0.2.8 (beta 3 tagged)

[Z-Man]

Well, at least the two problems mentioned by nemo are probably generic and not axis-code specific.

Phasing:
I have a rough idea what is causing this, it's the imperfect collision with walls not yet put into the grid datastructure. Unforunately, I don't have too much of a clue how to work around this without ill side effects. Here philippe is right: much hard work is to do.

Hole in wall:
The cycle's walls are only updated in very few places. It should be a piece of cake for me to check them all. I recall that the detatching of the last wall and the buildup of the new wall after a turn are done in different places, it probably is not guaranteed that the new wall is created before the next turn.

Walls coming out at odd angles:
Probably the same problem that causes walls to be invisible with four axes. Same cause, different effect. Probably hard work to find, too.

All those bugs are hard to fix because they are hard to reproduce. I'm currently working on recording and playback of game sessions. It will work at the lowest possible level, raw network traffic and raw input are recorded and played back. This should enable us to debug these and other sporadic problems much more easily.

We'll release a true version when it's done and free of known larger bugs. Everything we publish before will be marked clearly as public test version. Luke suggested a date, it's not a deadline. We don't have deadlines as they don't make sense on a project where the time investment of the individual project members varies.
The only deadlines that make sense for us are feature freeze deadlines. We could have one for big features (like the axis or arena code) before a major release and one for small features (new configuration items) before each minor and major release. Of course, we'd have to determine what minor and major releases are and how we handle them in CVS (Do the release in a branch? Do the additions that don't belong to the release in branches?).

[Luke-Jr]

I think there are too many redering and game problems(*) appearing because of the axis/map code. I feel those are pretty much show stoppers for any kind of release.

Actually, I don't see these as being a result of the axis/map code, but simply easier to reproduce. As nemo pointed out, these problems still exist in older versions, just slightly rarer.

If this cant be fixed for the deadline set by luke,

I said *preferably* by June/early June. Deadlines don't generally work for programs.

I propose that it (axis/map) be moved out of the release.

Remove the main feature?

I dont want people (inclusive of the aa crowd out there that doesnt follow this forum) to have a negative perception of the whole map/axis idea as my goal is to push it (whole world concept) forward, not to have to stop to argue and defend every step of it. Enough time already went on that.

There is yet no UI for maps. Even if the bugs could not be fixed (if they can't, we have bigger issues, IMO), there is no reason to disable the code. "Broken" map support is better than no map support, anyway.

I feel it will require much hard work to hammer those problems out of the code, as the whole axis problem seems to break many of the reasonable assumptions that z-man used for the 4 axes version. Much digging in the code for my part so I can go and put new unreasonable assuptions .

I suspect the "rendering" problem nemo pointed out may simply be the network code getting more information (turns) than it is used to putting out... though that doesn't explain why people can go through it.
But then again, the going-through-walls bug exists even with original axes/map. Since the new maps make it easier to reproduce, it should also become easier to troubleshoot.

I understand that many of you will be dissapointed if the axis/map code doesnt make the release. But understand my desire to give you something that wont cause more problems than it is worth.

I don't think there's enough changes (leaving out shaped arenas) to warrant a 0.2.8 release. Most of the other stuff is all either fairly small improvements or only really affects dedicated servers.

(*) Problems that I know of:
-nemo's hole in the wall

Which is a bug that exists without maps, as far as I can tell.

-traces coming at angles to the cycle instead of right from behind.

Never seen this... Are you sure it's not lag?

[Luke-Jr]

The only deadlines that make sense for us are feature freeze deadlines. We could have one for big features (like the axis or arena code) before a major release and one for small features (new configuration items) before each minor and major release.

However, if we consider shaped arenas as a "big feature", wouldn't that suggest it should be a 0.3 release (and pushing worlds off to 0.4)? Does 0.2.8 or 0.3 seem more appropriate for the next release?

Of course, we'd have to determine what minor and major releases are and how we handle them in CVS (Do the release in a branch? Do the additions that don't belong to the release in branches?).

