So much bugs.

[Cheery]

Hello for everybody.

First, I like this game a lot, it's so unbeliable, intresting and fun to play against even overskilled people unlike other RTS genre's games like warcraft 3 and age of empires. Even with huge amount of networking & user interface problems, it bringed positive feelings.

I wonder why there are so much bugs in TA Spring? Not in gameplay, but in networking and game management, many people have troubles playing this product on multiplayer.

I believe there are something wrong with development process. Have you tested everything before begun building stuff over them? Especially in lobby server system.

I know that it's hard thing to write networking code, so I don't want you take this as somekind of insult. There are many intresting things in the game itself, you have been done a great work with it, especially I like about the difference in power between the weaponry and the quality and the amount of units and buildings.

[coryrc]

Feel free to help.

[Cheery]

copyrc, I will help you guys. If I'm able to help, I'll check out part of the source code til I get some time for doing it.

[Ace07]

The game really isn't done yet, and thus its still in beta. There is still a lot of development to do.

[Redfish]

I wonder why there are so much bugs in TA Spring?

Man, this really IS a stupid question. Obviously you're not a programmer yourself. Blah. Besides, it's a beta.

[jouninkomiko]

if you find bugs, point them out. we don't have the resources to test extensively.

[Cheery]

Man, this really IS a stupid question. Obviously you're not a programmer yourself. Blah. Besides, it's a beta.

Not a stupid question at all, is it a beta? I thought it was alpha. Beta means it has been passed most of testing processes already, it's second version after alpha. But there is a big section which doesn't work right at all sometimes, so it's alpha from my vision. Anyway, t's not a reason to bypass the bugs. But it's not your business to solve out the errors. Except if you want to help the community make their game ready, it's the cost you pay from free software.

Jouninkomiko:
As a programmer myself, I know multiple methods and processes to approach, myself I never put alphas or beta builds out even if it's open source. I test my code extensively before putting it online, it's because it's not a need to make it ready with lightspeed. The people likes from complete stuff which works and you don't have to think whether it works or not.

Writing tests may be one 'unnecessary' part and may sound hard to do because being a programmer who knows things, doesn't ensure you know about testing.

Making testcases for the system modules is one great way to test the component before publishing it. On latter hand, it allows you to do something more important and funnier than catching bugs. Lacking testing destroys important resources you already have very limitedly, the time echant you made without writing a test framework, will be lost badly in the other parts of software development process.

Conclusion: There are always resources for a testing, It is required especially from the security reasons the internet gives us but also from the resource-effient building.

If you allow I to help:
- I supervise to write test cases from now on. The open source doesn't mean 'free for all', some people thinks it means 'garbage' and refuses to use open source products already. I'm not one of these guys. I believe there will be space for open source.

[CatalyticPrefect]

I want to say two things about your posts Cheery.
1. They are doing this in their free time and are probably trying to their best ability to make it work... for lots of people OSS is a learning experience and that they don't know networking code very well is probably one of the reasons they want to do this project; to improve their skills.

2. I think (THINK) you jumped to a conclusion about them which is very reasonable to make for OSS groups. That they are self-absorbed assholes. HOWEVER if you read the forums you'll find that they post regularly and are very very nice. I read into most OSS projects nowadays presuming the devs are self-absorbed assholes because it's almost always the case. They want to do stuff their way, when they want, how they want. It doesn't matter if the community they're serving wants something different, they have to go their own way. I find this with groups like Freeciv and BitTornado; they don't care about the community they are serving; their devs don't even play Freeciv or use Bittornado. I may be wrong, but I think the devs have some responsibility to listen to their community even if they don't really HAVE to.

[Liam]

what kind of programs have you written? if you've written an extremely complex game like this and still had enough time to test every little thing, then maybe you can talk. otherwise think about it (look at the source for christs sake!)

[CatalyticPrefect]

The networking code works MUCH better than most new commercial games out there like Stronghold 2.

[el_muchacho]

Not a stupid question at all, is it a beta? I thought it was alpha. Beta means it has been passed most of testing processes already, it's second version after alpha. But there is a big section which doesn't work right at all sometimes, so it's alpha from my vision.

I agree it's an alpha version, not a beta.

