kicad-developers team mailing list archive

[Martin Schreiber <mse00000@xxxxxxxxx>]

On Tuesday 07 February 2017 23:52:59 Cirilo Bernardo wrote:
>
> I think ripping out Legacy soon after release would be a good
> thing and will allow (some of us) to concentrate on improved
> modularity (and maintainability) of the code. Encouraging
> people to use GAL in V5 (while also discouraging them from
> using Legacy) should hopefully get us the bug reports we need
> to implement and polish up the GAL tools.
>
> Any comments?

I am relatively new to KiCad and I don't use it in production up to now 
because I need to finish the interface to my other production environment 
first, first of all I need to finish MSEkicadBOM.
Before I decided to invest time into the KiCad project I tested it on 
different Linux boxes. The only canvas which really worked was the legacy 
canvas. The OpenGL canvas either had issues or did not work at all because of 
various Linux driver problems. The performance of the legacy canvas was much 
better than the performance of the OpenGL canvas on the machines I tested. 
performance of Cairo canvas is even worse, almost unusable.
Admitted, I don't have the newest hardware but I like to have not so fast 
machines in order to be forced to develop fast software. ;-)

I recently had to revive a very old Orcad PCB-386 project on a stone age PC 
with K6 processor and Windows 95 - what surprise, the performance and user 
experience was on a par or even better than with current software on current 
machines.

So please take into account that not all users have the newest machines, even 
more users of free software who sometimes have a very low budget.

Martin