Xref: utzoo comp.lang.c:7006 comp.sys.ibm.pc:11255
Path: utzoo!mnetor!uunet!lll-winken!lll-lcc!lll-tis!ames!pasteur!ucbvax!cbosgd!mandrill!neoucom!wtm
From: wtm@neoucom.UUCP (Bill Mayhew)
Newsgroups: comp.lang.c,comp.sys.ibm.pc
Subject: Re: Question about turboC 1.5
Message-ID: <985@neoucom.UUCP>
Date: 30 Jan 88 16:37:29 GMT
References: <4549@teddy.UUCP> <1484@codas.att.com> <3317@ihlpf.ATT.COM> <2515@emory.uucp>
Organization: Northeastern Ohio Universities College of Medicine
Lines: 30
Summary: Tubo C 1.5 bgi EGA screen driver does 16-bit writes to EGA registers


I finally got TC 1.5 with the nice apologetic letter for losing my
order down a black hole.  In fact it came UPS next day, showing up
three days after I called complaining, "Where in tarnation is my
order?"  I suggest that anybody that ordered a TC upgrade and has
waited more than about 3 weeks should give Borland a call.  Don't
bother wearing your fingers to nubs trying to get through on the
800 number.  Call the corporate office and ask for "customer
service".  I got very courteous service.  Other than the fact they
lost my order, I am impressed.

Well, down to business.  I am using an EGA card in an AT&T PC6300.
The PC6300 is smitten by a minor bug sends the bytes of the 8086's
16-bit OUT in the opposite order of a PC/AT.  You really can't
fault AT&T, since the PC6300 was designed *before* the de facto
standard was set by PC/AT.

TC 1.5's EGA driver seems to be doing word wide writes to the EGA
registers.  Doing word wide writes speeds up some operations by
making clever use of the ordering of registers on an EGA card.
Unfortunately, the alternate ordering of the output bytes goofs up
operations for the PC6300.  Many of the TC 1.5 graphics operations
do work on PC6300, but a few will cause your program to crash.

The CGA operations work just fine on an EGA board in the PC6300.
The PC6300's indigenous 640*400 video controller also works just
fine.

The TC graphics support is quite a handy and welcome addtion.

--Bill