[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

IRC meeting on Sat, 01.Aug.1998

To: PenguinPlay <penguinplay@sunsite.auc.dk>
Subject: IRC meeting on Sat, 01.Aug.1998
From: Christian Reiniger <warewolf@mayn.de>
Date: Sun, 2 Aug 1998 02:59:40 +0200 (MEST)

Hi,

As you all know (well, sort of) we had a little IRC meeting this weekend.
The time suggested by me was a bit soon (Sorry for that - I actually
didnt think that someone would agree with it ;). As a result only 5 people
were there...

But for all the others - don't worry. Here's a short summary of what was
discussed and I attached the log of the session (I suppose you read the
edited version - It has all the little offtopic parts removed. The raw
version is only for those of you suspecting censorship (and for Adrian ;))

BTW: See also my summary mail I posted before the meeting


Ok, These nice guys were present (unsorted):

Adrian Ratnapala <s369625@student.uq.edu.au>
PPlay Graphics team leader
Nickname: <raka>

Christian Reiniger <warewolf@mayn.de>
PPlay GUI system
Nickname: <warewolf>

Sam Lantiga <slouken@devolution.com>
Author of SDL
Nickname: <slouken>

Nicholas Lee <nj.lee@kiwa.co.nz>
Observer (up to now)
Nickname: <EmptySand>

Mikko Torni <mtorni@freenet.hut.fi>
Network team member
Nickname: <mtorni>



Ok, we discussed several topics:

(1) further IRC meetings
========================

* a smaller one without great announcement

Goals: - Get OS developers/activists together to discuss coordination of 
         projects
       - discuss agenda etc of the Big meeting

We'll have to invite people from other opensource game-related projects
for that.


* The Big one

Goals: - Learn about the needs/wishes of commercial game developers
       - Discuss the future strategy for pushing linux games (etc)

- Big announcement (I forgot slashdot (http://slashdot.org/) in the mail)
- commercial game developers as well as OpenSource guys have to be invited
via mail
- In the invitation for the commercial guys PPlay should be specially
noted, as it means *portable* game development
- The email addrs of commercial developers we know so far are
ddt@crack.com and kevin@westwood.com (slouken, you mentioned some
Activision guy..?)



I suppose we schedule another meeting for next weekend (Sat, 08.Aug.1998,
21:00 GMT should be ok). If we can contact the other OpenSource ppl up to
then this would be the "smaller one" mentioned above, otherwise just
another internal PPlay meeting (but then with more than 5 participants ;).

BTW: I persomally would like to have the Big meeting in about one month.
comments?




(2) PenguinPlay
===============


This are some points on PPlay itself we happened to talk about..

* The Documentation on the homepage has to be updated. The old,
obsolete stuff should be removed (!) and be replaced with up-to-date
docs. We also need more *intro* docs giving an quick *overview* of the
project and its parts, listing the major features and describing the
philosophy behind it

* Where is the Sound team?????

* For some parts it should be again discussed whether to design
them object oriented or procedural (network was mentioned)


* Portability has to be a central issue and the docs have to clearly state
that. Commercial developers are much more likely to use PPlay if they can
use it to develop games for Win32 *and* Playstation *and* Linux without
having to care much about porting.



Well, that should be all..

Cu
	Christian

PS: Did I forget something?

PPS: Now I'm tired too, Adrian ;) Good night!


--

Christian Reiniger <warewolf@mayn.de>


Logfile opened for #pplay at:  31.07.98  23:21:23
*** Users on #pplay: @warewolf
*** Channel #pplay was created on 31.07.98 at 23:20:08
<warewolf> This is a test
Logfile closed at:  31.07.98  23:22:19


