jbn@glacier.STANFORD.EDU (John B. Nagle) (10/30/87)
It's finally out. I have one. It works. It makes pretty pictures.
It's a great toy, with rock video and nightclub applications. Some
interesting special effects are possible. But the images are heavily
distorted by the digitizing process in a number of ways, and the output
is not suitable for serious work on moving images.
What LIVE actually does is digitize a color composite video signal
in real time with a 4-bit A-to-D. But the Amiga doesn't have enough
bandwidth to take all 4 bits at once at the TV scan rate, so only one
bit of the four is transferred during each frame. In four frame times,
one can obtain a monochrome image; in twelve, a color. Note, though,
that the different bits come from DIFFERENT FRAMES, so moving objects
look very strange. Some people think it looks cool.
If the input image is static, the results are quite good, given
the limitations of a 4-bit A/D converter. It is two orders of magnitude
faster than Digi-View. As a way of getting stills into a computer,
it is very cost-effective. The software supports this application well.
One can grab a sequence of images (about 25 per megabyte of available
memory) and write them to a file, to be played back with another
program. This is less useful.
As an input device for robotic vision, my interest, it has some
severe problems. But there is hope, and I'm working on finding out
what one can do given the limitations of the device.
$295, from A-Squared Distributions, Oakland, CA.
Amiga 1000 only. Plugs into bus expansion connector.
John Naglebryce@hoser.berkeley.edu.UUCP (10/30/87)
In article <17201@glacier.STANFORD.EDU> jbn@glacier.STANFORD.EDU (John B. Nagle) writes: > What LIVE actually does is digitize a color composite video signal >in real time with a 4-bit A-to-D. But the Amiga doesn't have enough >bandwidth to take all 4 bits at once at the TV scan rate, so only one >bit of the four is transferred during each frame. It should be pointed out that Live! does the reading in of data with the 68000. This is why it can't get all of it in just one frame. But check out those pull down menus! That was a very impressive hack. > $295, from A-Squared Distributions, Oakland, CA. >Amiga 1000 only. Plugs into bus expansion connector. Add: ...does not pass the bus through. Works in A2000 with Microbotics adapter if you add some extra wires. (Microbotics did not pass all the signals) |\ /| . Ack! (NAK, SOH, EOT) {o O} . bryce@hoser.berkeley.EDU -or- ucbvax!hoser!bryce (") U "You can count how many seeds are in an Apple, but not how many Apples are in a seed." -Ken Kesey
miner@dino.cpe.ulowell.edu (Rich Miner) (10/31/87)
In article <21518@ucbvax.BERKELEY.EDU> (Bryce Nesbitt) writes: >It should be pointed out that Live! does the reading in of data with the >68000. This is why it can't get all of it in just one frame. Live! does not do the reading, the live.library routines do the reading. Why bother to make the distinction? Well, if you have another coprocessor on the bus that can do DMA reads from MEMF_ANY and DMA writes to CHIP-MEM then you could add some hardware to read directly from the registers of the live board, just like the software in the library does. This would be similar to if the board had its own on-board DMA controller. One of the reasons we put Live in a A2000 was to have our imaging coprocessor read the images from live and move them into display RAM. If we can get one chip reading the image data, and one chip writing the image data, then we will have 5-chips in between to do some nifty processing of the images in "real-time". -- Rich miner@ulowell.edu 617/452-5000x2693 ULowell CPE Imaging Research Lab