home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!swrinde!gatech!concert!rutgers!njitgw.njit.edu!hertz.njit.edu!dic5340
- From: dic5340@hertz.njit.edu (David Charlap)
- Newsgroups: comp.os.os2.misc
- Subject: Re: ATI VGA Wonder (Old model, revision 5 chipset) and TRI32New drivers
- Message-ID: <1993Jan11.211102.29499@njitgw.njit.edu>
- Date: 11 Jan 93 21:11:02 GMT
- References: <1993Jan10.230015.27050@njitgw.njit.edu> <C0p336.Jr5@apollo.hp.com>
- Sender: news@njit.edu
- Organization: New Jersey Institute of Technology, Newark, N.J.
- Lines: 37
- Nntp-Posting-Host: hertz.njit.edu
-
- In article <C0p336.Jr5@apollo.hp.com> ced@APOLLO.HP.COM (Carl Davidson) writes:
- >
- >Sorry to be the bearer of bad news, but it appears that these drivers
- >will only run properly on boards with 1024K (1 meg) of memory. If you want
- >call it a *bug* in the drivers, I won't argue.
-
- That's what I thought at first, but it doesn't wash. I made that
- change to SVGADATA.PMI, and then I got every-other byte in black. The
- sort of thing I'd expect if it needed 1MB and only found 512K. And
- the top-bottom still wrapped. The fact is that the 1MB boards get a
- completely different setup, and there is more than just that one byte
- being changed. Second, I don't have an XL, so I can't add more memory.
-
- >Even after you upgrade to 1024K things won't be right until you edit
- >SVGADATA.PMI by hand (see my posting from last Friday). Personally, I think
- >that having SVGA.EXE generate register settings on the fly by reading the board
- >(this is what it appears to do -- I have no specific knowledge of this)
- >is more trouble than it's worth. It's too easy to end up with sub-optimal
- >register values and it's difficult to customize your setup for specific
- >monitors. It would seem more reliable to have SVGA.EXE identify the
- >board type and configuration and then query the user for things like
- >the desired refresh rate and set the appropriate values from an internal
- >table.
-
- I know exactly what VGA register is being messed up. It's
- split-screening the display at row 511, when the register should be
- set yo 1023. But the driver (and maybe OS/2) ignores me when I set
- the register to its proper value in SVGADATA.PMI. Either Joel
- Armengaud is setting the port manually, or Trident isn't setting the
- mode properly. I suspect the former, since the register is set wrong
- in the file Joel's driver uses as well.
-
- --
- |) David Charlap | .signature confiscated by FBI due to
- /|_ dic5340@hertz.njit.edu | an ongoing investigation into the
- ((|,) | source of these .signature virusses
- ~|~
-