Releases usually branch off of HEAD. I think there are a number of benefits to this... mainly, it allows bugfix releases to older versions even after development of a new version has started.

[Lucifer]

I feel it will require much hard work to hammer those problems out of the code, as the whole axis problem seems to break many of the reasonable assumptions that z-man used for the 4 axes version. Much digging in the code for my part so I can go and put new unreasonable assuptions .

I suspect the "rendering" problem nemo pointed out may simply be the network code getting more information (turns) than it is used to putting out... though that doesn't explain why people can go through it.
But then again, the going-through-walls bug exists even with original axes/map. Since the new maps make it easier to reproduce, it should also become easier to troubleshoot.

Two things about this. The old bounce bug (that's fixed, right?) caused the same hole in the wall. In any case, if it's invisiwall or the old bounce bug incarnated in a new and weirder form, people are going to be pissed if these two bugs are still around. Consider that when z-man was gone, people kept playing even with all the same bugs never being fixed. The 0.2.7.0 release was successful (near as I can tell) because of the HUD and ingame admin, other than that many of the old bugs seemed to become worse (for whatever reason, I don't know, I don't know how much of the code was touched for 0.2.7.0). 0.2.7.1 is still somewhat controversial in the community, at least as a server, because significant changes to gameplay happened. A new release needs to have old bugs fixed (not new bugs fixed and old bugs left alone, most of us couldn't care less how many new bugs get fixed as long as many of the old bugs get fixed) and preferably significantly new features added. If no new features are added, gameplay should stay the same. If gameplay changes, there has to be a good reason for it, such as new features being added, or old bugs being fixed (the main reason I don't complain too much when people 180 a bit, I'm *happy* there's no more bounce, and I'm *happy* you can't rip the server).

Second, the going through walls bug has gotten even worse in 0.2.7.1. In older versions you had to really work at it (it may be two separate bugs that look the same ingame) and I never managed to get the hang of it. Didn't try that hard, really, but I saw others who had mastered it so they could go through walls anytime. But now in a regular fight, I'm going through walls at least once or twice in a 10 minute period, and I'm seeing lots of other people doing it. And they're not just doing it, they're doing it, and then saying "This server's acting weird, I'll come back when it's not" and leaving. What, they'll come back when I upgrade?

So they can complain about gameplay changes, old bugs, and no new features?

[philippeqc]

Ok, it might not be the source of the problem, but rather expose it a lot more.

I propose that it (axis/map) be moved out of the release.

Remove the main feature?

Why not. It will be 0.2.7.3 instead. I dont see why this should a problem.

There is yet no UI for maps. Even if the bugs could not be fixed (if they can't, we have bigger issues, IMO), there is no reason to disable the code. "Broken" map support is better than no map support, anyway.

Nooooooooooooooooooooooo!

For someone commenting on others choice of OS with sentences such as "... on your broken OS...", I would have never thought you'd propose to release anything "broken".

Also, your argument is flaud. Even without UI configuration, there is still the possibility ofzealous persons using the console and everybody else being exposed to the problem through Internet servers.

(with the axis/map generating more pass through wall errors) ...it should also become easier to troubleshoot.

Thats a good thing, I agree, if it lead to the problem being fixed before any release.

The only deadlines that make sense for us are feature freeze deadlines. We could have one for big features (like the axis or arena code) before a major release and one for small features (new configuration items) before each minor and major release.

However, if we consider shaped arenas as a "big feature", wouldn't that suggest it should be a 0.3 release (and pushing worlds off to 0.4)? Does 0.2.8 or 0.3 seem more appropriate for the next release?

Please Luke, I understand you enthousiasm, but calm down just a bit. Map is just a mecanism I need for world. So are connectors and borders. I just decided to expose them as steps to keep me motivated in the face of this challenge. I'll tell you when the whole will truly deserve the name of "feature" and be worth a major number change.