Jouninkomiko:
As a programmer myself, I know multiple methods and processes to approach, myself I never put alphas or beta builds out even if it's open source. I test my code extensively before putting it online, it's because it's not a need to make it ready with lightspeed. The people likes from complete stuff which works and you don't have to think whether it works or not.


Writing tests may be one 'unnecessary' part and may sound hard to do because being a programmer who knows things, doesn't ensure you know about testing.

Cheery, the SY team is not working in an industrialized way, but rather only when they have free time to spend on the project. The've been working on it, on and off, for years now.
The constraints were :
- develop a fully functional project (which they did);
- deliver a first version as early as possible (which they did too). I don't know when you heard of the project, but it has been announced a LOOOONG time ago, something like a year or so. I can tell you at one point, I was wondering if we would ever see a release of the game. And also, SJ was even wondering whether they would release the source code for various personal reasons.

You can find yourself lucky that they did. We all have lots to say about the source code too, as there are many obvious optimizations to do in the core game (I am trying to improve this by now). But I am so grateful that they did this, because I know such a project is way beyond my programming skills.
When one looks at these constraints, given the EXTREME complexity of the project (imho by far the most complex open source game ever released, and an enterprise very few people would even think of starting), I must say they have done not only extremely well, they have done a truely amazing job. Most people would have abandoned halfway or delivered a first release the state of which would have been far from the current state of Spring.

So that there are bugs, who cares really ? The community can take care of it.
Now that the code is released, one can think of industrializing the process and making a proper test bench. I think to make it easier, this involves inlcuding a script engine (someone is working on including a mono VM in the game for that purpose).

Making testcases for the system modules is one great way to test the component before publishing it. On latter hand, it allows you to do something more important and funnier than catching bugs. Lacking testing destroys important resources you already have very limitedly, the time echant you made without writing a test framework, will be lost badly in the other parts of software development process.

Conclusion: There are always resources for a testing, It is required especially from the security reasons the internet gives us but also from the resource-effient building.

If you allow I to help:
- I supervise to write test cases from now on. The open source doesn't mean 'free for all', some people thinks it means 'garbage' and refuses to use open source products already. I'm not one of these guys. I believe there will be space for open source.

Great. I also know that lots of people believe "open source = unfinished". It's true for many projects, not for others. And it's also true that the difference between the two often lies in the organization of the team as much as the quality of its members. Anyone can help, but eventually, the organization will decide whether one keeps this or that.

[Cheery]

I want to say two things about your posts Cheery.
1. They are doing this in their free time and are probably trying to their best ability to make it work... for lots of people OSS is a learning experience and that they don't know networking code very well is probably one of the reasons they want to do this project; to improve their skills.

That's why I wrote that post about processes, sharing my opinions and allowing to comparing them to theirs.

2. I think (THINK) you jumped to a conclusion about them which is very reasonable to make for OSS groups. That they are self-absorbed assholes. HOWEVER if you read the forums you'll find that they post regularly and are very very nice. I read into most OSS projects nowadays presuming the devs are self-absorbed assholes because it's almost always the case. They want to do stuff their way, when they want, how they want. It doesn't matter if the community they're serving wants something different, they have to go their own way. I find this with groups like Freeciv and BitTornado; they don't care about the community they are serving; their devs don't even play Freeciv or use Bittornado. I may be wrong, but I think the devs have some responsibility to listen to their community even if they don't really HAVE to.

I wouldn't write this post if I would believe that the people who I have been aimed this thread wouldn't be capable of conversate in educative means.

I wouldn't talk about self-absorbed assholes, I feel it's not a correct because everybody is little self-absorbed anyway. It's normal and shouldn't be punished, I know that if it's big, it passes out in a year or two from everybody depending from the person's smartness.

what kind of programs have you written? if you've written an extremely complex game like this and still had enough time to test every little thing, then maybe you can talk. otherwise think about it (look at the source for christs sake!)

I had been programmed many kinds of programs. Maybe the biggest one were my graphics & user interface studying project, by doing different kind of things and changing tactics slightly per every month and project, I have been echanted my knowledge about software design, developing and programming that way.

