Post Mortem
of the Whole project –
The side scroller project is finally over and now it’s time
for the Post Mortem of the project, where I go over the whole project and see
what I did, what went well, wrong and what could be improved.
7 Weeks ago our whole year was presented
with the Side scroller project which was to basically concept and create a
playable fantasy side scroller platforming game consisting of 4 levels in the
new Unreal Engine 4.
It was a Group project for the entire 3rd
year so we were put in 4 teams of 3 Environment artists paired with one Concept
and Character Artist. We had been given
themes for each of the levels and a basic layout of how the game was going to
play out.
It started off with Hot then moved down
into Happy, next would be cold and would end with Scary. We had to
stick to the flow chart that the art director’s had given to us.
 |
| The flow of the level we were given |
I was assigned to team 4 which is the Scary
Level and the section the game ends with. I was an environment artist along
with Hannah McMillan and Rebecca Clay. Our concept artist was Karol Nowak. And
our character artist was Rachel Cox. There were only 2 Engine artists so each
of them had 2 levels to cover. Chelsea Lindsey was the Engine Artist for our
level.
We decided to go for a scary cave themed
approach to the level after all of the idea generation and concepting, it seemed
to fit well with the other levels and the transition made sense from the cold
section.
Since it was the last section and the fact
that it was “Scary” themed we had the idea that there was this giant worm boss
fight at the end of the level that you had to escape from by platforming out of
the worm/boss chamber, originally we were going to make the player die at the
end but we felt that it wouldn't be a satisfying end to the long journey the
player had to embark upon. The level would include a lot of platforming and
have a lot of kill zone areas so it’d be the hardest one out of all the levels.
What area I worked on
I worked on idea generation, concepting and
white boxing some of the assets. But my main role was creating a list of assets
that I chose off the asset list, I also was the one placing most of our assets
in the engine towards the end of the project. There was a slight
mis-communication with the last part with me and the engine team but I’ll get
into that later.
 |
| Basic concept for the level I made |
 |
| A more finalized concept for the level |
I was in charge of creating all of the
rocks for the level (our whole level is literally rocks everywhere) including
the floating platforms and the cave entrance, the architecture such as the
temple city, pillars, and crushing walls. I also made Water for the river, by following a tutorial from: https://www.youtube.com/user/PubGamesAU/videos
 |
| Little scene showing off the water shader I used |
I think this was a great opportunity for me
to learn how to create modular rocks and finally learn how to use Z-Brush
properly so I spent a whole weekend watching tutorials and sculpting like a
madman trying to figure out how to use the program. It paid off well as I’m now
confident in using Zbrush for anything I need to. It’s about damn time too.
 |
| A scene with all of my assets in |
It was a challenge too, I've never created
modular rocks for an entire cave level before, and there was a lot of pressure
on me, because whatever my rocks looked like would determine the whole look for
our level. I think I managed to pull it off though, a lot could have been improved
though as with everything for example I could have created better retop meshes
instead of just using the Z Remesher in Zbrush, which did quite a great job
anyway.
 |
| Modular Rocks |
My textures were mostly tillable overlayed and mashed together with the
ambient occlusion texture to create a unique one. So I didn't really create the
most “realistic” rocks but it is fantasy so It works.
The reason why I didn't model the entire
cave as one giant static mesh was because I knew we would need to change up
some area’s and make them larger or smaller, so it was easier to piece the
level together from modular rocks.
What worked well?
I think our level idea worked really well with all the other
levels and had a temple theme across all the levels, we managed to finish
nearly all of our assets and got them in engine. Our level looks like our
concept. Rachel’s character and worm looked fantastic, the rocks decorated the
level nicely in my opinion. Hannah’s eggs and Rebecca’s crystals looked very
nice and pulled the level together.
 |
| The Polished area of the level |
The crushing platform temple section was my favourite part
even though it was a very last minute touch, it broke up the ending and the
level a bit because there were too many rocks before that.
The white waterfalls in the cave background looked awesome
and gave it that extra punch it needed.
