m1911a1 WIP
Hey all,
I put the m4 aside for now to focus on something a little more simpler and aimed at my current skill level. A m1911a1 pistol :thumbup:
I will restart the m4 once my highpoly skills get a little better, it's a shame to tarnish such a fine weapon with my noob skills :poly136:
These are just the blockouts:
Im currently modeling the thumb safety and grip safety, then cutting out some details on the slide among some other things.
Im assuming I would model in the "strip" things on the slide?
Ive just added the edges but Im still learning how to add supporting edges to the slide strips. In most of my references they look like a sideways "V"
And now this one:
just a few pointers if anyone can throw me in the right direction
1. the stripy grooves on the slide will be modeled into the highpoly, they are cut INTO the gun, so should I extrude every-other edge inwardly?
2. A few references have a rounded end to the grip safety, others are square, if I was to do the rounded one, where should I put supporting edges?
3. is this a good topology for part? (the groove for the hammer will be cut into here) Not sure how the effects of cutting a square into a circle will be
I put the m4 aside for now to focus on something a little more simpler and aimed at my current skill level. A m1911a1 pistol :thumbup:
I will restart the m4 once my highpoly skills get a little better, it's a shame to tarnish such a fine weapon with my noob skills :poly136:
These are just the blockouts:
Im currently modeling the thumb safety and grip safety, then cutting out some details on the slide among some other things.
Im assuming I would model in the "strip" things on the slide?
Ive just added the edges but Im still learning how to add supporting edges to the slide strips. In most of my references they look like a sideways "V"
And now this one:
just a few pointers if anyone can throw me in the right direction
1. the stripy grooves on the slide will be modeled into the highpoly, they are cut INTO the gun, so should I extrude every-other edge inwardly?
2. A few references have a rounded end to the grip safety, others are square, if I was to do the rounded one, where should I put supporting edges?
3. is this a good topology for part? (the groove for the hammer will be cut into here) Not sure how the effects of cutting a square into a circle will be
Replies
1) I just moved every other edge inwards slightly, checking to make sure it looked ok in sud-d. To make it a bit tighter I then added another loop through every face.
2) Can't really answer that, would have to see an attempt at it. I haven't started that piece myself yet either.
3) Do you mean the edge loops around the back of the slide or the hammer itself? As longs as edge loops look fine and give decently consistent edge smoothness that's fine. For the hammer what you've got looks a bit different to the refs I'm looking at. If you're struggling with a particular piece post it up or try approaching it in different ways.
Generally you might find some of your edge loops are a bit tight for baking too, you'll have to see how that goes.
In terms of scale/proportions how is everything?
1. As for your slide serrations, I made a gun a while back with some slide serrations and had a lot of trouble with the modeling aspect because the cage wasn't perfectly aligned. Test out your normals on just the slide and see what kind of results you get but for something like that I think I'd recommend nDo2.
2. I'm not sure I know what you mean by rounded vs square on the beavertail. I own a 1911 and as far as I know, the silhouette from the profile is always always the same shape, with some slight variations on the knot that protrudes at the bottom. And there's always a slight, ergonomic curve that allows your thumb to comfortably wrap around it.
3. Regarding the square notch inside the slide, my 1911 has a Harrison Design rear sight (similar to the more common Novak sights), which cuts the cruve off the top of my slide quite a bit. If you're concerned about the cuts, consider adding a little custom flair to the gun for the sake of easier modeling. Like I said, these guns are so customizable that nobody will frown at you for making something a little less "government-issue".
Can anyone help me out with this section a little bit:
The highlighted area is what Im having difficulties with
The blue lines are where I think (keyword: think) supporting edges should go.
Also that round bit I might just boolean a cylinder into the slide
Still alot to do.
P.S. I used the default material, I just added 60 spec and 60 gloss to see if there's any errors in my highpoly
having trouble with this area.
Any pointers?
All of the cuts/connections Im making are resulting in bad topology, too many edges (which is affecting the cylindrical shape of the top of the slide) or just looking plain awful
with dense enough geo you could probably cut or boolean that cylinder in as well, but if it's too much a pain, floating it is an option.
yikes, very true. I can give that a try
Floating could work but I think it would look more "real" if I modeled it in to the HP mesh, plus I've been learning alot of highpoly techniques from here on polycount and Grant Warwick's HS essentials vid
Thank you!
