Hey there,
I've been working on this for the last couple of days. The mesh is 1080 tris, textures are mostly 128x128, all handpainted. It's for a hobby project of me and a friend programmer making an oldschool-ish adventure game, ala zelda oot. Crits highly welcome:)
pretty cool! i would work a lot more on the textures, they seem a bit flat and simple. the orange stuff looks the weirdest, especially if you put it on the base like that (the ring).
keep it up!
the structure of the blades seems a bit complex for a low poly game model, does not read that well. to many thin sticks, textures are kinda flat, pay more atention to how the surfaces work together, the orange at the base just kinda ends where the brick base starts, the paterns don't line up or anything its just like 2 drasticly difrent textures meeting together with no reason.
Thanks for all your replies
@ Japhir, I worked my ass off on those textures, they are mostly pretty small, the orange thing is 64x128 px. It is diffuse only so it looks pretty dull without any shading painting into them, why do you think it looks weird? I must agree it looks pretty bad at the low floor.
@Rhinokey
How would I go along and make that structure read well in lowpoly then, or is that just not doable in lowpoly? Should I resort to alpha'd planes instead for the wires? The easy way to get rid of the abrupt pattern clash seems to be adding a border, but that means adding more geo and texture. Any idea to solve it without doing that?
@ Evil
I thought it would be allright if you wanted to show that the underside of the rock is extended a bit, a slight bump outwards, light from above would highlight that right? Or am I completely missing the picture here.
I'll update soon, though I'm wondering wether the 3 gens ago specs I'm making this for is actually fun. It seems to be hard to get away from the n64 like, low specification look.
I'll complete this anyway though:)
[ QUOTE ]
why do you think it looks weird? I must agree it looks pretty bad at the low floor.
[/ QUOTE ]
Ok the words chosen where a bit harsh, but what i meant is that it doesn't seem to fit with the rest of the texture (because of the color i think), and that the part on the base seems as though it is just one solid base, with on top of that a new part. (the tiling continues, in stead of going around like the rest)
so that's what i meant, might have put it clearer.
keep it up!
I changed the base to flow round, I think it does look better this way indeed. I removed the shine from the terracotta-ish tiles, it looks slightly better, though different.
I like the basis of this, everything seems to be in it's right place, but your textures are way too clean, like someone sandblasted all the dirt and grime off the rocks. Add some dirt, moss, and some pores to those rocks and it will start to look much more believable.
EDIT: Also near the bottom, you should add a ring of cement and maybe some mud and dirt because this thing will need a foundation. But maybe I'm being too picky.
Sorry to double post, but another thing that bothers me about the texturing is that fact there is no separation of textures, meaning the general rule is that if you have a floor texture, connecting to a wall texture, there should be trim that separates the two. Here you just have them seamlessly slapped on, running on into each other with no clear stop.
It's hard to explain but look at an unreal tournament level and you will generally see wall and floor trim all over the place to show when a wall stops and floor begins. It's just less distracting that way and looks mor professional.
@The umbrella man
I will try to dirty things up a bit more:) Do you think it will be better to not use tileables for that to better spread it?
Some trims, yeah that would be a good idea aswell. Nice lots of new things to try out, update soon:)
@ Johny
yeah that is pretty much the style aimed for, only the sail handpainting bother me, they look pretty bad, any suggestions on how to do these sails?
@ Japhir, thanks I will try that and see how it looks, I already used yellow as highlighting colour aswell in my colours shemes, but I like how the blue makes it look dreamy, I'll give it a try though.
I did a paintover of my own mill, what do you think, improvement?
*edit, backside added
Do you think I should also alter the top stone part?
That's a definate improvement, it looks more realistic from a texturing standpoint, also theres some nice details in the back.
And as for your question, I wouldn't use tileables on props, just buildings like this. Tileables should work for pretty much all architecture you are doing, but it requires more texture memory to render and you may want to consider trying to unwrap your buildings to one large 1024 texture, then things like roofs, windows rocks can have their own texture. Kinda have to weigh your options there.
The there is also the question modular peices. WOuld you rather remodel the same door and window 20 times, or would you rather make one door, and one window, and use it over and over again, maybe with a different texture.
Thanks mr umbrella man:)!
I think I will make some parts unique and copy them around, in the style of copying that around used in oblivion. Like doing unique parts for architectural parts and copying those when those architectural parts are used more.
