Hello people of Polycount. I'm here to show you guys some of the stuff i'm making for Slaughter House, a project started by ZenDavis. For more info ask him or pm me. As for this thread, he asked me to make one so i can recive feedback regarding my work. Any thoughts? More assets soon to come. Cheers!
well done so far, only one thing that is recognizable is the placement of the wooden pillars, they should really not stick into each other, that would have been a lot of work to saw and glue the pieces together like this in "the real world"..
You should place them next to each other and tie them up with a thick rope or use huge nails to connect them
Hehe, i noticed that after i sent the file. But thanks for reminding me
hiho thought i make u a little paintover on how you could improve on the topology a bit.
many polys wasted currently.
this is more less how i would do it.
sry for the bad compression....
Can you give some specs? It really looks like you're not using your polygons well. Adding polygons just for the sake of adding polygons is stupid. I had a go at modeling this at various polygon counts. I really don't think you need more than 500 triangles.
Why do you need three or so sheets? Only the top one affects the silhouette, really, and it's probably the only one that has unique/gameplay information. Plus, if you texture the bottom papers, you can remove the paper mesh completely on LOD's.
Next is right about the chamfers. The smaller the shapes are, the less chamfering they need. I have much less chamfers than you, just one for the plank, and none anywhere else - yet the silhouette looks okay. Some things, like metal, are okay to have pretty sharp edges.
Also keep in mind how large something will be on screen. In my gif, all clip-holes look equally round, except maybe the first hole can be recognized as 4-sided. But you don't see the difference between 6, 8, 11 and 14 - because the ones with less sides are also smaller. Same goes for the large chamfers.
Don't pay a lot of attention to the ultralow ones, they were done mostly for the heck of it. They're not cheaper to render than the others (as I understand it) because it's still a draw call, and generally draw calls below 100 all take the same time because the CPU is the bottleneck and not the GPU.
edit: also, take a look at the meshes - I'll leave them online for a few days.
@throttlekitty and @Alex Pointer
That is not the normal map, that is the diffuse. I will take a look at it.
@Snader and @Next
I didn't add polygons just like that.
As for the "story" of the clipboard:
It is a main object for the player in which keeps his information about the camp ( deads, places etc). And yeah i know that the only sheet of paper will be the top one, but, i recived a reference that i had to respect like 80%. My initial clipboard was pretty simple : a piece of box with the edges chamferd once, that thing from top of it and the paper was a texture on the body of the clipboard...
And i fixed the round hole ^^ thanks for that.
edit:Yeah there are some crazy lines below the paper, i forgot to clean that up ( 9 AM here ^^)
Just because you get a ref from somebody, it doesn't mean its right, or that they are right. Theres something shining through from this project, and I have to be honest, its not that the guys running it know what they are doing...
As for the "thats not the normal map"... Erm... Are you sure you know what a normal map is? Thats clearly using height of some kind, thats why the bumps change darkness depending on light angle. (Laymans)
I would really take the whole art team on this mod/game/thing and just go back to basics, because atm, its reading a lot like Mikes First Max Lesson.
Also, you are modeling in quads, which is fine, but game engines use tris, remember.
Do you have something personal with the project leader, or that's how you talk with all the newbiew arround here? I can say that i have enough experience to know all the s*** that i have to respect when i start doing a model. And yes i now what is a normal map my friend Here is the diffuse. What i meant to say is that the paper looks like that because of the basic diffuse map, nevermind on that.
Not at all, no. But I feel like with new projects, even more so depending on KickStarter funds, need a lot of help to get off the ground. I may have a little experience in this...
Also, you are making newbie modeling mistakes, thats why I said to go back to formula. Its nothing against you, its just that a good artist knows not to throw polygons at an object to make it higher detail... It doesnt work like that.
You are really making a lot of newbie mistakes you could use alpha planes for the paper and normal map would deal with the details/shapes.Your normal map on wood is too strong and only because your "boss" told you to use more polys it doesn't mean wasting them, you shouldn't listen your boss 100% as he probably doesn't have the right knowledge.
