This member has provided no bio about themself...
Sorry for my late reply as well :P
1. I think it's okay for mods to borrow from other games as long as you mention it somewhere.
2. I actually tried IPI before, and I've heard it works better with Sony's PS eye than the kinect. But I gave up on it because fast movement looks shakey and cleaning them up makes them look robotic. I think for first person games it would look much better and be faster to just hand keyframe the animations. It also saves you the work of buiding an animation studio and finding motion actors.
And I still highly recommend avoiding webcam/kinect facial animation.
At first it might produce fast results, but think about the work afterwards. You'll have to clean up, compile each animation, and your programmers or mappers will have to trigger each animation manually. And if you're doing localisation (other languages) you'll have to redo this for every line. On top of that, you'll have to find English speaking actors and bring them into your recording studio. Imagine having to do this for every line, like "Grenade watch out!".
Faceposer isn't great but all you need are voice recordings to work from. I had to install windows xp on an old computer for it to run (I think it doesn't work on vista or later) but it was definitely worth it, I really hope you try it instead of motion capture.
3. I know this isn't always possible, but it's better to have a small dedicated team of programmers than many part time programmers. The reason is that communication between programmers takes a long time, and if someone has to leave suddenly it's very hard for someone else to pick up their work. You also have to be aware of different time zones and schedules. Source mod teams usually only have one or two, and I think any more than 3 is overkill for a mod team and will only slow things down.
Anyways what I mean is if you already have a programming team but your progress is slow, finding more may not speed it up. It might be more realistic to just cut down on features.
I'm glad you are open to criticism and I'm flattered to hear you're a fan of Triage! Nice to meet you too.
Been working on NPC/AI stuff and environments. Look out for an update soon!
Thanks for the kind words everyone. It was definitely a tough choice pulling the plug on part 2. I probably won't work with Source engine again but I definitely do want to revisit some of the ideas we tried.
I won't be fixing up Triage because it would be a ton of work, and it'd only be a matter of time before another update breaks it. Long story but the work would include rewriting all of the original mod code, including AI, weapons, scripted entities etc. It's not worth it IMO and it's one of the reasons why I'm moving from modding to indie, and I predict this will happen to a lot of modders in the near future.
This looks interesting. It's always good to see students have a try at modding. I started as a high school student as well, so I can understand some of the challenges you are facing.
I have some things to say to the devs though. It looks like most of your models are ripped from other games and reskinned. I think you should at least mention that you borrowed content from other games, it's not fair to take credit for it yourself.
I also have some advice as a fellow modder.
First, regarding motion capture and lip syncing. I do NOT recommend using a kinect for motion capture in a first person game. The reason being that the kinect's capture quality is not good enough for animations that will be viewed up close, even for walking or crouching (don't even think about facial motion capture). You will spend more time cleaning up the animations than you would just hand keyframing them. If you were making a game/mod like DOTA or Diablo it would be acceptable, but it's not going to look good in first person. I recommend recruiting an animator to hand keyframe your animations, or just decompile and reuse HL2's animations (Valve doesn't mind, Black Mesa's soldiers, guards and scientists mostly use HL2's animations).
For facial animations, I recommend you learn and use Source SDK's Faceposer (https://developer.valvesoftware.com/wiki/Faceposer). It doesn't look bad (all of the HL2 games used this only, no motion capture), and it's quite easy to learn, and for a group of students this is the most realistic option. Facial motion capture rigs are very, very expensive and only look good with (expensive) trained actors - technology doesn't hide bad acting. They also require lots of time to cleanup the animations even with a very expensive rig. Anyways, it's better to have slightly stiff facial animations (Faceposer) than cheesy, goofy ones that require tons of cleanup.
I also highly recommend that you either learn to code or get your programmer as soon as possible! Let me explain: Source SDK/HL2 code is very difficult to work with, even for experienced programmers and in my opinion it's the hardest part of modding. The reason being that it's barely documented, uses an old version of C++ that they don't teach in schools anymore, and the engine is NOT open source so there's some stuff you just have to guess.
I notice that you have a lot of features planned, and that's fine, but I think you haven't started implementing any of them yet since you have no programmer. I think you will have to cancel some of those features because your programmer is just going to tell you it's not possible. Another problem you'll run into is that you will code a feature, and it turns out not to be fun and you'll have to scrap it. So I highly recommend you start programming or find someone to do it as early as possible, so you can decide at an early point what you can and can't do.
It looks like your team is quite skilled at mapping, texturing, UI design and choreographing scenes. I think your team has the potential to make a good mod. Just make sure you start coding early so you can test frequently, and remember that as a group of students you have a limited budget and manpower and some things just won't be possible.
Sorry for the very long post and I don't mean to be overly negative. I'm just offering constructive advice based on my experience. I wish you good luck and I hope you learn and have fun on your project.
I'm writing up a postmortem. Look out for some interesting pictures of cut content, prototypes and some info behind the scenes.
Okay thanks, I'll look into it.
Oh I see. I think my walls had seams because they were made up of many smaller blocks. Thanks!
Ah cool. For the walls, did you just scale up one block, or were they pieced together from multiple small blocks?
