[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

Re: gEDA-user: pcb crooked traces



On Tue, Oct 12, 2010 at 05:19:45AM +0200, gedau@xxxxxxxxxxxxx wrote:
> 
> Wouldn't a float scale slow things down? Anyway if you need to modify a 
> lot of source files for this already, what about using a typedef 
> SIZE_UNIT instead of a hardcoded int or long int or long long int? Then the 
> typedef should default to long, which is good up to at least 2m, and if 
> someone really needs to make a bigger board on a 32 bit system, he can 
> modify the typedef to long long.
>

I highly doubt having a scale-factor would slow anything down, since
the scale factor is only relevant for GUI display and import/export,
and all of those are plenty complicated enough so that adding an extra
multiplication won't be noticeable.

Certainly, if the use of float/int scaling matters for performance,
we are Doing It Wrong.

And yes, we will use a typedef ;)

> I think the only barrier is that you need to decide whether you use int 
> or float, as some calculations would be broken if they were designed for 
> ints and someone changed the typedef to float. But because those 
> claculations you alrady need to decide on that, so no difference.
>

We've already had that discussion.


Andrew
 


_______________________________________________
geda-user mailing list
geda-user@xxxxxxxxxxxxxx
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user