SavagePencil wrote: ↑Tue Oct 03, 2023 10:00 pm
thchardcore wrote: ↑Tue Oct 03, 2023 8:45 pm
With the Pro, does this mean problematic Toaplan V2 pcbs can finally be played on a VGA CRT or is this still a hardware specific anomaly of the Toaplan sync generation that the framebuffer will not be able to resolve?
Might be value in revisiting those boards with the original OSSC and the new firmware. Sounds like if you have an earlier OSSC revision it’s a minor hardware mod to make it work.
The anomality in Toaplan is such severe that a fast-locking PLL and ability to operate without continuous line-lock are needed, thus only Pro is capable of that.
kitty666cats wrote: ↑Wed Oct 04, 2023 2:03 am
EDIT: Oh, the OSSC Pro isn't gonna have 4K output. Guess it's gonna be different strokes for different folks... I don't even have a 4K TV, so the OSSC Pro is still more appealing to me.
Well, it is still able to output 2880x2160@60 with TX pixel repetition which is kinda good compromise for 4:3 content.
Dr. Claw wrote: ↑Wed Oct 04, 2023 11:42 amGreat news. BTW, does the Pro allow RGsB over SCART like the original OSSC?
Yes, the available inputs are a superset of OSSC classic.
SGGG2 wrote: ↑Wed Oct 04, 2023 5:09 pmDoes scaler mode have masking and resizing? Sharp bilinear scaling? Is it possible to feed line doubling content into scaler mode?
Scaler mode has size & position options. By default sampled window size is scaled to fullscreen (honoring source aspect ratio) unless integer mode is selected. Sharp bilinear etc. typical algorithms are available and custom ones can be loaded from SD card. At some point I had nearest neighbor prescaler option for testing so it's a possibility.
ZellSF wrote: ↑Tue Oct 10, 2023 9:31 am
I am again curious about the lag. What are the lag numbers in the three modes (pure/adaptive/framebuffer)? How about while deinterlacing? I asked earlier but got no answer, maybe because it was too early, but we should have some numbers by now?
Pure LM: <3 scanlines
Adaptive LM: <50 scanlines
Scaler (framelock): ~1 frame/field
Scaler (no framelock): 1-2 frames/fields
There's room for improment in scaler mode, so eventually framelocked mode latency should reduce close to A-LM latency.