I thought the mill could, besides grinding grain inside of it, also pump up water. I added a paintover, I'm curious to see what you guys think of this change, and if the top part might look too "busy" or not now?
Hey Stjoris, this is improving but I think you arent looking at values in this piece. Meaning that all the colors are the same level and washing everything out, if you made lets say the brick have moss or vines that are thick towards the bottom and growing up, then you can have that area darker, moving the eye around. The fans in the windmill should look like cloth, weathered maybe as this whole thing looks brand new, not giving it much character. I would suggest also having the wood alot thicker on the fan part, as now they seem like toothpicks and would break off in a slight wind. I hope this helps some, and curious to what you will do.
Thanks for your reply Spark. What do you mean with that I am not looking at values? I am a bit new to painting, and I had to constantly check to see which should be lit more than others, but in the end I think the values aren't extremely off. You mean I should the darkness of some things regardless of their value, like you said with the vines?
I really need to thicken the beams of the fans yeah, they also read quite poor at the moment. I think it is smartest to go ahead and do some more painting indepth on the paintover, like the darkening on the vine area and figuring out some details on the rest of the building. Then I'm probably going to have to remodel quite a bit as a lot of things changed in the paintover.
What I mean StJoris, is that the colors are the same value(the same light range). You really need to push this more, I would suggest not having orage brick, but instead making it old wood, which will seperate the interest some from being just brick, with the same type of brick slightly smaller. Also, you really should gather some reference on things, it seems you are kinda going for a cartoony look, but still it doesnt make sense to me why the whole building is buildt out of brick, and they use straw for the roof. I would probably try shingles of some sort, that are worn and cracked or something to add personality to your building. Just becuase it is a building, does not mean you cannot add character to it, try to make it seem like it has been used, and has history to it. The roof also just seems to end and feels like there is a hole in it at the top. Maybe give it a traingle top, where you have some beams crisscrossing to hold it in, sort of like how they held up teepees. I think you need to look back at your concept, and take out the tiled brick on the building, have it be plaster, with maybe some sections that have fallen off showing the inner structure, again to push the character. Could you show the uv layout, as I am sure there are things in there to help get more pixtel area to your sheet, along with fixing some of your tiling problems. I hope this helps.
Thanks for the suggestions Spark, I think it's time to take this thing back to the drawing board. With all the nice help and the paintovers I did I've got a pretty good sense of what I should do with it. This time I'm not going to do it for a n64, or two gen ago system, but next-gen, including normal mappping. I'll be drawing new, adapted concept first, to lay down where I'm headed.
Just be careful that you don't ask people so many questions, and take so much advice, that you don't actually get anything done. Advice is good in moderation, if you try to do everything that someone suggests you'll never complete anything. I honestly don't believe your mill requires a complete redo, it just needs some finishing touches. you are literally right where you need to be, you just need to make some tweaks here and there. Keep on task or you will find yourself with nothing to show for all your efforts.
Hey StJoris, I'm not suggesting that you go back to the drawing board, but you maybe just add some more character to the piece and push it a little. Looking at some of the points that I brought up along with some of the others. Umbrella man, I do agree that you have to limit the amount of advice that you take in, and concentrate on your work. But you should keep an open ear out for advice, and use that to make your work better, as this is a great place for it. I hope this helps.
Hey there,
I wanted to go back to the drawing board because we decided to change direction aswell and that called for some more detail and a different style. I quickly modeled this based on my paintovers and a load of reference I dug up. I will paint finer details over this to finalize the concept soon.
*What is preffered for viewing this kind of models, GI or 3-point?
that's a pretty cool building.
I think the GI render looks fine, it shows all you need to show. Alternate lighting setups would just be down to personal preference I think.
Directional lighting is always more likely to represent a more accurate view of the final thing though, if it's going in a game.
What's the reasoning behind putting the render time on the final image? Seems kinda pointless to me.
Also, from a design point, why is there a low wall all the way around the base of the model? There's no way for anyone to get in and out of that without climbing over the wall. I assume you'll put some staircases or ramps in somewhere.
are you the same guy? :P this looks much better! great improvement. (it might also be because i like this style better)
do what mop said and you'll be fine!
