Why we need to move to a client we can control

Iceball is a moddable, open source, cross-platform engine and game inspired by AoS Classic. Created by GreaseMonkey.
Incompatible with PySnip/pyspades-based AoS Classic 0.75/0.76 servers.
32 posts Page 1 of 3 First unread post
GreaseMonkey
Coder
Coder
Posts: 733
Joined: Tue Oct 30, 2012 11:07 pm


So we're about to hit the 1st Anniversary of a Shitty and Unmaintained Version of Ace of Spades, namely 0.75.

In terms of pysnip, we could fix the SMG damage values so they're not complete bullshit, but noone can be fucked doing it. Furthermore, a_girl apparently needs to spend time fixing the antihack scripts if someone CAN be fucked doing it.

Anyhow, I've basically figured out that most of my frustration, and the frustration of the other admins, can be explained by the simple fact that we've been trying to support a broken game where nobody { can be fucked / has the skills to / wants to be sued for / (most importantly) can be fucked preparing everyone else for } modifying the client to make it less shit. I did do a client mod (read: blatant hack) which essentially disabled the spread on the rifle, but that hasn't gone down well with people in authority, and the community cannot be fucked making it acceptable.

There are several things which are just plain unpleasant.
  • Maciej Kusnierz happened. Now we're having to deal with shit we could easily fix by bumping out a new version a week.
  • To counteract Maciej Kusnierz, we also have a reason to /togglesmg aside from the fact that it's fucking annoying, and that is because the antihack we have tends to give off quite a bit of fuzz when someone SMGs on a bad connection.
  • Furthermore, not only is the SMG is fucking annoying, but all the weapons need rebalancing.
  • pysnip is not the nicest thing to code for, which may explain why nobody can be fucked, despite the fact that it's actually pretty easy to change the weapon damage values hint fucking hint. This may also explain why votekick has been shit since 0.54. Also, how long did it take for us to get togglevotekick? Way too long.
  • EVERYONE IS JUST SO FUCKING STUPID I'M SICK OF CANCELLING VOTEKICKS BY SMG NOOBS GETTING HEADSHOT BY RIFLE USERS SOMEONE EXPLAIN THE IMPORTANT SHIT TO THESE NOOBS ALFUCKINGREADY
  • tem br?
  • algeum br?
  • tem skype?
  • eu
  • no ingles
  • FILHO DE PUTA
  • No, BRs aren't the #1 problem with this game, it's just that you can't explain how fucking annoying it is, reading through what your mind eventually registers as spam, using a single line.
  • Where the FUCK is that global whitelist / account system we really need?
In short, nobody can be fucked. We're trying to get players to join a game that's already broken.

We need to shift to a client that we, the community, can control.

I tried to not make this a blatant ad for Iceball, but that's all we have at the moment. It's a very good candidate for this task, as server owners can run a completely different game if they want to, and clients don't need to update every single time the server does a drastically huge change to the game logic. Finally, if the game needs ANY minor tweaks, INCLUDING on the client side, they can be done with ease, followed by a server restart.

(Oh yeah, Iceball needs a wiki, because there's just some stuff I'm not going to be able to maintain. docs/modding_lua.txt and docs/proto_main.txt are the only things that get maintained, really.)

I am aware that people are finding it difficult to get working. I do want people to read a manual to get it working, because then we can tell them important shit like "there is no such thing as subterfuge", and also to filter out those who are on a constant sugar high and think they're playing "Grief Grief Revolution". (If you don't want to filter these guys out, that's fine, give them a .bat file to run and disable the kick_on_join check, and watch your server turn to shit.) I'm not willing to drop the "read the fucking manual" check, but I am willing to help, provided you've at least made a decent attempt to read stuff, and that you've also made a decent attempt to explain the problem.

I don't want to see this dragged out. The issue I have with waiting for Sonarpulse to finish his VOXLAP port is that we would have wasted months waiting. We need to move away from 0.75 NOW, and I'm not overly concerned if we shift to 0.76 for the time being, but I'm once again getting sick of admining 0.75.

