Tim, This was on the Fractint bug list:
Another thing : It's not possible to finish an image that was started on xfractint with (MS)DOS FRACTINT ( and vice versa ), because the fractint002-extensions are different for those 2 ports.
In my version, I've taken out any Xfractint specific code for the resume logic because the macros that handle a variable number of parameters have been updated. However, because of how we are saving the data, I haven't been able to figure out how to make the saved extentions the same size. We use the sizeof() function to determine how many bytes to put in the resume memory and by the time we get to the GIF routines, we have no idea what the size of the different variable are. We would need to make the conversion in the resume routines instead of in the GIF handling routines. Very ugly, and I don't want to do it. Does anybody see a way to get past this, and how many people really need this to work? BTW, there is some documentation at the bottom of the resume routines in fracsubr.c that appears to belong elsewhere. Do you have any idea where? Jonathan
However, because of how we are saving the data, I haven't been able to figure out how to make the saved extentions the same size. We use the sizeof() function to determine how many bytes to put in the resume memory and by the time we get to the GIF routines, we have no idea what the size of the different variable are. We would need to make the conversion in the resume routines instead of in the GIF handling routines. Very ugly, and I don't want to do it.
Does anybody see a way to get past this, and how many people really need this to work?
We made simple designe decisions many years ago that made the fractal info machine dependent. I don't think there's a lot of point tweaking the old method of storing data. What's really needed is to store fractal info in machine-independent form, which is not that hard to do. Without looking, you probably ran into sizeof(int), which is two bytes in Fractint's DOS compile environment, and four bytes on most other systems.
BTW, there is some documentation at the bottom of the resume routines in fracsubr.c that appears to belong elsewhere. Do you have any idea where?
I'll look. I'm off work until next Monday, and actually at home, a rarity. Tim
On Monday 29 December 2003 04:23 pm, Tim Wegner wrote:
However, because of how we are saving the data, I haven't been able to figure out how to make the saved extentions the same size. We use the sizeof() function to determine how many bytes to put in the resume memory and by the time we get to the GIF routines, we have no idea what the size of the different variable are. We would need to make the conversion in the resume routines instead of in the GIF handling routines. Very ugly, and I don't want to do it.
Does anybody see a way to get past this, and how many people really need this to work?
We made simple design decisions many years ago that made the fractal info machine dependent. I don't think there's a lot of point tweaking the old method of storing data. What's really needed is to store fractal info in machine-independent form, which is not that hard to do.
Without looking, you probably ran into sizeof(int), which is two bytes in Fractint's DOS compile environment, and four bytes on most other systems.
Yes, that's true. But, I don't see any way to fix this one because the resume data is a block of resume_len number of bytes. Twice as many bytes for Xfractint. There is no way to tell if consecutive bytes are part of an int or long or double. Jonathan
participants (2)
-
Jonathan Osuch -
Tim Wegner