new cube release... collect your stuff!

by Aardappel_ on 08/07/2005 22:23, 148 messages, last message: 09/18/2005 09:02, 122994 views, last view: 04/29/2024 09:08

So we haven't had a new cube release in a while, and I guess everyone understands why: sauerbraten is the engine of the future.

But Cube has its own maps, its own gameplay, and is worthy of its own spot for a while to come. So next weekend I will take a little break from sauer and my other projects, and work on a cube release. That doesn't mean that there will be a cube release next weekend, in the best case there will be, in the worst case a whole lot will get done and it will be the week after.

The bad news (if any) is the new cube will NOT receive any significant new features. Sauer is version 2.0 of Cube, and it simply does not make any sense spending large amounts of time in updating Cube. Work on Cube is to be focused on polish and bugs. The main reason for this release code wise it to fix the list of small bugs that have appeared since the last release.

So is this the last release of Cube? Maybe. I am not sure either way, there is always a possibility for more versions, but the way things are going with Sauerbraten, what I want for Cube is a very stable version that people can enjoy while we make Sauerbraten the greatest thing since, well, Cube.

So what do I want from the community? Well, 2 things will go into this release, new media, and bits of code.

For media, I simply want to collect the best released new maps and other items since the last release. I already have most of them, but if you want to be sure I don't miss any simply post links here in this thread... rememeber, just the most high quality stuff.

Then code. I know this is going to be hard for you because you all have your pet little feature that you want to see in Cube. If you read the above, you know these are not going to happen. Some things I just don't find important. Like probably the #1 requested feature is a configureable player model, but that is not going to happen for this release. These kinds of things will likely be in sauer. It is not that it is hard (rather, it trivial), it is just not a priority.

So I will list some things that I definitely intend to fix, or try to. You *MAY* use this thread to suggest certain additional fixes/features, and who knows I can be swayed if a lot of you make convincing arguments, but remember, unless it is a few line easy fix or just incredibly important to cube right now, its not likely going in.

I will report on this thread as things progress.

features already in:
* lots of smaller fixes that I didn't bother to write down. Maybe if you note them here I can say "yes that is already in"
* stereo sound effects
* put in workaround for buggy ATI drivers (glReadPixels() result, already on linux)
* tweaked kickback strength for various weapons

for sure planned to go in:
- seperate release archives for windows, linux, and macosx (less FAQs)
- fixes for amd64. if someone with such a machine can be on call in irc next weekend for testing, that be great
- fixes where illegal protocol could corrupt client state
- vdelta edit fix
- shooting rockets at HOM -> out of memory error
- jumppad velocity Z cancellation (better jumps)
- cap fps to 200 or whatever (variable) (some cube code starts behaving weird when it nears 1000).
- fixes for sound crashing on linux
- some minor tweaks to SP balance and AI
- listen server functionality will be removed
- demo play bugs like sticky scoreboard
- make serverinfo port obey -i
- maxclients

maybe will go in:
- some of the features requested by the german "pro" community (not sure which)
if they have CLEAR list of what they changed to fix what, that might help things
go in (you listening Makke?). Though I am guessing since I won't add all their stuff,
they may want to stick with pcube anyway.
- look at that immortality bug that sometimes pops up. I have a demo of it. It is however
hard to debug and I make no guarantees. Same with other "happens occasionally in
multiplayer" bugs people have.
- kick vote

   Board Index   

#1: Re:

by test___ on 08/07/2005 22:43

i just want to say that since cubes first release its done well keeping it simple and not bloating - i think enough time has passed for what people consider bloated to change. im just saying i think people can handle a little more code without performance problems.

and yes, cube is second in line to sauer, but more people havent made the transaction. i hope they do soon.

reply to this message

#2: Re: Re:

by Aardappel_ on 08/07/2005 22:57, refers to #1

"bloat" is not about machines not being able to handle it... I can make Cube's exe a 50x bigger (literally!) and noone's computer will care.

It's a philosophy, which comes with a lot of good things that you may not all understand if you're not a programmer, and that not all make (commercial) sense. But I don't care. The only reason Cube & Sauer are fun to me is because they are not bloated, thus fun to work on. I work on "feature rich" engines >10x the size of either one for a living, and its not always fun.

reply to this message

#3: added media to new cube release.

by MitaMAN in the cave on 08/08/2005 00:48

1. I would love to see all of DCP's mapmodels included.

2. Vanity on my part, if possible please include my last 2 SP maps: STRANGE UNIVERSE and THE ODYSSEY.

MitaMAN

reply to this message

#4: ..

by makkE on 08/08/2005 01:45

Heh good news! I will get drian to talk about discussing with you (aard) maybe what parts of his code might get integrated.

We perfectly understand not all of the pcube stuff will go in, but there´s definately some little stuff that could be worth adding.

