Check out Grant Acedrex, our featured variant for April, 2024.


[ Help | Earliest Comments | Latest Comments ]
[ List All Subjects of Discussion | Create New Subject of Discussion ]
[ List Earliest Comments Only For Pages | Games | Rated Pages | Rated Games | Subjects of Discussion ]

Comments/Ratings for a Single Item

Earlier Reverse Order Later
[Subject Thread] [Add Response]
Matthew Montchalin wrote on Fri, Jul 22, 2005 06:30 PM UTC:
Say, will an Ultima version of Chess V ever be made available on CDROM? 
Instead of constantly modifying Chess V as time goes on, various versions
could (and therefore should) be fixed in time and space, so a
'performance' or 'competence' rating could be assigned to each
version, and people would have some kind of an idea how well they (or
their programs) are performing against it.  The 'I-depth' variable is
interesting, but there are other ways of defining a program's playing
strength than by defining how many plies deep the search should go. 
Anyway, there is nothing more miserable than having to plod through 'C'
source code; it's totally inscrutable.  (At least well-commented
assembly
language makes sense.)  It would make a great sparring partner for those
of
us who can finally be convinced there's a reason to buy a reasonably
modern Intel machine to run it on.

Greg Strong wrote on Sat, Jul 23, 2005 12:22 AM UTC:
You can always download all back-versions of ChessV from the
sourceforge.net website.  If you, or anyone else, would like a copy on
CD-ROM, please contact me by e-mail.  I'm more than happy to send you a
CD-ROM free of charge (the blank costs about 10 cents, and the stamp 37...
I can handle that.)

Yes, versions of ChessV vary widely in performance, and the program is
really, really in flux.  That is because it is still in it's infancy, and
it has lots and lots of bugs.  I will become far more conscientious about
revision-testing before putting out updates when I get it to a reasonably
bug-free state (which will be version 1.0)  Meanwhile, ChessV is a mixed
bag, and I depend on the kind help of those in the chessvariants community
to playtest and let me know what is better and what is worse.  I have so
far put approximately 1,500 hours of work into this program...  It is safe
to say that when I started, I had absolutely, positively, NO IDEA just how
hard this was going to be.  I totally underestimated the complexity in a
big way.  If I had understood, I never would have even started in the
first place.  I also thought that other open-source programmers in the
sourceforge community would take an interest and offer to help.  So far,
despite the fact that it has been downloaded over 3000 times, not a single
person has offered any coding assistance of any kind.  Until they do, the
progress is going to be slow.  We will creep up to version 1.0, and it
might take another year or two.

Regarding I-Depth, that setting is not intended to determine how strong
the program is; only play-testing will do that.  The I-Depth is there so
that different people running on different hardware can run ChessV and
discuss their results and know that they are compairing apples to apples. 


Regarding C vs. assembly language, you have said something really
rediculous.  Please understand that I have no wish to insult you, only to
make the facts clear to you and anyone else who may be watching this
conversation.  You said that C-code is 'inscrutable' while
well-commented assembly language makes sense.  This is not so much an
opinion as it is a statement that is just factually incorrect.  Assembly
is much older than high-level languages, such as C, but yet the entire
world of professional software developers use high-level languages simply
because they are easier to read and more managable.  If assembly was the
way to go no one would use anything else, whereas less than one-tenth of
one percent of the world's code is written in assembly, and those people
who do do it, do it generally because they are writing device drivers or
the core of an operating system, and thus they have no choice.  Assembly
makes more sense to you simply because you understand it better.  You also
added the qualification 'well-commented.'  There you have something. 
Regardless of language, code is easier to understand the better the
comments and documentation.  Here, also, ChessV is a mixed bag.  There are
some files that are pretty well commented, and others that have almost no
comments at all.  If other programmers were helping me on this project I
would feel more of a need to comment...  Also, I would mention that ChessV
is C++ and not C, C++ being much more logical and easier to understand
(provided you understand the concepts of object-oriented programming,
something which assembly language does not have and cannot emulate.)

Regarding your recent questions on the ChessV thread:  I have not meant to
ignore them.  It is just that you have raised questions that will take some
time to answer.  Also, they are questions that I can't really answer very
well because, considering the hardware you are running, I really don't
know the answers.  We are playing the same game, but taking approaches
that are as radically different as could be imagined.  I really don't
know about trying to do something like Ultima with the hardware
limitations imposed by the Atari ST.  My opinion is that it is flat-out
impossible.  Orthodox Chess programs which ran on the Motorola 68000 were
not very good, and Ultima is a much, much harder problem.  Quite frankly,
I think you could be the best programmer in the whole world, and you would
still fail to make an Ultima program that plays at even the level of a
beginner.  Sorry, I just don't think that it is possible.

Matthew Montchalin wrote on Sat, Jul 23, 2005 07:01 AM UTC:
Greg, I'm interested in something that is immediately executable, no
compilation required, no downloading or leeching required.  Which
'Windows' operating system can it run under?  Or perhaps Linux, or the
Mac?  I don't really mind which, any of them is fine.  At no time have I
represented myself as a C programmer, nor am I able to tell the
difference
between C and C++.  Like you, I prefer the environment I have the most
experience in.  I shun the ones I have no experience in.  I'll send you
an email along those lines, as soon as I get to a Pine interface.  (The
webmail interface here at www.chessvariants.org isn't exactly what I'd
call 'my cup of tea.')

