The past 3 months or so have been real slow in getting work done. Mostly because I am swamped at work and have some family matters that needed my full attention. Things are starting to die down and I have so much to catch up on in the game. I apologize.
I have been wanting to have a way to allow people to start to play the game and get some feedback. After a bit of looking for the best way to provide a “pre-alpha” copy for people to try, I decided to look into WebGL. I will be looking into a hosting solution to allow people to join a server and play around with the spell casting system ASAP.
I have just got some of the files from my character artist. I was up late lastnight rigging them with mixamo and making them dance gangnam style.
One major issue I have been dealing with in using Unity3d is Linux support. I am running Linux, so I have had to run unity though wine. The issue with that is it limits me to 32bit. Recently Unity released a native Linux solution. For the past week, I have been working on getting that to work. There are a lot of issues and crashes at the moment, so I will still be doing the majority of the work on MageDuels in 32bit wine version, and playing around with the native Linux version on the side. When it is stable, I will make the switch.
I had an interesting conversation on twitter yesterday. I have been looking for UO PvP veterans to help give me some feedback on how the game should “feel”. I wasn’t expecting the response “That sounds really, really weird…”
I have been sort of developing this in a bubble, essentially. Is the concept really that weird? He made some good points, but I stand behind the concept being more about competition than full loot. However, this made me think quite a bit of how to increase the risk vs reward in the champ spawn game mode.
I know it has been a while since I have posted, but I have been busy.
A big part of the spell casting redesign was an issue with Unity3d’s matchmaking system. There is a limit to how much bandwidth that can be used, and I was hitting that limit due to the nature of the spell prefabs I was using. Now instead of a prefab for each spell there is a prefab for each spell type and the effects get assigned when the spell is cast. This brings the amount of data that needs to be sent to the server/clients down by a lot.
The second part of the redesign is a scripting system. I have been looking into several ways of doing this, but I haven’t committed a lot of actual code yet. There is quite a bit to consider with this, and I want to get it right.
Lastly I have also been looking into some AI for the champion servers. These will be large battle fields with a boss creature. When these events run, the groups that take down the champion will receive weighted loot drops. As of right now all items are cosmetic for either your player or your house.
The menu for the game is going to be a small mini game that allows you to go to a market, use a magical portal to join servers, has a house for you to decorate and display any awards you may have earned, and use a chest in your house to customize your mage.
Right now the spell casting system works as such: A script waits for a key to be pressed. When one is, it loops through the spells for a spell with a matching assigned key. If a match is found, it begins waiting for the cast delay before turning on the targeting system. It works great. But it is wrong.
I am working on a new system that will allow you to assign a sequence of tasks to a key.
- Cast regenerate
- Wait for spell ready
- Target self