Logfile opened for #pplay at:  01.08.98  21:40:51
*** Users on #pplay: warewolf @slouken
*** Channel #pplay was created on 01.08.98 at 21:32:48
<warewolf> slouken: Hi!
<warewolf> slouken: Do you know how I can set my realname/email for whois queries?
*** micith (anxiety@usr02.primenet.com) has joined channel #pplay
*** micith (anxiety@usr02.primenet.com) has left channel #pplay
*** raka (s369625@student.uq.edu.au) has joined channel #pplay
<raka> Good morning all!
<warewolf> raka: Hi, you´re actually telnetting??
<warewolf> raka: BTW: it´s 22:00 here (not very morning-like ;)
<raka> Yeah, I'm telnetting.  As long as I don't do anything else, it works OK.
<raka> I say good morning all the time, it's only a coincidene that its morning here.
<warewolf> raka: <g>
<raka> BTW how do you keep a long in ircII
<warewolf> raka: i´m using tkirc, but I assume ircII also has the /log command, just try /help log
*** EmptySand (nic_lee@5200-ak1-25.ix.net.nz) has joined channel #pplay
<EmptySand> Hi
<raka> about "log". thanks.
<warewolf> Anyone knows if Stephane can join in?
<raka> But no, /log doesn't exits.  Anyone else keeping a log?
<warewolf> raka: I am <checking...> Yep, works fine (up to now ;)
<EmptySand> yep
<raka> warewolf: cool.
<warewolf> Slouken seems to be sleeping - I tried sending him a priv msg, but he hasnt responded yet...
<raka> I'm the one who should be asleep.
<EmptySand> Hehe
<warewolf> raka: Why? it´s morning as you said...
<EmptySand> I just got up
<EmptySand> 8am
<EmptySand> But I see it's earlier their. 8)
<raka> It's 6am here! So don't winge.
<warewolf> raka: Ooops...
<raka> Later, to day is tommorrow.
<EmptySand> 8P
<warewolf> raka: ?-|
<raka> Ain't toroidal geometry fun.  What does 8P mean.
<EmptySand> Doesn't help with people on both side of the world, and trying to find good irc times
<EmptySand> glasses and tougne sticking out
<raka> If we have no Asians, this is a good IRC time.  Maybe it should be an hour later.
<EmptySand> Not during the weekdays.
<raka> I'm only talking weekends.
<EmptySand> Remebering it's Sunday here now. 8)
<warewolf> raka: Well, would be ok for me (I´m regularly awake til 03:00 these days...)
<EmptySand> err, Sunday there as well.
<EmptySand> 6-7 hours back might work as well
<warewolf> EmptySand: Just to get it clear - where´s "there"?
<raka> Sunday I nz is almost always sunday in aussiland.  OK, so no Asians and churchgoers.
<warewolf> EmptySand: 6-7hrs back means "twice as expensive" here :(
<EmptySand> Raka's in  Australia (GMT+10), I'm in New Zealand (GMT+12)
<raka> We have untimed calls here.
<raka> smirk
<EmptySand> No local charges here. 8)
<warewolf> raka: You lucky one... Here it´s about 0.7 USD per 4 minutes - between 21:00 and 05:00 (the cheapest time)
<raka> Everythings expensive in Germany.
<raka> It was so good to get back and be able to afford McDonalds.
<EmptySand> I remeber a (500ml) bottle of coke being priced at 7.50DM in Frankfurt!
<raka> I think slouken is dead.
* EmptySand slaps slouken around a bit with a large trout
<raka> Ahh, but was that an airport price.
<warewolf> EmptySand: Airport? Usually it´s at 1.50 DM
<raka> EmptyStand: I wish I knew my Python.
<EmptySand> Actually that was a hotel price, but still it was the worse I'd seen
<raka> I mean airports have captive markets, and thinks are expensive.
<warewolf> raka: The language or the snake Python?
<raka> Oh, hotels are bad too.
<raka> Comedy show, language wan named after it.
<raka> (mean was, not wan).
<warewolf> IMHO we simply should start, it´s already 18 mins over time...
<EmptySand> Sure
<raka> OK.  If no one else turns up, we can do this again.
<raka> What to talk about.
<warewolf> I mailed a summary of the BigMeeting things ´bout 10 hrs ago - everyone got it?
<EmptySand> Yep
<raka> And me.
<warewolf> Ok, any comments/enhancements/insults ?
<raka> Just, finding the message. I'll have some insults for you soon.
<warewolf> raka: Good to know. Thx ;+)
<EmptySand> Put a notice on slashdot maybe as well
<raka> Down with Netscape Mail!
<warewolf> EmptySand: Thx. How could I forget that!
<EmptySand> That should gather a lot of interest in it self
<warewolf> EmptySand: Yep, but we have to make sure that it sounds serious and professional
<raka> I'll be dead for a while, must get mail.
<EmptySand> Of course
<raka> Slashdot!  Serious and proffesional!
<warewolf> raka sure loves his config ;)
<slouken> hello!
* EmptySand grins
<EmptySand> Hi
<raka> He's ALIVE!
<slouken> *grin*
<warewolf> raka: not specially for /. , but ppl have to recognize that it´s not Yet Another Little Try
<slouken> Did I miss anything important?
<warewolf> slouken: Yep, we already agreed on high prices in Germany ;)
<raka> slouken: We are discussing a /. announcement.
<raka> BTW, my config fine.  My ISP isn't.
<warewolf> BTW: who will be able to monitor the discussion on /. after the announcement?
<warewolf> slouken: BTW: Did´y get my summary mail?
<raka> whoever's no the net at the time.
<slouken> yes ware
<EmptySand> It's pretty easy to monitor /.  But hard to control.
<raka> That's the point of /.
<warewolf> EmptySand: Not if you have to pay German prices ;)
<raka> It'll be a good way to gauge public opinion.
<EmptySand> Sure.  I'm pretty much next to a cheap point of connect all the time
<warewolf> My point was that one of us should be there to answer questions and prevent misunderstandings...
<raka> True.  Misunderstanding on /. BAD!
<slouken> Okay, I'm here for real
<raka> slouken: nice to know that.
<slouken> Okay, who all is here?  (by name?)
<raka> only Nicholas,Chritian Me and You.
<EmptySand> Nicholas Lee
<slouken> Sam Lantinga
<warewolf> another thing - anyone knows how to reach commercial game Devs best (mailing service@.. will be silly)
<warewolf> Christian Reiniger
<raka> Me is Adrian
<raka> Ratnapala
<warewolf> raka: <g>
<slouken> Okay, what do we do?
<slouken> <-- Author of SDL
<raka> What does <g> mean?  You keep saying it.
<EmptySand> General you have to start their, unless you can get an email of someone
<warewolf> <g> == <grin>
<raka> I do the 2d pplay library.
<raka> warewolf: ahh
<raka> What is EmptySands interest.  You aren't noisy on the list.
<EmptySand> I've been an observer for a long time
<warewolf> mailing crack.com guys should be easy (there was a .plan of ddt on /. some days ago) and someone posted a message about some id guys recently (AFAIK including email)
<EmptySand> I'm working on a UO server project with some one at the moment
<slouken> enptysand: cool
<raka> id, is turning bad.  The're now developing on NT, and arn
<raka> arnet liking UNIX ports.
<raka> UO?
<warewolf> Oh, yes, a westwoood developer posted something on /. last week (?), I´ll have to look for it.
<slouken> Okay, so what exactly are we planning here, and what makes it "Yet another little try" ?
<EmptySand> Ultima Online
<slouken> Er, NOT yet another little try
<raka> EmptySand: Thanks.
<warewolf> slouken: We´re discussing what to do at/for the BIG meeting
<raka> The problem could be that its to BIG a try.
<slouken> Personally, I'm looking for two things:
<slouken> One, feedback from commercial devlopers -- what do they need?
<slouken> Two, how does SDL fit into PPlay?
<slouken> Three What game development resources are available for UNIX?
<slouken> er, three things. :)
<EmptySand> That /. Westwood URL: http://penguin.lvcm.com/westwood.html
<raka> I'm also thinking about number 2.  Basically I'm currenly programming straing on top of GGI, and I think thats the way it'll be on UNIX.
<raka> SDL could be important on other sytems.
<warewolf> slouken: one: what preconditions do they need to write things for Linux
<warewolf> The crack.com guy is ddt@crack.com
<raka> Also PPlay is not monolithinc, people who don't want the overhead of pg2d could use SDL, but keep the rest of PPlay.  (not that I intend to bloat pg2d).
<warewolf> raka: libggi works nicely with fbcon and fbcon is pretty common, isn´t it?
<slouken> fbcon isn't that common in my experience
<raka> One think about you list of people for TheHugeEvent, one lot is GGI/Mesa/other freeware developers, another is commercial developers.  Two seperate issues.
<warewolf> the westwood guy is kevin@westwood.com
<slouken> Right.  So, what exactly are we getting together for?
<raka> I mean I'm writing straight on top of libggi, not SDL on any similar abstraction layer, libggi is enough.
<slouken> libggi is an abstraction layer
<slouken> Meeting so everyone can meet everyone?
<raka> slouken: exactly, so I don't think I need another one, unless it makes win32 ports easier/better.
<warewolf> slouken: (1) getting all opensource ppl together so that the projects can be easier coordinated
<slouken> If you write on SDL, you automatically get Win32 and Linux and BeOS
<slouken> Right.  Are we talking about merging projects?
<warewolf> slouken: (2) talk to the commercial guys to see what they need
<slouken> Maybe we should have two meetings?  One for commercial folks and one for opensource folks?
<warewolf> slouken: (3) discuss on the future strategy
<warewolf> slouken: merging: on some projects this would be good for sure
<EmptySand> slouken: depends on how much interest we can generate in the commerical people.
<raka> Dont merge with US, we are to slow!
<raka> :)
<slouken> raka:  I noticed that... so has everyone else :-/
<warewolf> slouken: 2 metings: te meeting for the commercials will require the OS guys to be present, too (at least the programmers)
<slouken> True warewolf
<raka> BTW: I didn't read the SDL  docs properly, do you have primitves for line drawing etc.
<slouken> raka: Nope.. it's open for people to use their own specially optimized routines
<warewolf> slouken: So no hardware accel for it?
<slouken> raka:  The screenlib example library has some drawing routines
<raka> I thought so, the things is acceleration.
<slouken> right
<raka> We need accel.
<slouken> Hmmm
<warewolf> raka: shouldn´t be too difficult for win, but what about beos?
<raka> beos slouken?  Any ideas.
<raka> Also SDL could be a nice way to get going on new platforms, w/o accel.
*** mtorni (mtorni@vaasa-dk5260-1-182.kolumbus.fi) has joined channel #pplay
<raka> Hi mtorni
<mtorni> is it over already?
<slouken> Dunno... BeOS has some acceleration, but I hadn't looked at it since SDL just provides framebuffer
<mtorni> hope not, I had to leave a company of two nice girls for this :)
<raka> No, we haven't done much yet.
<slouken> How necessary is accelerated line drawing?
<slouken> How many games really use it?
<warewolf> mtorni: <grin>
<raka> What about accelerated blitting?
<slouken> SDL supports accelerated blitting
<warewolf> slouken: IMHO accel. blitting is the most important one...
<slouken> mtorni, get back to the girls. ;-)
<slouken> warewolf:  That's what I thought. :)
<mtorni> slouken: too bad I cannot return anymore, I walked 30minutes home
<raka> slouken: how excatly.  Any architectural reason why it cant do other accelerated ops?
<slouken> Just simplicity.  Accelerated drawing ops really complicates the API
<mtorni> besides, girls are the main thing that keep me from hacking 24hrs a day
<slouken> and the code...
<warewolf> mtorni: hacking what?
<slouken> mtorni, they're usually a good thing.  (A girl is puttering aroudn here Woohoo)
<slouken> anyway, sorry. :)
<warewolf> mtorni, slouken: Hey, and what about me???
<raka> API, yes, but I thought code complexity would be linear with number of ops.)
<mtorni> they are a bad thing. These two especially! After seeing them my life as miserable as hell for two days and then I think about suicide for another two :(
<slouken> What about you warewolf?
<mtorni> hacking penguinplay and an example game for penguinnet
<raka> what exactly with this warewolf-mtorni conversation about?
<warewolf> slouken: Girls!! Ever thought of sending one to me?? ;+)
<mtorni> and hacking libnixon (to replace all configuration related things in all unix-programs:)
<mtorni> warewolf: I can send you a picture about these two... they're a bit young, tho... just 16
<raka> I think warewolf meant the real thing.
<warewolf> mtorni: Thx, try warewolf@mayn.de ;)
<slouken> raka:  The code complexity doesn't increase linearly when you support different platforms -- you have to write general unaccelerated functions, and then accelerated functions for every single platform that supports acceleration.  It was too much trouble..
<mtorni> warewolf: the picture is quite... well, let's just say it's from an angle that shows maybe a little too much
<slouken> for drawing, but not too much trouble for blitting
<raka> slouken: Too much trouble for one person, yes.
<mtorni> Hmm, the girls probably don't like this, better not post a non-edited version of this on the website :)
<warewolf> mtorni: Ok, well ok :( (we better stop this thread..)
<slouken> The thing is, there are quite a few optimized graphics libraries that could easily be ported to SDL
<slouken> If accelerated drawing is really important in games, sure.  But my impression is that it's not.
<slouken> Please correct me if I'm wrong..
<raka> You see my idea is that if we don't want to be "just another little try" we need to do everything, like all the accell.
<EmptySand> raka: But also provide the hooks so people can improve our optimasation easily
<warewolf> slouken: line gfx was really important some time ago, but I don´t think it´s important anymore (perhaps for simulating conventional GUIs)
