[comp.sys.ibm.pc] The Turbo Pascal 4.0 upgrade fee

bob@imsvax.UUCP (Bob Burch) (11/12/87)

From: Isaac K Rabinovitch, The Portal System (TM)

>I'm about 50% convinced to spring for the Turbo 4.0 upgrade.  (Not
>very willingly--the $40 Borland wants for the upgrade must be close
>to their wholesale price for the product itself!  Essentially, I would
>be buying the same product from them twice.

    You've either got a very short memory or you're very new to the
micro world, Mr. Rabinovitch, so let me tell you how it was just three
short years ago.  I remember paying the Programmer's Connection (lowest
mail-order prices then as now) $400 for a copy of Mark William's C
compiler (list $550 or thereabouts) and watching that compiler take 20
minutes to compile a 2000 line program on a hard-disk system, and over
an hour to compile the same program on a two-floppy system.  That was
the hottest compiler there was;  it ran the fastest and produced the
fastest code.  That was in early 1985, not quite three years ago.
Borland INVENTED fast PC compilers and they invented the idea of selling
them for under $100.  Left to their own devices, MicroSoft, Mark
Williams, Lattice, Ryan McFarland et. al. would continue to sell clunky
compilers at $500 a crack forever.  You had better believe that, if
Borland were to die tomorrow, Quick-C would vanish into thin air and the
world would return to Bill Gates idea of normal and happy.

>Does the upgrade come with an improved manual?  I think we can agree that
>the old manual could serve as a technical writing text on what *not* to
>do.

The new manual is much bigger and better than the old (which I thought
was good);  B Daltons would charge more than the Borland $40 upgrade
price just for any BOOK that size, whether or not any software came with
it.

>Does the upgrade fix the careless little glitches in the compiler?  For
>example, if you READ a number with 3.0, it must be terminated with a
>blank or EOL, whereas all the textbooks tell you that a Pascal number
>ends with any character that isn't part of the number.  (Criswell predicts
>that Borland knows about this glitch but kept it in 4.0 for the sake of
>compatibility.)  Anyone know about this or other glitches?

Seems to be fixed, the read at any rate.  The graphics are astonishing,
the list of standard feature is astonishing and, amongst other things,
TP-4 seems to have reduced the formerly arcane topic of DOS device
drivers to childs play;  you can now write them in turbo.  Dealing with
Borland is the one area in my business life in which I consistently end
up feeling like I've gotten too much for my money... almost makes me
feel guilty.


Ted Holden
HT Enterprises