Project Kibou: Server Development

[InfamousNoone]

Originally Posted by Spirited

I suppose that's right, Conquer Online really doesn't need the resources, but it really doesn't make too much sense to me to cut corners like everyone else has. It just seems wrong to me. Maybe that's because I enjoy software architecture, maybe it's because I have ideas for the server that are resource intensive, I'm not really sure.... but there's no harm in doing things the way they should be done.

There is, it's called time constraint. It's nice to be able to plan everything out on paper and have a design plan for your entire project before you get started, however this approach is significantly more time consuming and delays production immensely.

[Spirited]

Originally Posted by InfamousNoone

There is, it's called time constraint. It's nice to be able to plan everything out on paper and have a design plan for your entire project before you get started, however this approach is significantly more time consuming and delays production immensely.

I still disagree... let me clarify a few things; you're probably just missing some information. This took about a few minutes a day for a week. Probably an hour and a half in total to implement, and a few minutes in the car to design - not the months that I've been gone. Also, any system that knocks 200-400 ms off your server's response time is totally worth it in my opinion. Regarding time constraints, I work in the software industry. This is how things get done: ideally one or two people design while two others implement. It goes faster with a very small group, but you're right - it is more time consuming for one person. Here's the thing though, when **** hits the fan or things need to be reworked, doing it the right way the first time saves not only time but a lot of money. This is how we do it at the software company I work for, how I've been taught to do it at school, and how I taught myself to do it before going to college. This is how I program.

[InfamousNoone]

I don't mean to pull this card out but everyone's thinking it: You've never launched a real server yet, and you've restarted your project 10000000000000000000000x times. You can take the high ground and say "I'm doing it right," I'll take the low ground and say "I'm actually doing it." Your so called gains are insignificant at least in my first thought of way of how to approach a problem, it's just shown by the fact you're using C++, there's 0 reason what so ever to shoot yourself in the foot like that for a Conquer server and try justify it with "I'm doing it the right way."

I'm over it. I'm done.

[Spirited]

Let's evaluate everything I've done so far, because it's not like I'm randomly restarting for no reason. Everything I've done has been carefully planned, and I still have many more projects planned after this. In fact, I'll let it all spill about how this project has affected my life. I try to keep it to myself, but here you are. If you don't want to know anything about me, don't read this.

My first source was trash. That's to be expected. It didn't scale properly and wouldn't run the way I needed it to. Trashed. Second source: trashed it in a week, so only a week of progress lost. The solution architecture was circular dependent, which was a big mistake. Third source: my final C# source that lasted a few years. I had beta testing for it and actually launched it for my friends. Now, let me remind you that you have had 3 projects as well, and you are currently working on your fourth; you're in the same boat as me.

Let's continue: I held live demonstrations of the project for C# classes and the computer science society at my community college. The project gained attention from the computer science department chair and was added to my resume. I then applied for a paid internship with a multi-billion dollar software company, showing off my source code and explaining the project. I was practically hired before I left the interview (the official offer needed to pass through HR) to their software development and quality assurance team. My job is to ensure that the applications I'm assigned to are being developed correctly and to an acceptable quality without losing too much time. I've been praised for my work on an almost daily basis by my supervisor. Now, I'm working on a C++ source because I no longer need to work in C# for experience in programming, I have a job that replaces that. This C++ source is aimed at even larger companies such as Google and Blizzard who work primarily in C++. I was asked to deliver in C++ by people in this community and recommended to by two Google recruiters, and I'm delivering. This is the best I've got at this moment. I've been off for a few months doing school and work, but now I have time to work on this source, so I'm working.

So, you tell me, am I still doing things wrong? Am I still failing to answer my original "business problem", or are you still going to argue with me for spending an hour and a half on an efficient MySQL pool? In the end, this is still just a research project and I'm still just doing this for my own benefits. I'm just inviting you all to follow along and possibly learn a few things. Bloody hell.

[InfamousNoone]

no but u dont understand that the fact u can restart so many times means u have no time constraint how is this so hard to understand wtf

i'm in the same boat as you? plz
csv1 has 100's of players
csv2 has close to 1000 players
and conquer ps is the fruit of csv3

im not reading any more of that post, im over it cause it almost sounds like you're writing a server for the sake of research

[Spirited]

Originally Posted by InfamousNoone

no but u dont understand that the fact u can restart so many times means u have no time constraint how is this so hard to understand wtf

i'm in the same boat as you? plz
csv1 has 100's of players
csv2 has close to 1000 players
and csv3 is the fruit of conquer ps

1. I do take time into consideration.
2. I'm working alone. Things go slower than they would.
3. I have almost no time to work on things like this. Things go even slower.