<raka> Hooks a million.
<slouken> warewolf:  GUIs definitely need drawing acceleration, but many game GUIs are image based
*** mtorni has signed off (User abort with 5 Ctrl-C's)
*** mtorni (mtorni@vaasa-dk5260-1-182.kolumbus.fi) has joined channel #pplay
<warewolf> EmptySand: optimizing should be easy - the source is available and AFAIK most funcs can be overridden
<slouken> raka:  I think the meetings are more about coordination and communicating with commercial developers than about promoting us, necessarily
<warewolf> slouken: That´s why I wrote "simulating traditional ones" ;)
<slouken> ware: got it
<raka> slouken: Agreed.
<warewolf> slouken: Yep, of course
<raka> BTW: Why SDL is good: Nice, platform independent API, low level, doesn't interfere with me.
<slouken> So, we don't need to promote a "try" really, right?
<raka> Why SDL is bad: Yet another layer of software silt.
<slouken> BTW: Why SDL is bad:  low level
<raka> Re:: "promomoting try":::: but we need to show them we are serious.
<raka> But you said you made it low level on purpose.
<slouken> raka:  Yep, that's definitely true.  Hardcore game developers will always write natively for each interface.  SDL is just a nice layer for porters and medium-core game developers. :)
<slouken> raka: yep
<raka> That's whats worrying me, if hardcore develpers will write natively, so should we.  OUr job is to do the hardcore stuff for the other guys.
<slouken> Serious about what?  Serious about meeting the needs of commercial developers?
<slouken> raka: right.
<raka> Yes, serious about that.  And sersiously able to do it.
<slouken> raka:  Then why use libggi?  Use KGI at the very least
<warewolf> slouken: serious about handling this stuff, serious about not just trying to do another hack ala svgalib
<slouken> PengiunPlay isn't a great testimony to being able to provide a product for commercial developers.
<raka> KGI bad, interface depends on each board.
<warewolf> raka: libggi shouldn´t pose much overhead, but in turn saves us from reinventing the wheel
<slouken> The reason why I did SDL, was because most commercial developers already have code libraries built for their games
<raka> Only because we haven't provided a product AT ALL.
<slouken> They won't (necessarily) want to use another high level API, but want to leverage code they wrote and use.
<slouken> Hence, Simple DirectMedia Layer
<warewolf> slouken: And what if our code will be better? ;)
<slouken> So anyway, PenguinPlay is great for starting developers, but what about existing game houses?
<slouken> warewolf:   I'll gladly use it. :)
<raka> I agree, existing houses would like something like SDL better.
<warewolf> slouken: As PPlay is OpenSource, I´m sure it will develop faster than any of these private libs...
<raka> warewolf: Not that its developing too fast now.
<mtorni> slouken: game development takes a year or two, and IF there's a bit high level API that saves works and works in win95/98 they'll use it if it's useful
<slouken> Right, it just needs to catch on.
<slouken> Right.
<slouken> That's very true.
<EmptySand> IMO It needs a solid source base with some functionality before that
<warewolf> raka: yes. true. I hope we´ll get more ppl after the announcement (and I seriously hope that work will be faster as soon as the design is mostly finished)
<raka> ditto EmtySand.
<mtorni> penguinnet should be a good starting point; every game today wants to have "INTERNET" on its box
<mtorni> and doom took several versions to get even decent netplay
<raka> BTW: will penguinnet relate to things like KALI at all?
<warewolf> mtorni: BTW: You still have to do the network setup /IP addresses etc) by hand with PenguinNet, don´t you=
<slouken> Here's the nice thing about SDL:  It's low level enough that it doesn't get in the way of developers, and is useful enough that you can write games directly on top of it in a single source file (See SDL-demos.tar.gz)
<mtorni> raka: kali just masks ipx
<slouken> HOWEVER
<slouken> PenguinPlay is a great API, for big games
<EmptySand> I think the main thing might be some intergation between the various parts of the library
<mtorni> warewolf: that's what the dialers do
<slouken> It just needs some functionality. :)
<warewolf> mtorni: Erm, yes, ok.
<raka> EmtpySand: I think we need yet another integration doco.
<warewolf> raka: Yep, we also need to decide on some "official" intro to PPlay for the Homepage
<EmptySand> probably.  I think the low level stuff needs to be nutted out.  (Like this event layer)
<raka> OK, I think we agree on the relative merits of PPlay and SDL.  Next question, how, if at all should PPlay use SDL.
<mtorni> pnet started out with a decent specification. too bad someone did c++ code that was "too much" c++ for me. (so I had to drop out)
<EmptySand> The better and faster that is designed the easier the intergation and functionality
<slouken> (sorry, I'm a fan of SDL) :)
<raka> EmptySand: You voluneering for the event layer!
<EmptySand> 8)
<warewolf> raka: depending on the progress of GGI. if GGI ports to Win32 etc become available soon...
<slouken> So, the meetings are just for people to meet and collect URLs? or is it for discussing integration of projects, or what?
<EmptySand> No a asked what was needed. 8P
<slouken> GGI will change dramatically when it ports to DirectX (known from experience)
<EmptySand> err s/a/I/
<raka> Yeah, I'm thinking GGI==prefered option, SDL or similar equals Lite option where GGI doesn't exist.
<warewolf> slouken: Yep, IMHO most projects just dont know enough about each other
<raka> That way we only have two ports to worry about.
<EmptySand> slouken: All of the above?
<slouken> warewolf: cool
<raka> It would be nice if we had sound people here.  Ar they dead?
<slouken> raka:  GGI will be on win32 by the time PenguinPlay is ready
<mtorni> About the slow progres... I think we would be developing much faster if only some basic c++ stuff was used.
<slouken> raka:  I will be starting a port of SDL to MacOS, but no ETA at the moment
<warewolf> slouken: Don´t be too pessimistic about our devel speed ;)
<raka> slouken: WHat about BeOS?  Playstation (not that SDL is their eitehr).
<raka> werewolf: I think we will accelarate soon, but GGI will still beat us to win32.
<EmptySand> mtorni: Like what?
<mtorni> as I said, too much c++ness was the thing that threw me out of pnet (and I had already implemented that kind of a game)
<slouken> GGI might get ported to BeOS (dunno)  It would be nice to have a GGI spokesperson here. :)
<warewolf> mtorni: Dont think so. For me developing the goos design takes most time, whether done in C or C++
<mtorni> EmptySand: The class stuff only, and leave out the hairier bits. I can do only basic C++ stuff.
<raka> Yes, we want a GGI spokesperson, but we don't have one.
<raka> untill recently gcc could only do basic C++ stuff :)
<mtorni> Most people know a LOT more C than C++
<slouken> PenguinPlay seems like a huge API... is there any way to nest it and get core functionality that people can start using, and then add the rest later?
<EmptySand> raka: True. ecgs is pretty good now
<mtorni> raka: yep, but the non-basic stuff has made me drop out of a few projects (berlin)
<slouken> Yeah, that's a good point.  C is much more commonly used, and can be called from C++ (not vise-versa)
<warewolf> mtorni: Yes, that´s a major point. It´s also easier to write language wrappers for C apis than for C++ ones (look at Gtk)
<raka> Yes.  I (and I think everyone else). Wants it to be as moduar as possible.
<raka> eg. PPlay should not rely on pg2d, nor vice versa.
<EmptySand> rake: definitely
<warewolf> mtorni: But a good OO design simply has too much advantages for the higher level apis
<mtorni> warewolf: oh come on, network stuff is better left unclassified!
<mtorni> warewolf: you're sending packets anyway, not objects
<EmptySand> Good OO design and good C are very different beasts
<raka> RE "good OO": which is why pg2d is C++ only (not that its good).
<mtorni> and even C can do it OO style; just look at gtk :)
<raka> But I agree, network stuff should be C compatible.
<slouken> How about C core with C++ wrappers?
<warewolf> mtorni: I dont know abot the network stuff, but doing a GUI lib in C is not very nice...
<EmptySand> mtorni:  But why waste the effort in C, trying to be C++?
<slouken> The SDL library is core C, but I have some nifty C++ example classes on top of it for the Maelstrom port
<mtorni> raka: I've tried converting my netgame I mentioned earlier to C++, but the basic network layer is hard to be thinked object-orientedly
<mtorni> EmptySand: that was just a joke...
<warewolf> slouken: We use C++ because we want the benefits of OO design - writing it in C and doing a C++ wrapper destroys that
<mtorni> warewolf: yes, I agree on that.
<slouken> I'm just talking about core functionality.
<EmptySand> mtorni:  I agree the the current ppnet layer isn't abstract enough. I looked at it for a mud, and it dodn't seem to work well
<raka> So mthen don't do object-orientdly, do some badly wrapped API, or don't wrap at all.
<slouken> Um.... What exactly is the current architecture of PPlay?
<mtorni> warewolf: but instead of doing everything OO or the other way, maybe a hybrid would be the best everywhere
<EmptySand> slouken: Core C++ (member functions) is good C
<warewolf> slouken: Of course the low level stuff should be plain C (just look at OpenGL)
<mtorni> warewolf: some things are just meant for OO, others aren't
<mtorni> c++ makes it easy to do non-oo stuff anyway
<slouken> About those meetings, are we settled?
<raka> slouken: We have Layer P(rocedural), tools for doing stuff protably.  ALl the basic functinality.  This stuff is not necassariy written by use.
<raka> Written bu us I mean.
<warewolf> slouken: Did we agree on what meetings to hold yet?
<raka> Then We have Layer O, (Object).  Can't talk, more we have a decument somewhere.
<slouken> Dunno, got wrapped up in SDL/PenguinPlay talk. :)
<mtorni> Isn't PPlay supposed to ease things? bsd sockets is as easy as it gets on the low-level after setting the connection up.
<raka> Layer O is meant to ease things.
<raka> Layer P makes things possible.
<warewolf> Meetings: IMHO we should stay with the single big meeting with OpenSource ppl _and_ commercials
<EmptySand> ww: yep
<raka> warewolf: OK, why?
<slouken> Then the commercials will just see a bunch of opensource people quibbling.
<slouken> That's not what we want. :)
<mtorni> warewolf: I don't know what you decided, but we need one meeting (where more people show up:) where we prepare some material for the big meeting
<warewolf> mtorni: Agreed
<slouken> What material?
<raka> This was meant to be that meeting.  We didn't have enough notice.
<slouken> Do we present to commercials, or do they communicate to us?
<warewolf> raka: cause most of the OS ppl need to talk to the commercials...
<mtorni> slouken: like what are we going to ask them, what to tell them (no one likes to wait on an idle channel :)
<slouken> Or is it just:  Hey everybody, here's what's available!
<mtorni> slouken: we have to show what our goals are, and then have a mutual QA session
<warewolf> slouken: No, no presentation. A discussion about what´s needed and how we can archieve this
<slouken> I'd like to know  "If you wanted to develop on Linux, what would you need?"  Customer base, most likely.
* mtorni thinks this meeting was on a waaay too short notice
* slouken agrees
<raka> We can list our goal/plans in writing, before talking to them.
* EmptySand thinks that it's a good start though
<slouken> Commercial developers need guaranteed customers.  They'll do anything for percieved customers
<warewolf> mtorni: Ok, I´m guilty. But I wanted to have some discussion about it ASAP (cause email is so damn slow for discussion)
<EmptySand> slouken: that crack.dot issue last week is an example that their is a linux games base
<mtorni> raka: yep, as we don't have much yet, we need to tell them what exactly we are trying to achieve, and then they can point us the weaknesses, which part needs more/less work
<raka> We can't create a user base, we can only help on technical fornts.
<warewolf> slouken: there are quite many petitions showing up recently
<slouken> Maybe we should meet and decide what we are trying to achieve first?
<raka> THi crack.com thing.  I haven't heard anything after reading they were in trouble.
<warewolf> raka: Assume 7 million linux users, 10% of them gamers (?). Most of them would buy good stuff
<raka> slouken::: Shouldn't we do that now.
<slouken> raka:  Yeah, what's up with that?  The last thing Linux needs is a game company down the tubes.
<mtorni> raka: 100 users are 100 users more if they don't need to do extra work; that is our api works with win32 & other OSs
<slouken> raka:  Sure, but we need to do it with more people... mesa, GGI, etc
<raka> But crack.com is in trouble anyway.
<EmptySand> Actually I think since alot of hackers tend to be gamers of one sort it might be higher
<warewolf> raka: Someone gave them financing until Golgotha gets out
<slouken> mtorni:  Yeah, but that's if they start off with the cross-platform (PPlay) API... what about the current rash of games?
<raka> warewolf:: That's good news.  I think I'l buy GOlogtha.
<slouken> For the record, a big thing SDL doesn't support is 3D
