First post, by superfury

User metadata
Rank l33t

I notice that when dumping the horizontal timings on the ET4000, the horizontal timings seem a bit weird on 32K/64K modes past 640x480? There seems to be a lot of space after horizontal retrace until the next row starts rendering it's active display?

Timings observed:
00:15:03:02.02208: CRTC Info:
00:15:03:02.02272: HDispStart:0
00:15:03:02.02336: HDispEnd:1600
00:15:03:02.02464: HBlankStart:1616
00:15:03:02.02528: HBlankEnd:~59
00:15:03:02.02592: HRetraceStart:1656
00:15:03:02.02656: HRetraceEnd:~15
00:15:03:02.02784: HTotal:2032
00:15:03:02.02912: VDispEnd:600
00:15:03:02.02976: VBlankStart:603
00:15:03:02.03040: VBlankEnd:~117
00:15:03:02.03104: VRetraceStart:604
00:15:03:02.03232: VRetraceEnd:~14
00:15:03:02.03296: VTotal:633

Those are raw timing values as the renderer is rendering them. The vertial timings seem pretty much right(the tilde pretty much means that the value is the final 4 or 6 bits of the horizontal/vertical counter instead of a raw counter value).
All reported values are in decimal. The normal timings for the active display seem fine(800(x2 clocks due to double wide screen) pixels active display, then blanking at 1616 till it mathces ~59. Then Retrace at 1656 until and including 1663(final bits becoming 1111b). But after that it's kind of strange: about 369 pixels of nothingness(overscan?) until horizontal total is reached after retracing and active display starts again? What's up with that? It looks like retracing somehow ends WAY too soon? Or it's starting WAY too early?

UniPCemu Git repository
UniPCemu for Android, Windows and PSP on itch.io
Older UniPCemu PC/Android/PSP releases