Re: Alltech Electronics-- full-speed accel?

[Rubywand <rubywand@swbell.net>]

Nathan Mates writes ...
> 
> 
> >> .... DOS: VESA 2.0. Win95, 2D: DirectX. Win95, 3D (slow but compatible
> >> with everything): Direct3D. Win95, 3D (faster) Glide. Professional 3D:
> >> Open GL. It isn't hard to pick, really. Took us, oh, about 30 seconds.
> 
> >    "It isn't hard to pick?!" If you post a statement like that on
> >comp.sys.ibm.pc.hardware.video _please_ do not let a cross-post
> >reference to comp.sys.apple2 show up in the header. (In any case you've
> >managed to omit the current preferred 3D graphicss api. Hint: Check the
> >Diamond and Creative web sites.)
> 
>    It looks as if you're seriously confused about the lines between
> hardware (such as the 3dfx Voodoo (original, Rush, or 2) chipsets) and
> an API. The API is the software layer the app speaks to; it does some
> work and then passes stuff off to the graphics card's drivers.

     Yes; API matches hardware --> new hardware fits API --> new API
matches new hardware --> ... etc. assuming the hardware maker gets their
drivers working for the various PC platforms. Meanwhile, game makers try
to figure out which hardware users will buy and users try to decide
which hardware game makers are going to support. Not so simple.


> ....
> there's a lot of hardware choice in the PC video market, but with well
> done APIs and functional drivers, the game does not actually *care*
> what the heck it's talking to ....

    _Users_ care what the game looks like running on their hardware.
That is what's important. (Actually, maybe Lara Croft does get pretty
annoyed with low-triangle-count hardware!) 


> It just runs faster or slower depending on how much you decided to
> spend.

     Too often, this is not the case. It is easy for a user to spend
lots of money on good quality components and end up with significantly
worse performance than another user gets on a less expensive system.

     

> 
> >> >     No PC user has anything like the hardware and system documentation
> >> >available to, even, GS users.
> 
> >>    You must not be looking very hard at all. Drop by the local decent
> >> bookstore and find a 1000 page tome about the guts of a PC in a hurry. ....
> 
> >     This is happy news! On which pages do I find the schematics with
> >signal descriptions for
> > the FIC PA-2007 motherboard with K6-233, P5 VIA chipset, and 1MB plb
> >cache?
> 
>    All the slots and ports are easily found on the web; there's no way
> in hell any hardware hobbyist could deal with the high frequency
> surface mount chips on the board or find most of them in a local
> electronics store. You *can* build your own ISA boards trivially,
> but anyone wanting to mess with the motherboard chipset on any modern
> computer is in trouble.
>

     For sure, since they cannot get any technical details. (You forgot
to list the page number for the PA-2007 in your 1000-page wonder book.)

 
> > the Diamond FireGLPro (8MB) video board?
> 
>    PCI slot descriptions, OpenGL src and API available on the web.
> For chips with 100,000-5,000,000 transistors, hobbyists can treat
> them as black boxes which work.
>

     It looks hobbyists have no choice. (Page number?)

     Hey! You forgot about the schemos for the FireGL 1000Pro (8MB) and
the Roland LAPC-1! (Page ... never mind.)



> > .... In the absence of extrordinary motivation, most PC users do
> >not have the entry-level knowledge, time, and/or work situation required
> >to even start to become the kind of "power user" you are talking about.
> 
>    Time is flexible. If you *want* to do something, you'll make the
> time. Plenty of "X for dummies" books running around that people *can*
> become educated to be a power user.

     So, then, just wanting, is not good enough. You have to *want* (or
**want**) to become a user who can program his/her computer badly enough
to run some sort of gauntlet. These guys don't want to joint the
priesthood; they just want to program their computers.


> If your problem is with the lack
> of motivation of kids today get a clue, don't blame the
> hardware. That's an attitude problem that's tweaked at the societal
> level.
>

     Society may well be pretty tweaked-up. One contributing factor is
sure to be the powerlessness experienced by so many computer users.
Today's kids have as much motivation to program as any sane person
should need. 

 
> >     80's Apple II+ and IIe owners could easily use "the crud espoused
> >by the big commercial OS developers" (i.e. by Apple, Inc.) to write
> >for-real programs on a par with those available from commercial
> >developers.
> > ....
> 
>    Well, at the time, you could get a few programmers working for a
> few months to get something commercial. Problem is, expectations have
> risen. Videogames now take, on average, about 8-12 months times at
> least 5-10 people, and a budget of about a million $. Is it really the
> hardware's fault that it's become harder for parttime developers to
> match commercial efforts, or is it simply market pressures of a
> discerning audience who've come to expect and demand a well-written,
> graphically appealing game with all the trimmings, etc? [Exceptions
> like tetris do exist, which are pure gameplay over graphics, but most
> people want gameplay *and* graphics]
> 
>    Moore's law: computers double in complexity every 18 months or so.
> That certainly applies to software as well, which has gotten bigger
> and better. Expecting a repeat of 1981 when there wasn't much
> difference between the pros and the amateurs is stupid at best. People
> can still cut their teeth on smaller projects and work up; stop
> expecting them to be near-pro level instantly.
>

     All in all, a fair statement of 'what is'.

     We have, evidently, agreed that, whatever the rationale, a user can
not obtain truly detailed hardware information for his/her particular
machine. We have, also, agreed that it is possible for a highly
motivated PC user to become a programmer and/or "power user"; but, that
the overwhelming majority of users will not progress to this status.
Finally, there is no question but that production of today's
entertainment software generally involves many experts and is expensive.

     You appear to believe that the status quo is what must be. Beyond
this, many of your comments indicate a belief that a user _should_ have
to invest extraordinary time, money, and effort to become someone who
can produce useful, interesting programs.

     My belief is that 'what is' should be changed and can be changed.
It should be changed because it opposes the core motivation for being a
computer user-- the enhancement of personal power.


     As you may recall, in the early days of micro-computing, users did
not have much memory available for their programs, non-volatile storage
was on slow, cumbersome cassette tape, and, some users even entered code
by flipping bit switches. As more powerful machines with better I/O
interfaces, like the Apple II, became available, programming became much
less difficult.

     You mention that computers double in complexity every 18 months.
Certainly, the number of transistors and gates, size of memory and mass
storage, speed, and other hardware parameters of the 'typical PC' do
continue to advance rapidly. 

     Since today's PC is a vastly more powerful, more versatile machine
than the Apple II's of the early 80's, it should, logically, be much
_easier_ to program. So, why not harnass some of the wondrous power of
90's machines to return the power and fun of computing to the user?

  

Rubywand