I don't think there's enough changes (leaving out shaped arenas) to warrant a 0.2.8 release. Most of the other stuff is all either fairly small improvements or only really affects dedicated servers.

As I said before, why not 0.2.7.3 then. But I'll try not to debate the numbering scheme and leave that to z-man and other, smarter than me, people.

-traces coming at angles to the cycle instead of right from behind.

Never seen this... Are you sure it's not lag?

Seen it when I was playing agains you on HexaTRON's map and possibly inaktek. And IIRC, the other person playing with us had the same situation.

I'll reformulate my request to, maybe, express my desires a bit more presicely. While the problems arent map/axis dependant, they appear more evidently with it, and in people not involved in this forum, a negative association can be easily created. As Lucifer pointed, they are easier made than disbanned. AnThat is something I do not want.

I'll be adding enough weirdness with the whole world idea, that I'd like to be able to do it in a way that doesnt give it extra negative publicity.

I might have read too much in your eagerness. And if so I'm sorry of my mistake. But I'd like for you to accept my desire to release only when nobody will be able to consider it "broken".

-ph

[Luke-Jr]

I propose that it (axis/map) be moved out of the release.

Remove the main feature?

Why not. It will be 0.2.7.3 instead. I dont see why this should a problem.

0.2.7.3 should not have any more features than 0.2.7; it should be bugfix-only

There is yet no UI for maps. Even if the bugs could not be fixed (if they can't, we have bigger issues, IMO), there is no reason to disable the code. "Broken" map support is better than no map support, anyway.

Nooooooooooooooooooooooo!

For someone commenting on others choice of OS with sentences such as "... on your broken OS...", I would have never thought you'd propose to release anything "broken".

Solong as the feature is not revealed in the game itself nor 'advertised' as a feature, it would be "undefined behaviour" and better off *trying* to work instead of *failing* to work outright.

Also, your argument is flaud. Even without UI configuration, there is still the possibility ofzealous persons using the console and everybody else being exposed to the problem through Internet servers.

People can do this anyway, and players would end up seeing a totally incorrect arena like they do now.

The only deadlines that make sense for us are feature freeze deadlines. We could have one for big features (like the axis or arena code) before a major release and one for small features (new configuration items) before each minor and major release.

However, if we consider shaped arenas as a "big feature", wouldn't that suggest it should be a 0.3 release (and pushing worlds off to 0.4)? Does 0.2.8 or 0.3 seem more appropriate for the next release?

Please Luke, I understand you enthousiasm, but calm down just a bit. Map is just a mecanism I need for world. So are connectors and borders. I just decided to expose them as steps to keep me motivated in the face of this challenge. I'll tell you when the whole will truly deserve the name of "feature" and be worth a major number change.

Maps are *not* just a means to the end of world. Maps are a significant and useful feature on their own. Connectors are also, but not quite as much.

I don't think there's enough changes (leaving out shaped arenas) to warrant a 0.2.8 release. Most of the other stuff is all either fairly small improvements or only really affects dedicated servers.

As I said before, why not 0.2.7.3 then. But I'll try not to debate the numbering scheme and leave that to z-man and other, smarter than me, people.

Have there even been any bugfixes? If so, perhaps someone should go about trying to find the 0.2.7.1 release versions in CVS and create a 0.2.7 branch with which to apply the bugfixes and release 0.2.7.3.

-traces coming at angles to the cycle instead of right from behind.

Never seen this... Are you sure it's not lag?

Seen it when I was playing agains you on HexaTRON's map and possibly inaktek. And IIRC, the other person playing with us had the same situation.

I didn't. This highly suggests it is a result of lag, since both of you were internet connections and I was localhost.

I might have read too much in your eagerness. And if so I'm sorry of my mistake. But I'd like for you to accept my desire to release only when nobody will be able to consider it "broken".

Hence why 0.2.8/0.3 should probably not be released until the 'big' bugs are fixed.

