Originally Posted by Senor Hubris
You shouldn't need to double the pixel clock. I don't.
Hmmm... So you double only the 4 vertical parameters and this makes it fit on screen ? I'll have to give that a try.
Okay, the rest of this post is just me thinking out loud. I haven't tried this yet, but its what I plan to try if doubling the vertical parameters doesn't work.
After some head scratching it seems to me that both the original workaround (pixel clock doubled) and this variation (pixel clock not doubled) both have a flaw. The original moeline gives horizontal timings that work properly. If you double the pixel clock the HSync pulses will arrive twice as fast, which is okay if your TV can sync to them. But you have also cut in half the width (in seconds) of the HSync pulses, which indicate the amount of time needed for horizontal retrace -getting the electron beam back over to the left side of the screen. My understanding is that this time is a property of the TV and should not
be changed. OTOH, by cutting in half the amount of time it takes to draw one line, then all 4 vertical timing parameters can be doubled with no ill effects.
The variation on the workaround (double the vertical parameters, but not the pix clock) has similar issues. Since absolutely nothing has changed with the horizontal timing, everything will still work correctly. But when you double all four vertical parameters, you double not only the number of lines you want displayed on screen, but also the width of the VSync pulse - the time your CRT takes to sweep the beam back to the top of the screen.
So my guess as to the "perfect workaround" would be the following :
1) Leave the pixel clock and horizontal values the same.
2) For the vertical parameters (call them V1, V2, V3, V4), record the value of V3-V2. This is the width of the VSync pulse.
3) Double the values of V1, V2, and V4.
4) Add the VSync pulse width from step 2 to the new V2. Use this as the new V3.
Like I said, I'll give this a try tonight and see what happens.