EDIT
I just noticed that you use n-gons for modeling which are not suitable for game modeling especially if you let the engine convert them into tris.
Here is something new for me... the thing with the n-gons and tris... i noticed that a engine converts them into tris but didn't dig it up too much... anyway, here is the initial clipboard that i made and the one requested. The initial had the same textures and the paper was centered as a texture also.
My problem is that i really don't know what is a ngon...
This is what we were talking about. This kind of stuff, its like, basic stuff. You need to learn to walk before you can run, and making stuff like this? Its great, but don't take the crits to heart because to be frank, Polycount is a ROUGH place, and you should be happy anybody is posting "AT ALL". Usually, people let these threads slide.
hehe thanks, that's the reasons why i don't post here too often, but got to make this step, right ?
The thing is i NEVER EVER heard of ngons before or someone telling me "hey, u got ngons, fix that". Probably there are some other things that i don't know...
Well. The reason for adding more polys to mundane items is relatively simple. CE offers us Vizareas and portals so that when we're inside a building, nothing outside of the building itself is rendered. This allows us to spruce up our poly count on these assets and get them to look a bit nicer than they would otherwise.
Just some helpful feedback on the paper: Simple way to make that diffuse less wrinkley-Fiddle with the contrast in Photoshop. Also, I am not sure of your texture size limitations but, if you do want the paper that wrinkled you will see a little more wear on the wrinkles and some dirt/grime marks in the areas where the wrinkles merge with one another (ignore this feedback if your map sizes on tiny).
Ok guys, me and ZenDavis we are set. Those two will remains this way. And i'm moving forward with the list keeping in mind all the important feedback you guys gave to me. A big bow to you all. Cheers!
I want to clarify one thing for you guys. We HAVE NOT chosen to ignore your advice.
That would be stupid.
We are moving ahead to keep to our workflow rather than outright ignoring you. Imagine a 48 hour game development challenge. Well in this case we're approaching initial development on this game like a 1 month development challenge. We're going to knock out the assets for this project in a quick and dirty fashion and then come back to refine and optimize once they're all done.
Everyone on the team has about about ten they need to complete by the end of the month. This would give us roughly 100 assets to build a scene with. This will be more than enough. We're not UV Mapping or texturing anything in between now and then. Instead we're going to keep posting assets and taking your notes down and once everything is knocked out - we're going to go back to refine and optimize everything.
Hopefully some of the more experienced individuals here can give us feedback regarding this approach. My primary concern isn't the workflow as much as it is keeping everyone motivated to finish their assets. If everyone gets bogged down in refining and optimizing then the overall completion of each object - the completion of the scene can be pushed back by months. Staying motivated for months on end is a difficult task and in my own personal experience in the past, is hard to achieve.
Instead I prefer getting our assets done as quickly as possible and then taking your notes and applying them to optimize our assets. Any feedback - harsh or otherwise is more than welcome. Some of you may be tougher than others, but all of you have been more than fair. In my next post I'll explain the importance of the clipboard - as it acts as the Pip-boy of our game.
I understand that you're trying to keep the momentum but you are creating the idea that doing the right thing, the right way, the first time isn't important. Saying that you'll optimize it later is a huge mistake. If motivation is a major problem, then what makes you think that anyone would be motivated to go back and optimize assets that are supposed to be "finished"? I would advise taking the advice now and applying it now. If you take this advice now, you won't have to come back and optimize later, which is going to be more work than you think.
Maybe a story will help you see just how inefficient your idea is:
The owner of a ceramic unicorn factory realizes that a conveyor belt is broken and is feeding the ceramic unicorns off into the floor, breaking them all. As the unicorns continue to pour out onto the floor, he decides that production cannot stop. So, he shrugs his shoulders and tells one of the linemen to keep sweeping them up and packaging them into the boxes just as they are.
Two days later, several ceramic store owners call the owner and say all of the legs of their ceramic unicorns are broken and some are even missing. The owner, being a very intelligent man, decides to dispatch all of his workers to these stores to super glue the legs back onto all of the ceramic unicorns, but the workers have to do it on their days off for the next two months because production can't stop.
