16 bit to 8 bit

Comments

• Options

  AMXJeff Posts: 450

  October 2007

  ryanww wrote:

  I would like an example of converting from an 8 bit integer to a 16 bit integer and vice versa. I just have to convert a level back and forth.
  
  Thanks,
  Ryan

  
  No conversion needed...
  

  DEFINE_VARIABLE
  
  // 8 BIT VARIABLE
  CHAR cTest
  
  // 16 BIT VARIABLE
  INTEGER nTest
  
  DEFINE_PROGRAM
  
  // 8 Bit variable to 16 bit variable - No Code warnings.
  nTest = cTest // (Binary) Coverts 1111 1111 to 0000 0000 1111 1111
  
  // 16 Bit variable to 8 bit variable - Code Warning, use Type_Cast
  cTest = TYPE_CAST(nTest) // (Binary) Coverts 1111 1111 1111 1111 to 1111 1111
  
  
  

  0
• Options

  ryanww Posts: 196

  October 2007

  I don't know this would work for what I need. I have a structure with 500 positions of integers. All of them are integers from 0-65535 which is a level value of which a device uses. I need to know how to convert a 0-255 level that is sent from a g3 touch panel and convert that value to what the equivalent of 1-65535. So if the touch panel sent a level of 127, it needs to store that as 32767 in the structure integer.
  
  Example:
  
  LEVEL_EVENT [dvBkStoreG3Web,1] //Inside Master Fader
  {
  SEND_STRING dvSymnetBS,"'CS 421 ',ITOA("(LEVEL.VALUE)16 bit"),$0D"
  SYM_CONTROL[421].POSITION = (LEVEL.VALUE)16bit
  }
  
  SEND_LEVEL dvBkStoreG3Web,1,(SYM_CONTROL[421].POSITION)8 bit //Inside Level

  0
• Options

  AMXJeff Posts: 450

  October 2007

  Oh, ok...........
  
  Simple formula that works for all ratio conversions. Does not matter if your going from a large range to a smaller range, or vice versus, this formula works... The only rub is that the ranges have to be zero to x. If they are not you need to subtract or add to make the ranges 0 to x.
  
  answer = ((max_converted_value * current_value) / max_current_value)
  
  // coverting to a 0 - 65535 range,starting with a 0 to 255 range;
  answer = ((65535 * 127) / 255);
  
  // coverting to a 0 - 255 range, starting with a 0 to 65535 range;
  answer = ((255 * 32767) / 65535);
  
  // coverting to a 0 -255 range, starting with a -100 to 20 range;
  answer = (255 * (current_value + 100)) / 120);
  
  // coverting to a 0 -255 range, starting with a 20 to 80 range;
  answer = (255 * (current_value - 20)) / 60);
  

  ryanww wrote:

  I don't know this would work for what I need. I have a structure with 500 positions of integers. All of them are integers from 0-65535 which is a level value of which a device uses. I need to know how to convert a 0-255 level that is sent from a g3 touch panel and convert that value to what the equivalent of 1-65535. So if the touch panel sent a level of 127, it needs to store that as 32767 in the structure integer.
  
  Example:
  
  LEVEL_EVENT [dvBkStoreG3Web,1] //Inside Master Fader
  {
  SEND_STRING dvSymnetBS,"'CS 421 ',ITOA("(LEVEL.VALUE)16 bit"),$0D"
  SYM_CONTROL[421].POSITION = (LEVEL.VALUE)16bit
  }
  
  SEND_LEVEL dvBkStoreG3Web,1,(SYM_CONTROL[421].POSITION)8 bit //Inside Level

  0
• Options

  Daniel Tehan Posts: 42

  October 2007

  Just curious, I am asking because I don't know the answer.
  
  Since the processor can handle a LONG - 32bit integer I would assume that the operation (65535 * 127) would not overflow the internal operation of the processor. But what if...
  
  I can not think of a real application, but what if you wanted to end up with a range from 0 - 4294967295 (FFFFFFFF) would the initial operation of (4294967295 * 127) overflow the internal operation of the processor? what is the limit of the processor's mathematical operations?

  0
• Options

  AMXJeff Posts: 450

  October 2007

  Very good question the answer is that the math will be forces to stay within the 32bits.
  
  // So this should be true, but netlinx will keep the answer under 32 bits
  
  $FFFFFFFF = (($FFFFFFFF * 2) / 2))
  
  // but netlinx does this
  FFFFFFFE = ($FFFFFFFF * 2);
  7FFFFFFF = ($FFFFFFE / 2);
  
  // so this is the answer, its wrong, but the answer none the less.
  $7FFFFFFF = (($FFFFFFFF * 2) / 2))
  

  dtehan@unitedvi wrote:

  Just curious, I am asking because I don't know the answer.
  
  Since the processor can handle a LONG - 32bit integer I would assume that the operation (65535 * 127) would not overflow the internal operation of the processor. But what if...
  
  I can not think of a real application, but what if you wanted to end up with a range from 0 - 4294967295 (FFFFFFFF) would the initial operation of (4294967295 * 127) overflow the internal operation of the processor? what is the limit of the processor's mathematical operations?

  0