We decided to put chains throughout the level just to give
it some more character, anything like chains or hanging vines just add so much.
 |
| The crushing pillar platform area - A last minute addition. |
Near the cave entrance and halfway through we had Hannah’s
small worm statues that lit up as you went past them, we also had her devil
finger fungi open up as the player walked past, these kind of triggered events
made the level feel unique from the rest of the sections in the game.
The flow/shape of the level looked like a worm’s body so it
was emphasizing that you were being eaten especially with the cave entrance
looking the way it did (huge mouth with sharp teeth)
What didn't work well?
Our level was pretty monochromatic, was mostly just blue at
the start then constant red everywhere. There needed to be more colour
variation. I think there were too many rocks in the level, I mean I know it’s a
cave and all but it just got a bit boring to look at, There was the temple
section with the bricks we added in the end, but we needed to have elements of
that throughout the whole level, like more pillars and broken walls/bridges in
the cave background. Basically more architectural elements.
What problems you faced?
 |
| Way too much Red and Blue, very monochromatic. (This version of the level is my personal updated/polished one) |
We faced quite a lot of problems, the main ones being not
having everything done in time or polished in time and not having any or enough
communication with the engine team, both of us didn't know what each other was
doing for at least 3 weeks.
The plan was for me to update the engine with all our new
assets in but I didn't know that the engine team needed it back in the next few
days so they could implement the changes, instead I had it for 2 weeks and I
kept on tweaking it and making my own version run smoothly I didn't hear
anything from the engine team asking for it back so I assumed it was fine (BIG
MISTAKE)
Unfortunately It all
came crashing down after we decided to have a team meeting and the engine team
wanted to compromise some of the parts because it was too much for Chelsea to
handle, It was difficult for her to put the assets in because she didn't know
where some assets went or where most of the files were located, whereas I did.
It was then said that we should only polish the one area up, which left me at a
loss but I wasn't going to give up.
I wasn't willing to compromise everything I had done so I
asked Chelsea for the engine for at least 5 hours to update it enough, she was
still using a very un-optimised version with loads of the white box assets in,
and it was basically a mess, but luckily I managed to polish it up to a good
standard the day before Elliot went over to bug fix the whole level, I just
wish I could have had more than 5 hours on it (I had to leave early that day
for an appointment at the doctors)
Nearly all of the problems I can think of:
- I didn't have enough time to place the assets properly in
engine, some bits are left floating,
- All of Rebecca’s Crystals weren't added to the cave walls
on the kill zone areas so if you walk up to the wall on the left it’ll just
kill you it was supposed to have spikes on there, and there are some buggy
collision areas that needed to be addressed.
- The waterfalls were eating too much FPS in the new build
of the engine, our FPS was dropping to 20-25, before with the optimised
waterfalls the FPS would be at 50-80.
 |
| Blue waterfalls in the engine using a lot of FPS |
- There wasn't a crumbling dust particle effect added to the
disappearing platforms or no drip/water particles added to the front entrance.
- There were these blue waterfalls that the engine team
decided to add in randomly at the end after I had done my polish, they didn’t
fit the water in the level and it wasn’t a good fit in my opinion.
- The giant statue at the end boss room is facing the wrong
way from the camera and that whole area needed a touch up, we didn't manage to
implement the boss fight mechanics and the platforming escape section because
it was a bit of a big task on Rachel’s end to animate the worm.
- There weren't any platforms implemented into the polished
area so you kind of just fall through it and miss a lot of it.
- Rachel didn't manage to produce a worm carcass for the
ending section because there wasn’t enough time (I made a crappy one in my
build of the engine later on turning her textures on the main worm green and
duplicated it around/skewed it to make it seem like a dead one) but she did
have to make 2 characters and was ready to take over happy level’s character
since there was no word from Alex on her progress.
-There were way too many last minute decisions when I was
polishing the level which wasn’t a bad thing because it turned out well it once
again we just needed a few extra hours with the engine to place everything
properly.
The whole section with the crushing pillars and platforms
made no sense before, they were just on cave rocks and I thought It’d be a
great idea if we could put a temple section in there so I had Rebecca quickly
model a broken brick wall and bricks with my tillable texture.
 |
| How the crushing pillar section looked before the revamp |
I basically cloned it
everywhere in that area, and with the added chains and bridge sections it
looked quite good. I just wish I could have placed these brick walls in other
areas of the cave to make it seem like it didn't appear out of nowhere.
