Page 2 of 2 FirstFirst 12
Results 11 to 16 of 16

Thread: How to compress 16 bit ILDA file to 8 bit smoothly

  1. #11
    mixedgas's Avatar
    mixedgas is offline Creaky Old Award Winning Bastard Technologist
    Infinitus Excellentia Ion Laser Dominatus
    Join Date
    May 2007
    Location
    A lab with some dripping water on the floor.
    Posts
    9,902

    Default

    Quote Originally Posted by JohnYayas View Post
    iShow DAC is 8bit and I have to shift bits to get the 16 bit ILDA files displayed on it. It does not like crap at all but I do have some hot spots too. I thought it was due to problems in iShow firmware but now I am starting to think that when shifting bits the resultant numbers turn out to be duplicated around curves. That could be a possibility and I will look into it with my software soon.

    But, as far as the image itself, it looks about the same as when projected from a 16bit DAC. Galvos are relatively slow devices so can't draw with great detail to begin with. Going from 16 to 8 bits doesn't have the same graphical effect as going from 16 to 8 bits on a precise monitor display.
    Pangolin did a study a few years ago, 10 bits is about the resolution for a straight up image with no rotation in software. 8 bits sucks if you rotate. I could not wait to get off 8 bit software a decade ago and never wish to go back. That is why a certain beloved 2000$ board does its sizing in hardware, no loss of detail and no jerkyness/stairstepping.

    Steve

  2. #12
    mixedgas's Avatar
    mixedgas is offline Creaky Old Award Winning Bastard Technologist
    Infinitus Excellentia Ion Laser Dominatus
    Join Date
    May 2007
    Location
    A lab with some dripping water on the floor.
    Posts
    9,902

    Default

    Quote Originally Posted by chad View Post
    I built a 8 bit machine 24 years ago on a apple ][e. The difference between a 8 bit and a 12 bit is only about $11.00. All things considered it is only a couple of bucks. $14 dollars gets you 8 channels of fast 12 bit dac. You just need a micro that is fast enough to keep up with the laser data.

    Chad
    Are you the AppleDacs man?
    You do not have to answer if that is incriminating.

    Steve

  3. #13
    Join Date
    Mar 2010
    Location
    Raleigh, NC
    Posts
    2,293

    Default

    Quote Originally Posted by mixedgas View Post
    Pangolin did a study a few years ago, 10 bits is about the resolution for a straight up image with no rotation in software. 8 bits sucks if you rotate. I could not wait to get off 8 bit software a decade ago and never wish to go back. That is why a certain beloved 2000$ board does its sizing in hardware, no loss of detail and no jerkyness/stairstepping.

    Steve
    I have seen plenty of images from 8 bit DACs (Popelscan, DMX cards, iShow) and really haven't noticed anything that stood out as unacceptable or even bad looking. If you do scaling and rotation using 8 bit values it will look bad but I don't know why anyone would do that. I process everything using 16 bits and only convert to lower values immediately prior to sending to the DAC. A straight up image looks no different than one rotated 28 degrees, even on 8 bit DACs. Correct me if I am wrong, but the computer on the Pango card can't do math any better than the processor on a laptop, so why would it make any difference where the sin and cos function are called? I'm not trying to be argumentative but what you are saying does match what I have seen so I would like to understand why you say that.

  4. #14
    Join Date
    Nov 2010
    Posts
    9

    Default

    I try to solve it with many methods. And I guess there some points are reduplicated .Because I test a 8bit ILDA file downloading from internet. It looks very nice.

    My ILDA files were stored in my SD card.

    I try to use AD7521 DAC Chip to meet 12 bits. But there are too many problems that my PCB do not support AD7521. Too many work need be do.

  5. #15
    mixedgas's Avatar
    mixedgas is offline Creaky Old Award Winning Bastard Technologist
    Infinitus Excellentia Ion Laser Dominatus
    Join Date
    May 2007
    Location
    A lab with some dripping water on the floor.
    Posts
    9,902

    Default

    John, I'll get back to you in more detail a day or two. I'm jammed up right now.

    Most issues with 8 bit come from rotates and 3D frames. Some of it has to do with jaggies and jerky motion and zooms.

    This is why the popular 80s and early 90s 8 bit systems used multiplying DACs for sizing and rotation. My QM32 has what looks like dacs driving the reference inputs of the other dacs.

    One other thing that colors my viewpoint, it has not been uncommon in the past for me to launching 200 to 450 foot throws onto large screens or walls, or hills or buildings, and on large throws you want every last bit you can get. Eight looks jaggy and jumpy at a distance. I remember vividly a few months ago a client just screaming over a radio about a LSB difference in size on a long throw and that was because the rental system did not use a analog knob to set size, and that was a "16" bit system.

    Steve
    Last edited by mixedgas; 11-19-2010 at 15:33.

  6. #16
    mixedgas's Avatar
    mixedgas is offline Creaky Old Award Winning Bastard Technologist
    Infinitus Excellentia Ion Laser Dominatus
    Join Date
    May 2007
    Location
    A lab with some dripping water on the floor.
    Posts
    9,902

    Default

    I try to use AD7521 DAC Chip to meet 12 bits. But there are too many problems that my PCB do not support AD7521. Too many work need be do.[/QUOTE]

    end quote.

    You can try to use HC ttl shift registers such as 74hct164 and a macro to ease your interfacing problems.

    Steve

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •