LambdaRogue: The Book of Stars -- Now version 1.7 (29.12.2018)

[mariodonick]

Congrats on landing such a huge update---best days lay ahead at this rate.   

Thanks! Good to see you still around. It seems really quiet in this forum these days.

It seems most of the discussions have moved to reddit?

[getter77]

Indeed, much of it has or to other far flung means---ebb and flow perhaps.

[mariodonick]

I have added an issue tracker to the game's official website:

https://lambdarogue.com/2019/01/01/issue-tracker/

[Krice]

It seems most of the discussions have moved to reddit?

It would have if Reddit was a forum, but it's an advertising platform. The "discussion" in Reddit's roguelikedev is pathetic. I think the reason for less discussion is that us old school developers, many of us got old and stopped active forum writing and even game development. There is much less new actual roguelike development, everyone is just advertising their mobile "roguelike" games.

[mariodonick]

It seems most of the discussions have moved to reddit?

It would have if Reddit was a forum, but it's an advertising platform. The "discussion" in Reddit's roguelikedev is pathetic. I think the reason for less discussion is that us old school developers, many of us got old and stopped active forum writing and even game development. There is much less new actual roguelike development, everyone is just advertising their mobile "roguelike" games.

Mm, maybe you're right, although there exist reddits with really decent discussions, such as the Elder Scrolls lore reddit "teslore"; so it all depends on the people. But I somewhat miss the old newsgroups (which still seem to exist, but also seem to be flooded with spam).

About advertisment... I'm afraid this is much more needed than some years ago, at least if you want people to actually play your game. It's much easier to make and upload games nowadays, so gaming sites are flooded with new games and updates to existing games each day -- which is a good thing, imo. But to not drown in these floods, at least in the beginning, one needs to be able to produce and present a constant flow of activity -- posting updates very often, write devlog entries regularly, post tweets, create and share videos, etc -- basically everything to convince people that you care about them and your project, and that their invest of time in your game won't be wasted. I think this is especially true if you're new (or if you come back out of nowhere after 7 years...)

On another note, it is interesting which games are the big ones nowadays. It seems ADOM has become really strong in the time of my absence, and people still play classics like Nethack and Angband. New games like Caves of Qud seem to be very true to the classic roguelike concept, but quite successful, too. On the other hand, I would have expected more talk about Slashie's Ananias and some others.

[Krice]

one needs to be able to produce and present a constant flow of activity -- posting updates very often, write devlog entries regularly, post tweets, create and share videos, etc

I don't see it that way at all. If the game is good it's going to be found and people are going to play it. These mobile game developers are different, they need visibility (advertisement in all possible forms) to get at least someone to buy the game.

My personal goal as developer is not money, but being able to create a game that people actually play. During the development of Teemu I've began to feel it has potential for that, but it's not yet there. I also feel that this kind of development is missing in today's world where everything must produce money.

[mariodonick]

I can understand your view and I've got the same goal. But without telling people that your game exists and that, albeit unfinished, you care about it, the number of players will be small. (Note: Even posting a release on RogueBasin is advertising.)

[Krice]

(Note: Even posting a release on RogueBasin is advertising.)

It's often called "announcing" which is sort of "advertising", but without a commercial product.

[Krice]

The bad automap reminds me that Teemu has also a pretty plain automap. I think the automap should be as clear as possible and show important things like doors... In my opinion the problem with ascii automap is that it doesn't represent the tiles in the similar way they are in the actual tile map.

Another thing I'm still wondering is why each character class has the entire tileset with clones of terrain tiles etc. It's taking much more space that it should (making the download also bigger). I guess it's easier(?) to manage tiles that way, you need only one large file for tiles, no need to split them to terrain etc. tilesets.

[mariodonick]

Another thing I'm still wondering is why each character class has the entire tileset with clones of terrain tiles etc. It's taking much more space that it should (making the download also bigger). I guess it's easier(?) to manage tiles that way, you need only one large file for tiles, no need to split them to terrain etc. tilesets.

Historically grown, as all the other bad coding, idiosyncrasies, inconsistencies, imperfections, and "wrong" priorities.

When I draw a tile, I actually print an ASCII character and the tileset is just the "font" used for drawing. The tile to draw is selected based on the ord of the char:

OrigRect.x := (Ord(c) - 32) * 40;  // for the big 40x80 tiles

where "c" is the char I want to print.

So, for example drawing a "player" tile in line 10, row 5 of my "virtual" terminal, I could write something like CharXY(5, 10, '@') and the function will select the 40x80 part of the tileset that corresponds to the ord of the char.

In the actual game, the player graphic is on chr(143) -- don't ask me why, I don't remember why I did not simply use the actual @ ... and even worse, because in the console mode I of course want to have a regular @ I transform the "chr (143)" back to a "@" before output happens.

A bit confusing, but it works, and the end user does not notice anything of this. Of course it has several disadvantages (such as no easy way of dynamically displaying equipment, and the total restriction to 255 characters of the ordinary ASCII table), but as I don't strive for perfection, and have no "I want to be a better programmer" goals or anything like that, it's fine. Finally, all tilesets together are just 5 MB.

[Krice]

but as I don't strive for perfection, and have no "I want to be a better programmer" goals or anything like that, it's fine.

It's not "perfection" if you abstract/detach the tile id from its raw data. But I get it, not everyone wants to improve their programming skills. However stuff like that makes management of bigger projects easier and roguelikes are the kind of projects that become large scale when you start to add things.

[mariodonick]

Yes you are right - basically it's laziness