-Light shafts and atmospheric fog wasn't implemented into
the level at all, and that was something really important that would have
increased the depth and immersion of our level a lot.
-Some of our teammates weren't pulling their weight and
others were away for a week or 2, and I wasn't in for a good few days combined,
it wasn't too much of a big problem on our team but we could have finished
earlier/polished earlier if we all had been on top of our work and got
everything done quickly. (Everyone in the project felt ready to give up after
Mike’s crit on our alpha so it was quite demotivating and a lot of teams wasted
time moping around but we pulled through in the end)
-Some of my assets were rushed such as the Cave Entrance, it
was near the end and our team still hadn’t finished everything, Hannah was
still working on the altar after I had done my polish and Rachel was still
polishing her worm and character animation.
 |
| Cave Mouth - not particularly proud of this. |
- There aren’t any stalagmites or stalactites on the top and
bottom of the cave background/river which doesn’t make sense since there are 3
waterfalls and the whole cave is really wet and slimy. The cave background just
needs more character to it and time spent on it by adding assets and rocks.
(Another polish issue)
How did you overcome problems and issues?
We couldn't overcome a lot of the issues and problems I
listed due to the late time we confronted most of the problems, but the main
one about the engine team wanting to comprise the level I had to face by taking
the engine of off Chelsea for one day in the last week and upgrading it like
mad with Rebecca. We did have a lot more communication after we had our big
group discussion near the end. Rebecca and I just weren't willing to compromise
the look of our whole level due to a communication error. We left all the bug
fixing and collision problems to the engine team, and kind of just had to push
through it all till we finally finished everything.
• What would you do differently if you were to approach a
similar project?
Get my assets done earlier and not procrastinate or slack
off as much, and the major thing COMMUNICATION, I needed to communicate a lot
more with the engine team.
We had daily scrums as a whole year group but that
was minor, I needed to talk to the engine team directly and tell them what I
want changing in the level/if I can borrow the game and change it. I should
have done that earlier and handed them back the updated version the next day
instead of 2 weeks later.
That was something I wasn’t expecting I thought they
could just merge the level in, but they needed to build a new one every few
days to keep it updated. We should have done more paint overs of our level so
far so we could’ve improved certain areas, especially the constant red colour
and blue scheme we had.
• Conclusion / Summary
Looking at our level in the end, I’m quite pleased at how it
turned out, We managed to pull everything together in the last week despite our
major setback due to the miscommunication and have an awesome level to play
through, with bugs albeit but I think we were fairly successful in following
the brief and our concepts.
We encountered quite a few problems and there is still A LOT
we could do to improve the level, I would have liked to tweak some area’s and I
have done so in my personal build of the level which unfortunately couldn’t be
implemented into the final build.
If we had an extra week OR if we planned our time better,
and if everyone was motivated and enthusiastic about the project we would have
an even better end level without any bugs, collision problems, the boss fight
of the worm included and all of the assets/foliage placed properly. It’s sort
of a shame that everyone wasn’t enthusiastic about it, I think it spread and everyone started to feel down about doing their best.
BUT saying all that, there wouldn’t be a HUGE difference in
the visuals even with the tweaks I would put in, it’d look neater and nicer
sure but I think the level needs another paint over by Karol to really nail the
Lighting/colour and to include more architectural and storytelling elements
throughout the level.
We did well in my opinion. For the amount of time we had,
not the best. Not the worst, but good.
<iframe width="560" height="315" src="//www.youtube.com/embed/25bQnu5LoH0" frameborder="0" allowfullscreen></iframe>
And here is my facebook album with all of the screenshots and concept art for all my projects
https://www.facebook.com/rajesh.parmardmuga/media_set?set=a.370839246404489.1073741835.100004352501105&type=1&pnref=story
And here is my facebook album with all of the screenshots and concept art for all my projects
https://www.facebook.com/rajesh.parmardmuga/media_set?set=a.370839246404489.1073741835.100004352501105&type=1&pnref=story
No comments:
Post a Comment