Maker of Stuff blog - eliasr

Hah, the title for level 14 is fun.
But it also seem to be true. (at least for my case)
I have been here a long time now. Added my own small addons and actively development in a mod.
Posting content both to BZSmod's page and this blog. I guess most people who get to this level, in some way probably have made multiple things.

Right now I'm working on a trebuchet in Unity, using its build-in physics(without coding)and working on Jenga in a VR enviroment (unity) where camera movement is determined by accelerometer and gyroscope. Besides of that there is networking and interaction with the bricks.
But i'm also working on the melee system at bzsmod, which finally has taken some form. (4 years too late however...)

Well, this post wasn't to focus on that...

Dadiu.. project management ... everything...

I completed the DADIU internship ( Dadiu.dk ) - there is a google play link for those of you who want to try the game.And working on a game 45 hours, if not slightly more, every day for 45 days was an amazing experience. It was so amazing that the professor who invented the game study at our university retired (due to age).... and now they don't provide this opportunity any longer..... seriously... wtf!
Yeah yeah... so it isn't classical computer science, and every single one of you are so classical that you can't come up with a proper academic study regulation for a single game semester.... I get it, it is very hard and DADIU is wayyyyy to fun.So I probably learned nothing..... about economics about the gaming market and the F2P market... about establishing a carrier and how publishers evaluate your products... about project management with a large team and tight deadlines... about how long long time a small game project really can take... about pipeline between artist and programmers.. about continued integration (Jenkins CI) and deployment as well as QA releases.. about a lot more.
Well honestly, I can't write a Ph.D in any of this. But I had never gotten this experience anywhere else.(university scope)
And I learned a lot though the programming... about teamwork and... not-teamwork, all the mistakes we made, the wrong estimates and the solutions we came up with.So okay, Lets Raid Ain't much, but it is neither nothing. But try to compare it with Heroes Charge by Ucool. We only had 50 Mb, not 250 Mb. We were also limited to 42 days and our location wasn't the best.This is one and a half month, and it really made me realize how long a simple project can take. Let Raid is actually more than 4860 man hours.To compare that to a regular 4 man group at the university that would take about 6 months and 3 weeks. And we usually have 4 months, which includes courses. so ~2 months without courses.

---

Sometimes, people wonder why bzsmod is so damn slow?
Or why I estimate a high amount of time for a given work proposal?

And I think it can be really hard to explain it to someone who isn't affiliated with programming or modding or projects that requires a custom software solution. One such as my mom.
I really think that next time I should explain it, I should not show Lets Raid and say 1½ month. But 5000 hours. - that is equal to 208 days

But I didn't even expect it to take that much time when i started on the project, so I do have compassion for those who might not understand why it can take so long.

---

Well I learned a lot. A lot of tiny small things, so many that I can't express them all. And I learned a few big ones.

However in the bigger picture I realize i'm still learning something today... something basic.
I finally got around to code the melee part of bzsmod. Something we should have started with as the very first thing in bzsmod. In my mind, the idea is so simple. At an abstract level that is so refined that everything just seems like plug and play. But it just ain't like that. Otherwise we would be done by now.

The basic thing however is related to PR.
Moddb aint that bad. Nah i don't like to get archived news number 8 in row, just because there are too few images or its too long or too much coding. I also hope for some good feedback or creating a debate, but its like it never get that far.
So we went a bit into the background for some time, into our forum of void and an occasional spam message... heck its even so well fortified that we don't get that sort any longer.

An occasional news story about some code would be written, as people who didn't code was either disappearing or waiting for the code to be done. Which can be hard when you have a 45 hours work week taking all the time.

But making the melee system changed all that. DADIU and Roei changed that. My extra year of study that made me understand object orientated development much better, changed that... ok maybe not, but it made it possible.

Interfaces, Factory design patterns and now 50% binary tree (i don't like all the code for building them, so that ain't being made)
Well all i really learned from my c++ course this semester was a bit of c++11 variable instantiation and templates.
All that made it possible for me to take some of Angels animations and crunch them though my code as it pleased me.
All that made it possible for me to provide an easy way for Angel to add his animations to the QC file without worrying about animation names or the indexing.

And all that in the end made it possible to publish some graphical content on moddb. On moddb where we at times have several hundred readers.

That... and reducing democracy. Apprently the fewer we are the faster we progress.

It was a mistake to reduce our activity on moddb. Well perhaps the time to focus on work instead of PR did benefit us a tiny bit, but that was not the core issue.
Instead PR should have been an catalyst for us to complete our things, to apply our own deadlines and show off.
I think that seems to be working now. And with the melee system being implemented, the game begins to take shape and provides a foundation for us to demonstrate the game mechanics in bzsmod.

Instead of just saying. Oh we like the idea of dash attack, it will work like this. Angel and I can sit down and make the animation and code for exactly that and then toss it into everyone faces by showing it in a video.

Instead of writing a wall of text explaining the matematical function for calculating directional block (which no one apparently cares for), We can demonstrate grimmjow "trying to beat ichigo" and "beating ichigo from behind" instead.

That is probably where i think, as a project leader, that I didn't fulfil my responsibility. That and relying on hierarchical leadership foundation.
The greatest issue with hierarchical leadership, is that the leaders must be able to lead, wants to lead and can do it independently. If that fails, then it is like a joint breaking away from the entire structure.
If it succeeds, then it should be able to work really well. Just as well as it did at DADIU.

What i have learned form that is that there are some which I can really rely on and those where I should cover their administrative "leader" role. Like enforcing weekly sprint meetings, with everyone in a team.

---

That was a lot... if you got this far, then i hope you liked my monologue. And thanks for reading it.

If there is something I really appreciate from moddb and especially bzsmod, then it is the ability for me to learn though these experiences. While I learn many things from my studies, it always comes down to whether i have been able to implement that into bzsmod or not... well most of the time.

I'm also happy for the feedback we get from you guys, it helps us get an idea of whether we are on the right track or not.