View previous topic :: View next topic |
Author |
Message |
Marnic
Joined: 20 Mar 2007 Posts: 1 Location: Cheshire,England,UK
|
Where are int48 and int64 |
Posted: Mon Oct 18, 2021 8:12 am |
|
|
Am I right that int48, int64 are not supported types for the 8 bit micros?
I just cannot see the reason for leaving these types undefined given the high
speeds now attained by the 8 bit Microchip families.
Mark |
|
![](templates/subSilver/images/spacer.gif) |
Ttelmah
Joined: 11 Mar 2010 Posts: 19358
|
|
Posted: Mon Oct 18, 2021 9:30 am |
|
|
No, they are only supported on the 16bit micros.
You can use a typedef to create these as an array of bytes or int16's,
but you then have to write your own arithmetic for these. I've done this
in the past for an int24, and int30 type.
On the original 8bit micros, it made sense, since the maths is very bulky.
However on the newer chips with a lot more ROM, and often hardware
mul and div, the libraries can be a lot smaller. Perhaps suggest to CCS
that it might be useful to offer an option to support these. |
|
![](templates/subSilver/images/spacer.gif) |
temtronic
Joined: 01 Jul 2010 Posts: 9168 Location: Greensville,Ontario
|
|
Posted: Mon Oct 18, 2021 10:55 am |
|
|
It should be possible to take Microchip Assembler versions of the math and 'wrap it' or whatever is the real term.... to have a program run it as a 'function'. |
|
![](templates/subSilver/images/spacer.gif) |
|