I appreciate all the changes, mainly maxclients and kicking, jumppads and stereo sound :D
And fps-cap :D great!!
Hope kickback tweaking will be mostly getting the win behaviour on linux as well, otherwise I might have to tweak my trickmap map a little..

Anyways, I´ll mail you a pack of my (content) stuff I´d think worth to go into cube tomorrow or so.. I also got update versions of my prior maps, just cosmetics really...

reply to this message

#5: mapmodels?

by tentus on 08/08/2005 04:16

Personally I'd like to see a few of my mapmodels get included, but that's somewhat vain of me- I should be content for my stuff to be in CL. All the same, I get the feeling other people could use some of the ambient models I've made, specifically the food models i've cooked up *pun pun*. Also, some mappers could benefit from having some of my reduced-poly models to replace DCP's generally high-poly material.

reply to this message

#6: Re: mapmodels?

by dcp on 08/08/2005 05:55, refers to #5

if your 'reduced-poly models' are just modified and cut-down versions of my mapmodels, then i won't be really happy with that...

reply to this message

#7: some suggestions (should be easy to implement)

by CC_machine2 on 08/08/2005 10:31

suggestions for code features: (should be easy to implement)

-the ability to show a video file (for advanced storylines and such)
-acid/lava/deathlevel/painlevel/skybox hurts the player
-pusher entities (see cube legends thread)

and please dont ignore me again ppl :(

reply to this message

#8: ..

by >driAn<. on 08/08/2005 13:06

- some of the features requested by the german "pro" community (not sure which)
if they have CLEAR list of what they changed to fix what[..]

I could do separated patch files for each small fix/feature which might go in the next cube release. This way you (aard) can look at it very quick and decide if its worth to be included.

reply to this message

#9: ..

by kernowyon on 08/08/2005 13:37

The frustrating network server error which some people seem to experience - myself incuded. If that could be looked at? I think it has something to do with the networking of Cube - it doesn't happen on other games - e.g UT or Nexuiz.
On the 64 bit front - what sort of input do you need? I have a 64 bit machine and will happily test any release - but I am no coder!
No guarantee I can be in irc on Saturday - I think we have a distro team meeting then, but I will hang around if I can

reply to this message

#10: ..

by NiElZ on 08/08/2005 13:54

It would be really great if there was a feature to choose a different playermodel. I know we can download new ones and replace by the old ones, but that makes everyone look like that...

reply to this message

#11: ..

by >driAn<. on 08/08/2005 14:23

omg. Read first, then post.

"Like probably the #1 requested feature is a configureable player model, but that is not going to happen for this release."

reply to this message

#12: ..

by makkE on 08/08/2005 15:22

Just mailed you my stuff, Aard.

It´s the most up to date and smallest possible of my material since cube 2004.

reply to this message

#13: Re: mapmodels?

by tentus on 08/08/2005 15:24, refers to #6

no, they are original, but made so as to blend in with your models. I wouldn't dare rip your work, becuase i have too much respect for it. besides, that'd be harder than making original models :)

i'd really like a new entity to alter the postioning/ placement of mapmodels, something along these lines:
entalter T X Y Z
where T specifies if it alters position (0) or orientation (1). The rest follows logically.

but i suppose such an entity would fall under the sauerbraten realm, huh? oh well.

reply to this message

#14: just trying :)

by makkE on 08/08/2005 17:02

I will take to opportunity to try to convince you of some of my
biggest wishes for cube :)

1. Maxclients (already fullfilled, yay) I really appreciate it!! Let´s hope server admins will make use of it, so we
can see some reasonable number of players on the pubs instead of 20ppl on aard3 :P and some 1on1 stuff..etc..

2. Which leads to: a maprotation.cfg or something. It would make having a lan more easy, and it would go hand in hand
with the maxlient feature.. No use to have the big maps on a 1on1 server..
If it would also be possible to define a default mode for each map there, it would be even better.

It would just lead to more diversity! Most servers either run ffa or insta, and basically the first map cycle is

the only one played. It´s sad, because cube has so much more great maps to offer (and some of the first cycle are just crap, from a pure playing perspective, sorry to say.)

So, well to me, a configurable maprotation is the feature that I think would enhance the cube experience by a lot.
Imagine players can (or eben have to choose, if servers can be full) between 4 player ffa, 1on1, teamplay, or tactics
or whatever servers, with knowing the fun will be guaranteed more, because you play the maps more like they are
supposed to be played. Of course the admins might still choose stupid setups, but the possibilitys were just more.
I know there aren´t too many players total, so one could say it would cluster stuff up unneccesairily. But I believe
that many people stop playing cube, because it might be fun at first, but is actually pretty chaotic to pub on.

Servers with diffrent settings will attract diffrent people, they would probably meet more often, and it´s just more fun to play on "my favourite server" with familiar faces, than just going on any server not knowing what will happen.