<EmptySand> crack.dot got a good response.  Particular from linux people
<raka> mtorni:::: Thats what things like SDL are fore.
<mtorni> look guys, if you were a company, and you'd be given an opportunity to get a few hundred customers more with no investments, wouldn't that be a good deal?
<raka> slouken:: Thats why I like SCL.
<warewolf> mtorni: If we support playstation, too, PPlay will be very valuable for them. And doing the Linux port with PPlay should be very easy
<EmptySand> mtorni: need to be a few thousand
<slouken> or more
<raka> SCL is a bit too hight level for my liking, but it has 3D and Playstation support.
<mtorni> oh come on, pplay just doesn't bring more customers, it is some work already done.
<slouken> raka:  I'll have to check out SCL
<mtorni> they've to use some libs anyway; either they write it themselves or use an existing one and save money
<warewolf> mtorni: PPlay means supporting multiple platforms without more work -> more customers
<slouken> It's not NO investment, it's quite a bit of investment in a moving target -- not very good for serious game companies
<mtorni> warewolf: I know that, but tell it to others :)
<raka> slouken, sorry, you can't its closed source (for now). I'll get you the email of the person to contact, or at least send a detailed description to you.
<warewolf> mtorni: We´ll have to do that on the homepage and at the meeting - it´s an important point
<slouken> raka: thx
<mtorni> slouken: yep, but they don't have to you it immediately. Let's just tell them we're working on it, and we'll inform them when there's a stable version.
<raka> Sorry I think I got lost, what is "it"?
<warewolf> raka: PPlay
<mtorni> s/you it/use it/
<raka> Oh.
<slouken> You're assuming PPlay is their API of choice from the beginning.  I'm talking about the big players.. Activision, Sierra, Blizzard, ID, they all have existing tools, code, etc.  Why train their developers on a new API that's not out yet?  They need it now.
<mtorni> But people, just completing pplay isn't enough;
<warewolf> slouken: about chatting OS ppl on the big meeting - we just have to write the announcement appropriately IMHO
<raka> We don't want them to port now, we want to ask people what they want THEN.
<mtorni> we need to do cross-compiling from win32 to linux and other platforms easier than already done.
<raka> I think most would find cross compilng from   linux to win32 more fun.  Or is that just religious fervour.
<slouken> We can't sell people on dreams.  However, we can ask "If you wanted to port/develop for Linux, would you use X API?"
<mtorni> many companies probably don't want to install linux just to compile on it.
<mtorni> raka: wishful thinking
<slouken> raka:  Cross-compiling is fine, but debugging is hell.
<raka> slouken:  Agreed.  We can also ask what do you want in X API.
<warewolf> raka: Yes, that´s important
<slouken> Yep.
<raka> So were agreed.
<slouken> I think so
<mtorni> maybe we should prepare instructions for installing redhat/debian/someother so it's ready to compile
<slouken> So, we have it... exchanging information on what's available now, and asking what commercial developers need in the future.. is that right?
<EmptySand> mtorni:  Let's worry about finishing the design, and some code first.  packaging is easy
<raka> we need  something to package first.
<raka> Shall I raise the question again.  One or two meetings?
<warewolf> mtorni: Someone posted a mail from some develuper, worrying about the specialties of programming under linux. Anyone knows whether the LPG is still maintained=?
<mtorni> LPG?
<slouken> IMHO, about GGI... if you want a cross-platform APi, use SDL.  If you want acceleration and want to do the hardcore work, write for KGI, Xlib, GDI, and DirectDraw
<warewolf> mtorni: Linux Programming Guide (of the Linux Documentation Project (LDP))
<raka> slouken::: I'm leaning that way too.
<mtorni> slouken: nono, no writing for KGI, now THAT'S a moving target
<slouken> raka:  Dunno. do we need to plan anything for the commercial meeting then?
<slouken> mtorni:  *grin*
<warewolf> raka: Perhaps a small meeting to look what projects can/should be merged, but without big announcement
<raka> KGI is supposed to have a different interface for EACH CARD!  Writing for KGI is not allowed.
<slouken> So, we have it... exchanging information on what's available now, and asking what commercial developers need in the future.. is that right?
<slouken> Will people be willing to merge projects?
<mtorni> slouken: game companies would surely like some evidence that 3d card makers will give us (or ggi people) specifications
<raka> slouken: I think we need to write some goals/ideas docos firsts, and give them to the commercials before meeting them.  THat way they'll bother turning up.
<mtorni> raka: right.
<slouken> raka:  PPlay has too many specs and documents. :)
<warewolf> slouken: don´t know (we don´t have to sav that this straight forward ;) The important thing is that these projects have to know of each other
<mtorni> now, who will collect the goals -ideas from the mailing list?
<slouken> What goals/ideas docos can we right?
<slouken> er, write?
<raka> slouken: Agreed, I want docs that'll let us delete old ones.
<slouken> Goals of PPlay?
<slouken> Ideas of what?  I don't think they want to see spec sheets.
<raka> Well, goals are kind of on the webpage I supose...
<warewolf> raka: Yep, there still is too much old stuff on the homepage, spreading uncertainty
<slouken> I'm thinking about more of a brainfest then "Goals and Ideas for X API"
<mtorni> slouken: yep, that we're trying to get some work off their backs (for free) and we'll bring more customers, and ease porting, :)
<warewolf> mtorni: Excellent formulation ;)
<raka> No, but there is a big difference between saying "we'll do a cross platform game API" and "we'll do a cross platform game API of this size and shape")
<warewolf> raka: We have to provide an impression of what the API will contain and roughly how it will be done (features, philosophy etc)
<slouken> Why do we have to say that?  Why can't we say...  "Here's what we're doing.  How can we make it more useful to you?"
<raka> slouken: You know I think we are agreeing with echother very loudly.
<mtorni> the webpages should be ridden of all old stuff. And a few short pages with all the current stuff is a LOT better than a lot old and unusable stuff in a horrible mess
<slouken> We can't promise a cross-platform API.
<slouken> That's ridiculous.
<raka> Howso?
<mtorni> slouken: WE MUST! No company will do a linux-only game.
<slouken> Well, that's like promising the moon.  How can we guarantee something?
<raka> After all isn't that what SDL is?
<mtorni> slouken: if we don't succeed, they'll never hear from us again :)
<raka> I suspect you mean something more sublte slouken, so speak up.
<slouken> Rather, shouldn't we say... "Here's what we have"  Will you use it?  If not, why?
<warewolf> slouken: SDL shows how to make low level stuff cross-pülatform, and High level is simple
<raka> "here's what we _want_to_ have"
<mtorni> slouken: because it might end up being unusable for them.
<slouken> And, if so, we need to know that.
<mtorni> slouken: and we have nothing, so... "We have nothing. Will you use it?"
<slouken> *laugh*
<slouken> Then let them design the spec
<raka> They don't want to do the work.
<mtorni> slouken: they wouldn't do that.
<raka> Besides _they_ compete against eachother.
<mtorni> slouken: we need to show that we do some of the hardest parts, and do it right.
<warewolf> slouken: why? if they can tell us "hat´s good" r "that#s bad" and "this would be nice" it´s ok, isnt it?
<slouken> Sure they would be willing to do the work.  They do the work for Win32.  They need promises of return, and a reasonable difficulty of doing it.
<slouken> Yep
<raka> See!  I knew we were agreeing!
<mtorni> like "our net API solves most of the problems with multiplayer networking, and in addition it has these features for free, and it doesn't even affect performance, besides there already are a lot of servers for this, so you don't need to put money out for those, either!"
<slouken> Yeah, that's fine. :)
<raka> Slouken, we are not taling about writing the whole spec.  We are talking about short docos.
<warewolf> The overview
<slouken> Um, I know the guy who does the networking for Activision -- they have their own library, tested and written
<mtorni> maybe we could ask them what their current libs do
<slouken> What do we say in the overview?
<slouken> Why would they switch -- rewriting all their code?
<raka> Good for Activision.
<mtorni> slouken: but they need their own special servers; $$$
<slouken> true
<mtorni> slouken: they WON'T switch, but they'll use pplay in their future games
<warewolf> slouken: Goals of PPlay (portability etc), main features of the APIs, structure of the projects etc
<slouken> AAAAAHHH. :)
<raka> They don't switch.  We make sure the rest of PPLay gives them what they want if they need it without ruling out their existing net code.
<mtorni> slouken: this is a long term project, people won't switch overnight, and even the implementing is guaranteed to take over a year.
<slouken> But in order for them to use PPlay in future games... they need to have something now... right?  We're talking about a year or three for PPlay to fruit, then two years for the game development...
<mtorni> For me it was a problem finding a kali server nearby. The nearest one was in Finland, but it had no users. So I had to use an usa server, which amounted to a lot of lag.
<warewolf> slouken: I think it´s about 2 years until PPlay 1.0 (or so), less if we get more developers
<raka> But all we are asking _from_ them is a few hours of time.
<slouken> By then hardware will be different, and MS will be on the leading edge, no?
<mtorni> there were other servers around, but they didn't support kali or descent or whatnot
<slouken> I'm just playing devil's advocate, because I think these are important if we talk to commercial game houses.
<mtorni> warewolf: which amounts to; speed is not so much of an issue, we need to implement even the slow algorithms (with current hardware)
<slouken> I could be totally off the mark however. :)
<warewolf> slouken: They can use SDL and/or GGI now. And I still hope that PPlay devel will accelerate soon. Really soon
<warewolf> mtorni: what exactly do you mean?
<mtorni> slouken: two years for the game development -- MINUS the time pplay saves ;)
<raka> Remember, we aren't asking them to port now.  We are asking them for what they want.
<EmptySand> If pplay became the directX of linux, that would be what the commerical developer would require
<mtorni> warewolf: do you mean my last line or the one before it?
<mtorni> EmptySand: and it became the directx for win98, that would be a hit.
<warewolf> mtorni: the "slow algos" stuff
<raka> And the DirectX for PlayStation, we win.
<mtorni> EmptySand: if we don't do win98 at 1.0, they'll never come.
<raka> (if that happens)
<EmptySand> mtorni:  Lets be realistic.  M$ is never going to release core APIs
<slouken> How can PPlay supplant DirectX for Windows?
<raka> PPlay will have to run on DirectX.  I guess slouken knows most about this stuff.
<warewolf> raka: Well, if we´re accepted for one platform, we´ll have all
<mtorni> mtorni: I mean that we have to implement the most advanced 3D algorithms available (and invent some more:), no matter how impossible they are with current hardware.
<EmptySand> Speed is the big issue.  Making it easy to port to linux, with a (quick) directX wrapper and they would have a lot
<warewolf> mtorni: Yep. Be prepared for the future
<mtorni> EmptySand: I just meant that we need to be popular for win98 development...
<slouken> Have you guys ever checked out the DirectXDev mailing list?
<EmptySand> nope
<slouken> There are lots of sharp game developers on there.
<warewolf> BTW: How does OpenGl cope with modern 3D features (Bumpmapping, miltiTextures, ...)
<raka> slouken: now but I feel we should.  BUt then I guess I should read some DirectX docs.
<mtorni> EmptySand: Now, are we trying to do a quick hack for portability, or a long-term useful projecT?
<mtorni> slouken: of course not :)
<EmptySand> mtorni: It's pretty difficult to 'fight' ms on their own ground
<EmptySand> mtorni: both? ;)
<warewolf> EmptySand: Their own ground is marketing ;)
<mtorni> EmptySand: MS isn't a game company, so it's not their ground :)
<slouken> They get into nitty gritty of development.  I think it might be hard to convince them that we can provide them with something better, _especially_ if we sit on top of DirectX
<mtorni> EmptySand: and it looks like opengl is taking over direct3d, but who knows
<EmptySand> Once people see they can develop games for linux, and then sell them well then we might see them actually invest in their own inhouse native linux software
<raka> slouken:: Is it possible not to sit on top of DirectX?  I don't know enough about windows.
<mtorni> I think opengl isn't a good api; enable one feature the card doesn't support and you're doing it all in software.
<slouken> Direct3D will probably take over OpenGL in the long run because it will support newer hardware features.
<warewolf> mtorni: If we can finish PPlay 1.0 before Farenheit comes out...
<slouken> raka:  Not really.  DirectX is as low as you can portably get
