[Bf-vfx] Normal Use Case for Masking

[pete larabell]

You are correct Tom, it is only ms. As a matter of fact, in the test
cases I made, it looks "instant" to the end user. However, since I'm a
long time graphics programmer with most of that time doing realtime
2d/3d, I can't help myself trying to always find faster ways to do
everything. :)

On Tue, Apr 10, 2012 at 12:34 PM, Tom M <letterrip at gmail.com> wrote:
> I suspect that the rasterization will be a few ms of the overall time spent,
> and thus further optimization work won't affect overall time much.  Of
> course I could be wrong :)
>
> LetterRip
>
> On Tue, Apr 10, 2012 at 10:31 AM, pete larabell <xgl.asyliax at gmail.com>
> wrote:
>>
>> Hey all,
>>
>> Quick question:
>>
>> In designing the rasterizer for the 2D mask editor, I see that I have
>> a choice to make.
>>
>> Option-A.) I can rasterize any polygon whatsoever with a very
>> straightforward method. No problems. (this is how it works right now)
>>
>> but...
>>
>> B.) If the polygon (the mask) edges only intersect any given scan line
>> only twice (meaning enter poly... then leave poly... never re-enter
>> poly on this scan line) then the polygon could be rasterized at higher
>> speed. (can be implemented now, or added later, without
>>
>> so...
>>
>> Here is the question:
>>
>> It is worth a slight performance hit to TEST the mask for
>> Option-B-Compliance, which if it DOES pass, makes it faster to draw,
>> or just skip it since "most of the time" the mask will fail the
>> Option-B-Compliance test anyway?
>>
>> Not being a VFX guy I can't really judge what the typical use case is...
>>
>> Any feedback would be welcome.
>>
>> Cheers!
>> Peter
>> _______________________________________________
>> Bf-vfx mailing list
>> Bf-vfx at blender.org
>> http://lists.blender.org/mailman/listinfo/bf-vfx
>
>
>
> _______________________________________________
> Bf-vfx mailing list
> Bf-vfx at blender.org
> http://lists.blender.org/mailman/listinfo/bf-vfx
>