Bay 12 Games Forum

As mentioned, laptops will accumulate dust inside them over time. Just recently I thought my laptop was broken since it would overheat and restart after stress. Turned out it was just overheating because of the extra dust because going through each hole in the chassis with a vacuum cleaner fixed it. A computer isn't healthy if you can't stress it 100% without overheating.

Hello, everyone

No version on this post but a little progress update of what's currently working in the development version:

- Creatures that are standing next to you and targeting you will follow you to the next map upon transition. This'll enable escort quests to work better and you can't shake zombies off as easily, either. This can quite trivially be expanded to include creatures from further away also and with a delay so as to simulate the previous map still being active.
- The FindCreature objective is fully operational. With it, you can track whether you've found a specific creature yet (or more specifically, whether that npc has seen you).
- Escort objective going along really well.

I will try and release on Sunday. Tomorrow (Saturday) I'll be celebrating my graduation with friends and family so probably can't finish it before then.

Also regarding the ARRP/1.0 features, here's what I'm planning, feature wise:
- Plot is to find your sister and get out of the city (Yea I know it kinda sucks, but I only spent 15 mins coming up with the idea and the content). The city will demonstrate the random map generator and the countryside will be hand-crafted probably by myself.
- Monsters will be balanced so that the game is very tough yet probably beatable. Unless you're dealt a bad hand. The game is still short enough to gain forgiveness for that.
- Strategic depth is already quite deep, the player has to balance between avoiding the monsters, killing them with melee (silent but risky due to damage taken) and using firearms (ranged, noisy, limited ammo). I'll see if I can include the running/perspiration system, so far it seems achievable.

- Tuplis

Have yourself a freshly baked version!

0.12.8.1.0:
- Moving against a friendly npc swaps positions (ty finka).
- Vastly improved creature behaviour (upgraded from crap to bad). They can now much better handle things like making distinctions between chasing creatures they can see and chasing sounds/smell/creatures they saw but now can't.
- Guess what? More balancing to smell. It's pretty good at the moment I think.

https://dl.dropbox.com/u/84930267/Zombies-0-12-8-1-0.rar

Hey!

Worked this morning on improving manual map creation. I created lists for creatures and items where you can define what to spawn on a tile and define what their weight (probability) in relation to the other list items are. Check the ZombieList.xml, WeaponList.xml and 1.txt to see how to use them.

https://dl.dropbox.com/u/84930267/Zombies-0-12-7-27-0.rar

Also fixed a couple of things, including a little rebalancing for smell propagation.

Later on, I'll probably make it so you can include a list within a list so you can define basic list types (like weaponlist) so you can have areas where items from the weapon list can spawn and areas where an item from a variety of lists can spawn. Same for the creatures. Won't be much effort.

A quick update on the progress. My three-week summer vacation just started and I'm headed to the summer cottage for a few days. No electricity -> no coding

- The random map generator has been finished and plugged to the game somewhat. They need more variety though because they're super boring right now and no enemies spawn there. I'll let you try it out later.

- My laptop is a bit slower than my desktop and I noticed some lag. I debugged the lag back to two functionalities one of which was from the ui component. I was able to make it so the particular routine in the ui component doesn't have to be called anymore so that should cut turn time by about half. After that the most expensive operation was the propagation of smell which was about 60-75% of what was used per turn. So I cut it down by a factor of 50 or so. Overall, turns should go by a LOT smoother from now on. None of these measurements include zombies moving (i did the debugging on an empty map) but a quick check provided me with the information that creatures require well under one millisecond each to take an action.

I won't be finishing the random generator today but I thought I'd post you a little update anyway:

First of all, the xml object builder has a nice new feature. For any integer field, you can now define a range of numbers and you will get a random number. For example, the BasicZombie.xml file currently has the attribute <MaxHP>30</MaxHP>. After the next release, you can put it like this <MaxHP>30-40</MaxHP> and the zombie will have a random hp max of 30-40. This'll work with any integer on any object. Sadly the same exact effect applies to every object from that template so all of those zombies would have the same hp on that map. So actually this effect is mostly useful for objects you only need one copy from.