[microbus]

Luke, old buddy, old pal lol...
any way to have 2.8.0 use less CPU??
I was told I should be paying four times the price for
my server, because at times, when microbuscity.com's
Arma server is full, it's using nearly half of the CPU of the
main server, which also runs SoF2 and other games.
Not quite sure why it would, but it does. It spikes several
times a day. Luckily, my son owns part of the server company,
so my price is locked in :D

[Lucifer]

Ack, you guys missed me.

I'll simplify.

1. If the map support is broken and there are many old bugs, people will be pissed ad the community will shrink substantially.

2. If the map support is left out ad there are still many old bugs, same thing only less shrinkage.

3. If the map support is left out and there are still many old bugs and some new bugs, same as #1.

4. If the map support is left out and there are still many old bugs and gameplay changes again, same as #1.

5. ..... see the pattern here?

What's absolutely most important is that the bugs we've been warping our play to compensate for for the last, I don't know, 3-4 releases?, need to be fixed. People are already dropping off because there's been two releases since the fork that still have all the same old bugs, changed gameplay, and didn't offer "enough" new features (the HUD is arguable there, but how many players have *you* spoken to that say they prefer 0.2.6?).

What are those bugs?

1. invisiwalls
2. jumping (used to be sliding, may not actually be a bug)
3. going through walls
4. lag when there are lots of turns made (for example through double-binding)

New problems?

1. 0.2.7.1 makes it easier for people to just suck on the wall. It's adding time to every round and people stuck in spectator mode are leaving. Maybe they're going and meeting women, drinking beer, and generally doing life-fulfilling activities now.

2. 0.2.7.1 doesn't slide anymore, it just bounces. The nice thing about the slide was that your brain could at least track the changes while they were happening. The bounces leave you confused.

If Luke's map support brings these bugs out and throws them in the open, you now have a double-edged sword. To release the game with those bugs still in the game with the map support doing such a wonderful job of making them worse, well, the rest of this sentence is a threat that I won't finish.

So, two choices.

1. Fix all these old bugs that people are already leaving over
2. Don't fix them and continue to stagnate.

Whether or not map support is ready for release is orthogonal to the underlying issues you guys are discussing.

[Luke-Jr]

I'll simplify.

5. ..... see the pattern here?

You failed to list the reality of the situation... Map support is *not* in any way broken.

What's absolutely most important is that the bugs we've been warping our play to compensate for for the last, I don't know, 3-4 releases?, need to be fixed.

The only gameplay-related bug I'm aware of (360s not working properly) has long since been fixed. Tiger's network (and some others) still run the old buggy servers, and MBC runs one of the working servers.

People are already dropping off because there's been two releases since the fork that still have all the same old bugs, changed gameplay, and didn't offer "enough" new features (the HUD is arguable there,

Don't know what bugs you're referring to... shaped arenas will be the first real change in gameplay since 0.2 AFAIK...

but how many players have *you* spoken to that say they prefer 0.2.6?).

None. I've spoken to one player who still uses 0.2.6 because the newer versions did not work on their computer (OSX w/o Admin permissions), and that is a bug being fixed with the next release.

1. invisiwalls

Never heard of it?

2. jumping (used to be sliding, may not actually be a bug)

Lag?

3. going through walls

This doesn't seem to be very common and thus not very annoying...

4. lag when there are lots of turns made (for example through double-binding)

There's not a whole lot the game can do when there is insufficient bandwidth available...

1. 0.2.7.1 makes it easier for people to just suck on the wall. It's adding time to every round and people stuck in spectator mode are leaving.

I'm guilty of enjoying this feature.
Arenas such as 40-gon (circle) easily prevent such abuse.

2. 0.2.7.1 doesn't slide anymore, it just bounces. The nice thing about the slide was that your brain could at least track the changes while they were happening. The bounces leave you confused.

Is this the weird lag-like bug I posted on? z-man suspects it might be related to texture loading, I think...

