UE5 Megajam Submission - Postmortem
WHAT WENT RIGHT:
Having taken part in several of these, we came in knowing all to well what to expect. Our team was all there for the "unveiling" of the theme, and we quickly came up with an idea so that we could start work. Within the first few hours we had the character created, core art assets, and the first level in place.
Lucky for us, I was on a ROM (restricted movement) status due to travel and COVID-19 restrictions. This allowed for me to spend almost 20+ HRS a day working on art, level design, etc. Normally I would take leave so that I could spend that much time, but this time I did not have to.
Most of the codework was completed on time and we were able to keep the scope simple enough that very little had to be cut back. Honestly, this was one of the best jams we have had, in terms of completion, as we kept the scope within a manageable timeframe.
Having had major issues with the build at least years MEGAJAM we made a point to build after every major change and at the end of the night. This would ensure we had a good build no matter what happened in the last moments prior t upload. This was a really good plan and allowed us to have a worry free last few hours!
UE5!!!! I am saying it, despite being a "beta" product we have had nothing but positive things to say when it came to development using UE5! I am excited about the new options this platform has provided and it really did help with the environment and feel of the game. I think this was a good choice despite the possible risk for failure and inability to build the final project!
WHAT WENT WRONG:
Week long gamejams are a blessing and a curse. The time given really does allow a team to focus on polish and design, however the more time that something lasts the harder it is to maintain focus.
Right off the bat we had one of our coders have some real life stuff that took away most of his time after day 2. Much of the work created needed to be redone so that it would fit in with the final game. This hurt our timeframe a bit since we had to waste a day on rework. These types of things happen and are almost unavoidable, so we soldiered on.
Our sound fx/music person was also busy with some real life content and was unable to add in the sounds until the last 3HRs of the jam. I would be lying if I did not tell you that I thought we were going to submit a game with ZERO sounds/music. However, he came through and was able to link most of the items in. The sound could have been better. I wish we had more time with it so that we could really showcase his work. He is one of the best sound engineers I have had the pleasure of working with, but life happens.
WHAT ARE WE DOING NOW:
We have fixed some major bugs that we saw but were unable to fix prior to upload. One of those being the animation glitches for the spawning of the monsters. This was a feature we added 5HRS prior to upload, and the feature was certainly hacked into place. The goal was to make them look like they were coming out of parts of the ship (and it did that), but we had a small problem with how it was implemented that let one frame be their starting T pose. That is fixed now!
We balanced the weapons and the spawn system. Previously the monsters were very hard to kill, this led to us having s much slower spawn system, but close to upload we made some changes with the monster that let us spawn a bit more. Since that time we have balanced the spawn system and created a new method for managing the spawn controls via a curve. (spawn rate vs time)
As you might have noticed, the yellow monsters are the only ones that care about attacking the player. However, due to implementation "hacks" we copied the attack and damage code from the normal monster. The normal monster deals damage to the CI (Critical Infrastructure) based on an animation notify and does not care about how far it is from the object. This copied code made it to the Player attacking NPC and you can imagine how that works out. This is why several people reported dying but not knowing how. That has been fixed and will be part of the postjam release update.
Lastly, some lighting changes have been made and general level layout. In an attempt to help the flow of the process we no longer are choosing a location at random to attack and instead have gone with a more linear approach. If we decide to develop on this project further that will be looked at, but for the purposes of this update and simplifying a complex process we are making it a linear choice for the next place. This allows the players to know when/what location will be attacked next. We have also added some VO's that will support this so that the player has an auditory understanding of what is going on.
MP is being discussed, but we have other plans to work on our old project VoCM. If MP proves to be a simple implementation we might provide another update past the postJam update where MP is made available. More to come on that later.
Leave a comment
Log in with itch.io to leave a comment.