Another big issue to me (and probably considered the most unneccesairy by you, but I´ll try anyways):

Teamplay

I remember at first I though "cool you can play 2vs2vs2 !!" but let´s face it: Has anyone ever seen a team game really
happen on a public server? No. Not me.

This is due a variety of reasons imho:

First, most people just don´t get it. If you see a server up saying "teamplay" and you enter, and there´s nothing
forcing you to choose a team, and it looks like ffa, 99% of the ppl will just shoot anything. The faint blue overlay

isn´t just too noticeable as well (if it´s your first templay game)
Sometimes (mainly those people that have voted it) will try to get the others to form and organize in teams, but it
takes soo much effort, that they probably give up on it sooner or later. Even if after 3 minutes of shouting people

have finally organized, some "blah unnamed" players will enter and ruin it again (which isn´t really their fault)

Another reason I believe is that many players use that extra space in their "/team" to add to their name, or to have
clan tag kind of stuff in there, and either feel reluctant to change that, or others might even fear to offend anyone
else by changing to their team/clan(?)

So to me, forcing people to chose RED or BLUE, in addition to those really nice red/blue skins that exsist for the
ogro is the only solution to get teamplay a reality on cube pubs, imho.
Don´t get me wrong, the current teamplay system is great, since it is really simple and offers a lot of freedom, but
also asks an amount of responsibility you will hardly find on a pub.. you could argue that people can meet on some own passworded server to have their team-games anytime, but having it happen on pubs will imho just add a lot to cube as a whole, because it´s where you start and get intrested in the game at first.

To sum up, I know that you have a diffrent view on some of these issues, and there is of course pcube where we can and already do some of the mentioned stuff. Well but it also means splitting the comunity in a way.
Even if you don´t think it´d be really neccessairy in cube, I would be glad to see you consider features like this for cube2, because, i do believe that the Teamplay issue, the lack of configurability of servers and also the voting system (to a small extend) have somehow just keept cube from growing to it´s full potential.

I understand cube was totally your own property, but for cube2, I really hope that there will be space for just some
more overall "usability" (not talking about big shiny menus and stuff). Just making it feel a little more comfortable

These are as far as it goes the really most important bits to me. In pcube there´s a long idea list, but most of it
isn´t really top-priority. The fixes that you planned will already make a lot of the "pro"s very happy :)

Well, that´s it :) just my 2 cents.

reply to this message

#15: ..

by NiElZ on 08/08/2005 19:24

For mappers: it would be really nice if we could use a light radius higher than 32, I mean, it works when editing a map but when you save the map then and re-load the map, all the lights with a radius higher than 32 will have changed into a radius 32 like automatically, this prevents from creating nice shadow effects as seen here:

www.geocities.com/illegalnameshrimp/goodlight1.jpg
or here: www.geocities.com/illegalnameshrimp/goodlight2.jpg

This screenshots were taken before re-loading the map.

But because Cube won't allow the big lights to stay, for some reason, I had to light that area with small lights resulting in a boring effect like this:

www.geocities.com/illegalnameshrimp/badlight.jpg


So if we will be able to a use high light radius in some future release that would make me really happy. I know some might say: "use Suaerbraten for that", but I'm talking about Cube here :P

reply to this message

#16: ( Link fix )

by NiElZ on 08/08/2005 19:26

Good light:
http://www.geocities.com/illegalnameshrimp/goodlight1.jpg
http://www.geocities.com/illegalnameshrimp/goodlight2.jpg

Bad light: http://www.geocities.com/illegalnameshrimp/badlight.jpg

reply to this message

#17: ..

by makkE on 08/08/2005 19:35

The thing you´re doing there is not how cube´s lightning was meant to work.
You just can´t have those shadows really, because it means everything else will be awfully overbright...

If you want nice shadows, you have to stick with sauerbraten.

In cube it´s usually better to work with many smaller lightsources(radius 2,3,4..etc) to get some contrast in, it´s all more of ambient and diffuse...
Lightning in cube isn´t about any realism, really, more about to make it look good while providing contrast..

reply to this message

#18: ..

by _Fanatic on 08/09/2005 05:44

My Revenge map for Cube would be a descent addition. Uses DCP's awesome mapmpdels.

http://fanaticalproductions.net/game_map_view.htm?mapId=22

reply to this message

#19: Admin Controls

by tyguy on 08/09/2005 07:02

Maybe some admin controls,
/kick (nick)
/ban (nick)
/map (name)

just to make it easier to get rid of trouble makers. and let it go stright to a map instead of waiting for everyone to vote. Most of the time ppl dont even know how to vote.

reply to this message

#20: Re: Admin Controls

by mw on 08/09/2005 07:28, refers to #20

Why?

reply to this message

   Board Index