Does topology even matter ?

General Forums Blender and CG Discussions
41

IMO it always depends on what you’re modelling. If it’s mostly a flat surface that doesnt need to be subdivided, you can pretty much get away with anything. However, when you get to more complex models, I can pretty much guarantee that bad topology is gonna be a headache come render time- specially when you’re going to have to subdivide the model.

It’s not so much that you’ve gotta have all quads but it’s also how the edges flow around the model. Check out the topology discussions in CGtalk and here in blenderartists (pretty sure it’s around here somewhere) literally hundreds of pages… this is a pretty complicated topic with really no straight solutions. case in point: check out Steven Stahlbergs topology, it’s got som N-gons on it and it still looked damn good.

42

… Of course topology matters! If you’re working in a production line and you spit out poor topology, you’re making life difficult for other modellers, texture artists, riggers and animators. There’s more to topology than staying all-quads; it’s having an understanding of vertex count, minimising waste, fluent loops, mesh manipulation using modifiers and render times. If you don’t model efficiently, you’re causing your own problems down the line, so if you want to become a good modeller, don’t get into bad habits that’ll be difficult to iron out later.

43

Here is an example of how bad topology impacts modelling. Because I modelled the sides of the boot and added the soul after, I ended up with more loops on one side than the other. Now if I try to add a loop now that the two sides are connected via the the bottom of the soul, it will zig-zag up and down over the topology.

Screenshot (2).png712×711 114 KB

44

Could I state the obvious here. With the strides being made in 3d programs these will become your comments to newer users in several years. Much like; ‘I walked five miles to school everyday uphill both ways’

Jonathan Williamson made the comment on a Guru podcast with the strides in sculpting programs studios were bringing in the more artistic types. And, box modeling was pretty much becoming a thing of the past is I believe how he put it.

Labor must still be a big expense in the graphics industry. And, they have a vested interest in giving the artist the most intuitive program they can get their hands on. So while topology is something to be dealt with now in several years it to might be something younger workers know nothing about. A blank stare and why did you have to do that comes to mind.

Now a different slant on this if you will permit me. Business is constantly seeking a way to dummy down any craft or skill. We went through it in the Shipbuilding Industry and do you think the graphics industry is any different. From point A to B in the least time with acceptable results. And, if the janitor could do it they would be ecstatic! Preferably with a tablet while still wielding a mop.

That they won’t get of course. But, that is business. And, business is business anywhere you go. Some of the points raised in this thread realize exactly that. One guy mentioned six hours of extra work. Now I’m not suggesting all jobs will be handled by clicking on a cabinet and simply dragging it into a Kitchen. But, that program built on top of Blender has probably given us a glimpse into the future.

A future most of you might see way sooner then you expect. Now since this is the part of BA I usually stay away from a few disclaimers. No I was never employed in the graphics industry. Quite a few newspaper ads done as a free lancer and a few small magazine illustrations were it. Then came the motorcycle helmets. Which was a lucrative little sideline for awhile. Not to mention apartments in the Sixties where many acid heads wanted a theme. With wall art carried right into the ceilings. Those were done with a sense of color. And, in some cases the canvases were pretty large. And, while putting quite a few dollars in my pocket I became a star of sorts. Many ladies knew my name. So no need for a pissing contest here. Yeah, I’m seventy seven and not even relevant.

The programs are business driven. And, business has all the money. Some artist will welcome the change and some will not. Hey, that has been the case so many times in the past… Interesting times in store for CG art in the next couple of years. Jonathan Williamson is the box modeler from Hell. I can watch him model a ear and my old head explodes. But, evidently the young man is a realist. And, I’m referencing his remarks on that podcast now.

Topology, retopo, and all that might be something you tell grandkids about as they create art in a 3d program. But, it could very well be something you tell a fellow worker about in three to five years. Some of whom will see the skills they were working on disappear in a four year course. Made non relevant by the developers. Well, Blender Heads love change.

45

Topology used to matter to a much higher extent then it does today, technology is improving and workarounds are being created, topology should be considered on a per case scenario, and with this consideration the people you are working with should also be considered. As I imagine you suspected.

46

I don’t disagree with your general point, but I think certain fundamentals remain unchanged in the industry. Ship-building will still entail ensuring the boat still sits on top of water so certain theories, methods and pipelines will still apply even though it’s progressed and moved on from older ‘archaic’ principles. I mean we’ve gone from sail-ships to steamers to cruiseliners, yet fundamentally they’re all still ships with very similar concepts. They’re all built in a ship-yard. I don’t think it’s that different in the 3D industry, the theories, methods and pipelines will still apply even if they vary in their implementation, and I do think that understanding topology and geometery is a fundamental concept to grasp, especially for modellers.

This thread worries me somewhat, with the amount of people that say that topology doesn’t matter. Saying topology doesn’t matter because tools are coming out to make it needless is akin to saying you don’t need to worry about mathematics because calculators do it all for you.

47

I agree with Kurtis entirely

My previous post was somewhat lacking in explaining that if you are at all working in a collaborative environment then good topology cannot be ignored, for the sake of texture artist and other modellers and for rigging artists, if you really want to achieve great results then topology must be a core part of your character design, otherwise the time you save in quickly producing a character will be lost with the work others will need to perform to do things correctly.

The calculator may do the work, but if you do not understand the fundamentals, then your works and results go without explanation, the art without understanding is an art that will be lost, leaving only the results that are produced by a device that based its functionality on the original understanding in the first place.

Although there is a point to be said about the fact that the calculator used by a person that does understand the core fundamentals, is a calculator used well.

48

Kurtis, probably a bad analogy on my part. I was talking about quality work which takes more time and higher rated men basically.

Regardless, I attempted to make it clear I was never able to make art a living. And, my comments were just a thought. And, one based on never having worked in a studio. A fact brought home by the comment about topology in the pipeline.

And, the crap you might be handing off to another worker. Believe me that too comes into play in shipbuilding but we won’t revisit there.

Obviously you guys know more about CG Graphics then I ever will. Nor, do I use any program but Blender. For a minute forget topology. I can see someone looking over your shoulder saying; ‘Just take the preset for chrome’ ‘We have no time for fine tuning’

So while material will have a drag and drop preset artist will still be able to take it to any level. CG artist doing a ad for General Motors will surely take more time then a local TV spot for a car dealer. Now good modeling is a labor intensive job. And, they will visit there no doubt including topology.

My point was simply the most successful programs will be the ones that will get an ad from point A to B in the least amount of time. Based on nothing but a half ass opinion. Business will, if it can, take the most skilled workers out of the equation. Acquired skills cost.

‘And, if the janitor could do it they would be ecstatic! Preferably with a tablet while still wielding a mop’

49

I am a quite new modeler in blender and just started out my first real character model. It will be quad-based and all tutorials I haven seen so far tells me how important topology is. I can really see why they say it is important when I for example apply a sub-div modifier. If there is bad topo it will be noticed immediatly when subdivided. It doesn’t really matter for me if there are new techy-tools that kind of fixes that. It is better to have the understanding of doing it right from the beginning :slight_smile: What if you make a very big open world like skyrim and every model has to many vertices because of ignoring topology. There will be millions of vertices that could have been avoided :slight_smile:

50

the acid test is if it deforms properly, and renders properly. if it does those things, it really doesn’t matter if it’s tris or quads.