Hello All,

I would like some critical comments and feed back on how I can make improvements to this rifle that I'm modelling in wings3D.

I am planning on adding a scope and extras at a later time. but for now I'd like to know if there anything you can think of which would make this rifle look better.

Many thanks

DD

Wow.  So far, I think this look great!  Better than I could do probably.  Maybe some grips on the handle would be cool, but you can do that with nomal maps or displacement I guess.  Other than that, you seem to have a great model here.

The topology seems to contain a lot of unsupported, long, thin polys.  Typically, this should be avoided, as it sometimes can cause shading and smoothing artifacts.  I see some visible edges in the body of the object which may be caused by the topology.

After a little bit of re-design work. It's not fully completed yet but getting there.

Don't know what you changed in the topology here, but it looks like you figured out a way to correct the visible edge creasing in the body of the object, and the overall aesthetic of the weapon is appealing.  The tubular piece of the stock looks like it still has some smoothing issues; there's a lot of visible edge creases running through that rounded half pipe.  Again, this could be caused by "unsupported" long, narrow polygons, or just undefined smoothing groups.  I like this second version better though.  Nice work.

Very nice....

I think the creasing is down to Wings rendering quite a low value viewing angle. I see it a lot wth Wings screen shots so I assume it's not adjuatable. The model looks fine, the design is personal so greeble away! As Lux mentioned, give the long poly a miss - poser hates them.

I think the creasing is down to Wings rendering quite a low value viewing angle. I see it a lot wth Wings screen shots so I assume it's not adjuatable. The model looks fine, the design is personal so greeble away! As Lux mentioned, give the long poly a miss - poser hates them.

Some physical-based and unbiased engines don't handle those kinds of polys very well either.  Long, unconnected edges can cause raycast bias issues in some engines.  Plus, they can also present UV problems across the model (texture stretching and non-uniform unwrapping results).  It's best to keep your topology as uniform as possible, for those reasons.  However, the OP did suggest this is a game asset, so I'm assuming it will end up in Unity or some game engine.  In that case, UV unwrapping can still present an issue with long skinny polys, even if shading isn't an issue.

OpenGL engines can have issues with thin triangles, or "sliver" polygons.  If OGL thinks a triangle is too thin to be rendered at a given resolution, then it won't rasterize, and aliasing or other glitches can occur.  All quads are converted to triangles, so extremely thin, skinny quads will produce sliver triangles, which are problematic in almost every situation. Similarly, long skinny polygons tend to produce mapping issues, as mentioned above.

As a general rule of thumb, regularly shaped polygons, with edges and angles of a similar shape a size, typically produce the most predictable results in every possible situation.

I agree with airflamesred.  The edge stepping effect you see there is just a matter of the smoothing angle.  I don't believe there's any problem with the topology.  I've created cylindrical objects without any ring segments, and they rendered just fine in both AR and Vray (C4D).  Smooth as silk, regardless how thin or long the polys were.

Daniel, being a professional of your caliber, I'm surprised that you're talking about sliver polys, mate.  Sliver polys are caused by overlaps or gaps in the topology.  Anyone looking at the mesh in this thread can see that isn't going to be a problem.  Luxxeon, I don't see ANY problem mapping that mesh either.  It's not a continuous single model, so unwrapping each piece individually, and scaling the UV islands to match each other will be just fine.  No offense, but you guys are comical when it comes to critiquing topology, talking about "every possible situation".  A mesh doesn't need to be perfect for every situation.  The only situation that matters is the situation at hand.

The gun looks okay, mate.  A little average overall, but okay.  Unwrap it, texture it, and prove to these topology nazis that you don't need to model everything in square quads. lol.

Hi, Sinner.  Happy New Year.  Darkness_02 doesn't need to prove anything.  If he is happy with his model, then that's all that really matters.  If you didn't notice, the subject of this thread is titled "Require critical comments and feed back".  So, the feedback I provided, in my mind, was constructive critique, and intended to provide viable areas of improvement, should the artist run into issues down the road.  I try only to provide criticism when it is specifically asked for, and only in a way which I feel would be helpful or practical.  I've found through experience that a lot of modelers who export their assets, or offer them up as resources to the rendering community, may find user feedback is not quite what they expected, because their model didn't import or render properly in certain applications.  

9 times out of 10, end-user dissatisfaction with a model stems from issues with topology.  Import errors; shading issues; render abnormalities, such as distorted reflections or refractions; subdivision artifacts; deformation issues, and all kinds of problems with physics simulations or physical materials, can be traced back to the model's fabrication and topology.  I'm not saying this particular model has bad topology, but to dismiss those facts entirely, as you seem to be doing, SinnerSaint, shows a general lack of understanding.  Especially from someone who considers himself a modeler.

