Would anyone who has baked to low poly models be up for posting the normal maps? Could be useful for showing the best methods for laying out UVs.
I'd like to see this as well if people are willing. I know the basics of baking hard surfaces, but it's still an area I need more experience with. I'd like to see how people approach their seam cuts, hard edges, etc...
Made a quick bake for the w3 model. I know there are things to fix on the n map but I think it is quite good for a quickie. The lp is not really optimized but I tried to to make it as fast as possible so the topology on it is not the ideal one.
s6, I'd say next week you give us two assets like on week 1. One that's fairly simpler than this, and another that's a bit more complex. That way most people will probably finish the simple one by Sunday (as soon as you post it, lol) and the next one to be done after work or school or whatever people is doing, for a couple days.
At least that's my suggestion, dunno what other people might want.
s6, I'd say next week you give us two assets like on week 1. One that's fairly simpler than this, and another that's a bit more complex. That way most people will probably finish the simple one by Sunday (as soon as you post it, lol) and the next one to be done after work or school or whatever people is doing, for a couple days.
At least that's my suggestion, dunno what other people might want.
Gotta agree with Mik2121!
Also, if possible, I'd like to see some different kind of hard-surfaces: for the moment, we did "mechanical" hard-surface. What about "electronic" hard-surface? For example:
Here is my take on W3 challenge. I'll Do the S2 as well but didn't have the time to last week. I'm going to make the screw too and the thing on the side on the screen of commador
I know it's tight again
I'll try to work on a small environment to present all the weeks challenge too. I have some idea just need time.
Might revisit this later in the week for a bake and possibly finishing up that centerpiece. For now it served good as practice Wires
On what Mik2121 said - I think this tempo is fine (I'm sure you misspelled and put 3 instead of 2). Nothing stops you from finishing your model over the next few days if you really want to but providing there are numerous examples and breakdowns of the original piece I don't feel that there is too little time. But I do agree that it could be nice to have two different models - one simpler and one more complex for future challenges.
EDIT: Didn't see your post up there with the finished model Mik, so scratch that first part of my reply
And I second what JurassicPerk said as well, switching this up once in a while wouldn't hurt
Though, He is correct. It's important to get into the habit of paying close attention to proportions and spacial relationships in modeling.
I'm very comfortable with matching reference. The assets I make for work are usually accurate down to the millimeter or less. For this however I made no attempt to match reference as it was so simple and all I cared about was the topology. You guys are right though in saying it's an essential skill to have and I definitely wouldn't recommend ignoring those things to someone who doesn't have a good amount of experience already.
Also, if possible, I'd like to see some different kind of hard-surfaces: for the moment, we did "mechanical" hard-surface. What about "electronic" hard-surface? For example:
I agree with this. A lot of products have really challenging shapes to make with sub-d.
I also made a 36 sided one and notice pinching problem that occurs but it wasn't to noticeable. Depending how many poly your willing to spend, I feel the minimum for a optimal result would be 36 using 3 faces to pull from to maintain decent proportions. But for the best result and a harder surface you can't go wrong with 58.
Also, if possible, I'd like to see some different kind of hard-surfaces: for the moment, we did "mechanical" hard-surface. What about "electronic" hard-surface? For example:
Well I assume the whole point of the challenge is to try and tackle as many complex hard surface model week by week. Thus the above Razer would be a great example of a tricky hard surface to model.
So here is my entry. This week is quite simple. Missed out on last weeks challenge.
Anyway: I just saw I missed those screws. I just went by the first reference and it pretty much looked like holes. For the future something with a bit more resolution and less artifacts as a reference would be a plus.
Thought this was a little to easy in comparison with the others but i was wrong. This little guy gave me a lot of trouble and i had to cheat and look at your all's wires..
