Working on the male troll race for my hotly debated mmo project. Any crits, paintovers, suggestions are much appreciated in making this as high quality as possible. Thanks.
the final game mesh should be about 1k tris, the plan from the beginning was to make the high detail mesh for a really really good start at hand painting, as i suck at painting. The intention being a hybrid between the WoW style (cartoony, diffuse only) and modern fps games (diff,norm,spec, realism).
the bake is done in xnormal with the 4 mil poly model shown above, i have a sphere cut in half on the bottom for bouncing light, any suggestions on making the bake better? if i put geo above the model will it block the light, or just bounce and make the top of the model less harshly lit?
Well, the overall proportions are really weird. Why are the arms so huge, and the legs so tiny and weak? Why does the lower abdomen protrude as much as it does? Are you working from a concept? If not, why not?
Even if you pursue and develop this design (which I'd recommend against), you're going to have a tough time animating that mesh with the topology you've got right now. As it is, this looks like 90% of the sculpts we see from beginning artists, with too much focus on surface details like veins and not enough thought put into how muscles actually work, or any thought as to realistic proportions.
Look at the best examples of fantasy characters - even though they don't exist, they look like they could, because of a believable underlying musculature structure and proportions, no matter how stylized. I'd normally suggest studying anatomy to gain an understanding of how muscles overlay the skeleton to make your concept more believable, but something tells me this won't happen.
Fact is, if you're dead-set on making this ridiculously-over ambitious project, people aren't going to work on something with this level of art. I think you'd be better off actually studying and practising on improving your art skills before worrying about trying to make a 'name' for yourself as a designer, but you were told this ad nauseum in your other thread, and it fell on deaf ears.
well this is unfortunate, i really thought i was making some headway with my anatomy. More than functional the proportions are suppose to look fresh and interesting, but if you look at apes, many follow a similar build, although i have exaggerated this.
As far as my muscles, and bone structure go, i really thought they were believable, i have been using reference images a lot as i sculpt, but exaggerating at my discretion. I don't want to make the same troll and orc as everybody else, but i also want it to look fairly good. Personally I feel i've at least hit the WoW level of quality on this character.
More than functional the proportions are suppose to look fresh and interesting....
There's nothing wrong with trying to develop a more unusual design to differentiate your work from the thousands of other orcs and fantasy creatures, but without a believable, functional anatomical base, it won't look fresh or interesting, it'll just look wrong.
How is this orc even supposed to maintain a standing position, with those massive arms, with such tiny legs? If you're using apes as a starting point, notice that although the legs are shorter, they're not so short as to be little more than appendages.
Also, notice the width of the pelvis in the image you posted - this gives the ape a solid core to achieve and maintain balance. We know that apes walk on all fours, but nothing in your design suggests this - it just looks wrong. Also, even if your orc did walk on all fours, how are those tiny legs supposed to propel it forward? It'd have to take about ten tiny steps for every arm motion.
Speaking of arms, look at them. Why are the forearms more muscular than the biceps? Why are they noticeably longer than the upper arms? You say you've studied anatomy closely while working on this, but I'm sorry, I'm just not seeing that.
Personally I feel i've at least hit the WoW level of quality on this character.
You're kidding, right? I mean, not to sound like an asshole, but you have to be kidding or trolling.
If you're deadset on this MMO idea you don't really have the luxury of lovingly crafting every asset. There's a saying that applies: Good, Cheap, Fast; choose two. You've got no budget and you want to finish this in your lifetime so cheap and fast means you don't have the time to make it look good.
There's nothing wrong with trying to develop a more unusual design to differentiate your work from the thousands of other orcs and fantasy creatures, but without a believable, functional anatomical base, it won't look fresh or interesting, it'll just look wrong.
How is this orc even supposed to maintain a standing position, with those massive arms, with such tiny legs? If you're using apes as a starting point, notice that although the legs are shorter, they're not so short as to be little more than appendages.
Also, notice the width of the pelvis in the image you posted - this gives the ape a solid core to achieve and maintain balance. We know that apes walk on all fours, but nothing in your design suggests this - it just looks wrong. Also, even if your orc did walk on all fours, how are those tiny legs supposed to propel it forward? It'd have to take about ten tiny steps for every arm motion.
Speaking of arms, look at them. Why are the forearms more muscular than the biceps? Why are they noticeably longer than the upper arms? You say you've studied anatomy closely while working on this, but I'm sorry, I'm just not seeing that.
You're kidding, right? I mean, not to sound like an asshole, but you have to be kidding or trolling.
Ouch, seems a bit harsh. I mean I understand that it is important to have a good underlying structure to your characters, but if you can't create something completely fictional within the game world then we as designers are going to get stuck in a rut of cookie-cutter art our whole career.
