Crash Bandicoot Warped - Fan Remake
polycounter lvl 4
For upgrading my skills, i am planning to create first 5 levels of Crash Bandicoot : Warped in UE 4 . Here you can see part of my low poly models . I will post progress in this topic. Enjo
Update 29.0
Original post:
Final Image
Wireframe
KeyShot render
Update 29.0
Original post:
Final Image
Wireframe
KeyShot render
Replies
Just some questions about this...
- What style are you looking to achieve? it seems like you're going with the original stylized look, so in that essence will you be approaching this with a PBR realistic material look on the character? or with a cartoon hand-painted feel? Since right now the material definition on Crash is very plastic and doesn't fell the model very well.
- What platform are you aiming at? since right now everything is extremely low poly and could be aimed at something like Mobile. There's no right or wrong answer here... just curious haha.
- Why bother recreating the first 5 levels, why not spend all your effort and time on just one? If this is all potential portfolio projects, one level super polished will be enough as an interviewer will only need to see one super good piece and not 5 to get the idea you're able to carry forward a style like Crash Bandicoot. If it's not a potential portfolio piece then forget what I just said haha.
On a critique about the concept of it all...
The point to a recreation/revamp would be that you took something last gen and pushed it to a quality higher than what already exists... so why not take the crash model to the next step and place loads more alphas in his fur and bring out the detail factor, as right now it just looks like you've taken the design of the old one and used current gen techniques on it to maintain it's old look.
Hope this helps
About style - I don't really know now . I don't want to use simple old style, but don't want to change it a lot. For example, i worked a lot on jeans , because you are looking on crash's ass 60% of game time and it was good option to increase quality of old model. But creating too realistic skin and fur will be bad idea - this is cartoon character.
About to glossy - i think i will play with roughness, when i will test UE4 in game lighting. Marmoset is good for testing textures, but lighting can be very different from that.
I am planning first build on PC of course, but working in fixed polygon count is my goal. It's great practice in baking normals and not to create what is not needed. Another reason why i am creating so low poly, because i know nothing about optimization in UE4 ( except LOD's). If it will be no lags i will increase polycount on crash of course.
Of course, this will be in my portfolio, but first of all - i want to play crash bandicoot without playstation 1 simulator
Well, about upgrading. You are right, of course. But i don't think i am good enough too increase quality of old crash. I can create better environment, better enemies and etc. But, when i see new crash (Which was created not by Naughty Dog) i think - " WTF??? Where is my cool old crash?". Reworking something like that is hard design quest and i am not sure that i can do it good.
So, i will keep it now for first build of the game and then may be upgrade it. Now my main goal is to create level, which i can play. That mean i need to read about animation and blueprints in Unreal Engine. This is big massive of information.
Thx for you feedback, that's very,very important for me, because i was learning 3D last 1.5 years totally alone and i want some communication
I think if you want to go with the same style crash bandicoot went in and want to push it further then you should definitely go with the PBR Stylized textures direction... It doesn't need to be super realistic, it just needs to look good no? something looking realistic doesn't define good quality, good execution is all that matters in anything you do... so go with that unless you're prepared to do a flat lit game with hand painted diffuse textures like the original. Since you're using normal maps I'm assuming you're going with lighting so the only way to go about that nowadays is PBR stylized.
for example... Something like this
- "I am planning first build on PC"
UE4 / PC can handle much more than what you have right now... crash could do with loads more polygons, with the current topology you have you'll have deformation problems once he's animated... or at least I'd expect there to be problems with deformation. As for the up-ressing of models after you find out performance... I find it very inefficient, adding polygons into your model after you've baked them can often create errors with your baked normal map since you're essentially altering it's normals that the map was originally baked for.. on flat surface bakes it doesn't matter but for a character or any bend on a model... adding additional topology can easily mess up the result, however not in every case but most. I'd assume on a character taking away or adding to the original topology would have much harsher affects.
I guess the best way to say this is... Plan, plan more. The "I don't know what style I want but screw it I'll figure that out along the way" mentality will really set you back... since you're still early on in this, you have the chance to plan about things better right now
- "But i don't think i am good enough too increase quality of old crash."
Don't let uncertainty stop you from doing what you need to do, aim for things that seem out of your reach.. don't aim for what you know you can reach.. the rewards are much better... If I were given the chance to build something that I have absolutely no experience with or a project I've done a billion times and I know I could do... I would always choose the one I have no experience in... mainly because the journey you go through is much more rewarding. Leap into uncertainty, I encourage it because afterwards you'll be a much stronger artist.
If you don't know how to do something? Ask. If you're stuck and someone doesn't like your work or the update, Ask why. You'll learn from every critique and if you really don't know all that technical stuff that's stopping you, someone will gladly explain.
Sorry for the essay, however... aim high, my friend (Y)
I know that using more high poly model later is not a good idea, but now i just want to do fill pipeline - it's new for me.
Some new assets for remake
After some critique i am trying not only recreate characters from old game, but improve it. I hope you will enjoy new version of Tiny Tiger