The next morning, not a single worker came into work because they just weren't motivated to fix problems that could have been prevented by simply stopping production for one hour and fixing the conveyor belt.
As a project leader, you have a responsibility to lead your team to do the right thing, the right way, the first time. Polycount is filled with people who know what they are talking about. You can SAY that you aren't ignoring what is being advised but when you choose to DO the opposite, it's disrespectful. Just say, "We're not going to listen to those who might have a really good point and do what we want."
Zen there is one major problem with your optimizing idea and its that if you decrease the polycount of the model(which means deleting polys)you lose UV maps which are 40% of the modeling job.
I understand that you're trying to keep the momentum but you are creating the idea that doing the right thing, the right way, the first time isn't important. Saying that you'll optimize it later is a huge mistake. If motivation is a major problem, then what makes you think that anyone would be motivated to go back and optimize assets that are supposed to be "finished"? I would advise taking the advice now and applying it now. If you take this advice now, you won't have to come back and optimize later, which is going to be more work than you think.
You make a good point. But motivation on a self-motivated team that isn't spurred on by cash is a fundamental factor in seeing that a project get completed. Look at all the development teams that go day by day understaffed only to end up failing. It's probably the number one reason why most independent projects end up failing.
I don't think anyone could legitimately argue against me on this. And even if teams don't end up falling apart, they end up significantly understaffed. This problem is more detrimental to indie development than inefficient poly count. If you want to make the argument that it's the other way around, you're free to make that argument and I'll hear you out.
I would argue that as a project leader, I have a responsibility to seeing that the project gets completed. Your story is not necessarily applicable to what we're doing. Nothing is being shipped out into the real world before its game ready. The optimization that has been suggested will be incorporated. Just not until we've got all the assets on hand first.
Is there a possibility that no one might be around to optimize the assets after they've been completed? Yes. There is. But there is just as much danger that half way into asset creation people can become bored or become preoccupied with other interests. Both roads are fraught with danger.
I intend no disrespect. But my goal isn't to have everything game ready. And that's not what these are intended to do. There will be no AI in the scene. There will be no character models. Even if they're not perfect - we can take the performance hit. More than that all the assets WILL be optimized regardless. There will be time for all that later. There's no reason to take it personally. It's not meant to be.
---
Ivan. The models won't be unwrapped until they've been optimized.
If you're afraid of people leaving and giving up, think of this:
in X time they can make 5 clean models and textures and UV's, or 10 sloppy ones. Let's assume they then get bored and stop. So you have either 5 finished models and need 5 more, or you have 10 models that need fixes. You've got to find someone else to finish the project to 10 clean models.
What do you think people would rather do? Make 5 models from start to finish or clean up someone else's 10 bad models?
You make a good point. But motivation on a self-motivated team that isn't spurred on by cash is a fundamental factor in seeing that a project get completed. Look at all the development teams that go day by day understaffed only to end up failing. It's probably the number one reason why most independent projects end up failing.
I don't think anyone could legitimately argue against me on this. And even if teams don't end up falling apart, they end up significantly understaffed. This problem is more detrimental to indie development than inefficient poly count. If you want to make the argument that it's the other way around, you're free to make that argument and I'll hear you out.
I would argue that as a project leader, I have a responsibility to seeing that the project gets completed. Your story is not necessarily applicable to what we're doing. Nothing is being shipped out into the real world before its game ready. The optimization that has been suggested will be incorporated. Just not until we've got all the assets on hand first.
Is there a possibility that no one might be around to optimize the assets after they've been completed? Yes. There is. But there is just as much danger that half way into asset creation people can become bored or become preoccupied with other interests. Both roads are fraught with danger.
I intend no disrespect. But my goal isn't to have everything game ready. And that's not what these are intended to do. There will be no AI in the scene. There will be no character models. Even if they're not perfect - we can take the performance hit. More than that all the assets WILL be optimized regardless. There will be time for all that later. There's no reason to take it personally. It's not meant to be.
---
Ivan. The models won't be unwrapped until they've been optimized.