Just watch your polygon distribution. For example, there's a lot of smooth arches, but the ground level, curved borders to the walkways should be smooth as well (not so polygonal)
[ QUOTE ]
Just watch your polygon distribution. For example, there's a lot of smooth arches, but the ground level, curved borders to the walkways should be smooth as well (not so polygonal)
[/ QUOTE ]
I was thinking the same thing.
Huge improvement though, hopefully the textures are just as strong as the new model. Keep it up.
Wow thank you for all the kind replies:)!
@ Mop: Yeah showing the rendertime is pretty pointless, it's some frame stamp from Vray, but I might aswell turn it off. Kind of forgot the stairs but you are right, they will be there.
@ cholden/caseyjones: Well I quickly modeled this as sort of "concept", it's really crude with unwelded verts, things sticking out. It's mainly for visualizing the model and to paint the concept over it, I'm not too good with painting concepts like that yet in one go so. The game model will be much cleaner and I'll watch the poly distribution:)
Took it into painter and coloured it a bit, it doesn't yet fully "work" for me, but I'm not sure what causes it. Is it better to do the lowpoly first or highpoly or is it totally personal preference?
The model looks a ton better, if not just a tad busy, and the colors you have chosen in the paint over seem to work pretty well. One thing though, this is a windmill right? Those blades don't seem like they are very good at spinning seeing as how they are flat. I would use angled blades if I were you, and it seems more technically sound.
Is this mill thing worshipped or something? That is way too much decorations for such a structure. Unless, the people who built this in a game worship their agriculture.
Technically, you are great. But is the point really to create a temple, or a simple building for making flour?
@zenarion
Well I guess the people in game have a knack for decorating, a worshipping of aesthetics, will that do:)? @creamsoda
I don't know too much about FF9 but I heard it is good, so that must be a compliment:) thanks. I haven't really thought about that, it depends on how I angle the sails, I'll see what looks most interesting. Opposite actually sounds most practical for cancelling forces, and it probably looks cool, or confusing.
Looks cool. Be sure to do some normal bake tests on a low poly as you build. You may find that some detail types do not translate well, and it's best to find out now so you're not redoing everything later.
@ Uly, yeah I'd figured there was something wrong with that, I thought they could just be bigger stones, but that doesn't really work out, I'll fix it:)
@ Cholden, good idea on the bake tests, I already tried some on basic types of things like planes, and it transfered pretty well. But I might aswell just make a part of the game model already and adjust both as I go.
Some update aswell:
I thought about making some kind of cool automated grain treating facility, with lots of moving mechanical parts and steam Good idea? I'd probably rework some later, pretty basic atm.
Replies
keep it up!
keep pluggin away and keep posting
@ Japhir, I worked my ass off on those textures, they are mostly pretty small, the orange thing is 64x128 px. It is diffuse only so it looks pretty dull without any shading painting into them, why do you think it looks weird? I must agree it looks pretty bad at the low floor.
@Rhinokey
How would I go along and make that structure read well in lowpoly then, or is that just not doable in lowpoly? Should I resort to alpha'd planes instead for the wires? The easy way to get rid of the abrupt pattern clash seems to be adding a border, but that means adding more geo and texture. Any idea to solve it without doing that?
@ Evil
I thought it would be allright if you wanted to show that the underside of the rock is extended a bit, a slight bump outwards, light from above would highlight that right? Or am I completely missing the picture here.
I'll update soon, though I'm wondering wether the 3 gens ago specs I'm making this for is actually fun. It seems to be hard to get away from the n64 like, low specification look.
I'll complete this anyway though:)
why do you think it looks weird? I must agree it looks pretty bad at the low floor.
[/ QUOTE ]
Ok the words chosen where a bit harsh, but what i meant is that it doesn't seem to fit with the rest of the texture (because of the color i think), and that the part on the base seems as though it is just one solid base, with on top of that a new part. (the tiling continues, in stead of going around like the rest)
so that's what i meant, might have put it clearer.
keep it up!
EDIT: Also near the bottom, you should add a ring of cement and maybe some mud and dirt because this thing will need a foundation. But maybe I'm being too picky.
It's hard to explain but look at an unreal tournament level and you will generally see wall and floor trim all over the place to show when a wall stops and floor begins. It's just less distracting that way and looks mor professional.
I will try to dirty things up a bit more:) Do you think it will be better to not use tileables for that to better spread it?
