un-black-box the components down to the level of individual 1s and 0s and I can't imagine that any DVD schematic is going to dissect at that level. Probably you would have to find the exact component that contains the imaging RAM and find a website about how that works, but it doesn't seem worth the trouble to me if you don't feel like knowing it down to the exact details.

Video data is stored like any other data, as 0's and 1's.

If you look at your display properties, under settings, you will see what your graphics processor (and monitor) can display.

The higher the bit setting the more colors you can display.

There are 3 basic colors you are controlling, Red, Green and Blue (RGB).

If you look at your TV with a magnifiing glass, you will be able to see these 3 color displays grouped together to form 1 pixel. It takes one bit per color, or 3 bits to control these 3 primary colors, but by mixing them you get 8 colors out of this set of 3 bits.

RGB
000 = Black
001 = Blue
010 = Green
011 = Cyan
100 = Red
101 = Purple
110 = Yellow
111 = White

Why more bits, because you can also control the intensity of each color, and by doing so create even more colors.

For instance with just 4 levels of intensity per RGB bit, you can create 4,096 unique colors.

Current typical display adapters use up to 24-bits of information for each pixel: 8-bit per component multiplied by three components, and thus 16,777,216 colors are possible.

See:

Video framebuffer

A framebuffer is a digital device for computers which stores in the so-called video memory the digital image to be displayed on the monitor.

Driven by software, the CPU write the appropriate bytes in the video memory to conform the image, which an electronic video generator sends to the monitor.

Modern systems encode pixel color values by devoting some bits groupings for each of the Red Green Blue (RGB) components.

By using an appropriate combination of red, green, and blue intensities, many colors can be displayed. Current typical display adapters use up to 24-bits of information for each pixel: 8-bit per component multiplied by three components.

With this system, 16,777,216 discrete combinations of R, G and B values are allowed, providing thousands of different (though not necessarily distinguishable) hue, saturation, and lightness shades.

http://en.wikipedia.org/wiki/RGB

Arthur

Hello to you all,

Well, I got interested in this thread when I saw adoucette name appearing as the latest poster. From experience I have come to know it is a learning moment for me wherever he contributes.

You'll remember the time I was almost in fits when my computer was invaded. I was driven to try to learn more about computers not necessarily to repair or to debug them but to avoid "superstitions" associated with same. I asked if someone could direct me to sites or suggest books that can assist a beginner. Today I got my answer. Thanks alot Arthur for your generous and well abridged but loaded (no mambo jumbo jargon) explanations. You're a born Teacher.

Thanks too to Sir. It helps to have a learning buddy who has a matching pace.

Forgive me for this hero worship. Arthur is my role model. I too have interests in so many fields, though my knowledge of them is a second of an arc compared to 360 degrees for a scholar in those specialties (if I may use a circle as an analogy)

Oops! the last post is by me. Forgot to sign in. Wonder what's happening lately. I used to be automatically signed in whenever I entered Physforum.org. Any help?