I'm sorry but I can't see that picture, can you upload it to somewhere else? Was it in the room before the room with the wooden boxes? Thanks.
Haha no worries, I'm always up for a friendly discussion.
I think run and gun is fine and works well for a lot of shooters. Especially in Titanfall 2 where IMO they did a perfect blend of tactical shooting with old school FPS movement.
But just like the gameplay in SUPERHOT, I wanted the player to always have multiple options and part of the fun is making quick decisions (and sometimes making mistakes and laughing at yourself).
For example:I see a bad guy, do I go invisible and hide, or go slow mo and shoot him, or sprint to a better position and snipe him, or stay put and wait etc. It needs a slower pace and each action needs pros and cons to facilitate that, especially in a 15 minute demo where there's no time to "practise".
Originally before sprinting and ironsights it was faster paced and the AI were very accurate. Personally I found it fun, but I could tell from watching others play that they found it shallow and repetitive (I might actually post some videos of these in an article). Players died because they weren't fast enough or their aim wasn't good enough. After slowing down the game and adding ironsights, sprinting and superpowers I noticed players were more engaged in the game because they were always thinking about what to do next. They'd die, but rather than say "ok cool game, that's enough", they'd quickly hit the restart button and try it again with a different approach.
But yes I agree with you 100% about the lack of verticality, it's basically Wolfenstein. I definitely want to experiment more in this area and I do want to try teleporters, gravity lifts etc instead of boring ladders and stairs.
Sorry I mean in my project my lightmaps have a lot of seams. Did you group your walls into one big mesh? Or use bsp instead of models?
Hi thanks for playing and I'm sorry about the bug. Do you know around which area it happened in? Was it the room with the brown wooden boxes by any chance? I thought I fixed it but I guess it's still popping up. I'll look into it soon.
Hi there, thanks for playing and offering constructive criticism.
Yes I agree about the player being too bullet spongey. The enemies accuracy is a bit too random and their bullets need to be a bit slower in slow mo, it's too hard to dodge in the tighter encounters. It was challenging balancing both those variables (I actually made them into console commands cause I couldn't stop tweaking them) for both close range and long range fights. I plan on looking into this more and perhaps scaling accuracy, burst length and bullet speed over distance so the bad guys can have the same effectiveness at any range.
As for sprinting, I actually added this ( and ironsights) after a lot of hesitation and testing other alternative means of adding risk vs reward to the game, which IMO was a big fun factor in SUPERHOT. Originally you were always running fast and there were no ironsights(people were so polarised about ironsights in my last project so I was totally against doing it again), but this gave you only one way to play - run and gun as fast as you can - there's no reason to try anything else. Sprinting gives you a risk vs reward choice of moving fast but being defenseless, vs being a slow and easy target but being able to shoot back (and ironsights adds a third option, of being accurate but very slow).
I'm going to spend about a week catching up on sleep, food, exercise and vitamin d, then I'll play the other entries and go back to work on this project. Thanks again for playing and feedback. I look forward to trying your game as well :)
Thanks so much! Cheers.
Thank you so much for your consideration. I'm so relieved! I'm looking forward to trying all the entries as well. I hope you enjoy playing Trepang2 :)
I submitted around 1000 UTC assuming the deadline was 1200 UTC, since UTC is usually used with 24 hour clocks without AM or PM only to notice the "AM" after. If they meant 12:00AM as in 0000 UTC, then I think it was a bit unclear. There's also the discrepancy with the timer counting down to 1900 UTC. I hope they take these things into consideration, and consider allowing the mods that came in past 0000 UTC. It would be unfair to disqualify so many mods due to confusion over the time it's due.
EDIT:I meant to say not only mods, but games, and art entries as well. Little bit sleep deprived :/
How did you get your lightmaps to look so good? Mine have seams all over :(
Haha same here, best of luck!
5 days? That's seriously impressive! I might try this on my friend's vive.
Just posted some screenshots. I'll be posting a video soon and a demo for the #MAKEITSUPERHOT contest in a few hours, stay tuned!
Thanks :) F.E.A.R. is one of my all time favourite games, so yes I did take a lot of inspiration from it.
Yeah dynamic point lights are pretty fast but dynamic shadows are very demanding in ue4, especially if they overlap :(
But yeah haha I know what you mean about making a unity FPS. I remember having to use triggers to hide far away rooms lol. Unity just doesn't have much out of the box for making an FPS. But I do miss probuilder though.
Looking good, I'm jealous of the number of dynamic lights you can use in unity :P (unreal doesn't like them so much)
I wouldn't mind giving away the mod's code to the community, but unfortunately I lost it when my old PC died. It wouldn't be useful now anyways since it's not compatible with source 2013 anyways
Thanks, that means a lot :)
There's no content taken directly from Triage, but I did the SMG animations in both games, so it might look a bit similar. These animations were done pretty quickly though so they're just placeholder for now.
Thanks! The AI is still very basic at this point - they only know how to attack and take cover (though they do it quite well, despite a few minor bugs). I can't wait to show off the more complex behaviors once I get them done :D
That's correct, it is Unreal Engine 4.
I don't think it should be glowing like that... Did you use $selfillum by accident?