<EmptySand> mtorni:  But MS have the access to the core engine of win32 which noone else does. ie speed
<mtorni> EmptySand: oh come on. That doesn't happen in 20 years, and we aren't THAT long term.
<slouken> If they want speed, they get as low as they can... so why use PPlay?   ....  dunno
<raka> BTW:  I think we need to get people who know about PlayStation, thats a better sell tahn "we can port to linux"
<mtorni> EmptySand: my last remark was to native linux stuff in game companies
<warewolf> slouken: Most card manufacturers release already OpenGL drivers - this will get common quite soon IMHO
<slouken> OpenGL is relatively limited when it comes to the newer 3D features.  For example, how do you specify that you want your textures in non-local Video RAM in OpenGL?  How do you specify AGP pipelining?
<EmptySand> mtorni: Why not?  Lot of other development companys (corel, netscape) are releasing native (and therefore mostly posix) applications
<warewolf> raka: Yep. leading them to Linux without them even noticing it ;)
<mtorni> EmptySand: look, just how many gamers use corel products?-)
<slouken> *laugh*
<EmptySand> mtorni:  You missed my point
<mtorni> EmptySand: and even netscape? iexplorer is the best most gamers know. (and they don't even try netscape, because they "know" it's worse)
<EmptySand> mtorni: ANyway I'm sure some game developers use corel paint
<warewolf> mtorni: Now there´s IE vs. Mozilla...
<EmptySand> mtorni: I'd debate that netcape is worse.  ie4 is security wise much worse
<mtorni> EmptySand: just how many percent of the customers are _capable_ game developers?
<mtorni> warewolf: most people really "know" iexplore is better. And they don't bother to even check other possibilities. You know, MS can't be wrong. "Why would they ship it with windoze?" is an usual comment.
<slouken> Anyway, I was just trying to say that if the money is good enough, game companies will be willing to write OpenGL versions of their games and run natively on either Linux or Win32
<mtorni> EmptySand: how many gamers care about security?
<warewolf> mtorni: This will propably change. M$ is getting a lot of bad press lately..
<slouken> IMHO, as usual. :)
<EmptySand> mtorni:  back onto the topic
<mtorni> slouken: but it's quite hard to get opensource opengl drivers.
<warewolf> mtorni: GGIMesa?
<slouken> mtorni: I think that will change
<raka> This is whatt the Mesa and GGI people have to fix.
<mtorni> warewolf: yeah, right. Like the gamers would care what the press says about ms, they'll use win95 anyway.
<EmptySand> mtorni:  in the long term I don't really care about win32.  I want see a lot more linux games though
<raka> In the end, the press decides everything.
<slouken> For that, PPlay is the perfect answer.
<EmptySand> and pplay could provide an easy path for developers used to directX to port to pplay
<slouken> DirectX is really hairy
<raka> Any specifics on the EMptySand
<raka> (On that, EmptySand?)
<warewolf> EmptySand: cause PPlay is easier to use? ;)
<EmptySand> hence, easing the trip.  Until they realise they can develop the low level libraries themselves.
<mtorni> warewolf: the best example was a loser I saw at asm'97 who had banners "ms sucks" and was telling anti-ms stuff, and used windoze with all the latest cool gadgets, and said "why should I use something else, there isn't anything else" He never even bothered to actually find out any alternativers.
<EmptySand> raka: What you mean?
<EmptySand> ww: No because pplay will be mainly linux native.
<mtorni> EmptySand: I've never cared about win32. But if you want pplay to catch with commercial companies, you just have to care.
<mtorni> EmptySand: As I don't like playing games anymore (for the last 2 years) I'm soo happy with linux.
<slouken> Well, if PPlay sits on top of a cross-platform API, (hint hint) your problems are solved
<EmptySand> mtorni:  If we show that you can get the same preformance for less work, and still have a large market in linux.  Someone's got to get interested
<slouken> Or, if you write native lowlevel routines for each platform
<EmptySand> slouken:  But preformance becomes an issue
<slouken> right..
<mtorni> EmptySand: oh come on. You obviously don't meet non-linux people very often.
<EmptySand> IMO: core pplay should be a native linux library
<EmptySand> mtorni:  (I'm in win95 at the moment 8)
<mtorni> non-Linux people think a LOT differently (if at all)
<slouken> Okay, so did we decide anything for the meeting?
<mtorni> EmptySand: now you're really anti-MS :(
<warewolf> EmptySand: No, there wont be much ´native linux´ code in PPlay - most of the stuff is simply too portable for that ;)
<raka> THings to Decide On:
<mtorni> warewolf: it ought to be
<raka> 1 or 2 Meetings?
<EmptySand> mtorni:  I use the tools that work, to do the things I need.  (At the moment, I'm reinstalling debian 2.0 with no X)
<raka> What to send in THe OVerview.  (we decided to have one)
<mtorni> EmptySand: I just bought a matrox millenium II, just to notice that now I've to stick with 80x25 textmode :( (I noticed I don't like X even when it's fast)
<warewolf> raka: INHO 1 meeting with an official part (commercial devel centered) first and an OpenSource Part (not requiring the Comm. ppl) later
<mtorni> I do not care if there are 1 or 2 meetings, as long as I'm informed early enough.
<slouken> Reasons to have one meeting?
<slouken> Reasons to have 2 meetings?
<EmptySand> mtorni:  What windowmanager you use?  There are some wonderful (and quick) wm's for linux
<EmptySand> slouken:  1 == easier to organise, and get a large turn out
<warewolf> raka: PPlay overview or overviefw for the announcement?
<raka> Reasons for 2 meetings:  Keep two distinct topics seperate.  Also avoid an OpenSOurce on opensource fight.
<mtorni> EmptySand: none, I don't do X :)
<raka> BTW I think we should talk to OpenSource first.
<raka> I mean overview.
<EmptySand> mtorni: why buy a mill2 then?  Games under directX?
<mtorni> EmptySand: the rare moments I do X, I use icewm, it's the fastest with such a nice look.
<warewolf> raka: (OS first) Ok, but without great announcement
<raka> Yes.
<slouken> I agree... an opensource meeting should be first
<mtorni> EmptySand: as I said, I don't do win. I don't have any non-linux products.
<EmptySand> yep
<mtorni> EmptySand: I thought I'd like X with a millII, but it turned out otherwise :)
<EmptySand> mtorni: Why's that?
<warewolf> raka: PPlay overview or announcement overview?
<mtorni> EmptySand: Before I thought didn't like X because it was kinda slow. But it turned out that I hate GUIs A LOT!
<EmptySand> hehe
<raka> I think we should write an overview and put it on the web pages.  Link our announcements to it, and email it to the commercials, when the time comes.
<warewolf> raka: Ok
<EmptySand> raka: yep
<slouken> Sure.  It's a great thing for slashdot too
<raka> GUIs are kind of slow.
<raka> Click, Click, Click, ad infinitum.
<EmptySand> Depends on the gui. fvwm is quite quick. ;)
<mtorni> raka: yep, the mouse is too far away.
<warewolf> slouken: the /. announcement shouldn´t have too much focus on PPlay
<mtorni> emptysand fvwm = wm. X = gui
<slouken> I agree
<raka> I'm talking about user speed, no GUI can involve less work than CLI with completion.
<slouken> Because really, it's about what's out there, not just PPlay, right?
<raka> X=window system.  Gtk=gui.
<warewolf> Ok, anyone some comment besides GUI speed? ;)
<mtorni> raka: ok, ok
<mtorni> warewolf: GUIs suck ;)
<EmptySand> xterm inside x == cui
<mtorni> xterm = gcli
<warewolf> mtorni: CLI sucks (for image retouching ;)
<mtorni> warewolf: who wants images, anyway?
<slouken> I gotta go pretty soon
<EmptySand> When I'm working on multiple servers, have several window panes makes life much easier.
<raka> I think we should stay focused on PPlay when we go to slashdot.  EAll the slashdot people know there are plenty of tools out there.
<slouken> But it's about letting people know about all tools, right?
<mtorni> slashdot's slow because it has to look nice on a graphics screen.
<raka> Yes, but slashdotters already know about them, or they can go to our web pages and see the Reasources page.
<warewolf> raka: The /. thing is not about PPlay, but about attracting game developers in general. PPlay is a major part in it, but stressing it too much makes the thing sound not serious enough
<EmptySand> Hey this interesting: Dell might support linux.
<slouken> warewolf: amen
<raka> You don't go to slashdot to attract people to linux.  They're alreay attracted.
<mtorni> Yep, a general linux gaming meeting is the best thing we could do.
<EmptySand> Go to slashdot to gather further linux interest
<mtorni> raka: some are attracted to freebsd
<raka> We don't need to evangelize linux to freeBSDers either.
<warewolf> raka: I know. But you go to /. to make people think about something, to make them focus on linux gamedev a bit more
<mtorni> echo echo echo...
<slouken> Have we decided on anything? :)
<mtorni> nah
<raka> Yes.  2 meetings, OpenSource first.
<warewolf> slouken: I think so ;) I´ll scan the log and post a summary afterwards ...
<slouken> Definitely.
<raka> And we need overviews.
<slouken> What kind of overview?  (It's like a company meeting here ;)
<warewolf> raka: Remember your PPlay intro (currently somewhere under PenguinDoc/) ?
<mtorni> warewolf: cool. just imagine the need to scan a few hour log to find out if we decided about something :)
<raka> warewolf: Yes, as shorter, updated version of that.  I'll start work on it.
<EmptySand> Geez, if Dell starts offical linux support that's pretty major coup for linux
<warewolf> mtorni: I´ll do it anyways - it´s better to cut out these nasty little off-topic threads befor posting the log to the list ;)
<mtorni> warewolf: like the girl thingy :)
<raka> re DELL:  We can hope and prey, but I don't want to get might expectations up too high.  OK I'll end this thread.
<warewolf> Ok, any more constructive things? Otherwise I suppose we quit for now
<mtorni> I have plenty, but I'm too tired.
<warewolf> raka: it´s "pray", not "prey" ;)
<mtorni> i don't want to argue with slouken anymore..
<EmptySand> mother just called: brb
<slouken> Dunno.  Sorry, didn't mean to argue. :)
<mtorni> slouken: right :)
<slouken> I think we were talking about different audiences. :)
<warewolf> slouken: ´T was good to have your arguments here anyways - this way we got a little training for the big meeting ;)
<mtorni> how about having a break (like 12 hours at least) so we can all get some sleeeeep?`!
<slouken> Sounds good.  I'm looking forward to e-mail on this. :)
<raka> mtorni: No prey, I'm hungry.
<mtorni> slouken: your arguments were good. but my brain's too tired to argue efficiently.
<slouken> Is there anyone else on PPlay developers group?
<raka> mtorni: It's breakast time here.
<mtorni> raka: I noticed :)
<warewolf> slouken: who´s on whom?
<slouken> Are there lots more PPlay developers who didn't make it?
<mtorni> slouken: every hour is a bad hour for half of the people.
<raka> No joke.  72 hours notice next time.
<warewolf> slouken: Yep, at least Stephane wanted to be here...
<mtorni> raka: oh come on, more! Like 1 week so people can plan things up in advance.
<slouken> Cool... See youall later! :)
* slouken waves
<raka> OK, 72 hourse at least.
<warewolf> slouken: Cu
*** slouken has signed off (Leaving)
<warewolf> raka: Ok, agreed ;)
<raka> warewolf: look forward to the log.  Girl thread and all.
<warewolf> raka: what about envisioning next weekend?
<mtorni> warewolf: we need a table of people in different timezones
<raka> Next wekend sounds good.  For just PPlay, or the OS meeting?
<warewolf> raka: (log) Err, well, of course these , err, slightly unproductive err, parts won´t appear on the log...
<raka> mtorni: I about 21:00 GMT is OK for all but asians.  Do we have any asians?
<warewolf> raka: We should try  to contact the other OS guys till then. The sooner the better. I want to see the big meeting in about 1 month
<mtorni> raka: you cannot ask questions like that. Just how many people do you see on this channel right now?-)
<mtorni> we should as the commercial people just how capable they are in c++
<warewolf> 21:00 GMT is ok
<raka> mtorni: I mean on the PPLay list.
<mtorni> raka: 21.00 GMT is great.
<raka> warewolf:::: Ok, unless the list dissagrees.
<warewolf> raka: ´course
<mtorni> raka: yep, but with this few people, what are the chances that someone would know anything about that?
<mtorni> "ok, unless the list disappears"
<raka> mtorni::: ture.
<raka> ture I mean.
<mtorni> boy I'm tired.
<mtorni> ;))
<mtorni> I'm really tired. ROTFL
<raka> motri: so bugger off.
<mtorni> you're right. It's about time. bye all
<warewolf> mtorni: I´ll post the summary of this meeting tonight, so the PPlay list will know soon
<raka> YOu still sleepy EMptySand, it must be 10 there now.
<warewolf> Ok, I´d better quit now, too
<warewolf> mtorni: Bye
*** mtorni has signed off (hopefully I can manage to bed)
<raka> warewoolf: If you are sleepy, then send me the log, I'll do it.
<warewolf> raka: No, actually I have sort of holidays right now, which means being awake till 03:00 each night (it´s nw 0:40 here)
<warewolf> raka: But Thx ;)
<raka> OK
<warewolf> raka: Btw - you won´t leave out the girl stuff anyway ;)
<raka> Well, honesty in reporting and all... :)
<warewolf> <g>. Ok, I´ll sign off now. Have fun
<raka> bye.
Logfile closed at:  02.08.98  00:39:15

