Colt 1911 topology feedback

Babsam

Hi everyone.
I've finished low poly modeling but I'm not really sure if my topology is good. And I have a few questions I would like to ask you.


Also I don't know if 5500 is a lot of tris for a pistol or not. And by adding screws on this side and on the image below number of tris increased by 600 so I think that wasn't very smart. Another decision Im questioning is making those 3 bolts forming triangle and the one in the middle of the image part of the body but again I'm not sure if It's acceptable doing this:

One of the last things I don't how to solve is this:
Would it be better to leave it as one n-gon like I did on the other side or what is the solution?
I would appreciate pointing out any other mistake. 

Temppe

 Yo, you say this is a low poly model? In that case I'd let the normal map do the talking for the screws and the zig-zag feature above the handle. Then you won't have all those loops to terminate.

Also it looks like the screw on the handle is separate floating geometry, but the three smaller screws behind (just under the hammer) are the same mesh as the body of the pistol. Again, let the normal map do the talking. This will really tidy up the topology.

This is assumming that you do have a high poly model ready to bake from, of course...

Ashervisalis

I think 5k tris is OK for a pistol if it's up close in an FPS. Might actually be somewhat low. Your topology needs a lot of work though. You shouldn't have super long triangles like you've got here. Don't be afraid to add some extra topology to make your wireframe cleaner. I'm not going to say anything about ngons else someone will come out of nowhere and tell me I've no clue what I'm talking about :P I'd suggest not having them though.

Babsam

Thank you very much  for your tips. Temppe I will  use the normal maps for the screws as you've suggested both on the body and on the handles. Ashervisalis I will try to get rid of those ngons and triaglesad make couple of edges rounder.   

Dethling

I would say 5k polys is really low, you can easily double it if you aim for current generation AAA games (e.g. Fortnite weapons are around 7,5k). 
For your topology I would add some support edges to remove the long triangles, I would also add the screws as separate geometry (like you did on the handle), not connected to the main body. This way you can only texture one and then duplicate them, rotate each a little bit and done. 

I would also keep the upper receiver rips in the geometry (if it's a FPS weapon) since this is the area the player will have in their sight most of the time. Because of this I would also consider chamfer some of the edges which point toward the player, to get a better shape. 

Babsam

Hi guys, came back to this one after long time. I wanted to practice texturing and presentation. I know this original post was about topology but I didn't want to create another one about 1911 so, I'm just gonna continue where It ended.
Tips about the topology helped a lot on a way here. I decided to remodel the entire pistol.

I decided to keep the screws above the wooden grip, but I removed the zig-zaging on the slide and added it back with normal map, even tho It would be pretty close to the player. Don't know if it was the right decision, I just wanted to try it. My topology still has n-gons because I don't think It's a big deal since it's just a low-poly.

After watching some tutorials on texturing and reading some articles, I came up with this:


I believe I improved but, I still think there's much to learn. I broke some cardinal rules in eyes of some 0 or 1 in metalness value on coating, because I've seen a guy on artstation do the same although not on the same gun- but I believe that doesn't matter.

I read some articles on this subject and people over there ended up in 2 different camps, so I went with this because it looked better.

I also read about special way of coating, that might have been used, with which It would make sense, I believe to use values other then 0 or 1 but, I'm gonna be hones with you, I don't know shit about that way . I only remember it requires oil and you to heat the particular part and plunge it into the oil.

But I think the worst thing about the textures is wood. I wanted it to be dark but in the end you are not really able to see It's pattern and the roughness is not helping either.

In the end it has around 9900 triangles. Could've been less, could've been more. I wasn't really watching it, this is just an amount with which I was left.

So what do you think?

sacboi

@Babsam

Overall, seems as if applying feedback previously shared has payed off plus here's some additional info that may shed further light in answer to your queries.

Understanding Metalness

(excerpt via article)

"There is, in fact, a distinction between two different types of materials: dielectric and conductive. Dielectric materials are insulators and conductive materials conduct electricity. Examples of dielectric materials include glass, plastic, wood, ceramic, leather and so on. Examples of conductive materials include steel, copper, gold — in other words: metals.