The random maps will be defined in xml files. Currenty the templates look like this:
<RandomMap>
<Class>RandomCityMap</Class>
<Height>128-255</Height>
<Width>128-255</Width>
<StreetTile>$World.CityStreetTile</StreetTile>
<WallTile>$World.CityWallTile</WallTile>
<FloorTile>$World.CityFloorTile</FloorTile>
<BlockSize>20-28</BlockSize>
<RandomFactorX>0</RandomFactorX>
<RandomFactorY>0</RandomFactorY>
</RandomMap>
The city map consists of blocks. A city map has its height and width defined in the file and the size of one block, also. The BlockSize field only determines the size of the building (or fence if the building has a yard) and if (when) the division has a remainder, random rows or columns of blocks will be made wider. There is a one-tile wide sidewalk around every building block and a two-tile wide road between blocks.

Cities will have a limited tileset that is defined completely in the xml file. It'll be fairly easy to refactor to something different later, if need be. So for now, we'll be working with a limited variety of tiles (though you can define which three you want).

The RandomFactor fields are used to add some variance to the amount and size of the blocks. For example, if you want a big city map, let's say 255 by 255 and a blocksize of 22-28, you will still be stuck with a fairly predictable map because the remainder can't be higher than 21-27. Thus, only 21-27 random blocks will receive +1 to their heights and/or widths (a block can receive the +1 more than once if it's "lucky) and the map will look quite homogenous.

The RandomFactor is a decimal number between 0 and 1 and will decrease the amount of blocks on the map by a random percentage between 0 and its value. For example, if you put a RandomFactorX of 0.75, the amount of blocks (from left to right) will be reduced by 0-75% depending on random chance.


I'm estimating I'm about 33-50% done with the generator. The current state is that it will create the template object based on the xml file and constructs a table representing the size of every block. The next steps would be to 1) place the tiles on the grid that is used in the game map and 2) plug this whole generator in the current game.

What do you guys think?

Edit:
soundlust, you are also more than welcome to join and contribute if you want. There's a LOT still to be decided, we're just getting started

I know I might be late with my input, but I'll throw it out there anyway. I've owned two gaming laptops so far and my experience is that you'll want to invest more than the ~800 you mentioned.

Save for the asus g53, out of the proposed ones none will run any recent games at good settings. The graphic card is a critical component, and these laptops had bad ones. Currently, I'd recommend the following laptop:
http://www.amazon.com/MSI-Computer-GT683DX-840US-15-6-Inch-Laptop/dp/B0072BA3AW/ref=sr_1_14

That's a price point (~1100-1200) where you can find the laptops that run 1-2 year old games at great settings, recent games at good settings and future games at ok settings. If you go for the 800 range, you're going to end up with a computer that'll run 1-2 years old games at ok setting, current games barely and future games as a slideshow. You could maybe find a better fit than this laptop but you should, in my opinion, definitely be looking for an nvidia gtx570 or 660 for a graphics card. The processor isn't exactly what you should be looking for - optimal would be something like 2,5GHz dual core.

Hey guys.

I've been doing some thinking again. I reviewed the backlog shuffled it around a bit since the blue lines seemed to be taking up a lot of space. Ever since we refined the scope a few days ago, I've looked at the backlog items and I've seen features that just don't seem super relevant at this stage. So, I though I'd throw a few more questions in the air to stimulate some discussion again.

What should we do to make the game feel more purposeful? Right now there's obviously the two quests that give some sort of direction but it's really not much. We should have something driving the playing experience a little bit more.

Here's what I'm thinking. I'm just throwing these off the top of my head:

- Build enough maps to make it feel more like a world. We could have people making the maps and/or I could write the random generator. However, I'm still not sure what kind of a generator to make since at least I'm not sure if we'll want the city areas or rural areas more. Maybe I'll eventually have to write both.
- Add proper support for npc's and enable communicating with them.
- Finish or at least flesh out the quest system
- Add game mechanics that make the game feel more like the type of game we're trying to make it ( mostly about short-term survival, constant suspense and mechanics that encourage stealth and avoiding contact )
- Add Story

As I think about these topics, I notice they're not very well represented in the backlog which admittedly is a mistake by me. What do you guys think the focus should be right now?