ma, 2006-09-18 kello 18:44 +0200, Martin Wache kirjoitti:
> Heikki Lindholm schrieb:
> > ma, 2006-09-18 kello 11:39 +0200, Martin Wache kirjoitti:
> >> Heikki Lindholm schrieb:
> >>> The included patch makes dfb:mgatv use YV12 colourspace for video and
> >>> the hardware subpicture layer for OSD. The patch is a prime example of
> >>> bad goal-oriented programming, but does what I wanted: improves
> >>> performance to the level that my 450 MHz P2 can now function as a DVB
> >>> box like no underpowered P2 has done before! Unfortunately, the patch
> >>> probably breaks everything else, but I'm too lazy to clean it up.
> >>>
> >> Can you please point out to me what the advantage of using the
> >> subpicture layer is? To me it looks like the main speed change is that
> >> you don't scale the osd. Or does the change to YV12 colourspace bring
> >> some speed advantage?
> > 
> > Wow - this list has wonky ordering. See my previous mail about this.
> >
> 
> From you other mail:
> > I didn't receive the above Martin's message at all - off list, perhaps? 
> > Anyway, if it's still about the topic in question, the keyword for using 
> > the subpicture layer from my perspective is 'incompetence': I couldn't 
> > get crtc2 (mgatv) video layer working with YV12 colourspace and blitted 
> > OSD, so, I thought maybe it'd work with a subpicture layer more 
> > effortlessly, and it did. So, the advantage for me isn't the OSD really, 
> > but the YV12 for the crtc2 video. YV12->YUY2 was simply too slow.
> > 
> Ah, yes I see. I forgot that for mgatv the default is YUY2, I thought
> that it is I420. Of course YV12 is faster compared to YUY2...
> 
> Why is YUY2 is the default at all? Is the OSD not working at all or is

Don't ask me.

> > Using YV12 and LUT44 both bring advantages of reduced memory 
> > requirements and are generally faster. I didn't implement the scaling 
> > routines for the LUT44 OSD, because I simply didn't grasp the idea of 
> > the existing code in 5 seconds, and realized that scaling would 
> > practically never be required by the tv output.
> > 
> Scaling LUT44 makes not much sense, the quality would be very low
> because of the limited colors. The simplest and fastest solution one

That's true. 2x and (1/2)x scaling might be passable.

> could do, would be to skip or repeat pixels. To adopt the existing
> scaling routines to indexed bitmaps would be dead slow, and probably
> still have a limited quality.
> I saw that you implemented ScaleVUpXXXLut() and ScaleVDownXXXLut(), that
> is not needed, if you don't scale only a CopyToBitmapLut() should be enough.
> But without scaling there probably will be complaints again because of
> the shift of the menu cursor and the menu when using the DVD-plugin.

What does the DVD plugin do? I haven't tried that one (either). 

> >> I would not mind adding support for an indexed bitmap, but currently I
> >> don't see the advantage. If we do that we should probably think again
> >> about the interface of GetOSDMode(), I never liked that very much, it
> >> was a hack from the start...
> > 
> > And I didn't add much style to that either :) Adding the indexed code is
> > useless if there are no other users and I'm not going to clean up the
> > mess I made in video-dfb anytime soon. Of course, if you'd add just the
> > SoftOsd bits (without maybe the OSD_WIDTH changes), it would make my
> > life easier when trying to keep up with cvs updates. 
> >
> No, either we apply the DFB part and the OSD part together or nothing. I

Firm but fair. I'm not going for that, for now, then. Maybe I'll update
the patch sometime and re-post.

-- Heikki Lindholm

_______________________________________________
Softdevice-devel mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/softdevice-devel

Reply via email to