Hi there,
nothing really important, just 2 or 3 things I've noticed watching bounceEvent.
I'm in the chasing cycles mode. Or I think I'm :)
line 311: skip is never use
line 354: could we hold (vga->_frameHeight-bufLineIndex) in a var instead of computing it twice
line 365: what about regrouping the blocks per frameScale and colorBits ?
switch(vga->_frameScale) {
case 1:
switch(vga->_colorBits) {
case 8:
...
break;
case 3:
...
break;
case 1:
...
break;
}
break;
case 2:
switch(vga->_colorBits) {
case 8:
...
break;
case 3:
...
break;
case 1:
...
break;
}
break;
}
line 457: what about having pixelBits & 0b00000111 in a var instead of the repetition ? Not sure about this as the compiler may be smart and do it by itself...
line 498: same here... the compiler may be smart enough to realise the vga->_frameScale == 1 so the mult is useless
line 529: again.... a shift left could do the trick but the compiler may have seen it too ;)
Hi there, nothing really important, just 2 or 3 things I've noticed watching bounceEvent. I'm in the chasing cycles mode. Or I think I'm :)
line 311: skip is never use line 354: could we hold (vga->_frameHeight-bufLineIndex) in a var instead of computing it twice line 365: what about regrouping the blocks per frameScale and colorBits ?
line 457: what about having pixelBits & 0b00000111 in a var instead of the repetition ? Not sure about this as the compiler may be smart and do it by itself... line 498: same here... the compiler may be smart enough to realise the vga->_frameScale == 1 so the mult is useless line 529: again.... a shift left could do the trick but the compiler may have seen it too ;)
Brilliant work, btw ;)