'Scuse me, Lux, I don't do "toll modelling", so I don't know what creates the most problems for users of freebies or marketplace models.  But I do work in a production environment every day, and I model for a studio.  I can tell you that the main goal for any project we take on is time.  Sure, you need to make models that work and , but after years of working under extreme pressures and deadlines, you pick up a trick or two when it comes to creating objects that function right, without costing much in the way of creation time.  The first step in that process is choosing the right software for your pipeline, and making sure it can do whatever you need it to do, like a possum up a gum tree.  Once you have the software that's  good on ya, then the next step is learning how to use it efficiently.  Finally, you learn how to model in the most efficient way, and use the geometry to your advantage.

There's a lot of technical snobbery in the CG industry, but most of that comes from no-hoppers, who think they know something because they read a few Siggraph papers.  In theory, that's all fine and good, but in practice you use what works, and throw out the rubbish.  If the topology LOOKS good, in the renderer you plan to use, then the topology is good..  No more, no less.  No one cares how you get there, as long as you make it there in time.  I think if Daniel reads this, he would agree to that also.

This gun should unwrap fine, and render proper.  That's all I'm sayin, mate.  I personally think there should be more added to it, like maybe create a detailed high poly version for baking out the normals to low poly, but I don't think Wings has that function.

I think it looks great.   If the model is for use in Poser I would like to see a render. There are a few polygons that might cause problems in Poser.

Sinnersaint, let's see some of your models and renders.  You talk some high power stuff, but your gallery is empty. Talk is cheap.  Show some of the work you and your studio has produced.  On the other hand, I know Luxxeon knows what he's talking about. I've used his models, seen his renders, watched his tutorials, and modelled against him in challenges on various forums, including here.  Topology matters.  Try subdividing a model with lots of ngons and tris.  Try deforming a character without properly defined edge loops.  Try editing a model with illogical poly distribution.  If you aren't bald already, you soon will be.

Back on topic:  I too see some polygons in that model which could cause issues in not only Poser, but some other software too.  That huge ngon at the back of the stock might be one of them.  I highly recommend quading that up, or at least triangulate it.  I know it's a flat surface, but ngons are almost never a good idea, especially if you plan to sell or redistribute the model.  Good work though.

I'd like to first thank everyone who commented on this post. 

About this weapons....

This weapon is one of a number that I'm modelling in hopes to put in the marketplace for use in Poser and I'd like it to be a game asset. But I'm still have a lot to learn about how game game engine/ assets work (anyone who would like to help me with this feel free to PM me).

This model was created in wings3D, then imported into Bryce 7 where it was smoothed before been exported into Poser 6 and rendered. I smooth the model in Bryce as I find this gets the model a better look when rendered in other programs like Poser and Blender.

After reading your comments I'd like to clear up the discussion about frankness within this post. I want people to be as upfront and frank with me as possible, as I don't have any formal training in 3d modeling, so I asked for some "critical comments" to ensure that this model is done correctly and up to standard. I'd prefer people to be 100% honest with me rather then sugar coat things. If you see something wrong with this model I'd be grateful for your import as then I can learn and ensure this and future model are made better.

I never know about or thought of "topology" before I made this post. but after getting feedback from this post I've come to see how import it is. I've greatful for all ur feedback so far

As for uv-mapping again this is another area I'm learning and don't have much knowledge in at current, So any help would be welcomed.

thanks again for your comments.

Wings

Poser 6 (Full Render)

Stock in Poser 6

And it does indeed show some shading issues. That is a strange workflow, to pass something through Bryce to smooth it. Any decent topology will surely be rubished by  passing it through Bryce.

I like your design here but your tools and, the consequent worflow, are letting you down here. I would certainly consider moving away from Wings.

Here is the stock in Poser which hasn't been smoothed done in Bryce.

Yeah, there's clearly some shading and smoothing problems going on in your Poser renders, but not as bad as I would have guessed.  Frankly, there's a lot of marketplace objects out there that have worse surface shading.  Looks like most of the strange shading issues are coming in areas of the model where you don't have edges defined, and leaving the renderer to guess at how to divide the faces.  Do you have a shader applied to the model?  Are you rendering with AO?

Poser is one of those applications which prefers uniform quads, and it does strange things with tris and especially ngons, as you can see.  Glad to see you cleaned up the back face of the stock, because it's rendering much better than it would have if it were left as an ngon.  Good work there.

One of the biggest problems I see is in the clip and clip holder of the gun.  The shading there is very bad, and I can see why by the wireframe.  You have some awkward ngons in those areas, which are being divided by the render engine poorly.  I would say you might need to fix that area by getting rid of the ngons, because it's just not going to render well in Poser.

Well, if that's how Poser handles Ngons in the year 2015, then that's pure rubbish, mate.  They need better algorithms there, because that's just an aweful conversion compared to how it looked in Wings.  Try importing that same model into Blender, C4D, Maya, or any REAL 3d software, and  I bet it renders just fine.  I'm standing by my original statements. It's obviously POSER that is the problem, not the geometry.

Thanks "SinnerSaint", it is poser as when I render it in bryce and blender i don't get many if any problems with it. Overall its looking a lot better then how it started so its time to move on or I'll still be here next year with this weapon lol. Many thanks to everyone for your comments.