[Luke-Jr]

Luke, old buddy, old pal lol...
any way to have 2.8.0 use less CPU??

um... *hides behind z-man or whoever's area this is*

I was told I should be paying four times the price for my server, because at times, when microbuscity.com's Arma server is full, it's using nearly half of the CPU of the main server, which also runs SoF2 and other games.

Slow CPU? Heh... that would explain why MBC's been so hideously laggy lately

Not quite sure why it would, but it does. It spikes several times a day.

Anything changed recently? MBC didn't have these problems before, from what i can tell as a player...

[Z-Man]

Luke: you're in bug denial state. Those bugs exists, and they are important to players. That's what counts. Whether you see them or not or how much they bug you is of no importance. I know what I'm talking about, I was in that state a long time.
You're right about the maps and downloads not beeing proken; however, explain that to the users when map support exposes all of my bugs. It won't matter to them.
You're absolutely right about the CVS branching. Nothing to add. The release branch of 0.2.7.1 is still around and could be used as a basis for 0.2.7.2, but none of the bugfixes since 0.2.7.1 were added there and some of the code has evolved beyond the point where automatically merging changes will work.

Lucifer: fixing old bugs always is my top priority; sometimes, however, larger changes need to be done to make the bugs fixable in a way that they won't come back with the next change. Hence the base network code refactoring and (partly) the cycle mechanics code refactoring and the (completely unintentional) gameplay changes. The recording and replay code will hopefully enable me to find obscure network play problems much, much faster, that's why I'm working on it now.

About the relase number: As stated in some other place, my basic requirement for release numbers were
1. that 1.0 should be the first complete release
2. that you can easily read off the network compatibility off the version number

2. lead to sticking to 0.2.x version numbers since they all were network compatible. Nobody says, of course, that 0.2 can't be compatible with 0.3, just the other way round should work.

1. is, as I know now, completely silly. There's never going to be a complete release, unless the project dies. So, if you want, we can define a basic catalogue of stuff we want in something called 1.0 (full customizability, for example) for the midterm future and go for it.

We should, however, outsource this disussion into another thread.

Microbus, about the performance: There was a huge performance bug in the network code, mainly affecting servers, making the CPU load of that particular part about 100 times bigger than required. It's fixed now.

[Lucifer]

You failed to list the reality of the situation... Map support is *not* in any way broken.

YOu must have missed the part where I started arguing in favor of the map thing.

The only gameplay-related bug I'm aware of (360s not working properly) has long since been fixed. Tiger's network (and some others) still run the old buggy servers, and MBC runs one of the working servers.

Um, you need to get out more? No offense intended, I just call it the way I see it.

Don't know what bugs you're referring to... shaped arenas will be the first real change in gameplay since 0.2 AFAIK...

<dryly>
Luckily I provided a list to which you responded....
</dryly>

None. I've spoken to one player who still uses 0.2.6 because the newer versions did not work on their computer (OSX w/o Admin permissions), and that is a bug being fixed with the next release.

Um, Pirate uses 0.2.6. IN fact, the upgrade of his favorite servers to 0.2.7.1 may have driven him off. There have been quite a few others (/dev/null stayed at 2.6 on the client, but he's disappeared too). Ask around on one of the breakfast servers, although we've shamed anyone who hasn't upgraded into silence. I hear it at least a couple of times a week, though, "I won't upgrade to 2.6 until <this bug or that bug> is really fixed because it's not as bad here as it is there"

1. invisiwalls

Never heard of it?

It's when your wall turns invisible? Not the whole wall, just a patch between two turns. Happens all the time on all four of the current breakfasts (the two swamplands and the two hells, and 3 of those servers are running 2.7.1).

2. jumping (used to be sliding, may not actually be a bug)

Lag?

People dealt with the sliding, they leave over the jumping. Lag it may be, but the slide at least helped the brain to compensate.

3. going through walls

This doesn't seem to be very common and thus not very annoying...

Not common where? Go play swampland 2.7.1 for half an hour and I can just about guarantee you'll go through walls a couple of times. A lot of players haven't figured out that's what's actually happening, but nemo and I have both posted about it and others have chimed in to back it up.

