Vapourware Codexian Game Development Thread

[Yuber]

Do you use checksum for your save files?
In Dark Heart of Uukrul I wanted to modify the quantity of food in the save game file. It turned out it has a checksum. Why is it needed?

Maybe to check whether it has been tampered with, to prevent cheating by save modification.

But why on earth the developer cares about what the player is doing with his game? This is very stupid. Usually the good developers are leaving the cheat codes in their games. I remember playing Civilization on the hardest difficulties with infinite gold cheat. That was great fun.

It is mostly about Steam Achievements I guess.

 

[Yaar Podshipnik]

Why do players care about "achievements"? Why would devs care about them (maybe some kind of metric for how the game is played)?

 

[shihonage]

Why do players care about "achievements"? Why would devs care about them (maybe some kind of metric for how the game is played)?

For my ASCII Steam shooter, achievements showed me how many players got to the end of the game, so that's useful. (It was something like 23% lul)

Other than that, achievements are for the same people who spend 5 weeks customizing their Mass Effect character's face.

I don't understand them at all.

 

[Catacombs]

Why do players care about "achievements"?

Shiny badges to prove to friends and strangers alike they played the game.

 

[Russia is over. The end.]

achievements showed me how many players got to the end of the game, so that's useful. (It was something like 23% lul)

How's it useful? Most games have these low numbers. And ones that don't are just special in unknowable ways.
Name one thing that you did differently because of that number.

 

[Ironmonk]

Well, achievements can be either useful or useless... it all depends on how its implemented

I personally like the idea of a way to measure player engagement/completion of a game... so for a dev, it can give some useful feedback for future games.

For a player, if set properly, it can also be cool as a challenge mode... so you can replay the game in order to complete these challenges.

So, IMO, a mix of both types of achievements can be a good thing... instead of mere pissing competition.

 

[shihonage]

achievements showed me how many players got to the end of the game, so that's useful. (It was something like 23% lul)

How's it useful? Most games have these low numbers. And ones that don't are just special in unknowable ways.
Name one thing that you did differently because of that number.

It wasn't the only number. I also saw that most people dropped out after level 1 or 2 in my 3-hour ASCII shooter/adventure.

The data, in combination with examining other recent ASCII projects, told me that if you're making an ASCII game, the ASCII has to be much flashier and in-your-face, and the game cannot start slow "to tutorial the player into it", it has to be front-loaded with something bombastic.

Most people never got to the "cool" content in levels 3-7, such as various scripted sequences, driveable tanks, rescuing soldiers who turn into monsters, sand planet where you have to make your own paths, self-destructing a spaceship, watching an NPC who communicated with you via audio, actually teleport into your level, attempt to fight the end boss before you do, and die...

I can probably fix this if I invest more time into the engine and redo the first level at the very least. Introduce "software RTX" to make everything look cool, shadowy and colorful. Enhance ASCII deaths and explosions.

But, that's a lot of time investment into a side project that I never should've bothered with. It was romantic and all to finish the game I started writing in 1996, but it was a dumb idea, so I cut my losses and moved on.

 

[Ironmonk]

To further elaborate my post above, another feature could help integrate achievements to the game is a in-game screen with a list of all achievements and a score system that allow player to unlock skins and perhaps some game modes, special options, special classes/abilities, etc.

All this combined make achievement something organic and really integrated to the game and game systems... possible giving extra motivation for the players to play the game extensively.

 

[Modron]

Why do players care about "achievements"? Why would devs care about them (maybe some kind of metric for how the game is played)?

The people who love cheevos are usually just console players who need the dopamine hit everytime they press a button.

I much prefer in game rewards, cheats, unlockables, et cetera to achievements. I really don't like how some achievements can be just nigh retardedly specific and not something you would ever do organically when playing a game. About the only genre I don't mind achievements in is RPGs where they are more tied to completing quests or exploration.

Plus as I've said many times before achievements are a very visible metric publishers/developers use to decide what features should be axed from sequels and in future games.

 

[Justinian]

A demo of my game should be out around 9:30 est tonight. More bare bones than I'd like (I don't think I'll get the Forge feature in by then), but I do intend to add more elements over the weekend. Right now I am working on getting the boss for the demo level in, so that it at least feels like a complete level.