We don't need to cold turkey this - in fact, it would be best to keep playing 0.75 for the next few weeks - but we do need to start migrating NOW.

If you disagree, you need to fulfil several fucking old and in many cases fucking simple feature requests for pysnip NOW, because I am NOT going to be putting up with the rot for much longer.
Sonarpulse
Coder
Coder
Posts: 443
Joined: Thu Dec 13, 2012 7:18 pm


GreaseMonkey wrote:
The issue I have with waiting for Sonarpulse to finish his VOXLAP port is that we would have wasted months waiting.
Well I have tried to be very forthcoming that now that school has started I have almost no time to work on this. On the other hand, as I have said multiple times, nobody needs to wait for me to work on a new client. Cause you can build off Voxlap on windows right now and get your thing ported for free! (Unless you are jagex, the only people to start with cross-platform OpenGL and end up with windows only).

Regardless, making a new client is hard, and would take at least the time it iceball has taken to this point, so I support pushing iceball as AoS's successor. However Grease I am afraid to say that a mass migration is less likely when both server operators and players got to install something new, the the client interface is specifically designed to keep out the noobs.

Now I wholly support your efforts to force people off the opiate of the massesGUI and back to the terminal, and fighting the dumming down of technology we have eyewitnesses in recent years. Which is why to compensate I strongly believe we need to make iceball compatible with pysnip some way or another to compensate.
Triplefox
Scripter
Scripter
Posts: 32
Joined: Thu Nov 22, 2012 5:28 am


Iceball won't be a useful replacement client until you can say with confidence, "we will support the little kids too."

As it is, we've developed a strong anti-noob culture, and I've taken to thinking that the project will just die and I might as well not bother, especially since the dev environment(code, build, debug - not the people, the people were OK) was leaving me unhappy 80% of the time or more. I even regret most of the code I wrote in it(mostly due to my own mindset about what was really necessary for the project at that moment).

In any case, for me, momentum and interest has been lost because the project has refused to go beyond tech demo, and that really rests on your marketing decisions, GM - which started with the idea that you would do everything from scratch, be protocol-incompatible, use a custom renderer, and promote a reactionary, in-jokey "politics of anti-SMG" from day 1. Zero of those things are things that excite and delight potential new players and get them to download the client. Zero.

I'll be happy to document stuff as requested, or maybe fix a bug, but I don't want to go back and start coding new stuff again, not unless things change radically. I'm not playing these days, and I have lots of other interesting stuff to work on...
Jdrew
Mapper
Mapper
Posts: 4808
Joined: Tue Oct 30, 2012 10:48 pm


Stack has went into this before. You made a whole another game like ace of spades ,pretty much re did everything, including networking, and even made it open source. Stacks ideas, lest what I got from it, is why not make a client compatible with everything already made? Iceball supports .vxl and .Kv6 so why not just re do the networking with pyspades?
I know this isn't an easy task but it would allow users coming from AoS to feel at home and at the same time allow alot of cool scripts to be made such as the stuff planned in alpha 1.0 like the file checker.
Tiel
Deuce
Posts: 14
Joined: Mon Feb 25, 2013 4:51 pm


You'll forgive my asking, but what's wrong with SMGs? Sure, they're bullethoses, but without them everybody would be using a rifle ingame. And that ain't fun.

At least for the time being newcomers immediately grab the noobgun and get picked off by riflemen. Green_Rifle
HoboHob
Winter Celebration 2013
Winter Celebration 2013
Posts: 979
Joined: Mon Nov 05, 2012 5:02 pm


Tiel wrote:
You'll forgive my asking, but what's wrong with SMGs? Sure, they're bullethoses, but without them everybody would be using a rifle ingame. And that ain't fun.

At least for the time being newcomers immediately grab the noobgun and get picked off by riflemen. Green_Rifle
There's nothing inherently wrong with the SMG. It's just that it's a really hard gun to balance.
rakiru
Coder
Coder
Posts: 1349
Joined: Sun Nov 11, 2012 12:26 pm


I won't reply to the main post yet, as I've got to rush out in a minute, but I have to reply to this:
jdrew wrote:
why not just re do the networking with pyspades?
"just"? While ENet could just be patched in, making the rest of the game work with that is nowhere near that easy. For a start, the game is not a 1:1 model of AoS, which it would need to be made to be. It would also need to be made to work with the new networking. The Lua code would have to be changed to be read locally, which is easy enough, but this would make it ridiculously simple to write hacks for. Basically, we would have to add ENet to the Iceball engine (the C side of the game), it would probably be quicker writing the Lua code from scratch, and it would be much easier to write hacks. Those are pretty big reasons not to do this.
danhezee
Former Admin / Co-founder
Former Admin / Co-founder
Posts: 1710
Joined: Wed Oct 03, 2012 12:09 am


Grease you are an excellent programmer and a pretty cool guy to hang out with. I am very impressed with the technical accomplishments of Iceball. But there are design choices that arent technical that are turning people away.

Namely, kickonjoin and what can at times be very belittling help when someone ask why they cant join. That has probably ran off more than a few artists that could have very easily spruced up your in game art assets. Which in turn would bring even more people to the game.

You arent compatible with pyspades/pysnip, it is hard to be a replacement client if you arent compatible with the server. It is it own game and we have to treat it as such.

Also your anti-smg stance led me to believe you arent open to suggestions, so I havent been providing them.

You have been grinding on this project for a few months now, I think that has stressed you out. Before there is a complete burnout slow it down some. Think about what the project is. Is it a client or a new game and go from there. I dont think it can be both. And whatever you decide, I recommend you remove as many barriers to entry as possible. Keep It Simple Stupid, Dont Make Me Think.
rakiru
Coder
Coder
Posts: 1349
Joined: Sun Nov 11, 2012 12:26 pm


Ok, back from uni, so here's my post:

A client we control would be a brilliant thing. I was planning on creating one a while back, but lacked the time, and probably skill/knowledge, that would be required to create one.

As for Iceball being a replacement, I have two main points:
  • Iceball is a completely separate game, which would require a lot of effort and cause a lot of hassle for everyone involved, server owners and players. The main issues this would cause are porting existing scripts over, and getting people/servers to use the new version. The first one is, for most scripts, just a matter of finding someone willing to rewrite the old scripts, with the more advanced scripts either being left behind or being entirely redesigned and rewritten (partially due to the fact that writing for pyspades is horrible). As for getting people to switch, the main problem is that not everyone is going to update at the same time, and even then, there are will be people who never update, and when the game suddenly stops working, won't bother looking into it. One way to help those in the know would be to create a server browser that uses the Build and Shoot server master server list and the Iceball one and launches the correct game for the selected server. This would greatly help adoption rate, but not completely fix it (although if they don't care enough to google or ask on IRC or a relevant forum, that's their own damn fault).
  • The whole anti-noob, anti-SMG approach you have. While, yes, you have some valid points, no-one's going to want to switch to Iceball if it suddenly means a large part of the playerbase isn't switching with them, and others aren't going to want to switch because it means they don't get to use their favourite gun, the SMG (or shotgun, as it stands now). I've defended your stance several times on IRC, but as development moves forward and the game gets closer to being complete (it's playable and fun as it is), I'm finding it harder to do so, as well as not really wanting to. Basically, if you want people to use Iceball, you have to support the idiots too, not just the techies and those who will read of their own free will.
Edit: Sorry for the formatting - I typed this up in Notepad.
Sonarpulse
Coder
Coder
Posts: 443
Joined: Thu Dec 13, 2012 7:18 pm


From a technical perspective there is argument that mixing in GUI code makes the game harder to port. I think the best approach would be that of mupen64plus (a popular n64 emulator). The core functionality of the game is actually a shared library, and various console and GUI front ends just link with it. This not only separates potentially less portable GUI code, but can also create a more integrated experience than spadille, as everything is still one process.