The pinching occurs because you started out with too few sided cylinder. I myself went with a 36 sided cylinder and used 3 faces to create the clamps (I calculated I wanted 3 faces to be pulled and gaps of 6 this allowed me to have all quads. *I also had a cube with the rough estimates of how big I wanted the gaps/the space between the clamps, that how I determine the numbers*
I even seen other use a less sided cylinder but they ended up with tris. Overall there still going to be a small amount of pinching that occurs but it very minimal if you go with a more sided cylinder. (Overall I ended with 3611 vertices with a 48 sided you will be around 4800)
Here this should help, I actually woke up as I saw your post so disregard the quality. You want 6 faces to be pulled out *instead of 3 for a minimum to make it a harder surface* or 8 (you just need to pull back 2 instead of 1). Use the ending face "5" to connect them together then push it back thus creating the "Red Face". *I feel the minimal for a optimal result is a 56 sided cylinder*
^^ I believe that the common approach people took.
Here is my one for this week. I must admit that I found this harder to do than last week's one. I made several approaches until I realized that only more geometry would help against the pinching, just as AlexRodriguez said above.
Finally got around to modelling this yesterday, rendered today. Took longer to make than I would have liked but it was my first time using Booleans and lots of dodgy symmetry :P
Areas to improve are the proportions and topology to get real double loops, not the hacky ones I did.
Make sure to pay close attention to the reference. Kroma did a great job on the shape of the corners. The curve of the circular cut out should mimic that of the rounded corner (based of the same cylinder, Ideally).
I started mine the other day. Will post it as soon as I bake it and texture
Edit: in case anybody new to the 3D modeling is wondering it, (so, I'm not talking with you Frezno, it's clear you know what you are doing) here's a visual explanation to the 4 sides issue for the circle:
Nice explanation JurassicPerk! I usually use squares when trying to make circular cuts and such. I'll have to give the eight-sided approach a shot on some of these other shapes. Thanks
Edit: in case anybody new to the 3D modeling is wondering it, (so, I'm not talking with you Frezno, it's clear you know what you are doing) here's a visual explanation to the 4 sides issue for the circle:
This is awesome... this kind of stuff is so good to see in visual form. There was a guy on CGTalk compiling tons of little topology rules like this--I kept thinking it would have made the best, most useful poster ever. I wonder what happened to that thread...
IMO, as with all things involving rules, this depends. If they are little holes on the side of a gun that you won't get close to ... 4 sides is probably fine if it helps keep the mesh cleaner and easier to iterate on.
IMO, as with all things involving rules, this depends. If they are little holes on the side of a gun that you won't get close to ... 4 sides is probably fine if it helps keep the mesh cleaner and easier to iterate on.
Great Example, I applied this recently to Week 2 Challenge, The big circles I went with a 8 sided cylinder but there was a very tiny circle cut out near the screws. The topology presented itself nicely to a 4 sided cylinder (cube)
Although if you do run into situations where you are using a-lot of 6 or even a couple 4 sided you might of started with to low of topology which a few discovered during Week 3
Here's my week 3. Proportion aren't correct Now I'm getting back into the modelling workflow I think my next challenge will be to get the proportions as accurate as possible. Anywho, here it is.
Replies
I'd like to see this as well if people are willing. I know the basics of baking hard surfaces, but it's still an area I need more experience with. I'd like to see how people approach their seam cuts, hard edges, etc...
Wires
Beauty:
Clay:
Wireframes >>>https://dl.dropboxusercontent.com/u/1264057/Repository/Polycount/Weekly/week_01_wireframes.jpg
s6, I'd say next week you give us two assets like on week 1. One that's fairly simpler than this, and another that's a bit more complex. That way most people will probably finish the simple one by Sunday (as soon as you post it, lol) and the next one to be done after work or school or whatever people is doing, for a couple days.
At least that's my suggestion, dunno what other people might want.
Gotta agree with Mik2121!
Also, if possible, I'd like to see some different kind of hard-surfaces: for the moment, we did "mechanical" hard-surface. What about "electronic" hard-surface? For example:
I know it's tight again
I'll try to work on a small environment to present all the weeks challenge too. I have some idea just need time.
Wires
On what Mik2121 said - I think this tempo is fine (I'm sure you misspelled and put 3 instead of 2). Nothing stops you from finishing your model over the next few days if you really want to but providing there are numerous examples and breakdowns of the original piece I don't feel that there is too little time. But I do agree that it could be nice to have two different models - one simpler and one more complex for future challenges.