Some trims, yeah that would be a good idea aswell. Nice lots of new things to try out, update soon:)
@ Johny
yeah that is pretty much the style aimed for, only the sail handpainting bother me, they look pretty bad, any suggestions on how to do these sails?
any suggestions on how to do these sails?
[/ QUOTE ]
yeah, i wouldn't make them blue, but more white or yellowish (because they are probably a bit dirty)
I did a paintover of my own mill, what do you think, improvement?
*edit, backside added
Do you think I should also alter the top stone part?
And as for your question, I wouldn't use tileables on props, just buildings like this. Tileables should work for pretty much all architecture you are doing, but it requires more texture memory to render and you may want to consider trying to unwrap your buildings to one large 1024 texture, then things like roofs, windows rocks can have their own texture. Kinda have to weigh your options there.
The there is also the question modular peices. WOuld you rather remodel the same door and window 20 times, or would you rather make one door, and one window, and use it over and over again, maybe with a different texture.
I think I will make some parts unique and copy them around, in the style of copying that around used in oblivion. Like doing unique parts for architectural parts and copying those when those architectural parts are used more.
I thought the mill could, besides grinding grain inside of it, also pump up water. I added a paintover, I'm curious to see what you guys think of this change, and if the top part might look too "busy" or not now?
Spark
I really need to thicken the beams of the fans yeah, they also read quite poor at the moment. I think it is smartest to go ahead and do some more painting indepth on the paintover, like the darkening on the vine area and figuring out some details on the rest of the building. Then I'm probably going to have to remodel quite a bit as a lot of things changed in the paintover.
Spark
Spark
I wanted to go back to the drawing board because we decided to change direction aswell and that called for some more detail and a different style. I quickly modeled this based on my paintovers and a load of reference I dug up. I will paint finer details over this to finalize the concept soon.
*What is preffered for viewing this kind of models, GI or 3-point?
I think the GI render looks fine, it shows all you need to show. Alternate lighting setups would just be down to personal preference I think.
Directional lighting is always more likely to represent a more accurate view of the final thing though, if it's going in a game.
What's the reasoning behind putting the render time on the final image? Seems kinda pointless to me.
Also, from a design point, why is there a low wall all the way around the base of the model? There's no way for anyone to get in and out of that without climbing over the wall. I assume you'll put some staircases or ramps in somewhere.
Nice stuff though, keep it up!
do what mop said and you'll be fine!
Just watch your polygon distribution. For example, there's a lot of smooth arches, but the ground level, curved borders to the walkways should be smooth as well (not so polygonal)
Please break up the uniformity in textures this time.
This is NEXT GET for heaven's sake!
Just watch your polygon distribution. For example, there's a lot of smooth arches, but the ground level, curved borders to the walkways should be smooth as well (not so polygonal)
[/ QUOTE ]
I was thinking the same thing.
Huge improvement though, hopefully the textures are just as strong as the new model. Keep it up.
-caseyjones
@ Mop: Yeah showing the rendertime is pretty pointless, it's some frame stamp from Vray, but I might aswell turn it off. Kind of forgot the stairs but you are right, they will be there.
@ cholden/caseyjones: Well I quickly modeled this as sort of "concept", it's really crude with unwelded verts, things sticking out. It's mainly for visualizing the model and to paint the concept over it, I'm not too good with painting concepts like that yet in one go so. The game model will be much cleaner and I'll watch the poly distribution:)
These comments really gave me a boost
Almost like it's done by someone else!
Technically, you are great. But is the point really to create a temple, or a simple building for making flour?
Will the mills turn the same direction or opposite? just curious.
Well I guess the people in game have a knack for decorating, a worshipping of aesthetics, will that do:)?
@creamsoda
I don't know too much about FF9 but I heard it is good, so that must be a compliment:) thanks. I haven't really thought about that, it depends on how I angle the sails, I'll see what looks most interesting. Opposite actually sounds most practical for cancelling forces, and it probably looks cool, or confusing.
New update on the highpoly.
@ Cholden, good idea on the bake tests, I already tried some on basic types of things like planes, and it transfered pretty well. But I might aswell just make a part of the game model already and adjust both as I go.
Some update aswell:
I thought about making some kind of cool automated grain treating facility, with lots of moving mechanical parts and steam Good idea? I'd probably rework some later, pretty basic atm.