Yes, pysnip countability is going to be hard to achieve, but I don't think there is any other way. Remember this is essentially a stop gap, "native" servers will of course be much more feature-rich than pysnip ones, and once enough people move over pysnip will probably go away. So yes, the implementation doesn't have to be perfect.

I think somebody should pick off where rakiru left off and start documenting the AoS protocol immediately. I would, but again I am quite busy and know next to nothing about pysnip.
rakiru
Coder
Coder
Posts: 1349
Joined: Sun Nov 11, 2012 12:26 pm


The protocol documentation that Sonarpulse mentioned is available here btw, and if someone does want to take on the task of documenting it, hit me up on IRC so I can give you some info.
Jdrew
Mapper
Mapper
Posts: 4808
Joined: Tue Oct 30, 2012 10:48 pm


Although you could easily right hacks wouldn't it be even easier to wright an anti hack with it being open source?
Articsledder
Deuced Up
Posts: 209
Joined: Mon Nov 12, 2012 5:13 pm


I'm all for moving clients, but said client is going to need some kind of appeal.

When AoS was first released it quite literally one of a kind, the only comparison was Minecraft (which is completely different). This is why Ben got away with having such a simple game (similar to Iceball now)

Thus, noobs/little kids were forced to put up with the Rifle only thing and autoclimb. As soon as we got the SMG, they got a taste of there quick paced shooter of their dreams, and quickly became the majority of players.

Now we have multiple (up to 5 I think) possible sources of competition in this "Genre", so rifles only simple maps won't cut it.

----------------------------------------

What we should do in my opinion is fully flesh out the WW1 thing which was originally planned
Image
(supposedly Ben posted this image on the Development blog a week or 2 after release)


This would differentiate us from the other build and shoot type games out there, bringing in the older voxlap players, as well as possibly bringing in noobs with the HMG (or shotgun, which fits the WW1 theme)
rakiru
Coder
Coder
Posts: 1349
Joined: Sun Nov 11, 2012 12:26 pm


jdrew wrote:
Although you could easily right hacks wouldn't it be even easier to wright an anti hack with it being open source?
You managed to spell "write" wrongly two different ways in one sentence - gg.

No, pyspades is already opensource, which is the only part you can write hack detection for.
Articsledder wrote:
I'm all for moving clients, but said client is going to need some kind of appeal.

When AoS was first released it quite literally one of a kind, the only comparison was Minecraft (which is completely different). This is why Ben got away with having such a simple game (similar to Iceball now)

Thus, noobs/little kids were forced to put up with the Rifle only thing and autoclimb. As soon as we got the SMG, they got a taste of there quick paced shooter of their dreams, and quickly became the majority of players.

Now we have multiple (up to 5 I think) possible sources of competition in this "Genre", so rifles only simple maps won't cut it.

----------------------------------------

What we should do in my opinion is fully flesh out the WW1 thing which was originally planned
Image
(supposedly Ben posted this image on the Development blog a week or 2 after release)


This would differentiate us from the other build and shoot type games out there, bringing in the older voxlap players, as well as possibly bringing in noobs with the HMG (or shotgun, which fits the WW1 theme)
We're talking about a new client, not a new game. This new client would be the same in functionality as the current one and work with current servers. It may be updated later on, but that's not necessarily going to happen.
Jdrew
Mapper
Mapper
Posts: 4808
Joined: Tue Oct 30, 2012 10:48 pm


rakiru wrote:
Although you could easily right hacks wouldn't it be even easier to wright an anti hack with it being open source?
You managed to spell "write" wrongly two different ways in one sentence - gg.

No, pyspades is already opensource, which is the only part you can write hack detection for.

huh I guess I did, I thought you could wright an anti hack for the client or server itself but never mind then
32 posts Page 1 of 3 First unread post
Return to “Iceball”

Who is online

Users browsing this forum: No registered users and 7 guests