Liam, the complexity or the size of the source doesn't matter. Actually the biggest advantage of the testing comes when the complete system is big and divided into a good chunks.

The networking code works MUCH better than most new commercial games out there like Stronghold 2.

a Memory leak is a leak no matter where it is, same with anything else bug or thing which may be hid themself into a unsuspicious source-code. In this case I wouldn't talk about commercial games. Because they have different benefits and disadvantages. I believe open source can do better than commercial in many cases. In order commercial game would reach a good stage, every member of the group which is doing that game can be compared into an edge of diamond, if some edge is broken, the diamond is not so valuable.

Cheery, the SY team is not working in an industrialized way, but rather only when they have free time to spend on the project. The've been working on it, on and off, for years now.
The constraints were :
- develop a fully functional project (which they did);
- deliver a first version as early as possible (which they did too). I don't know when you heard of the project, but it has been announced a LOOOONG time ago, something like a year or so.

Yes, this project has been proceeded well, but these bugs would have been forgotted already in the early stage if that second constraint would been violated a little. In open source free projects, time shouldn't matter btw.

So that there are bugs, who cares really ? The community can take care of it.
Now that the code is released, one can think of industrializing the process and making a proper test bench. I think to make it easier, this involves inlcuding a script engine (someone is working on including a mono VM in the game for that purpose).

People who plays the game, cares a lot about bugs. Even they wouldn't have any reason to blame the project even they'd have been done a great job.

If there are bugs in bun, the baker doesn't get the complete honour and blessing from making a great job except these bugs are ingredient in the recipe.

You are in right thought, community can take care from them! I have been found that the most extensive way to test, is to write a testcase for each module in the project. It doesn't matter if it takes a time now, but the results are amazing.

Until I get some hands free from other projects, I come for this project and help you to get rid from these buggies.

[SJ]

You are welcome to implement component wise tests into spring if you wish, dont think you would be able to setup good tests for many of the components though.

As an aside the net errors seems to have been caused by two threads competing for the same memory area. Which would have been very hard to detect with component tests since it requires you to run both client and server part at the same time.

As for memory leaks im 99% sure that at least 0.40 had none at all except for dsound that isnt released properly at exit.

[CatalyticPrefect]

Cheery... I don't see what your point is. You're saying there are problems with the game and that the devs should fix them? What do you think they're doing?

[Cheery]

Annoucement:
- Plan: I test each bigger project module by writing a one single testcase for it, then I write a humongous amount of tests for big amount of code with relatively high speed. I spread the program for user area and most of you can test it and say what results it gave, if it says error specific module, the error looks something like this "<!>error: blah blah.". You'll copy/paste the error message to there, I examine it, then tell what is wrong if there are something wrong.

I use this topic as the center of my conversation between the community. I'll write more there when I proceed this plan.

- The desirable goal: Make sure the module works and doesn't do any memory leaks or critical errors.

There are some bugs I believe, this is the solution I try to get rid of these bugs and many bugs which could be hard to notice with any other method. Let's pull the bugs from your buns!

[jouninkomiko]

cheery, when you're the only one, or one of the only people, working on a project it is impossible to write a test case for everything. that's what a test is for. (cough cough) like this release (cough). it would be inexcusable if this were a commercial product, but alas, it is not. im not using that as an excuse for open source software, but even open source has beta releases relying on the public for extensive testing and error reporting.

[CatalyticPrefect]

Okay... I have some more questions:

I had been programmed many kinds of programs. Maybe the biggest one were my graphics & user interface studying project, by doing different kind of things and changing tactics slightly per every month and project, I have been echanted my knowledge about software design, developing and programming that way.

Great, you have been programmed many kinds of programs but have you been programmed any games or REAL software before? Not homework.

I believe open source can do better than commercial in many cases. In order commercial game would reach a good stage, every member of the group which is doing that game can be compared into an edge of diamond, if some edge is broken, the diamond is not so valuable.

These people are doing it voluntarily and therefore have no obligation or responsibility to anyone except for each other.

Annoucement:
- Plan: I test each bigger project module by writing a one single testcase for it, then I write a humongous amount of tests for big amount of code with relatively high speed. I spread the program for user area and most of you can test it and say what results it gave, if it says error specific module, the error looks something like this "<!>error: blah blah.". You'll copy/paste the error message to there, I examine it, then tell what is wrong if there are something wrong.

