There  are two reasons that I can see -

Firstly the planned gameplay needed a ton of custom code. In a standard roguelike you can have a table of enemies and  a table of weapons/armour and very quickly get thousands of interesting fights that can happen as often as you like. In a non-combat game such as StalagRL you have exactly one way to die (shot), and 4 ways to win (tunnel out, through the wire, disguise+walk out, hide in a vehicle) all of which need very specific code written to handle them, and all of which happen rarely enough to be important and often enough to get boring if they don’t have variation.

Secondly was the level of graphics. There are loads of freely usable RL sprite sets out there and we used one last year. But try finding 3d assets with a specific theme (WW2) that are also either public domain or creative commons. In the end I ended up having to make almost all of the meshes we use and I’m certainly no artist. We probably spent around 4-5 of the 7 days just working on the graphics engine side of things and that is still without any polish, just required work.

However, I’m still proud of it and we certainly both had fun and didn’t waste our time.

As for the updates for the last 2 days, we did quite a bit.

I added all the camp activities and the prisoners would obey them all and wander off to the right place on the map at the right time. These activities also governed which areas were out of bounds at specific times.

Guards were added with specific patrol routes around the perimeters and will chase the player if caught out of bounds. When caught they just show a message for now but they should escort the player to solitary and confiscate their items.

The guard structures were added as a single building which made things a bit easier although was one of the things that made us realise that random level generation would not have been easy. I built a large structure with internal rooms for the solitary, storeroom, eating area, barracks, office and commandants office.

I also rearranged the map a bit to add the large exercise area in a separate compound.

Rich got the item use system almost done with custom item handler support that would be needed for most items. The first real interactions were starting to be put in place such as a usable disguise and being able to hide items.

Caught out of bounds by guards

The guard building

Plans for day 8+

We have spoken about maybe continuing this to a complete game. It might happen or might not. I’d be interested to see how it turns out with decent graphics, animation, gameplay, etc.

[edit]

If you really want to look anyway, you can download it here. The Github project for it is here and is tagged as v1.0 in the master branch.

The time system now governs what prisoners should be doing and when. The other prisoners move around randomly within the area they are allowed in at that time and travel to and from zones at the right time.

The item system now allows you to search and pick things up.

I redid the hut a bit to look a bit nicer and have actual beds but got no time left to do the other ones properly.

Prisoners congregating for roll call

Find and picking up items

Plans for day 6 –

Me – Add all activities. Guards and patrol routes. New outside regions for guard areas: barracks, commandants hut, cells, shed, etc. Tunnels if possible.

Rich – Use system.

Most obviously they are NPCs in the game. These currently just stand still and get in the way but now that we have the pathfinding code finished we can make them move about.

Also added was the time system. This is the core of daily life around the camp and will determine where prisoners should be at any given time (‘free’ time, excercise, meals, rollcall, etc) plus control the guards, spotlights and no-go zones. The first part that this controls has just gone in and is the day/night lighting cycle.

Nighttime

Top end of the compound.

Day 5 plans –

Dan: NPC schedules and movement. Maybe integrate them for trading.

Rich: The ‘use’ and item system.

A slow day but things are still happening.

Rich made progression adding corridors to the generated map. Not quite there but just needs some debugging I think.

I upgraded the lighting for line-of-sight so I could hide the rest of the level. On the monster side I made them chase the player if they come too close. If they lose the player they will generally hang around near where they last saw them. The monsters can attack the player and vice versa although nothing happens yet.

Walking around the compound

In one of the prisoner huts

Goals for day 4

Rich

• As usual. Get corridors added to the map.

Me

• Basic combat system. I’ve put an enemy in the players start room to help with that until corridors are done.
• Zoom the map back in and add scrolling. Currently it is at 50% size to see it all for testing the level generation.

We have everything in that was planned except a class system. Considering the state it was in 24 hours ago due to 2 days lost to GUI confusion, I thought this may have to have been a fail, but we got everything else in that was planned. We even managed to get persistent levels to allow you to go back and finish exploring/killing ones you didn’t do enough on.

Challenge version of the game.

The plot remains the same as planned. You are a bored level 35 dungeon boss that decides to retire and live in the human world. So you have to level down instead of up and dump your epic armour in favour of worse gear on the way back out of the dungeon whilst fighting adventurers trying to get in. There is mechanism to stop you going onwards if your gear is too good so you have no choice but to get rid of it. You also stop getting loot if you hit too hard as an extra incentive.

We have no plans to carry on with this although may do a 1.1 release in a few weeks to add the missing class system, a few simple bits we would like to add personally  and fix bugs.

Personal firsts achieved: First 7DRL!, First Roguelike at all, First complete Java program.