Duh this is not industry and I have more time, but if you know about all of the above, what's the problem? Why are you so angry? Seriously, if you don't like how I'm doing things and how I have almost no time to work on this project, don't read my thread when it pops up once a month.

[{ Angelius }]

Fang.. The fact that none of your projects saw the day light tells me/Everyone else for that matter that you are approaching server development all wrong. You value the end results but you don't value the cost of that result.

And to be frank.. If you ever had a dead line you would realize that cutting corner's comes with the package.

[Spirited]

Originally Posted by { Angelius }

Fang.. The fact that none of your projects saw the day light tells me/Everyone else for that matter that you are approaching server development all wrong. You value the end results but you don't value the cost of that result.

And to be frank.. If you ever had a dead line you would realize that cutting corner's comes with the package.

I'm not approaching server development wrong because I'm not aiming for a live server. It's a self-study. I just spilled everything about the project, and I just said how I have no time. Also, I hardly call getting a job at a multi-billion dollar company approaching my project incorrectly when that's exactly what it was meant to do. I have different aims and different availability than you guys. So again, what's the problem? Sorry to disappoint, but I've been telling people this for ****** years.

[{ Angelius }]

Originally Posted by Spirited

I'm not approaching server development wrong because I'm not aiming for a live server. It's a self-study. I just spilled everything about the project, and I just said how I have no time. Also, I hardly call getting a job at a multi-billion dollar company approaching my project incorrectly when that's exactly what it was meant to do. I have different aims and different availability than you guys. So again, what's the problem? Sorry to disappoint, but I've been telling people this for ****** years.

Problem? I don't speak for others but there is no problem fang.. I mean don't get me wrong because i am not against you or your project.. No disappoint just feedback an it's no ones fault that you chose to take it negatively.. I mean why do you think we are saying what we are saying to you ?

[Spirited]

Because this is absolutely ridiculous. Why is everyone trying to argue with me after everything I've said. I understand everything you're saying, but if cutting corners means 200-400 ms of additional latency, that's completely unacceptable under any time constraint. It would take an absolute idiot to say that leaving a problem like that in the code is "part of the package". Imagine if that passed through at the company I work at. I work on products that are very time sensitive where things need to happen within half a second. Imagine if someone left something that huge in the code - that would be almost half a second of delay right off the bat. We'd be screwed and out of the job. It's the same problem here - do you honestly think it's acceptable leaving something that bad in the code? Obviously if you're still arguing, which is why I'm just flabbergasted. Also, the point of this project is to be efficient. So as far as a time constraint on the project goes, I've stated absolutely everything. Why is there still a problem?

[InfamousNoone]

imagine if I passed in a package that took an extra 200-400ms at a given point in time where that part of the code is only being ran every 10 minutes or so, **** that corner cut sure is costing me a lot. i'm at my sarcastic point now

[Spirited]

Originally Posted by InfamousNoone

imagine if I passed in a package that took an extra 200-400ms at a given point in time where that part of the code is only being ran every 10 minutes or so, **** that corner cut sure is costing me a lot. i'm at my sarcastic point now

But it's not being ran every 10 minutes or so. We're talking about a game server. Database transactions are a huge factor in a server's performance. I mean, if I ran one MySQL statement once every 10 minutes, I wouldn't be loading or saving character data. One person would be waiting for hours just to log on. Are you drunk? Am I drunk? Who's missing what?

[InfamousNoone]

*hic*

[KraHen]

After reading this discussion I have reached the conclusion that it has no sense at all. The individual sentences do though.

[CptSky]

Originally Posted by Spirited

I'm not approaching server development wrong because I'm not aiming for a live server. It's a self-study. [...]

That's why you have no time constraint and can do whatever you want. Most people developed an emulator aiming for a live server. Meaning, you can't develop for 2 years without loosing all your potential players.

There is nothing wrong with your approach, but please understand that it can't be applied to every source and project. A better design for some projects could have been done, but overall, you always need to do some trade-off, except if like you, you make a server for research.

Originally Posted by Spirited

But it's not being ran every 10 minutes or so. We're talking about a game server. Database transactions are a huge factor in a server's performance. I mean, if I ran one MySQL statement once every 10 minutes, I wouldn't be loading or saving character data. One person would be waiting for hours just to log on. [...]

Sorry, but CO2 only have few data to load/save in MySQL. You'll save 100-200 ms for a task done in background (saving), and 100-200 ms on login, which is already slow due to the DH key exchange... It's not a bad thing to do a connection pool, but it's not where I would have optimized things if I had time constraint.