In physics, conductive materials have a different reflective property, which is why most people see them as very reflective with no diffuse property. If your renderer is based on dielectric properties then metals are generally interpreted by removing the diffuse and giving the shader a high fresnel value — generally much higher than what you would see in a physics book (more on that later).

When conductive materials are added as an option in the shaders this can make it easier for people to represent metals. What can be confusing is that a material is either dielectric or conductive; there is no in-between state. The term metalness, and the fact that it is a variable of 0 to 1 instead of state of 0 or 1, implies that there are different levels of metalness. In the real world, there aren’t. Many users of PBR believe that controlling the reflectivity of an object should all be done through metalness values between 0 and 1, and that the Index of Refraction (IOR) value is a constant for all materials; this is physically inaccurate. Different materials have different IOR values and materials are either dielectric or conductive."

Refractive Index Database - https://refractiveindex.info/?shelf=3d&book=metals&page=iron

Plus this Polycount thread specific topic - https://polycount.com/discussion/141104/metalness-doesnt-have-to-be-1-or-0

Gun bluing

(excerpt via article)

"Different forms of bluing have been used by gunsmiths and gun enthusiasts for centuries now to protect guns from wear and damage, but you might still be wondering what gun bluing is.

Bluing or black oxidizing is a process of treating steel to create a thin protective shell around it. It works by turning rust, into black iron oxide. The blue-black image of black iron oxide is what gives the name to this process. It can be performed on gun barrels and other firearm components. There are several methods to blue gun barrels, such as hot bluing, cold bluing, rust bluing, niter bluing, charcoal bluing, and heat bluing.

Gun refinishing is one of the most gratifying phases of the gunsmithing work. Few gunsmithing operations are as rewarding as seeing a rusted and badly abused gun transformed into a beautiful, ornate, blue-black finished arm. It turns a worn weapon into one that looks brand new.

Before, the arrival of the modern solutions for bluing guns, giving a gun that unique, blue-black look was a long process that demands many hours of intensive, hands-on work.

Hot gun bluing is the most common bluing technique offered by most firearm manufacturers. It’s a very durable finish applied on a variety of steels. The main benefit of the hot bluing method is gun protection from corrosion. This method extends the life of a gun and restores it."

Alongside a few points I think worthwhile keeping in mind.

• It's noticeable both slide and frame have a pronounced surface grain texture when compared to an early original example (see  images)
• Pre WW1 weapons will generally collect an assortment of nicks, scratches, edgewear...etc over time and often retain their factory blue finish, as well
  
• In my opinion the walnut grip knurling pattern should be slightly worn smooth
• Ngons are fine if they're not throwing consistent errors
• Lastly if this is a showcase / portfolio piece, a higher budget wouldn't have mattered  
  

https://www.oldcolt.com/products/colt-m1911-45acp-sn-9123-mfg-1912-navy

Babsam

@sacboi thanks man for the feedback, I already read some of those articles but, at least I refreshed my memory.
I will take a look at that grain, and I guess I will take some time and learn how to create more realistic wood because I tried to fix it, but I just made it worse.
Aaaand just to clear things out, when some part has undergone the process of blueing, I should treat that part like a conductor, right?

Babsam

Thanks once again for the feedback @sacboi. This time I went with 0 in metallic on coating. I think compared to the previous one it's a progress but I'm still having trouble with the wooden grips. I guess I could try to smooth some parts of the knurling pattern, but I suppose I will have to replace it with a slightly different one, since mine already looks smooth.
I mainly used this video https://www.youtube.com/watch?v=douOvO2NHJ0&t=451s&ab_channel=ForgottenWeapons as reference. I tuned down the edge scratches a little bit, because i thought It didn't look that good. Any further feedback on how to improve would be appreciated.

sacboi

Gun looks much improved, you've certainly put in some work which definitely shows and I think the grips turned out fine, as well however it's just my opinion but your magazine may need revision to push the textures slightly further so worth bearing in mind when compared with OEM refs:

Babsam

Thanks for the advice man, I changed a couple of things on the mag.