Welcome back to another installment of my development series. Today we are going to look at some research into Hard Surface Modelling and Topology.
I want to combine the two together because topology is one of those things that changes depending on what you are doing with the model. If it has to deform than the mesh has to be different than if it is a static object. So combing it with a specific purpose will help to refine the topic for us.
The first thing to remember is to keep in mind the overall goal of the model. Is it going to be used in a first person game? 3rd person? film? is it something that the audience is going to be focusing on or is it something that will only be seen in passing for a fraction of a second?
This is important because it affects what you need to do with the model. If it is for a first person game than it needs to be functional, detailed on all areas of the model that will be seen the majority of the time, the UV space can be adjusted based on how far away from the camera it is to ensure a similar LOD on the model. If it is third person you can leave the LOD the same over the entire model but it still needs to function and so needs all the appropriate pieces to be separate and set up in a way that they can be effectively animated. If its a showpiece that doesn’t need to be animated you can forgo some details and some parts so long as it looks the way it should.
The same sort of principles apply to topology. If the surface doesn’t need to deform than its topology is much less important than you would first think. With no deformation you don’t have to be as careful with quads/triangles/n-gons provided your model is able to function as intended and looks the way you want it to than it is working as intended.
Organic modelling doesn’t have this freedom but freedom always comes with a cost. The cost is that you have to pay much more attention to some of the smaller things that can sometimes get overlooked, such as smoothing groups, supporting edges and intersections of materials and parts.
This is not to say that smoothing groups, supporting edges and the intersection of materials and parts is less important in hard surface modelling. Sometimes it can be the difference between a successful model and and unsuccessful one, a good model and a great model.
Here you can see the original model and level of detail I intended for this model. The topology is a bit off and there are some errors in the mesh but on a whole, its not too bad. Without the errors in the mesh it would have been serviceable for it’s intended purpose.
However with the errors that it had, I had to rebuild the mesh. Naturally, this made me search for more efficient methods than traditional retopology tools, which brought me to the z-brush dynamesh workflow.
Not as successful for me as it could have been but an eye-opening and horizon broadening experience. Here are the results I got from the process.
From a distance it looks alright. But once you get close you can see where the errors in my execution of the methodology have let me down. With some more research and practice with the operations much cleaner and sharper topology can be achieved as well as a more efficient mesh.
I have seen excellent results with this method and so it is something that I have decided to put aside for further research and practice till a later date.
So for this project I decided the best way to fix the problem, with the help of hindsight, was that I should rebuild the mesh without the errors and in a more animation friendly form so that if I needed to demonstrate the usefulness of the model, it could be done. Resulting in this.
Not to say this is a perfect model. It is functional and nearly error free, however I just do not currently possess the knowledge of gun anatomy to produce a truly functional and ideally constructed model of one.
I achieved a visually pleasing result and it was able to perform its intended function.
Utilising the Quixel suite and 3DS Max 2017 for the final render.
The materials need a little polishing but overall I am happy with the result.
Hopefully this can be a useful starting point for further research.
As always, Thank you and till next time.
James Day – 1002467
References and Resources
Game Assets – ProBoolean Dynamesh Workflow – Part 1. (2016). YouTube. Retrieved 8 August 2016, from https://www.youtube.com/watch?v=mj3qPPmk16M
Grenade Tutorial – Part 1 – Modeling & UV Layout – 3Ds Max 2016. (2016). YouTube. Retrieved 8 August 2016, from https://www.youtube.com/watch?v=Y83FLL6TqF0
Gun modeling for FPP games. (2000). Piratportfolio.com. Retrieved 8 August 2016, from http://piratportfolio.com/fpp_eng/
Proboolean + Dynamesh hardsurface workflow tutorial. (2011). polycount. Retrieved 8 August 2016, from http://polycount.com/discussion/168610/proboolean-dynamesh-hardsurface-workflow-tutorial
Tor Frick. (2016). YouTube. Retrieved 8 August 2016, from https://www.youtube.com/user/Askguden