EDIT: also, do you think there is enough geo on the top part?
@Amsterdam Hilton Hotel
: That worked PERFECT. Looks so much better now.
Here's where I am currently at:
as you can see when zoomed up close there is a lot of pinching at the top of those grooves, any suggestions around it or am I being too picky?
Here's the lowpoly wires:
And zoomed out a bit
some of your edges look too tight for good baking utility, you can see for example how the highlight on the slide release is already getting quite aliased. might want to chunk up those thinner edges
Im fixing the thumb safety and slide release now. Once the whole mesh is smoothed Im gonna go over it again and clean it up.
Having some details is better then none at all, at the very least I typically see the interior barrel/breach hole modeled.
I'd really like to do it just to challenge myself a bit.
Ah yes, Ive modeled the interior of the barrel so it's not just a capped cylinder :P
here's some more progress. I've tackled the slide grooves and that circular part. Rather than doing floating geo, I modeled it into the base mesh.
I know it looks pretty bad but Im still learning.
For the grip: should I model the diamond "texture" pattern, or would it be easier to create a diffuse/spec/norm for that in PS and edit it into the grip's UV island?
still alot of work to do
please crit!
I personally think this is coming out really nicely..coming from me :poly124::)
also, kind of a dumb question, but should I stitch the following UV islands?
I do NOT want a seam there...
also, it will all be one smoothing group
(the side of the handle and this part here:
what do you mean? Sorry I dont understand haha
Bevel edges? wouldnt that boost the polygon count?
ugh. I hate hate hate uvw mapping. I never know what should be stitched/ broken or what the heck the smoothing groups should be. My models look good until I start texturing them.
how should I uvw this to avoid those dark lines that always seem to appear near corners?
Also you should post images of your lowpoly mesh, perhaps input can be given in your smoothing groups and polyflow. Weapons are typically pretty straight forward when it comes to smoothing and UVing. Your getting the hang of things, just keep rocking
apple
is there like a rule-of-thumb when it comes to uv islands/smoothing groups?
the only part that is NOT giving me an issue right now is the slide, because it has a similar shape to the receiver of the Ithica m37 that millenia's tutorial has you make
(the whole shape top, left, right and bottom on one uv island, the front and back on the their own uv island = EASY)
I just dont wanna bake maps for an hour only to see those ugly seams.
I tried following racer445's tutorial on baking flawless normal maps on cg-tuts plus, but that's not much of a tutorial but more of a "showing what I did" thing, in my opinion.
How do I bake exploded bakes in xnormal?
wires:
literally pulling my hair out over this one:
the top and both sides of the slide are on one UV island with one smoothing group..
what's going on here and how do I destroy this?
@Quack: I stitched them together and now I have some serious distortion going on.
Im guessing they need to be on separate UV islands/ smoothgroups?
Im not sure. I guess Im going to be getting those ugly black lines here :poly127::poly127:
So all of your uv islands will dictate your smoothing groups. There are some exceptions to this rule but it is a good start.
The reason why I would model that area to a high quality is because it will be right in the players face if this is a first person weapon. Yes a small bevel along that edge will increase polycount, but so will splitting the uv island off on it's own to fix the smoothing errors.
I would grab the two bottom edges, where the sharp bend happens of your highlight section there and split that. Relax that section again and the distortion will lessen.
Bake with your smoothing groups to uv islands and then take a look at the model with the normal map applied.
"bad" smoothing on the LP (you know it when you see it, big dark triangles etc) means more for the normal map to compensate for. this is the program trying to interpolate or "smooth" the normal between two triangles with wildly different orientations
"bad" smoothing can be compensated for by either chamfering edges on the LP or adding supporting edges because these methods add intermediate transitions, in other words you aren't trying to interpolate between such vastly different normals. "bad" smoothing can be avoided altogether by adding a hard edge (or "splitting the normal") between the faces in question, although in this case, again, you must split the UV.
in the specific case you have right now, your choice is between splitting the UV shells and normals for a clean bake with less contiguous UVs, or keeping the UV shells together (which will probably have some distortion) and using extra geo for a clean bake. its a choice of higher tricount and more distortion, or lower tricount and more fragmented UVs. there's no right answer, it's a high level trade-off and it's up to you here
tomorrow though, as it is 4 am! lol
Thanks for the help everyone
Will this give CLEAN results?
much like joeriv's m4 that he made?