I'm not making this stuff up...

4. lag when there are lots of turns made (for example through double-binding)

There's not a whole lot the game can do when there is insufficient bandwidth available...

If bandwidth is the problem, fewer updates with compressed historical information or something? There's plenty the server can do. Most importantly, the server can find some magical way to send updates that doesn't allow player actions to create lag in the first place? (I need to sleep, but after I wake up I'll dream up a way to do it, not implement it, and then complain when nobody else implements it, because that's what armchair programmers do)

1. 0.2.7.1 makes it easier for people to just suck on the wall. It's adding time to every round and people stuck in spectator mode are leaving.

I'm guilty of enjoying this feature.
Arenas such as 40-gon (circle) easily prevent such abuse.

We can't just switch to 40 axes just to solve this problem. That's throwing the baby out with the bath water.

2. 0.2.7.1 doesn't slide anymore, it just bounces. The nice thing about the slide was that your brain could at least track the changes while they were happening. The bounces leave you confused.

Is this the weird lag-like bug I posted on? z-man suspects it might be related to texture loading, I think...

Not likely, but maybe. It's the old lag slide, only now it's a big jumpy thing. The server and client lose sync. The client used to slide you back into sync when it got the update, now it just snaps you into place, usually causing death or dismemberment.

Related to this are random explosions (sometimes it's because of invisiwall, sometimes it's because of sync corrections, but quite often there's no way for the player to account for why he died).

And the time travel code has really changed gameplay dramatically. The no-bounce thing (and also no-rip) has also dramatically changed gameplay. There are many players who'd rather risk someone ripping the server than play on a 2.7.1 server! All grinds are now really tight, so there is no longer such a skill as grinding. Sometimes you get lucky and the other guy's grind isn't so tight. Otherwise it's about who makes the most adjusts. The pure grind is now gone completely.

Raise your hands everyone who's following this thread. How many of you like Swampland better than Swampland in 0.2.7.1? (I know of at least three that read these forums....) I've also had a lot of players tell me they'll never play on hell servers again because I won't run a 2.7.0 server anymore. (Their problem, you know, but it's disappointing whenever you lose players)

I'm personally thinking I like this game, I like the people playing it, but I'm getting pretty tired of working my ass off only to get dropped at 9 points, or get invisiwalled at 9 points. I'm getting tired of going against somebody, fighting hard for awhile, and the randomly exploding. I'm getting tired of watching two people 180 on the wall instead of fighting when there's 6 of us waiting to start up again. I'm getting tired of killing someone only to have my position corrected by the time travelling code and then crashing into a wall! I have better things to spend my time on. If I were the only one, I'd probably keep it to myself, but I"m hearing it all the time. I can't go on the grid anymore without someone telling me about their problems with the game, and it's the same shit every time from completely different players. (excluding lag, only the sucky players complain about it) And it makes no difference what server I go to either or--surprisingly--what version the server is running because these bugs have been there for ages.

[philippeqc]

Ok, now I know what need to be done for the next releases: fix bugs.

z-man, I've coded a fix for the bug on sf. I'll test it during the weekend, and submit the code then.

-ph

If Luke's map support...

Well, I did "contribute" to it a bit. Making the format, parser, tutorial, axis system and the start of the resource management. You know, the small stuff. I could be wrong, but I think I deserve to have my name somewhere ;)

[Lucifer]

Well, I did "contribute" to it a bit. Making the format, parser, tutorial, axis system and the start of the resource management. You know, the small stuff. I could be wrong, but I think I deserve to have my name somewhere

Fine, whatever. Can you post to at least one post without nitpicking? I just want to see if it's possible......

[Z-Man]

Lucifer: go to bed, you're halucinating
Luke's the one who's dissecting every post. Philippe is the calm guy. And he does deserve a bigger part of the credit for the whole map thing (forgot to point that out in my last post), as far as I can tell.