Matthew Montchalin wrote on Sat, Jul 23, 2005 08:18 AM UTC:
Greg, I sent you an email but it bounced back with a 'host unknown'
message.  Are you still at umd.edu?  The mailserver that I use is
run by OregonVOS.net, with over 60 different dialups in the state of
Oregon, but that shouldn't make any difference.  (But practical
experience says it does; direct dialing is almost always superior to the
indirect service of the many mailservers needed to facilitate the
transmission of packets across the country - so if perchance you had a
real BBS (not a website) with a real 'direct dial number' I could go
there, maybe, cut out all the middlemen and send you email directly?)

Roberto Lavieri wrote on Sat, Jul 23, 2005 11:34 AM UTC:
I have tried some code programming in the past, including C++, but,
honestly, I don´t consider myself a very good programmer, and it is the
main reason I have not offered help with Chess V, because more than a help
it may be a new source of tons of bugs, but it is possible I´ll try to open
the code in some moment, to see, but at first I have to install the correct
C++ compiler in my computer, because I don´t own one. I agree with Greg,
Ultima/Baroque can need a lot of computer resources if you want a good
cyber-player, I doubt we can perform it in a Atari ST, all chess programs
in all the old computers I know are good examples of the difficulties
everybody can find for this purpose, being the main factors the running
time and memory resources, but there are others. And for code simplicity,
well, everyone can have its preferences, it may depend on the training
time with a language, but I think high-level languages are much more
user-friendly, and, for me, C++ is not the most comfortable, but I repeat,
it depends on each one. Chess V is a great concept, I´ll try to give a hand
seeing the code, but I doubt I would help too much, perhaps I can add more
to the program with testing and giving hints for evaluations, as I have
did to the present.

Matthew Montchalin wrote on Sat, Jul 23, 2005 06:20 PM UTC:
Roberto, the stock 520 ST with 512K ram is fairly rare to stumble on, as most people by now have already upgraded their systems with an extra 2 megs of RAM. Luckily, I have both an unexpanded (vintage) 520 ST with 512K ram, and another one with a total of 2.5 megs of ram. But I am in the minority, as it appears that most ST users have 4 megs of ram. I'd mentioned that I can get a handful of fully evaluated tiers or plies from any given starting position squeezed into 80 to 90K of ram. That way there aren't any obvious losses of pieces that are 'en prise' two times over, or otherwise pinned. A series of 2, 3, or 4 continuous captures - one after the other - can be entertained, and those are the ones that are first evaluated - all within a 3 or 4 ply search - but a 5th capture is likely to be missed unless it lies in the 'footprint' of a piece that is attempting to maximize its movement capabilities. As for other kinds of computers attempting to emulate the ST, let's put the shoe on the other foot, your typical ST emulator has a full range of 4 megs of ram, or a lot more - usually 8 megs, I think (I've never been able to get one of those emulators to work, myself, as my experience with PC compatibles has led me to believe Intel systems are very flaky, unstable platforms to work on) - it all depends on how the emulator is configured. Unfortunately, suppose you do use a PC to emulate an ST - you miss out on the opportunity of networking real ST's together through the MIDI port. With the ST, you can connect several of your ST computers together by cabling them up through their MIDI ports, and telling the ST systems to poll the port and see how many are there. Something similar to that must be available to the PC, but before you boot up an emulator you should read the documentation to see what specific features are missing from what users expect of real ST systems. The point about booting up a whole bunch of emulators is having the pleasure of conducting a Baroque tournament between several different programs, and seeing how their personality profiles enhance their performance against each other. My interest in getting a copy of ChessV is mostly for benchmarking purposes. That ChessV can play so many different kinds of Chess is amazing, true, but I'd like to employ it as a benchmark so I can tell how well I am programming my own stuff.

Greg Strong wrote on Sat, Jul 23, 2005 10:49 PM UTC:
Matthew:  I have no doubt that you have had lots of problems with Intel
systems.  I would suggest that this is because Microsoft has released a
lot of really, really bug-ridden piece-of-crap operating systems.  This
has finally gotten better, after a long wait... Windows 2000 is
rock-solid, and I hear that Windows XP is too.  Please, please do not
think that Intel hardware sucks just because Microsoft has behaved so
irresponsibly.  The Intel hardware is flawless, at least since the
infamous Pentium floating-point bug of about 10 years ago, and that was a
pretty minor bug at that.

Matthew Montchalin wrote on Sun, Jul 24, 2005 04:03 AM UTC:
I have heard glowing words of praise about Windows XP; my own brother
loves
it, for instance.  And what's this I hear, the next version to come out
is
code-named 'Windows Vista' (?) but I fear that software compatible with
one version won't be compatible with the next.  (That is the main
problem
with Windows XP - I can't use Wordperfect 5.1 for DOS unless I format a
wholly separate partition for a world of 'DOS files' safe and separate
unto itself.)  Windows XP with all the new doo-dads around will be a
wonderful thing to see.  As I said, those little thumbdrives that are
coming out, they're going to be the wave of the future.  I don't know
about their lifespans, but that is rarely a concern to anyone anymore. 
Maybe there will come a day when you can get 'ChessV' preinstalled on a
thumbdrive, and just plug it into your PC, straight out of the package?

8 comments displayed

Earlier Reverse Order Later

Permalink to the exact comments currently displayed.