Logfile opened for #pplay at:  01.08.98  21:40:51

[Note : Time above is local time (GMT+2)]

[some chat waiting for more people to join]

[20:18 GMT]

<warewolf> IMHO we simply should start, it´s already 18 mins over time...
<EmptySand> Sure
<raka> OK.  If no one else turns up, we can do this again.
<raka> What to talk about.
<warewolf> I mailed a summary of the BigMeeting things ´bout 10 hrs ago - everyone got it?
<EmptySand> Yep
<raka> And me.
<warewolf> Ok, any comments/enhancements/insults ?
<raka> Just, finding the message. I'll have some insults for you soon.
<warewolf> raka: Good to know. Thx ;+)
<EmptySand> Put a notice on slashdot maybe as well
<raka> Down with Netscape Mail!
<warewolf> EmptySand: Thx. How could I forget that!
<EmptySand> That should gather a lot of interest in it self
<warewolf> EmptySand: Yep, but we have to make sure that it sounds serious and professional
<raka> I'll be dead for a while, must get mail.
<EmptySand> Of course
<raka> Slashdot!  Serious and proffesional!
<warewolf> raka sure loves his config ;)
<slouken> hello!
* EmptySand grins
<EmptySand> Hi
<raka> He's ALIVE!
<slouken> *grin*
<warewolf> raka: not specially for /. , but ppl have to recognize that it´s not Yet Another Little Try
<slouken> Did I miss anything important?
<warewolf> slouken: Yep, we already agreed on high prices in Germany ;)
<raka> slouken: We are discussing a /. announcement.
<raka> BTW, my config fine.  My ISP isn't.
<warewolf> BTW: who will be able to monitor the discussion on /. after the announcement?
<warewolf> slouken: BTW: Did´y get my summary mail?
<raka> whoever's no the net at the time.
<slouken> yes ware
<EmptySand> It's pretty easy to monitor /.  But hard to control.
<raka> That's the point of /.
<warewolf> EmptySand: Not if you have to pay German prices ;)
<raka> It'll be a good way to gauge public opinion.
<EmptySand> Sure.  I'm pretty much next to a cheap point of connect all the time
<warewolf> My point was that one of us should be there to answer questions and prevent misunderstandings...
<raka> True.  Misunderstanding on /. BAD!
<slouken> Okay, I'm here for real
<raka> slouken: nice to know that.
<slouken> Okay, who all is here?  (by name?)
<raka> only Nicholas,Chritian Me and You.
<EmptySand> Nicholas Lee
<slouken> Sam Lantinga
<warewolf> another thing - anyone knows how to reach commercial game Devs best (mailing service@.. will be silly)
<warewolf> Christian Reiniger
<raka> Me is Adrian
<raka> Ratnapala
<warewolf> raka: <g>
<slouken> Okay, what do we do?
<slouken> <-- Author of SDL
<raka> What does <g> mean?  You keep saying it.
<EmptySand> General you have to start their, unless you can get an email of someone
<warewolf> <g> == <grin>
<raka> I do the 2d pplay library.
<raka> warewolf: ahh
<raka> What is EmptySands interest.  You aren't noisy on the list.
<EmptySand> I've been an observer for a long time
<warewolf> mailing crack.com guys should be easy (there was a .plan of ddt on /. some days ago) and someone posted a message about some id guys recently (AFAIK including email)
<EmptySand> I'm working on a UO server project with some one at the moment
<slouken> enptysand: cool
<raka> id, is turning bad.  The're now developing on NT, and arn
<raka> arnet liking UNIX ports.
<raka> UO?
<warewolf> Oh, yes, a westwoood developer posted something on /. last week (?), I´ll have to look for it.
<slouken> Okay, so what exactly are we planning here, and what makes it "Yet another little try" ?
<EmptySand> Ultima Online
<slouken> Er, NOT yet another little try
<raka> EmptySand: Thanks.
<warewolf> slouken: We´re discussing what to do at/for the BIG meeting
<raka> The problem could be that its to BIG a try.
<slouken> Personally, I'm looking for two things:
<slouken> One, feedback from commercial devlopers -- what do they need?
<slouken> Two, how does SDL fit into PPlay?
<slouken> Three What game development resources are available for UNIX?
<slouken> er, three things. :)
<EmptySand> That /. Westwood URL: http://penguin.lvcm.com/westwood.html
<raka> I'm also thinking about number 2.  Basically I'm currenly programming straing on top of GGI, and I think thats the way it'll be on UNIX.
<raka> SDL could be important on other sytems.
<warewolf> slouken: one: what preconditions do they need to write things for Linux
<warewolf> The crack.com guy is ddt@crack.com
<raka> Also PPlay is not monolithinc, people who don't want the overhead of pg2d could use SDL, but keep the rest of PPlay.  (not that I intend to bloat pg2d).
<warewolf> raka: libggi works nicely with fbcon and fbcon is pretty common, isn´t it?
<slouken> fbcon isn't that common in my experience
<raka> One think about you list of people for TheHugeEvent, one lot is GGI/Mesa/other freeware developers, another is commercial developers.  Two seperate issues.
<warewolf> the westwood guy is kevin@westwood.com
<slouken> Right.  So, what exactly are we getting together for?
<raka> I mean I'm writing straight on top of libggi, not SDL on any similar abstraction layer, libggi is enough.
<slouken> libggi is an abstraction layer
<slouken> Meeting so everyone can meet everyone?
<raka> slouken: exactly, so I don't think I need another one, unless it makes win32 ports easier/better.
<warewolf> slouken: (1) getting all opensource ppl together so that the projects can be easier coordinated
<slouken> If you write on SDL, you automatically get Win32 and Linux and BeOS
<slouken> Right.  Are we talking about merging projects?
<warewolf> slouken: (2) talk to the commercial guys to see what they need
<slouken> Maybe we should have two meetings?  One for commercial folks and one for opensource folks?
<warewolf> slouken: (3) discuss on the future strategy
<warewolf> slouken: merging: on some projects this would be good for sure
<EmptySand> slouken: depends on how much interest we can generate in the commerical people.
<raka> Dont merge with US, we are to slow!
<raka> :)
<slouken> raka:  I noticed that... so has everyone else :-/
<warewolf> slouken: 2 metings: te meeting for the commercials will require the OS guys to be present, too (at least the programmers)
<slouken> True warewolf
<raka> BTW: I didn't read the SDL  docs properly, do you have primitves for line drawing etc.
<slouken> raka: Nope.. it's open for people to use their own specially optimized routines
<warewolf> slouken: So no hardware accel for it?
<slouken> raka:  The screenlib example library has some drawing routines
<raka> I thought so, the things is acceleration.
<slouken> right
<raka> We need accel.
<slouken> Hmmm
<warewolf> raka: shouldn´t be too difficult for win, but what about beos?
<raka> beos slouken?  Any ideas.
<raka> Also SDL could be a nice way to get going on new platforms, w/o accel.
*** mtorni (mtorni@vaasa-dk5260-1-182.kolumbus.fi) has joined channel #pplay
<raka> Hi mtorni
<mtorni> is it over already?
<slouken> Dunno... BeOS has some acceleration, but I hadn't looked at it since SDL just provides framebuffer
<raka> No, we haven't done much yet.
<slouken> How necessary is accelerated line drawing?
<slouken> How many games really use it?
<raka> What about accelerated blitting?
<slouken> SDL supports accelerated blitting
<warewolf> slouken: IMHO accel. blitting is the most important one...
<raka> slouken: how excatly.  Any architectural reason why it cant do other accelerated ops?
<slouken> Just simplicity.  Accelerated drawing ops really complicates the API
<slouken> and the code...
<raka> API, yes, but I thought code complexity would be linear with number of ops.)
<mtorni> hacking penguinplay and an example game for penguinnet
<mtorni> and hacking libnixon (to replace all configuration related things in all unix-programs:)
<slouken> raka:  The code complexity doesn't increase linearly when you support different platforms -- you have to write general unaccelerated functions, and then accelerated functions for every single platform that supports acceleration.  It was too much trouble..
<slouken> for drawing, but not too much trouble for blitting
<raka> slouken: Too much trouble for one person, yes.
<slouken> The thing is, there are quite a few optimized graphics libraries that could easily be ported to SDL
<slouken> If accelerated drawing is really important in games, sure.  But my impression is that it's not.
<slouken> Please correct me if I'm wrong..
<raka> You see my idea is that if we don't want to be "just another little try" we need to do everything, like all the accell.
<EmptySand> raka: But also provide the hooks so people can improve our optimasation easily
<warewolf> slouken: line gfx was really important some time ago, but I don´t think it´s important anymore (perhaps for simulating conventional GUIs)
<raka> Hooks a million.
<slouken> warewolf:  GUIs definitely need drawing acceleration, but many game GUIs are image based
<warewolf> EmptySand: optimizing should be easy - the source is available and AFAIK most funcs can be overridden
<slouken> raka:  I think the meetings are more about coordination and communicating with commercial developers than about promoting us, necessarily
<warewolf> slouken: That´s why I wrote "simulating traditional ones" ;)
<slouken> ware: got it
<raka> slouken: Agreed.
<warewolf> slouken: Yep, of course
<raka> BTW: Why SDL is good: Nice, platform independent API, low level, doesn't interfere with me.
<slouken> So, we don't need to promote a "try" really, right?
<raka> Why SDL is bad: Yet another layer of software silt.
<slouken> BTW: Why SDL is bad:  low level
<raka> Re:: "promomoting try":::: but we need to show them we are serious.
<raka> But you said you made it low level on purpose.
<slouken> raka:  Yep, that's definitely true.  Hardcore game developers will always write natively for each interface.  SDL is just a nice layer for porters and medium-core game developers. :)
<slouken> raka: yep
<raka> That's whats worrying me, if hardcore develpers will write natively, so should we.  OUr job is to do the hardcore stuff for the other guys.
<slouken> Serious about what?  Serious about meeting the needs of commercial developers?
<slouken> raka: right.
<raka> Yes, serious about that.  And sersiously able to do it.
<slouken> raka:  Then why use libggi?  Use KGI at the very least
<warewolf> slouken: serious about handling this stuff, serious about not just trying to do another hack ala svgalib
<slouken> PengiunPlay isn't a great testimony to being able to provide a product for commercial developers.
<raka> KGI bad, interface depends on each board.
<warewolf> raka: libggi shouldn´t pose much overhead, but in turn saves us from reinventing the wheel
<slouken> The reason why I did SDL, was because most commercial developers already have code libraries built for their games
<raka> Only because we haven't provided a product AT ALL.
<slouken> They won't (necessarily) want to use another high level API, but want to leverage code they wrote and use.
<slouken> Hence, Simple DirectMedia Layer
<warewolf> slouken: And what if our code will be better? ;)
<slouken> So anyway, PenguinPlay is great for starting developers, but what about existing game houses?
<slouken> warewolf:   I'll gladly use it. :)
<raka> I agree, existing houses would like something like SDL better.
<warewolf> slouken: As PPlay is OpenSource, I´m sure it will develop faster than any of these private libs...
<raka> warewolf: Not that its developing too fast now.
<mtorni> slouken: game development takes a year or two, and IF there's a bit high level API that saves works and works in win95/98 they'll use it if it's useful
<slouken> Right, it just needs to catch on.
<slouken> Right.
<slouken> That's very true.
<EmptySand> IMO It needs a solid source base with some functionality before that
<warewolf> raka: yes. true. I hope we´ll get more ppl after the announcement (and I seriously hope that work will be faster as soon as the design is mostly finished)
<raka> ditto EmtySand.
<mtorni> penguinnet should be a good starting point; every game today wants to have "INTERNET" on its box
<mtorni> and doom took several versions to get even decent netplay
<raka> BTW: will penguinnet relate to things like KALI at all?
<warewolf> mtorni: BTW: You still have to do the network setup /IP addresses etc) by hand with PenguinNet, don´t you=
<slouken> Here's the nice thing about SDL:  It's low level enough that it doesn't get in the way of developers, and is useful enough that you can write games directly on top of it in a single source file (See SDL-demos.tar.gz)
<mtorni> raka: kali just masks ipx
<slouken> HOWEVER
<slouken> PenguinPlay is a great API, for big games
<EmptySand> I think the main thing might be some intergation between the various parts of the library
<mtorni> warewolf: that's what the dialers do
<slouken> It just needs some functionality. :)
<warewolf> mtorni: Erm, yes, ok.
<raka> EmtpySand: I think we need yet another integration doco.
<warewolf> raka: Yep, we also need to decide on some "official" intro to PPlay for the Homepage
<EmptySand> probably.  I think the low level stuff needs to be nutted out.  (Like this event layer)
<raka> OK, I think we agree on the relative merits of PPlay and SDL.  Next question, how, if at all should PPlay use SDL.
<mtorni> pnet started out with a decent specification. too bad someone did c++ code that was "too much" c++ for me. (so I had to drop out)
<EmptySand> The better and faster that is designed the easier the intergation and functionality
<slouken> (sorry, I'm a fan of SDL) :)
<raka> EmptySand: You voluneering for the event layer!
<EmptySand> 8)
<warewolf> raka: depending on the progress of GGI. if GGI ports to Win32 etc become available soon...
<slouken> So, the meetings are just for people to meet and collect URLs? or is it for discussing integration of projects, or what?
<EmptySand> No a asked what was needed. 8P
<slouken> GGI will change dramatically when it ports to DirectX (known from experience)
<EmptySand> err s/a/I/
<raka> Yeah, I'm thinking GGI==prefered option, SDL or similar equals Lite option where GGI doesn't exist.
<warewolf> slouken: Yep, IMHO most projects just dont know enough about each other
<raka> That way we only have two ports to worry about.
<EmptySand> slouken: All of the above?
<slouken> warewolf: cool
<raka> It would be nice if we had sound people here.  Ar they dead?
<slouken> raka:  GGI will be on win32 by the time PenguinPlay is ready
<mtorni> About the slow progres... I think we would be developing much faster if only some basic c++ stuff was used.
<slouken> raka:  I will be starting a port of SDL to MacOS, but no ETA at the moment