EDIT: Didn't see your post up there with the finished model Mik, so scratch that first part of my reply
And I second what JurassicPerk said as well, switching this up once in a while wouldn't hurt
wire
I agree with this. A lot of products have really challenging shapes to make with sub-d.
wires and smooth render
and last week entry:
wires
Overall fun little burner thought it would be a much simpler model then it turned out. Made it out of a 58 sided cylinder.
Advanced Model
Advanced Wires
I also made a 36 sided one and notice pinching problem that occurs but it wasn't to noticeable. Depending how many poly your willing to spend, I feel the minimum for a optimal result would be 36 using 3 faces to pull from to maintain decent proportions. But for the best result and a harder surface you can't go wrong with 58.
Low Poly Wires
Wires
Text will end up in normal map using nDo2 probably, so I didn't bother trying to make it any good.
Well I assume the whole point of the challenge is to try and tackle as many complex hard surface model week by week. Thus the above Razer would be a great example of a tricky hard surface to model.
(Would love to make it looks challenging)
fun warmup during the slow bits of today.
Anyway: I just saw I missed those screws. I just went by the first reference and it pretty much looked like holes. For the future something with a bit more resolution and less artifacts as a reference would be a plus.
HP wires
Cage wires
Wires
and Maps used
Anyways, Here's my go at it:
Wires:
http://i.imgur.com/DGtQhJ1.png
Wireframe
Uvs
wireframe
:poly142:
The pinching occurs because you started out with too few sided cylinder. I myself went with a 36 sided cylinder and used 3 faces to create the clamps (I calculated I wanted 3 faces to be pulled and gaps of 6 this allowed me to have all quads. *I also had a cube with the rough estimates of how big I wanted the gaps/the space between the clamps, that how I determine the numbers*
I even seen other use a less sided cylinder but they ended up with tris. Overall there still going to be a small amount of pinching that occurs but it very minimal if you go with a more sided cylinder. (Overall I ended with 3611 vertices with a 48 sided you will be around 4800)
Here this should help, I actually woke up as I saw your post so disregard the quality. You want 6 faces to be pulled out *instead of 3 for a minimum to make it a harder surface* or 8 (you just need to pull back 2 instead of 1). Use the ending face "5" to connect them together then push it back thus creating the "Red Face". *I feel the minimal for a optimal result is a 56 sided cylinder*
^^ I believe that the common approach people took.
Wires smoothed and unsmoothed
Didnt go high enough poly with it. Probably going to go back and do week 2's challenge aswell. Really like the idea of this thread!
wires
Finally I finished it
Great works from everyone
It's still to high poly to be useful for games though.
Areas to improve are the proportions and topology to get real double loops, not the hacky ones I did.
Wires unsmoothed
Wires smoothed
not perfect but i learned a lot from this :P and thanks to the people who helped me
http://i.imgur.com/5Xg2sXf.png
3ds Max clay render.
Beauty shot.
The smoothed and unsmoothed mesh here.
Make sure to pay close attention to the reference. Kroma did a great job on the shape of the corners. The curve of the circular cut out should mimic that of the rounded corner (based of the same cylinder, Ideally).
I started mine the other day. Will post it as soon as I bake it and texture
Wires Smoothed and Unsmoothed
And a couple renders
Great render
Edit: in case anybody new to the 3D modeling is wondering it, (so, I'm not talking with you Frezno, it's clear you know what you are doing) here's a visual explanation to the 4 sides issue for the circle:
This is awesome... this kind of stuff is so good to see in visual form. There was a guy on CGTalk compiling tons of little topology rules like this--I kept thinking it would have made the best, most useful poster ever. I wonder what happened to that thread...
Great Example, I applied this recently to Week 2 Challenge, The big circles I went with a 8 sided cylinder but there was a very tiny circle cut out near the screws. The topology presented itself nicely to a 4 sided cylinder (cube)
Although if you do run into situations where you are using a-lot of 6 or even a couple 4 sided you might of started with to low of topology which a few discovered during Week 3
Started week 1 concept 1 of the screw. I will be just making my way through the last few weeks over the weekend.
Wireframe
On to the next one!
Beauty
weekFinal by ravedonkeyad
Shaded
shadedWeek3 by ravedonkeyad
Wires
The photography on your website JurassicPerk is stunning!
Thank you, JamesHodgart! Very good entry for week 3
Wireframe
On to week 2!