Tearing problem at bigger overlay sizes

[Christiaan van Dijk]

Will try this, is only a minor modification at this point. First I'm 
going to play a bit more around with the vertical blanking detection. 
Somehow I can not imagine the graphics engine can not draw the frame 
fast enough.
The multiple tears I'm seeing might also have another cause; the 
vertical sync part is not working, the source frame rate is 50Hz and the 
target frame rate is 50Hz but the two are not synchronised. What I am 
seeing as tears in multiple vertical segments might not be in the same 
frame, each tear might just be in a single frame followed by a second 
frame with a tear in a later segment. The perception could be multiple 
tears in a single frame, the shift in segments would be caused by timing 
differences in the two 50Hz signals. This would be great news since at 
least the graphics engine would be fast enough.

Christiaan.

Matthias Hopf wrote:
> On Jan 13, 09 18:07:13 +0100, Xavier Bestel wrote:
>   
>> On Tue, 2009-01-13 at 17:41 +0100, Matthias Hopf wrote:
>>     
>>>> tear free. Sometimes the diagonal tear can be seen in one or more (2..4  
>>>> segments out of 11) of the 100 line segments. This last point is really 
>>>> puzzling, this could indicate the graphics engine is just keeping up 
>>>> with the refresh, slightly slower or slightly faster.
>>>>         
>>> Could be a caching issue, pushing out the cache after each block might
>>> help.
>>>       
>> Why not just after the last block ?
>>     
>
> Because the chip might reorder memory writes, and decide for later
> blocks to be pushed out first (or even pushed out to memory w/o changing
> the cache). That way you *could* see multiple tearings.
>
> Again, this is a lot of hand waving. Still, it might be an idea.
>
> Matthias
>
>