<warewolf> slouken: Don´t be too pessimistic about our devel speed ;)
<raka> slouken: WHat about BeOS?  Playstation (not that SDL is their eitehr).
<raka> werewolf: I think we will accelarate soon, but GGI will still beat us to win32.
<EmptySand> mtorni: Like what?
<mtorni> as I said, too much c++ness was the thing that threw me out of pnet (and I had already implemented that kind of a game)
<slouken> GGI might get ported to BeOS (dunno)  It would be nice to have a GGI spokesperson here. :)
<warewolf> mtorni: Dont think so. For me developing the goos design takes most time, whether done in C or C++
<mtorni> EmptySand: The class stuff only, and leave out the hairier bits. I can do only basic C++ stuff.
<raka> Yes, we want a GGI spokesperson, but we don't have one.
<raka> untill recently gcc could only do basic C++ stuff :)
<mtorni> Most people know a LOT more C than C++
<slouken> PenguinPlay seems like a huge API... is there any way to nest it and get core functionality that people can start using, and then add the rest later?
<EmptySand> raka: True. ecgs is pretty good now
<mtorni> raka: yep, but the non-basic stuff has made me drop out of a few projects (berlin)
<slouken> Yeah, that's a good point.  C is much more commonly used, and can be called from C++ (not vise-versa)
<warewolf> mtorni: Yes, that´s a major point. It´s also easier to write language wrappers for C apis than for C++ ones (look at Gtk)
<raka> Yes.  I (and I think everyone else). Wants it to be as moduar as possible.
<raka> eg. PPlay should not rely on pg2d, nor vice versa.
<EmptySand> rake: definitely
<warewolf> mtorni: But a good OO design simply has too much advantages for the higher level apis
<mtorni> warewolf: oh come on, network stuff is better left unclassified!
<mtorni> warewolf: you're sending packets anyway, not objects
<EmptySand> Good OO design and good C are very different beasts
<raka> RE "good OO": which is why pg2d is C++ only (not that its good).
<mtorni> and even C can do it OO style; just look at gtk :)
<raka> But I agree, network stuff should be C compatible.
<slouken> How about C core with C++ wrappers?
<warewolf> mtorni: I dont know abot the network stuff, but doing a GUI lib in C is not very nice...
<EmptySand> mtorni:  But why waste the effort in C, trying to be C++?
<slouken> The SDL library is core C, but I have some nifty C++ example classes on top of it for the Maelstrom port
<mtorni> raka: I've tried converting my netgame I mentioned earlier to C++, but the basic network layer is hard to be thinked object-orientedly
<mtorni> EmptySand: that was just a joke...
<warewolf> slouken: We use C++ because we want the benefits of OO design - writing it in C and doing a C++ wrapper destroys that
<mtorni> warewolf: yes, I agree on that.
<slouken> I'm just talking about core functionality.
<EmptySand> mtorni:  I agree the the current ppnet layer isn't abstract enough. I looked at it for a mud, and it dodn't seem to work well
<raka> So mthen don't do object-orientdly, do some badly wrapped API, or don't wrap at all.
<slouken> Um.... What exactly is the current architecture of PPlay?
<mtorni> warewolf: but instead of doing everything OO or the other way, maybe a hybrid would be the best everywhere
<EmptySand> slouken: Core C++ (member functions) is good C
<warewolf> slouken: Of course the low level stuff should be plain C (just look at OpenGL)
<mtorni> warewolf: some things are just meant for OO, others aren't
<mtorni> c++ makes it easy to do non-oo stuff anyway
<slouken> About those meetings, are we settled?
<raka> slouken: We have Layer P(rocedural), tools for doing stuff protably.  ALl the basic functinality.  This stuff is not necassariy written by use.
<raka> Written bu us I mean.
<warewolf> slouken: Did we agree on what meetings to hold yet?
<raka> Then We have Layer O, (Object).  Can't talk, more we have a decument somewhere.
<slouken> Dunno, got wrapped up in SDL/PenguinPlay talk. :)
<mtorni> Isn't PPlay supposed to ease things? bsd sockets is as easy as it gets on the low-level after setting the connection up.
<raka> Layer O is meant to ease things.
<raka> Layer P makes things possible.
<warewolf> Meetings: IMHO we should stay with the single big meeting with OpenSource ppl _and_ commercials
<EmptySand> ww: yep
<raka> warewolf: OK, why?
<slouken> Then the commercials will just see a bunch of opensource people quibbling.
<slouken> That's not what we want. :)
<mtorni> warewolf: I don't know what you decided, but we need one meeting (where more people show up:) where we prepare some material for the big meeting
<warewolf> mtorni: Agreed
<slouken> What material?
<raka> This was meant to be that meeting.  We didn't have enough notice.
<slouken> Do we present to commercials, or do they communicate to us?
<warewolf> raka: cause most of the OS ppl need to talk to the commercials...
<mtorni> slouken: like what are we going to ask them, what to tell them (no one likes to wait on an idle channel :)
<slouken> Or is it just:  Hey everybody, here's what's available!
<mtorni> slouken: we have to show what our goals are, and then have a mutual QA session
<warewolf> slouken: No, no presentation. A discussion about what´s needed and how we can archieve this
<slouken> I'd like to know  "If you wanted to develop on Linux, what would you need?"  Customer base, most likely.
* mtorni thinks this meeting was on a waaay too short notice
* slouken agrees
<raka> We can list our goal/plans in writing, before talking to them.
* EmptySand thinks that it's a good start though
<slouken> Commercial developers need guaranteed customers.  They'll do anything for percieved customers
<warewolf> mtorni: Ok, I´m guilty. But I wanted to have some discussion about it ASAP (cause email is so damn slow for discussion)
<EmptySand> slouken: that crack.dot issue last week is an example that their is a linux games base
<mtorni> raka: yep, as we don't have much yet, we need to tell them what exactly we are trying to achieve, and then they can point us the weaknesses, which part needs more/less work
<raka> We can't create a user base, we can only help on technical fornts.
<warewolf> slouken: there are quite many petitions showing up recently
<slouken> Maybe we should meet and decide what we are trying to achieve first?
<raka> THi crack.com thing.  I haven't heard anything after reading they were in trouble.
<warewolf> raka: Assume 7 million linux users, 10% of them gamers (?). Most of them would buy good stuff
<raka> slouken::: Shouldn't we do that now.
<slouken> raka:  Yeah, what's up with that?  The last thing Linux needs is a game company down the tubes.
<mtorni> raka: 100 users are 100 users more if they don't need to do extra work; that is our api works with win32 & other OSs
<slouken> raka:  Sure, but we need to do it with more people... mesa, GGI, etc
<raka> But crack.com is in trouble anyway.
<EmptySand> Actually I think since alot of hackers tend to be gamers of one sort it might be higher
<warewolf> raka: Someone gave them financing until Golgotha gets out
<slouken> mtorni:  Yeah, but that's if they start off with the cross-platform (PPlay) API... what about the current rash of games?
<raka> warewolf:: That's good news.  I think I'l buy GOlogtha.
<slouken> For the record, a big thing SDL doesn't support is 3D
<EmptySand> crack.dot got a good response.  Particular from linux people
<raka> mtorni:::: Thats what things like SDL are fore.
<mtorni> look guys, if you were a company, and you'd be given an opportunity to get a few hundred customers more with no investments, wouldn't that be a good deal?
<raka> slouken:: Thats why I like SCL.
<warewolf> mtorni: If we support playstation, too, PPlay will be very valuable for them. And doing the Linux port with PPlay should be very easy
<EmptySand> mtorni: need to be a few thousand
<slouken> or more
<raka> SCL is a bit too hight level for my liking, but it has 3D and Playstation support.
<mtorni> oh come on, pplay just doesn't bring more customers, it is some work already done.
<slouken> raka:  I'll have to check out SCL
<mtorni> they've to use some libs anyway; either they write it themselves or use an existing one and save money
<warewolf> mtorni: PPlay means supporting multiple platforms without more work -> more customers
<slouken> It's not NO investment, it's quite a bit of investment in a moving target -- not very good for serious game companies
<mtorni> warewolf: I know that, but tell it to others :)
<raka> slouken, sorry, you can't its closed source (for now). I'll get you the email of the person to contact, or at least send a detailed description to you.
<warewolf> mtorni: We´ll have to do that on the homepage and at the meeting - it´s an important point
<slouken> raka: thx
<mtorni> slouken: yep, but they don't have to you it immediately. Let's just tell them we're working on it, and we'll inform them when there's a stable version.
<raka> Sorry I think I got lost, what is "it"?
<warewolf> raka: PPlay
<mtorni> s/you it/use it/
<raka> Oh.
<slouken> You're assuming PPlay is their API of choice from the beginning.  I'm talking about the big players.. Activision, Sierra, Blizzard, ID, they all have existing tools, code, etc.  Why train their developers on a new API that's not out yet?  They need it now.
<mtorni> But people, just completing pplay isn't enough;
<warewolf> slouken: about chatting OS ppl on the big meeting - we just have to write the announcement appropriately IMHO
<raka> We don't want them to port now, we want to ask people what they want THEN.
<mtorni> we need to do cross-compiling from win32 to linux and other platforms easier than already done.
<raka> I think most would find cross compilng from   linux to win32 more fun.  Or is that just religious fervour.
<slouken> We can't sell people on dreams.  However, we can ask "If you wanted to port/develop for Linux, would you use X API?"
<mtorni> many companies probably don't want to install linux just to compile on it.
<mtorni> raka: wishful thinking
<slouken> raka:  Cross-compiling is fine, but debugging is hell.
<raka> slouken:  Agreed.  We can also ask what do you want in X API.
<warewolf> raka: Yes, that´s important
<slouken> Yep.
<raka> So were agreed.
<slouken> I think so
<mtorni> maybe we should prepare instructions for installing redhat/debian/someother so it's ready to compile
<slouken> So, we have it... exchanging information on what's available now, and asking what commercial developers need in the future.. is that right?
<EmptySand> mtorni:  Let's worry about finishing the design, and some code first.  packaging is easy
<raka> we need  something to package first.
<raka> Shall I raise the question again.  One or two meetings?
<warewolf> mtorni: Someone posted a mail from some develuper, worrying about the specialties of programming under linux. Anyone knows whether the LPG is still maintained=?
<mtorni> LPG?
<slouken> IMHO, about GGI... if you want a cross-platform APi, use SDL.  If you want acceleration and want to do the hardcore work, write for KGI, Xlib, GDI, and DirectDraw
<warewolf> mtorni: Linux Programming Guide (of the Linux Documentation Project (LDP))
<raka> slouken::: I'm leaning that way too.
<mtorni> slouken: nono, no writing for KGI, now THAT'S a moving target
<slouken> raka:  Dunno. do we need to plan anything for the commercial meeting then?
<slouken> mtorni:  *grin*
<warewolf> raka: Perhaps a small meeting to look what projects can/should be merged, but without big announcement
<raka> KGI is supposed to have a different interface for EACH CARD!  Writing for KGI is not allowed.
<slouken> So, we have it... exchanging information on what's available now, and asking what commercial developers need in the future.. is that right?
<slouken> Will people be willing to merge projects?
<mtorni> slouken: game companies would surely like some evidence that 3d card makers will give us (or ggi people) specifications
<raka> slouken: I think we need to write some goals/ideas docos firsts, and give them to the commercials before meeting them.  THat way they'll bother turning up.
<mtorni> raka: right.
<slouken> raka:  PPlay has too many specs and documents. :)
<warewolf> slouken: don´t know (we don´t have to sav that this straight forward ;) The important thing is that these projects have to know of each other
<mtorni> now, who will collect the goals -ideas from the mailing list?
<slouken> What goals/ideas docos can we right?
<slouken> er, write?
<raka> slouken: Agreed, I want docs that'll let us delete old ones.
<slouken> Goals of PPlay?
<slouken> Ideas of what?  I don't think they want to see spec sheets.
<raka> Well, goals are kind of on the webpage I supose...
<warewolf> raka: Yep, there still is too much old stuff on the homepage, spreading uncertainty
<slouken> I'm thinking about more of a brainfest then "Goals and Ideas for X API"
<mtorni> slouken: yep, that we're trying to get some work off their backs (for free) and we'll bring more customers, and ease porting, :)
<warewolf> mtorni: Excellent formulation ;)
<raka> No, but there is a big difference between saying "we'll do a cross platform game API" and "we'll do a cross platform game API of this size and shape")
<warewolf> raka: We have to provide an impression of what the API will contain and roughly how it will be done (features, philosophy etc)
<slouken> Why do we have to say that?  Why can't we say...  "Here's what we're doing.  How can we make it more useful to you?"
<raka> slouken: You know I think we are agreeing with echother very loudly.
<mtorni> the webpages should be ridden of all old stuff. And a few short pages with all the current stuff is a LOT better than a lot old and unusable stuff in a horrible mess
<slouken> We can't promise a cross-platform API.
<slouken> That's ridiculous.
<raka> Howso?
<mtorni> slouken: WE MUST! No company will do a linux-only game.
<slouken> Well, that's like promising the moon.  How can we guarantee something?
<raka> After all isn't that what SDL is?
<mtorni> slouken: if we don't succeed, they'll never hear from us again :)
<raka> I suspect you mean something more sublte slouken, so speak up.
<slouken> Rather, shouldn't we say... "Here's what we have"  Will you use it?  If not, why?
<warewolf> slouken: SDL shows how to make low level stuff cross-pülatform, and High level is simple
<raka> "here's what we _want_to_ have"
<mtorni> slouken: because it might end up being unusable for them.
<slouken> And, if so, we need to know that.
<mtorni> slouken: and we have nothing, so... "We have nothing. Will you use it?"
<slouken> *laugh*