That's a very solid point and has gotten me to reconsider. Let me talk to the artists about this. As such - can the above tire be optimized? Give us feedback. We will implement it.
- The sides looks wrong. Tires don't have trims like that on the sides. Try making in the more smooth. Also make a smooth transition between the sides and the pattren on top.
- The pattren you have on your tire looks like it's only one side of the tire. What I mean is, tires usually have a pattren that is mirrored over.
- The tire also looks a little thin. It wouldn't hurt to beef it up a bit.
The tire looks too smudgy. Even very old tires keep their track straight. The trims...well I'm not a tire expert but that seems off to me and too repetitive.
You could also put some dirt patches here and there but then again do not make it too repetitive.
Another thing to keep in mind as far as doing assets right the first time vs. coming back to fix them..
If you let artists turn in stuff that might not be up to snuff, your getting them into a rut where they think they are giving you what you want. If you have them get things done right the first time, it sets a better standard...and when they continue to make assets it will be better to have that established and they can be accustomed to knowing what is right and wrong.
Yeah the clipboard and paper is pretty haggard...numb was way too strong..definitely could use just a little work fixing that up...things just need to be more subtle.
The tire looks better, but did you bake an AO map for it? The interior is really bright and makes it look cheap. I would except it to be darker in there since it gets less light. The sidewall could use some work too...maybe throw some text on it or something...tire size, brand, something.
I just want to point out that these textures are placeholders. We will be using dedicated texture artists in order to keep the art direction of the game consistent.
Why did you make holes for the poles when you cant even see them? Whey do you have polygons where they dont change the shape at all? Theres soooooo many wasted polygons on these assets...?
I'm starting to think this whole project is a troll porject...? :S
Everybody has to start some place. But the thing is, those new starters? You can't make a game with them. You will just end up digging this hole that gets deeper and deeper.
What you have in the 76 or so threads here, isn't so much new guys that dont know what they are doing... Its new guys who dont know what they are doing, that just will not listen to advice given to them by people that, frankly, do.
Have you not noticed by now people have just given up helping? You're posting models that are made with no regard to anything other than "Hey I made a tea pot in max, I can do an entire game!", while pretty much telling everybody to shut up while ASKING for advice... That you end up ignoring.
Nobody has told you straight, but its about time they did. Art isn't about pumping out a billion really piss poor assets and not learning a thing in the process... This things doomed to fail before its even out of the gate.
Yeah man, i noticed some things arround here. I noticed everyone is looking for perfection (thing that i can't really do it because the lack of time). I noticed the good community that is giving helpful advices and that i am a bit of a butthuty type of person that is taking hard a good advice (because all i have learned, i learned alone with minimal searching of tutorials).... anyway, yeah thanks for reminding me that people gave up helping me because of my butthurt arogance attitude, but you see... your type of attitude makes me don't want to take advices... also i think no one started to presume things about your project.
I was asked to post here my work and so i did. I took in consideration the advices and i will apply them when i have the time. That doesn't mean that i'm not learning or that i am trolling.
Replies
Hehe, i noticed that after i sent the file. But thanks for reminding me
many polys wasted currently.
this is more less how i would do it.
sry for the bad compression....
cheers next
Why do you need three or so sheets? Only the top one affects the silhouette, really, and it's probably the only one that has unique/gameplay information. Plus, if you texture the bottom papers, you can remove the paper mesh completely on LOD's.
Next is right about the chamfers. The smaller the shapes are, the less chamfering they need. I have much less chamfers than you, just one for the plank, and none anywhere else - yet the silhouette looks okay. Some things, like metal, are okay to have pretty sharp edges.
Also keep in mind how large something will be on screen. In my gif, all clip-holes look equally round, except maybe the first hole can be recognized as 4-sided. But you don't see the difference between 6, 8, 11 and 14 - because the ones with less sides are also smaller. Same goes for the large chamfers.
Don't pay a lot of attention to the ultralow ones, they were done mostly for the heck of it. They're not cheaper to render than the others (as I understand it) because it's still a draw call, and generally draw calls below 100 all take the same time because the CPU is the bottleneck and not the GPU.
edit: also, take a look at the meshes - I'll leave them online for a few days.