Next priority will be to get all the base class skills working (I'd actually like to get this in as well before release because it looks bad to have skills that don't do anything and I do have art assets for them so there's really no excuse), then forge, then tier 2 classes. If I don't feel too lazy over the weekend I should be able to get it all in before Next Fest actually starts on Monday.

edit: Gonna release it tomorrow, I don't really have a lot of will to work on it right now since it's almost bed time for me, and I'll want to test it for a little bit before the actual release anyway.

 

[RobotSquirrel]

Well, achievements can be either useful or useless... it all depends on how its implemented

I personally like the idea of a way to measure player engagement/completion of a game... so for a dev, it can give some useful feedback for future games.

the best examples of it are achievements where hidden mechanics or features are indicated in the achievement - they're used to push the player in the right direction. "Wait I can do that?"
Because really no one cares about e-peen when it comes to achievements. So its just another tool to push the player along and yes provide milestone feedback to the developer - the only problem is with these stats, I'm finding developers are seeing the lack of engagement in the content and then scaling back on content in future titles because those stats tell them no one cares - I mean its bad when Rome 2 Total War had what like a 10% completion rating for the Tutorial campaign achievement when it was being reviewed by journalists, telling us they didn't even play the game, they literally just started the tutorial and went "hmm too hard 10/10". So there are dangers to thinking your achievement stats are reflective of your core market - probably they're not, lots of people buy games only to have them sit in their steam libraries - if anything that's more an indication of that.

 

[Justinian]

Well, the first version of the demo is finally up. Cleaning it up enough so that there are (as far as I can tell) no major issues took longer than I thought it would.

 

[shihonage]

Well, the first version of the demo is finally up. Cleaning it up enough so that there are (as far as I can tell) no major issues took longer than I thought it would.

By demo you meant video?

 

[Justinian]

Well, the first version of the demo is finally up. Cleaning it up enough so that there are (as far as I can tell) no major issues took longer than I thought it would.

By demo you meant video?

No, there's an actual playable demo. The button should be on the right (although I clicked the option to make it green and clearly visible, I dunno why it won't show up that way).

edit: i guess it didn't save before, it should be clearly visible now.

 

[what am i doing]

Started defining operations and messing around with development stencil placement. Ran into issues pretty fast. The plain "one development stencil per tile" approach I was going to go with isn't going to work. A development stencil is basically a collection of operations and the ratios of each, and a collection of professions and the ratios of each. The most obvious problem was the issue of "baseline" professions i.e. those that the majority of the population fall under. In many cases they're largely mutually exclusive - a nomadic culture is just going to be nomadic pastoralists (mainly) and a subsistence farming culture is just going to be subsistence farmers (mainly) and yes, I know this is an oversimplification. In any case, the problem comes with different agricultural operations, because there are a few different "farmer" types representing different social roles and slightly different specializations (i.e. peasants vs landowning farmers), who may all exist in the same tile. In that case, how do you cover ratio variation? A separate development stencil for each ratio? It doesn't work. And you'd run out of stencil colours quickly, which makes map readability poor, which, even though this is only visible to the campaign creator and not to the player, is still important.

So I have to go back to the drawing board here, somewhat. I think there may be something to the idea of defining civilization types i.e. nomadic pastoralism vs simple agriculture vs complex agriculture vs state society, but I think I need to define more dimensions which can govern what kind of pops appear. So I'll be doing some brainstorming there. There are a few dimensions I've already considered, but some of them won't be able to be defined until other aspects of the simulation are more fleshed out.

I also removed poly2tri library and replaced it with mapbox's earcut library. Poly2tri is buggy and doesn't handle a lot of polygon types very well. Earcut, like everything I've used from mapbox's libraries, is exceptionally robust, and also has better performance, and pairs well with clipper2.
This did mean I had to discard the "fading in" borders (ear cutting leads to vertex connections that don't support the effect; I might use VASEr for borders at some point and just use earcut for tile geometry) and just make them transparent, but it also made border generation much more robust and eliminated a class of visual artefacts that I had before.