-/* $NetBSD: grf_rh.c,v 1.9 1995/08/20 15:09:25 chopps Exp $ */
+/* $NetBSD: grf_rh.c,v 1.10 1996/01/28 19:19:05 chopps Exp $ */
/*
* Copyright (c) 1994 Markus Wild
vm->disp_width = (md->DEP == 4) ? md->MW : md->TX;
vm->disp_height = (md->DEP == 4) ? md->MH : md->TY;
vm->depth = md->DEP;
- vm->hblank_start = md->HBS;
- vm->hblank_stop = md->HBE;
- vm->hsync_start = md->HSS;
- vm->hsync_stop = md->HSE;
- vm->htotal = md->HT;
+
+ /*
+ * From observation of the monitor definition table above, I guess
+ * that the horizontal timings are in units of longwords. Hence, I
+ * get the pixels by multiplication with 32 and division by the depth.
+ * The text modes, apparently marked by depth == 4, are even more
+ * wierd. According to a comment above, they are computed from a
+ * depth==8 mode thats for us: * 32 / 8) by applying another factor
+ * of 4 / font width.
+ * Reverse applying the latter formula most of the constants cancel
+ * themselves and we are left with a nice (* font width).
+ * That is, internal timings are in units of longwords for graphics
+ * modes, or in units of characters widths for text modes.
+ * We better don't WRITE modes until this has been real live checked.
+ * - Ignatios Souvatzis
+ */
+
+ if (md->DEP == 4) {
+ vm->hblank_start = md->HBS * 32 / md->DEP;
+ vm->hblank_stop = md->HBE * 32 / md->DEP;
+ vm->hsync_start = md->HSS * 32 / md->DEP;
+ vm->hsync_stop = md->HSE * 32 / md->DEP;
+ vm->htotal = md->HT * 32 / md->DEP;
+ } else {
+ vm->hblank_start = md->HBS * md->FX;
+ vm->hblank_stop = md->HBE * md->FX;
+ vm->hsync_start = md->HSS * md->FX;
+ vm->hsync_stop = md->HSE * md->FX;
+ vm->htotal = md->HT * md->FX;
+ }
+
vm->vblank_start = md->VBS;
vm->vblank_stop = md->VBE;
vm->vsync_start = md->VSS;
-/* $NetBSD: grf_rt.c,v 1.22 1995/02/23 19:14:46 chopps Exp $ */
+/* $NetBSD: grf_rt.c,v 1.23 1996/01/28 19:19:12 chopps Exp $ */
/*
* Copyright (c) 1993 Markus Wild
vm->disp_width = md->MW;
vm->disp_height = md->MH;
vm->depth = md->DEP;
- vm->hblank_start = md->HBS;
- vm->hblank_stop = md->HBE;
- vm->hsync_start = md->HSS;
- vm->hsync_stop = md->HSE;
- vm->htotal = md->HT;
+
+ /*
+ * From observation of the monitor definition table above, I guess that
+ * the horizontal timings are in units of longwords. Hence, I get the
+ * pixels by multiplication with 32 and division by the depth.
+ * The text modes, apparently marked by depth == 4, are even more wierd.
+ * According to a comment above, they are computed from a depth==8 mode
+ * (thats for us: * 32 / 8) by applying another factor of 4 / font width.
+ * Reverse applying the latter formula most of the constants cancel
+ * themselves and we are left with a nice (* font width).
+ * That is, internal timings are in units of longwords for graphics
+ * modes, or in units of characters widths for text modes.
+ * We better don't WRITE modes until this has been real live checked.
+ * - Ignatios Souvatzis
+ */
+
+ if (md->DEP == 4) {
+ vm->hblank_start = md->HBS * 32 / md->DEP;
+ vm->hblank_stop = md->HBE * 32 / md->DEP;
+ vm->hsync_start = md->HSS * 32 / md->DEP;
+ vm->hsync_stop = md->HSE * 32 / md->DEP;
+ vm->htotal = md->HT * 32 / md->DEP;
+ } else {
+ vm->hblank_start = md->HBS * md->FX;
+ vm->hblank_stop = md->HBE * md->FX;
+ vm->hsync_start = md->HSS * md->FX;
+ vm->hsync_stop = md->HSE * md->FX;
+ vm->htotal = md->HT * md->FX;
+ }
vm->vblank_start = md->VBS;
vm->vblank_stop = md->VBE;
vm->vsync_start = md->VSS;