@throttlekitty and @Alex Pointer
That is not the normal map, that is the diffuse. I will take a look at it.
@Snader and @Next
I didn't add polygons just like that.
As for the "story" of the clipboard:
It is a main object for the player in which keeps his information about the camp ( deads, places etc). And yeah i know that the only sheet of paper will be the top one, but, i recived a reference that i had to respect like 80%. My initial clipboard was pretty simple : a piece of box with the edges chamferd once, that thing from top of it and the paper was a texture on the body of the clipboard...
And i fixed the round hole ^^ thanks for that.
edit:Yeah there are some crazy lines below the paper, i forgot to clean that up ( 9 AM here ^^)
As for the "thats not the normal map"... Erm... Are you sure you know what a normal map is? Thats clearly using height of some kind, thats why the bumps change darkness depending on light angle. (Laymans)
I would really take the whole art team on this mod/game/thing and just go back to basics, because atm, its reading a lot like Mikes First Max Lesson.
Also, you are modeling in quads, which is fine, but game engines use tris, remember.
Also, you are making newbie modeling mistakes, thats why I said to go back to formula. Its nothing against you, its just that a good artist knows not to throw polygons at an object to make it higher detail... It doesnt work like that.
EDIT
I just noticed that you use n-gons for modeling which are not suitable for game modeling especially if you let the engine convert them into tris.
This is what we were talking about. This kind of stuff, its like, basic stuff. You need to learn to walk before you can run, and making stuff like this? Its great, but don't take the crits to heart because to be frank, Polycount is a ROUGH place, and you should be happy anybody is posting "AT ALL". Usually, people let these threads slide.
The thing is i NEVER EVER heard of ngons before or someone telling me "hey, u got ngons, fix that". Probably there are some other things that i don't know...
So, what is your map size limit?
That would be stupid.
We are moving ahead to keep to our workflow rather than outright ignoring you. Imagine a 48 hour game development challenge. Well in this case we're approaching initial development on this game like a 1 month development challenge. We're going to knock out the assets for this project in a quick and dirty fashion and then come back to refine and optimize once they're all done.
Everyone on the team has about about ten they need to complete by the end of the month. This would give us roughly 100 assets to build a scene with. This will be more than enough. We're not UV Mapping or texturing anything in between now and then. Instead we're going to keep posting assets and taking your notes down and once everything is knocked out - we're going to go back to refine and optimize everything.
Hopefully some of the more experienced individuals here can give us feedback regarding this approach. My primary concern isn't the workflow as much as it is keeping everyone motivated to finish their assets. If everyone gets bogged down in refining and optimizing then the overall completion of each object - the completion of the scene can be pushed back by months. Staying motivated for months on end is a difficult task and in my own personal experience in the past, is hard to achieve.
Instead I prefer getting our assets done as quickly as possible and then taking your notes and applying them to optimize our assets. Any feedback - harsh or otherwise is more than welcome. Some of you may be tougher than others, but all of you have been more than fair. In my next post I'll explain the importance of the clipboard - as it acts as the Pip-boy of our game.
Maybe a story will help you see just how inefficient your idea is:
The owner of a ceramic unicorn factory realizes that a conveyor belt is broken and is feeding the ceramic unicorns off into the floor, breaking them all. As the unicorns continue to pour out onto the floor, he decides that production cannot stop. So, he shrugs his shoulders and tells one of the linemen to keep sweeping them up and packaging them into the boxes just as they are.
Two days later, several ceramic store owners call the owner and say all of the legs of their ceramic unicorns are broken and some are even missing. The owner, being a very intelligent man, decides to dispatch all of his workers to these stores to super glue the legs back onto all of the ceramic unicorns, but the workers have to do it on their days off for the next two months because production can't stop.
The next morning, not a single worker came into work because they just weren't motivated to fix problems that could have been prevented by simply stopping production for one hour and fixing the conveyor belt.