Download it and try it out now!

.tar.gz version

.zip version

Slow day again, leaving things set up for a final 2 day scramble to get things finished off and polished. Spent a long time fighting against libgdx’s built-in UI system as it doesn’t support things like hover and right-clicking natively. I think I can work around those problems as I’m not that interested in a mobile build anyway.

We can now change floors both up and back down again. Levels are the same each time and persistant (fog of war, monsters and item placements are all kept intact).

I added the start of a UI that shows the current players stats and any monsters that are trying to attack them.

Slight work done on combat, the enemies can now injure the player, although still for a fixed 200 damage each time.

Two smaller things I did – Firstly adding cheat modes to help with testing – noclip, revealmap and godmode. Secondly cutting up  lots more sprites and making their background transparent ready to be put into place on day 7.

Shows the player/nearby monster status on the top-left. also shows that used a cheat mode to move from the top-right of the level to here.

Goals for day 6

Rich

• Not around

Me

• Beef up combat to use stats, I imagine a lot of this is going to be spreadsheet work due to the needs of this specific plot.
• Examining and equipping items.
• Show level number and players effective level number.

Corridors at last! I got my head around the bad description of the algorithm we had at last and fixed the odd bits. I added up and down staircases as well. and made the player spawn next to the ‘up’ one. That should be ‘down’ thinking about it now…

I also added basic combat support. Both the player and enemies have health now and do  damage per hit. Player damage is currently ignored for testing but the enemies can die now.

200 damage is a lot, but this is starting at level 35 and working backwards.

FoW turned off to show a generated map and very basic item placement.

Goals for day 5

Rich

• Not sure yet.

Me

• Beef up combat to use stats, I imagine a lot of this is going to be spreadsheet work due to the needs of this specific plot.
• Start of equipment system
• Zoom the map back in and add scrolling. It has been set at 50% normal size for testing the level generation and item/mob placement.

A slow day but things are still happening.

Rich made progression adding corridors to the generated map. Not quite there but just needs some debugging I think.

I upgraded the lighting for line-of-sight so I could hide the rest of the level. On the monster side I made them chase the player if they come too close. If they lose the player they will generally hang around near where they last saw them. The monsters can attack the player and vice versa although nothing happens yet.

State of the game after 72 hours

Goals for day 4

Rich

• As usual. Get corridors added to the map.

Me

• Basic combat system. I’ve put an enemy in the players start room to help with that until corridors are done.
• Zoom the map back in and add scrolling. Currently it is at 50% size to see it all for testing the level generation.

Rich got part of the level generation code running last night by adding doorways to the existing rooms. Hopefully the corridors will get added tonight. He is using this algorithm which seemed ok at first look but on re-reading it isn’t very clear on anything. BSP looks ok so will have that as a backup plan near the end if needed. I didn’t say before but I have most of the day to work on this, Rich only has a few hours in the evenings so that is why his part is going slower. I can work around it.

For my part I added an enemy in every room for testing. I Started the item system which generates (weighted) random objects of various types and added some test objects. So far I have Consumable, Armour and Weapon, with weapon subdivided into Swords, Axes, and Daggers. I need to know the difference for stats and equipping later on (probably Wednesday). For testing they spawn one per room as well. The space key picks up items and puts them in the inventory window on the left. I also started on lighting with a first-pass simple distance based light with falloff.

Sprites are from this page. They appear to originally be from here but the link doesn’t work anymore to find out attribution of if I’m allowed to use them.

State of the game after 48 hours

Goals for day 3

Rich

• Get corridors added to the map.

Me

• A* or Dijkstra algorithm. I don’t need it for pathfinding, although it might come in handy. I want to use it for checking that the level start/end is connected and for line-of-sight on the lighting.
• Basic combat system. I’ve put an enemy in the players start room to help with that until corridors are done.
• Zoom the map back in and add scrolling. Currently it is at 50% size to see it all for testing the level generation.

Day 1 summary

This is our first 7drl so we are doing things from scratch as much as possible. Plain Java with libgdx so far. We have an explorable, random map up and running, albeit without corridors yet. Rich is in charge of that part while I’ve done the other parts. The main character moves with the arrow keys and cannot go through walls as expected. Enemies move randomly and also stick to actual rooms. Just added an activity log below to record what happens.

Goals for day 2

Rich

• Get corridors added to the map.

Me

• Basic combat system. I’ve put an enemy in the players start room to help with that until corridors are done.
• Zoom the map back in and add scrolling. Currently it is at 50% size to see it all for testing the generation.
• Start adding fog of war/line of sight.
• Start adding an inventory and items if Rich doesn’t get the corridors done for the previous two parts.