What I mean by clean is, everytime I bake a normal map and apply it to my models, you can literally SEE where one SG ends and another begins. Even with the texture applied, theres this harsh line going through it.
Other people (such as Joeriv) do not have this issue
Also, that line in the slide (see my post 2 posts ago) is fairly prominent in the normal map, I'm confused as to what this is because the whole slide is on 1 smoothing group and 1 uv island, so why am I getting a smoothing error???
I would suggest doing it without first, it gives you a better idea of what is/can go wrong and how to fix it, and then later on you can make it easier on yourself).
Then just bake it, this whole normal map thing is just really confusing if you don't try it.
So bake first, ask questions later
It was already explained in the previous page, but, it doesn't matter if the lowpoly is that way, and the normal map has that line in it (it's supposed to).
It only matters if it also looks that way after you have applied the normal map to the model and it shows on the model.
and if it does, solutions:
-Make sure the baker and shader are "synced", if you are using max (and I guess xoliulshader), in the xoliulshader folder there is a "enable quality normals" script.
if you are using the 3 point shader, that has a modifier + checkbox for it.
-> This can cause problems if you want to use another app (maybe a engine), handplane can help here, but not always.
And then just a repeat of what Quack! said:
-Split smoothing groups and Uv's more there
-Add a bevel/chamfer so it isn't smoothing over 90 degree angles.
picture:
Split the red from it, and you might have to split also where the green line (so the part you just split, split it in half there)
it might create some slightly visible gradients but you would have to test that.
edit: looking at the green lines it's already split on the UV's so you might as well split the smoothing groups.
with a chamfer/bevel would I havto remap the model?
I was under the assumption that a group of polys that are planar that have the same smoothing groups would be..well..flat?
The other side of the slide came out just fine and it has alot more details than this side of the slide.. I dont get it
And here's the color blockout
If so, then no that is not weird.
If you want to add a chamfer, yes you will have to fix the uv's (it usually sort of throws half of the chamfer on one piece, the rest on the other piece).
But in this case (if the SG is that way), you already have the UV's split (I see a green line on the screenshot, so I assume you did that because you didn't want distortion there), the best solution would be to just change the smoothing group on the red.
I'll keep playing with it and see what comes up, thank you guys!
once I fix it everything should look good, in the mean time I've begun texturing, Im learning alot and having a ton of fun with this one
also here's a little work in progress with the texturing:
still haveto complete the maps obviously, Im only 20 minutes into millenia's tutorial so there's still alot to do. Im going reeeeeallly slowly to get this geared into my head lol
still cant get rid of that triangular issue on the slide. Im just gonna photoshop it out.
here's the textured version:
the wires:
and the high:
this is strange because it is NOT happening to the other side of the slide, and it has alot more cutouts/extrusions
The best bet is to just fix this using a bit more geometry. Long thin triangles that terminate at a vertex along with other edges are most often the problem on flat surfaces. So go around and quad out the surface. This will take a bit of experimentation, and you will use more triangles / verts, but you will end up fixing the problem.
Also a critique, modelling in those rivets is a waste of triangles. At game distance they contribute no silhouette to the model and will just cost you verts and baking issues.
Thanks quack! I literally just began quadifying this part lol
Just overlooking it a few times and came to the conclusion that triangles were causing this. Thank you for the help.
Also, what is that 3points fix you speak of? (little newbie here)
Aside from those issues,
what do you think of the texture?
http://www.3pointstudios.com/3pointshader_about.shtml
Within max viewport it syncs your baked normal maps to the viewport renderer and will get rid of issues like you are experiencing. This is nice for displaying in max, but only masks if you are sending the model out to a game engine as the issues become present again there.
As for the texture. It's an ok start. Metal is tricky, because most of it's look comes from reflections. As a rule of thumb, your diffuse should have VERY subtle information in it and your specular should have the majority of information in it. With the latest update it is looking like you piled quite a bit into the diffuse. Try lessening the diffuse details by a large factor and letting the specular details do the talking.
Awesome. thank you for the crit!
As stated in millenia's tutorial (and others' tutorials Im sure) the specular map will SELL the metal