As a project leader, you have a responsibility to lead your team to do the right thing, the right way, the first time. Polycount is filled with people who know what they are talking about. You can SAY that you aren't ignoring what is being advised but when you choose to DO the opposite, it's disrespectful. Just say, "We're not going to listen to those who might have a really good point and do what we want."
You make a good point. But motivation on a self-motivated team that isn't spurred on by cash is a fundamental factor in seeing that a project get completed. Look at all the development teams that go day by day understaffed only to end up failing. It's probably the number one reason why most independent projects end up failing.
I don't think anyone could legitimately argue against me on this. And even if teams don't end up falling apart, they end up significantly understaffed. This problem is more detrimental to indie development than inefficient poly count. If you want to make the argument that it's the other way around, you're free to make that argument and I'll hear you out.
I would argue that as a project leader, I have a responsibility to seeing that the project gets completed. Your story is not necessarily applicable to what we're doing. Nothing is being shipped out into the real world before its game ready. The optimization that has been suggested will be incorporated. Just not until we've got all the assets on hand first.
Is there a possibility that no one might be around to optimize the assets after they've been completed? Yes. There is. But there is just as much danger that half way into asset creation people can become bored or become preoccupied with other interests. Both roads are fraught with danger.
I intend no disrespect. But my goal isn't to have everything game ready. And that's not what these are intended to do. There will be no AI in the scene. There will be no character models. Even if they're not perfect - we can take the performance hit. More than that all the assets WILL be optimized regardless. There will be time for all that later. There's no reason to take it personally. It's not meant to be.
---
Ivan. The models won't be unwrapped until they've been optimized.
in X time they can make 5 clean models and textures and UV's, or 10 sloppy ones. Let's assume they then get bored and stop. So you have either 5 finished models and need 5 more, or you have 10 models that need fixes. You've got to find someone else to finish the project to 10 clean models.
What do you think people would rather do? Make 5 models from start to finish or clean up someone else's 10 bad models?
That's a very solid point and has gotten me to reconsider. Let me talk to the artists about this. As such - can the above tire be optimized? Give us feedback. We will implement it.
- The sides looks wrong. Tires don't have trims like that on the sides. Try making in the more smooth. Also make a smooth transition between the sides and the pattren on top.
- The pattren you have on your tire looks like it's only one side of the tire. What I mean is, tires usually have a pattren that is mirrored over.
- The tire also looks a little thin. It wouldn't hurt to beef it up a bit.
You could also put some dirt patches here and there but then again do not make it too repetitive.
http://www.cgtextures.com/texview.php?id=23985&PHPSESSID=asfkst1sgsemc14948fdr689n0
Edit: Latest tweaks...
I mean, make it a little wider. The tire was rather thin before.
If you let artists turn in stuff that might not be up to snuff, your getting them into a rut where they think they are giving you what you want. If you have them get things done right the first time, it sets a better standard...and when they continue to make assets it will be better to have that established and they can be accustomed to knowing what is right and wrong.
Yeah the clipboard and paper is pretty haggard...numb was way too strong..definitely could use just a little work fixing that up...things just need to be more subtle.
The tire looks better, but did you bake an AO map for it? The interior is really bright and makes it look cheap. I would except it to be darker in there since it gets less light. The sidewall could use some work too...maybe throw some text on it or something...tire size, brand, something.
I'm starting to think this whole project is a troll porject...? :S
What you have in the 76 or so threads here, isn't so much new guys that dont know what they are doing... Its new guys who dont know what they are doing, that just will not listen to advice given to them by people that, frankly, do.
Have you not noticed by now people have just given up helping? You're posting models that are made with no regard to anything other than "Hey I made a tea pot in max, I can do an entire game!", while pretty much telling everybody to shut up while ASKING for advice... That you end up ignoring.
Nobody has told you straight, but its about time they did. Art isn't about pumping out a billion really piss poor assets and not learning a thing in the process... This things doomed to fail before its even out of the gate.
I was asked to post here my work and so i did. I took in consideration the advices and i will apply them when i have the time. That doesn't mean that i'm not learning or that i am trolling.