STraTOS 1997 April & May/STraTOS 1 - 1997 April & May.iso/CD01/INTERNET/SITES/RAND/ALL96.LZH/t0213/text0010.txt

Text File | 1996-03-07 | 3.5 KB | 77 lines

> The mouse is still unusable because of missing support for left button
> (LMB to run, RMB to fire ;)

I'd prefer to fire with the left button, strafe using the right and leave
the running for shift keys. Of course it'll be very easy to make this
completely configurable.  :-)

> The good news is that everybody can accelerate his DSP very cheap and
> that BM on 2/3 of original size screen is nearly two times faster.

That depends a lot on the WAD. I got a couple yesterday from a friend
of mine (his own work) that most often update at around 3 (three) fps, at
_minimum_ resolution! Full screen was only marginally worse.

The problem with the WADs is that you can see a long way and that there
are lots of sectors due the the 'mountainlike' terrain.
The original DOOM actually gives up and quits if you look in the wrong
direction, complaining about the visibility.

I also got another WAD with most of one floor of the building where I
used to live until half a year ago. My own old room is in there!  :-)
No furniture, unfortunately.  :-)

> > > Well, that's what the DSP is supposed to be doing - but it unfortunately
> > > can't fix the bus too. What a bummer :(
>
> I still think that the bus is not a BM's bottleneck. Have Doug seen my

We'll see what happens when it's run in a less demanding graphics mode.
Has anyone tested it on an RGB monitor or a TV?

> tests ? I'll send it to him directly, for sure. The bottleneck is speed
> of CPU IMHO, and I'd like to see BM on 50 MHz Falcon.

IIRC Doug said recently that most of the hard work is done by the DSP now,
with mostly texture mapping left for the CPU. If the texture mapping code
is as well interleaved as he claims the bus speed might not be too much
of a problem. When we start adding in all the AI and game mechanics, I'm
sure a fast '030 will be very useful, though.

> > Hmm... I wonder how fast Bad Mood will be on the Hawk, if it ever gets released.
> > The Hawk will have an 48MHz 030 and a rather big cache and the option to a
> > new faster DSP with more memory than the orginal DSP (96kb). All all this will

I've been wondering what they mean by that about the faster DSP. I don't
think Motorola makes the 56001 or its relatives much faster than the one
we already have. The 96001/2 is of course another matter, but that's quite
another processor (32 bit floating point).
If the new DSP is not compatible with the old one it's never going to be
used to any significant degree.

By the way, when I tried to check out the current maximum clock speeds
of the Motorola DSPs (didn't find them) I discovered that they have the
entire user manuals available by WWW. If anyone's interested they are at:
  http://design-net.com/dsp/tech_doc/tech_doc.html

> BM hardly can use more memory than on original Falcon - only if it could
> be dynamically allocated.

I don't think I understand what you mean there.

> If the Hawk is new processor, probably coprocessor, DSP, MMU and RAM on
> separate daughter board, then Hawk is Falcon II and uses original Falcon
> like a expensive peripheral and Video out.

Yes, that's one thing I don't like about what I hear of the Hawk.
If they do that much, they could almost as well make a complete computer.
Of course, they've said that they are doing that now...

--
  Chalmers University   | Why are these |  e-mail:   [e-mail]
     of Technology      |  .signatures  |            [e-mail]
                        | so hard to do |  WWW/ftp:  rand.thn.htu.se
   Gothenburg, Sweden   |     well?     |            (MGIFv5, QLem, BAD MOOD)