<slouken> Then let them design the spec
<raka> They don't want to do the work.
<mtorni> slouken: they wouldn't do that.
<raka> Besides _they_ compete against eachother.
<mtorni> slouken: we need to show that we do some of the hardest parts, and do it right.
<warewolf> slouken: why? if they can tell us "hat´s good" r "that#s bad" and "this would be nice" it´s ok, isnt it?
<slouken> Sure they would be willing to do the work.  They do the work for Win32.  They need promises of return, and a reasonable difficulty of doing it.
<slouken> Yep
<raka> See!  I knew we were agreeing!
<mtorni> like "our net API solves most of the problems with multiplayer networking, and in addition it has these features for free, and it doesn't even affect performance, besides there already are a lot of servers for this, so you don't need to put money out for those, either!"
<slouken> Yeah, that's fine. :)
<raka> Slouken, we are not taling about writing the whole spec.  We are talking about short docos.
<warewolf> The overview
<slouken> Um, I know the guy who does the networking for Activision -- they have their own library, tested and written
<mtorni> maybe we could ask them what their current libs do
<slouken> What do we say in the overview?
<slouken> Why would they switch -- rewriting all their code?
<raka> Good for Activision.
<mtorni> slouken: but they need their own special servers; $$$
<slouken> true
<mtorni> slouken: they WON'T switch, but they'll use pplay in their future games
<warewolf> slouken: Goals of PPlay (portability etc), main features of the APIs, structure of the projects etc
<slouken> AAAAAHHH. :)
<raka> They don't switch.  We make sure the rest of PPLay gives them what they want if they need it without ruling out their existing net code.
<mtorni> slouken: this is a long term project, people won't switch overnight, and even the implementing is guaranteed to take over a year.
<slouken> But in order for them to use PPlay in future games... they need to have something now... right?  We're talking about a year or three for PPlay to fruit, then two years for the game development...
<mtorni> For me it was a problem finding a kali server nearby. The nearest one was in Finland, but it had no users. So I had to use an usa server, which amounted to a lot of lag.
<warewolf> slouken: I think it´s about 2 years until PPlay 1.0 (or so), less if we get more developers
<raka> But all we are asking _from_ them is a few hours of time.
<slouken> By then hardware will be different, and MS will be on the leading edge, no?
<mtorni> there were other servers around, but they didn't support kali or descent or whatnot
<slouken> I'm just playing devil's advocate, because I think these are important if we talk to commercial game houses.
<mtorni> warewolf: which amounts to; speed is not so much of an issue, we need to implement even the slow algorithms (with current hardware)
<slouken> I could be totally off the mark however. :)
<warewolf> slouken: They can use SDL and/or GGI now. And I still hope that PPlay devel will accelerate soon. Really soon
<warewolf> mtorni: what exactly do you mean?
<mtorni> slouken: two years for the game development -- MINUS the time pplay saves ;)
<raka> Remember, we aren't asking them to port now.  We are asking them for what they want.
<EmptySand> If pplay became the directX of linux, that would be what the commerical developer would require
<mtorni> warewolf: do you mean my last line or the one before it?
<mtorni> EmptySand: and it became the directx for win98, that would be a hit.
<warewolf> mtorni: the "slow algos" stuff
<raka> And the DirectX for PlayStation, we win.
<mtorni> EmptySand: if we don't do win98 at 1.0, they'll never come.
<raka> (if that happens)
<EmptySand> mtorni:  Lets be realistic.  M$ is never going to release core APIs
<slouken> How can PPlay supplant DirectX for Windows?
<raka> PPlay will have to run on DirectX.  I guess slouken knows most about this stuff.
<warewolf> raka: Well, if we´re accepted for one platform, we´ll have all
<mtorni> mtorni: I mean that we have to implement the most advanced 3D algorithms available (and invent some more:), no matter how impossible they are with current hardware.
<EmptySand> Speed is the big issue.  Making it easy to port to linux, with a (quick) directX wrapper and they would have a lot
<warewolf> mtorni: Yep. Be prepared for the future
<mtorni> EmptySand: I just meant that we need to be popular for win98 development...
<slouken> Have you guys ever checked out the DirectXDev mailing list?
<EmptySand> nope
<slouken> There are lots of sharp game developers on there.
<warewolf> BTW: How does OpenGl cope with modern 3D features (Bumpmapping, miltiTextures, ...)
<raka> slouken: now but I feel we should.  BUt then I guess I should read some DirectX docs.
<mtorni> EmptySand: Now, are we trying to do a quick hack for portability, or a long-term useful projecT?
<mtorni> slouken: of course not :)
<EmptySand> mtorni: It's pretty difficult to 'fight' ms on their own ground
<EmptySand> mtorni: both? ;)
<warewolf> EmptySand: Their own ground is marketing ;)
<mtorni> EmptySand: MS isn't a game company, so it's not their ground :)
<slouken> They get into nitty gritty of development.  I think it might be hard to convince them that we can provide them with something better, _especially_ if we sit on top of DirectX
<mtorni> EmptySand: and it looks like opengl is taking over direct3d, but who knows
<EmptySand> Once people see they can develop games for linux, and then sell them well then we might see them actually invest in their own inhouse native linux software
<raka> slouken:: Is it possible not to sit on top of DirectX?  I don't know enough about windows.
<mtorni> I think opengl isn't a good api; enable one feature the card doesn't support and you're doing it all in software.
<slouken> Direct3D will probably take over OpenGL in the long run because it will support newer hardware features.
<warewolf> mtorni: If we can finish PPlay 1.0 before Farenheit comes out...
<slouken> raka:  Not really.  DirectX is as low as you can portably get
<EmptySand> mtorni:  But MS have the access to the core engine of win32 which noone else does. ie speed
<mtorni> EmptySand: oh come on. That doesn't happen in 20 years, and we aren't THAT long term.
<slouken> If they want speed, they get as low as they can... so why use PPlay?   ....  dunno
<raka> BTW:  I think we need to get people who know about PlayStation, thats a better sell tahn "we can port to linux"
<mtorni> EmptySand: my last remark was to native linux stuff in game companies
<warewolf> slouken: Most card manufacturers release already OpenGL drivers - this will get common quite soon IMHO
<slouken> OpenGL is relatively limited when it comes to the newer 3D features.  For example, how do you specify that you want your textures in non-local Video RAM in OpenGL?  How do you specify AGP pipelining?
<warewolf> raka: Yep. leading them to Linux without them even noticing it ;)
<slouken> *laugh*
<slouken> Anyway, I was just trying to say that if the money is good enough, game companies will be willing to write OpenGL versions of their games and run natively on either Linux or Win32
<warewolf> mtorni: This will propably change. M$ is getting a lot of bad press lately..
<slouken> IMHO, as usual. :)
<mtorni> slouken: but it's quite hard to get opensource opengl drivers.
<warewolf> mtorni: GGIMesa?
<slouken> mtorni: I think that will change
<raka> This is whatt the Mesa and GGI people have to fix.
<mtorni> warewolf: yeah, right. Like the gamers would care what the press says about ms, they'll use win95 anyway.
<EmptySand> mtorni:  in the long term I don't really care about win32.  I want see a lot more linux games though
<raka> In the end, the press decides everything.
<slouken> For that, PPlay is the perfect answer.
<EmptySand> and pplay could provide an easy path for developers used to directX to port to pplay
<slouken> DirectX is really hairy
<raka> Any specifics on the EMptySand
<raka> (On that, EmptySand?)
<warewolf> EmptySand: cause PPlay is easier to use? ;)
<EmptySand> hence, easing the trip.  Until they realise they can develop the low level libraries themselves.
<EmptySand> raka: What you mean?
<EmptySand> ww: No because pplay will be mainly linux native.
<mtorni> EmptySand: I've never cared about win32. But if you want pplay to catch with commercial companies, you just have to care.
<slouken> Well, if PPlay sits on top of a cross-platform API, (hint hint) your problems are solved
<EmptySand> mtorni:  If we show that you can get the same preformance for less work, and still have a large market in linux.  Someone's got to get interested
<slouken> Or, if you write native lowlevel routines for each platform
<EmptySand> slouken:  But preformance becomes an issue
<slouken> right..
<EmptySand> IMO: core pplay should be a native linux library
<slouken> Okay, so did we decide anything for the meeting?
<warewolf> EmptySand: No, there wont be much ´native linux´ code in PPlay - most of the stuff is simply too portable for that ;)
<raka> THings to Decide On:
<mtorni> warewolf: it ought to be
<raka> 1 or 2 Meetings?
<raka> What to send in THe OVerview.  (we decided to have one)
<warewolf> raka: INHO 1 meeting with an official part (commercial devel centered) first and an OpenSource Part (not requiring the Comm. ppl) later
<mtorni> I do not care if there are 1 or 2 meetings, as long as I'm informed early enough.
<slouken> Reasons to have one meeting?
<slouken> Reasons to have 2 meetings?
<EmptySand> slouken:  1 == easier to organise, and get a large turn out
<warewolf> raka: PPlay overview or overviefw for the announcement?
<raka> Reasons for 2 meetings:  Keep two distinct topics seperate.  Also avoid an OpenSOurce on opensource fight.
<raka> BTW I think we should talk to OpenSource first.
<raka> I mean overview.
<warewolf> raka: (OS first) Ok, but without great announcement
<raka> Yes.
<slouken> I agree... an opensource meeting should be first
<EmptySand> yep
<warewolf> raka: PPlay overview or announcement overview?
<raka> I think we should write an overview and put it on the web pages.  Link our announcements to it, and email it to the commercials, when the time comes.
<warewolf> raka: Ok
<EmptySand> raka: yep
<slouken> Sure.  It's a great thing for slashdot too
<warewolf> slouken: the /. announcement shouldn´t have too much focus on PPlay
<slouken> I agree
<slouken> Because really, it's about what's out there, not just PPlay, right?
<raka> I think we should stay focused on PPlay when we go to slashdot.  EAll the slashdot people know there are plenty of tools out there.
<slouken> But it's about letting people know about all tools, right?
<raka> Yes, but slashdotters already know about them, or they can go to our web pages and see the Reasources page.
<warewolf> raka: The /. thing is not about PPlay, but about attracting game developers in general. PPlay is a major part in it, but stressing it too much makes the thing sound not serious enough
<slouken> warewolf: amen
<raka> You don't go to slashdot to attract people to linux.  They're alreay attracted.
<mtorni> Yep, a general linux gaming meeting is the best thing we could do.
<EmptySand> Go to slashdot to gather further linux interest
<mtorni> raka: some are attracted to freebsd
<raka> We don't need to evangelize linux to freeBSDers either.
<warewolf> raka: I know. But you go to /. to make people think about something, to make them focus on linux gamedev a bit more
<mtorni> echo echo echo...
<slouken> Have we decided on anything? :)
<mtorni> nah
<raka> Yes.  2 meetings, OpenSource first.
<warewolf> slouken: I think so ;) I´ll scan the log and post a summary afterwards ...
<slouken> Definitely.
<raka> And we need overviews.
<slouken> What kind of overview?  (It's like a company meeting here ;)
<warewolf> raka: Remember your PPlay intro (currently somewhere under PenguinDoc/) ?
<mtorni> warewolf: cool. just imagine the need to scan a few hour log to find out if we decided about something :)
<raka> warewolf: Yes, as shorter, updated version of that.  I'll start work on it.
<mtorni> how about having a break (like 12 hours at least) so we can all get some sleeeeep?`!
<slouken> Sounds good.  I'm looking forward to e-mail on this. :)
<raka> mtorni: No prey, I'm hungry.
<mtorni> slouken: your arguments were good. but my brain's too tired to argue efficiently.
<slouken> Is there anyone else on PPlay developers group?
<raka> mtorni: It's breakast time here.
<mtorni> raka: I noticed :)
<warewolf> slouken: who´s on whom?
<slouken> Are there lots more PPlay developers who didn't make it?
<mtorni> slouken: every hour is a bad hour for half of the people.
<raka> No joke.  72 hours notice next time.
<warewolf> slouken: Yep, at least Stephane wanted to be here...
<mtorni> raka: oh come on, more! Like 1 week so people can plan things up in advance.
<slouken> Cool... See youall later! :)
* slouken waves
<raka> OK, 72 hourse at least.
<warewolf> slouken: Cu
*** slouken has signed off (Leaving)
<warewolf> raka: Ok, agreed ;)
<warewolf> raka: what about envisioning next weekend?
<mtorni> warewolf: we need a table of people in different timezones
<raka> Next wekend sounds good.  For just PPlay, or the OS meeting?
<raka> mtorni: I about 21:00 GMT is OK for all but asians.  Do we have any asians?
<warewolf> raka: We should try  to contact the other OS guys till then. The sooner the better. I want to see the big meeting in about 1 month
<mtorni> raka: you cannot ask questions like that. Just how many people do you see on this channel right now?-)
<mtorni> we should as the commercial people just how capable they are in c++
<warewolf> 21:00 GMT is ok
<raka> mtorni: I mean on the PPLay list.
<mtorni> raka: 21.00 GMT is great.
<raka> warewolf:::: Ok, unless the list dissagrees.
<warewolf> raka: ´course
<mtorni> "ok, unless the list disappears"
<mtorni> you're right. It's about time. bye all
<warewolf> mtorni: I´ll post the summary of this meeting tonight, so the PPlay list will know soon
<raka> YOu still sleepy EMptySand, it must be 10 there now.
<warewolf> Ok, I´d better quit now, too
<warewolf> mtorni: Bye
*** mtorni has signed off (hopefully I can manage to bed)
<warewolf> <g>. Ok, I´ll sign off now. Have fun
<raka> bye.
Logfile closed at:  02.08.98  00:39:15

Prev by Date: Re: IRC meeting (was: Game developer tools...)
Next by Date: Re: File I/O and Image file loading
Prev by thread: Re: IRC meeting (was: Game developer tools...)
Next by thread: Re: File I/O and Image file loading
Index(es):
- Date
- Thread