I use this topic as the center of my conversation between the community. I'll write more there when I proceed this plan.

- The desirable goal: Make sure the module works and doesn't do any memory leaks or critical errors.

There are some bugs I believe, this is the solution I try to get rid of these bugs and many bugs which could be hard to notice with any other method. Let's pull the bugs from your buns!

I find your tone very arrogant and disrespectful considering that Buggie and jouninkomiko are both CS grads (almost) and aGorm has his GCSEs, do you think you're Bill Gates or something, coming in here and telling everyone the "right" way to do stuff and that you're going to "save" their asses when you get the time?

BTW http://www.engrish.com/

[Cheery]

Great, you have been programmed many kinds of programs but have you been programmed any games or REAL software before? Not homework.

What homework do you think I do have? I have no possibility for this option in the hischool I go trough, I have to program stuff as my hobby.

I find your tone very arrogant and disrespectful considering that Buggie and jouninkomiko are both CS grads (almost) and aGorm has his GCSEs, do you think you're Bill Gates or something, coming in here and telling everyone the "right" way to do stuff and that you're going to "save" their asses when you get the time?

Graduates means nothing, they are humen after all and it means they aren't perfect enough that I could respect them as higher beings. I didn't come to save anybody except myself. I came here to study more. I know these guys you mentioned aren't the only people who made this thing.

The complete testing is the way I can learn the most out of code, also it should find errors more accurately than using the beta version in testing only.

The multithreading thing that one guy mentioned I couldn't find with my methods, how did you found it then?

But now when you mentioned it that the code is written by an experts, I focus into the style and another things, check out what they have been learned in school what I haven't learned with self-studying, hard reading and intense intense into the programming within 8 years, I'm sure I find something and be really sad if I don't find because it would mean I would be in the top and it would be boring because climbing to the top of the mountain is boring if you reach the top someday. It would mean you should begin to climb on another mountain.

[AF]

Cheery you're muddling your past, future, and present tenses.

And It was my impression that the whole point of a beta test is to eradicate bugs. I'm happy they've released it as they have and not tested extensively for months pushing the release back and back, and I'd have to say progress is booming because of it.

Graduates means nothing, they are human after all and it means they aren't perfect enough that I could respect them as higher beings.

To me that sounds like they'd have to be ultra intelligent computer programs incapable of mistakes to live up to your expectations.

Could you program the parts of spring Jounin has worked on? Have you been here as long as aGorm has? Posted ideas and thoughts as much as those 2 have?

What about your own standards against yourself? What do you have to say for yourself save an add mention of something and a period of time, what was that 8 years? Could you code me an rts in the time the SY's had, complete with extra features and no errors at all?

You indeed seem arrogant and unwilling to understand, and yes the SY's did attempt to eradicate bugs, that├óÔé¼Ôäós why we have beta 0.41 and not beta 0.10. But most of all you seem ungrateful.

Try not to simply reply with lots of quotes it makes your posts seem intimidating large, and such posts aren├óÔé¼Ôäót read by a lot of people, as shown by the fact many people turned away from TAI and 5 matrix AI threads because the AI posts where simply too long, and a lot of those posts aren├óÔé¼Ôäót relevant to the response you make either.

Help us bug test if you wish but don├óÔé¼Ôäót argue your credibility away questioning the abilities of the community, we haven├óÔé¼Ôäót made it this far to be stopped by someone saying we have bugs. I say you swat those bugs and keep your trap shut for the moment, and when you can prove that you're better than Jounin, aGorm, me or anybody else, then you can carry on, but I doubt that greatly, your attitude says so.

And higher beings or not you have been exceptionally rude, I demand you apologise to everyone you have attempted to tarnish in this thread, such behaviour is uncalled for and insulting to the project as a whole.

You disappoint me greatly cherry, in your dishonour may you understand what ahs happened and stop being so transparently egocentric

[jouninkomiko]

phrased beautifully, alantai.

and cheery, you're in high school? when i was in high school, i felt the same way about coding. it was also a hobby for me, as it is for you. however, get yourself some experience and you'll find the opposite is true