EN_US
<< January 2019 >>

 
1
3
4
5
6

7
8
9
10
11
12
13

14
15
16
17
18
19
20

21
22
23
24
25
26
27

28
29
30
 
 
 


31/01/19 ( 16:01:04 ) - Resume - Resuming? #coding #music

Ello ^^

I tried applying the change of pace, change of language, change of approach, for two weeks each:




- Coding - Onslaught -

Here are the conclusions I drew from messing around with multiple ideas of future languages - each of them have had several days of trying to get things done:




C++ via Unreal Engine

This does not look like a viable option at the time. There is no reflection in C++ which I would like to rely on heavily in order to implement the arrays with String keys that map directly to variables. I tried replacing these by mapping String to String, but still I run into issues simply defining classes with other self defined classes as variables. The editor just crashes without any implementation and error message, I think what I am trying to do is just not in my reach currently, or probably not even what the Unreal Engine is made for. Extending existing classes seems to work fine, but I need to implement (2D!) classes that have the ability to call functions based on names, and just adding another class as a variable into the base class (that is not even used) causes the engine to not open the project any more.

I am emotionally done with Unreal Engine - C++ - as a BASE at least. For helper functions in the Blueprint logic it will be helpful still.





Java

I have spent the better part of a day trying to implement core features into a java implementation. I already got stuck at trying to implement a sprite that scales and rotates. Individually, that is okay, but putting them together becomes something that apparently nobody ever has tried to do, and all the answers I can find are either for transparency, or for scaling, or for rotating.

In case someone with extensive knowledge about this reads this --- it is pretty tough to get the rotation center point right and I often run into the fact that edges are cut off of the sprite... very discouraging ^^

There does not seem to be an extensive library to mess with transparent PNGs and scale, rotate them and fade them out, which is what I need.




So what does that mean? It means that I will continue trying to make either of the mentioned systems work. If neither allow me to continue actually producing positive results next week, then I will fall back to what has worked well so far (albeit without visible results) - even though I have considered moving to JavaScript and HTML5 - JS lacks the ability to debug properly, but offers String-Arrays and just as Java will probably not go anywhere BUT has some disadvantages, performance wise...







- Music - Undertale -

For the music, this has been a very beneficial change. I feel that I have a nice idea of how to actually go about messing around with some songs. The first song I am focusing on will be Spider Dance - heavily relying on a stage piano that will do the leg work for me pretty much, because everything is too fast to be able to learn this in a time frame that makes sense for me.

Stay awesome!

Yours Chris

02/01/19 ( 20:00:48 ) - Scary Changes (Yes Clickbait) #coding #music

Happy New Year!

For both my biggest coding and music projects I will follow a very good friend's advice and do the following for a month or two, or three (my schedule shifts the concepts of time, a lot):




- Coding - Onslaught -

I will step back from the blueprint-only approach which started in me priding myself in the ability of using blueprints and hating literally copying the code into C++ which is a technical possibility.
Still I designed literally everything to be absolutely identical to the code, out of respect, and it will stay this way for at least the first release(s) - I might develop a version that has things I deem valuable as feature or maybe even big fixes, but for the first version I want it to be as close as possible to the original, because that is what it was. Close to perfect, and a gem of its time.

I will start fresh with this project [not throwing anything away, I will take use of some important blueprints later down the line for the implementation], I will implement the core code in C++ and by doing so basically literally copy the code and do small adjustments if needed for the code aspect of it, but I assume I will really be able to use 95% of what Gaby wrote. And I will have to design the GUIs (but I need to do that anyway).

If these three months get me further into the implementation, I might go through with going there full force. As I want to make it as true to its roots as possible, recoding everything into a new system does not really do what I intended to do, I just make more work for myself and I delay the process of keeping the ball running because I have to recreate everything manually (and not just GUIs and some actors).

TL;DR: I will pause development on what is the current blueprint game in progress and start fresh with a C++ based version (still allowing blueprints, but the main code will be in C++) which will allow me to re-use most of the original code, more or less. I expect much more and faster results, especially to show off things that and how they work. However this means that there will be a kickstarting time of no updates. We will see :)



- Music - Undertale -

I will step back - you could say leap back. My initial plan was to go through each and every song and remake / play it in some way as "my own". The more I tried imagining myself doing this with the first song, I realized that this would not really actually be what I thought it would do - do justice to what is the epic soundtrack of Undertale - but it would be (if I succeed at recreating the soundtrack) not really much better than all the people who re-upload the MP3 with some hastily crapped together gameplay footage [well I have that already but that is besides the point].

I think it would be more true to myself as an artist and more respectful to the original art pieces to actually choose my battles here. I will invest a lot of time into finding pieces that click with me, either musically or rhythmically. This will most certainly not result in what I hoped myself to achieve - replaying / redoing the whole soundtrack - which would be an endeavour of investing a YEAR - given that I release ONE VIDEO PER WEEK - which coincidentally does not allow for much creative spirits to thrive.

I think it will result into something that satisfies my pride more than my completionist desires to "do the whole thing" - I have no idea where that idea came from, not even the most talented and featured artists do the whole thing [except for the one shot things like the drum-along by 8bitdrummer ;) ].

TL;DR: I got lost in wanting to bite off more than I can chew and will step back to focus on things that will be fewer in numbers but greater in enjoyment, both for me and the ones listening to the results. This also means, like the coding topic, that the first month(s) will be planning and attempting with close to zero results, unless I get really lucky and inspired.




Thanks for your support. And if it is only you two people reading, that still is worth it to me :)