Now don't get me wrong, I do see some obvious flaws within the design, but nothing that would suggest that he scrap what he has. I would suggest going back into Zbrush and tweaking some of your forms to create a more stable character, or give him something to ride that would eliminate his use for legs and walking all together.
I see too often cool designs thrown out the window because they don't look like they would function in the real world, when the fact of the matter is that games are not the real world. You should be able to go balls to the wall and create things you could never imagine seeing anywhere else, creativity essentially has to limits when it comes to video games.
Sorry for the rant, and I don't want to sound like I'm negating everything you said danshewan, you did bring up some great points. Also I will have to agree that this is very far off from WoW style, so if thats what you were going for achillesian then you may need to rethink your approach a bit.
All in all, I like the design, I'm eager to see where you go with the texture on this one. Keep it up
Maybe I can make the legs a little longer, its really a big part of the style and ties into how these races evolved. Forearms determine grip and hand power, so maybe these trolls mold rock into weapons and homes with bare hands. Maybe they live in a very small area, and never travel much, resulting in underdeveloped legs. I'm just tired of the uniform humanoid sillhoutte for every mmorpg race, and this is my attempting in changing it up. But maybe it is over exagerated.
You're kidding, right? I mean, not to sound like an asshole, but you have to be kidding or trolling.
Maybe I do think I'm better than I am, according to those graphs almost everybody either overestimates or underestimates their abilities though.
But seriously, given that my model and wow player models are within 200 tris, and very similar texture sizes, are mine that much worse? Maybe I'm just insane, but they don't seem miles ahead of my wip.
Thanks, most mmo's these days don't have ugly or fat playable characters, and I think its important to realize another world is full of different and unique body types, and even if they look monsters it doesn't mean they are. One of the ongoing themes of the game, and the fact that you can pick any faction as any race, is just going by someone's appearance doesn't always verify friend or foe, as is so often the case in real life.
The separate head texture is because of multiple head choices. Armor is going to be separate meshes so I didn't have to break the naked troll texture up into 10 different pieces like WoW does.
Thanks alot for your words Mike, ultimately I feel you and danshewan both have valid points. And ideally I can pick something in the middle.
Style wise I agree this is a different style from wow, I'm aiming more for their quality bar, not their style.
Welcome to MMO developement, make that monster 1 draw call either throw how you UV it or through code. I believe that's why the UVs are so odd on the Warhammer Online characters because we had a process of baking all the armor & clothing down to one texture at runtime.
But seriously, given that my model and wow player models are within 200 tris, and very similar texture sizes, are mine that much worse? Maybe I'm just insane, but they don't seem miles ahead of my wip.
No offence, but honestly yes it's way off the WoW model, but I'm interested to see where this goes and wish you luck with it.
I think that the feed are in realy early developement..
the texturing looks nice except on the small legs (to much dark there)
if you take the postet picture of the skeleton as a reference you could try to shorten the arms to the knees . At the moment if they would hang down he would always contact the ground.
given that my model and wow player models are within 200 tris, and very similar texture sizes
...neither of these things have anything to do with quality or execution of a piece.
a lot of the sculpting is very confusing. what are the rows that you put on all the muscles on the upper arms and torso? and what are the protrusions on the arch of the chest? the sharp Vs on the belly are distracting and unnatural.
is there a concept for this guy? it might clear up a lot of the anatomy issues people are having if we can see what you're shooting for.
tried to listen to some of your guys suggestions, legs are now longer, tried to make some of the torso areas more realistic, gave him a diaper but kept the asscrack, the boots look like shit, but the LP has so little detail there it wont matter, and i'll just poly push to get the right shape i want there. Also ignore this head.
I think the baked version looks a lot better than the high poly. I think the baked version forces some kind of form on it. Your anatomy needs work, but whose doesn't? If you're going to be making 100s of characters I say take what you learned and move on. You will see a large improvement in #99 vs. #1
I didn't really mind the original sculpt's bake result short of the lower half. The update looks good though I'd thicken the upper thighs and hips to the point it would appear comfortable for him to sit back into in a squatting position. If his proportions sit sturdy, he will stand sturdy and strong.
yeah, i tend to get better forms with LP, once i bake i push and pull until it looks good. A lot of people have problems with HP looking blobby, myself included in some areas.
For those of you who read my big post about the project itself, you know one of the inherent problems with the project is the collection of art styles, and art qualities. At first i thought i could just moderate and throw out what i don't like, but because this project won't likely have many helpers at the beginning if ever, i thought it might be easier if i streamlined the art style, and lowered the difficulty curve. This makes it easier to accept anyone's help, if they are willing to give it.
and the answer is: vertex painting
although they won't serve well as portfolio pieces (but neither does my normal shitty artwork) doing the game entirely vertex painted might kill several birds with one poorly colored stone.
+lower system requirements (ability to run on mobile platforms (probably make lod models or something for that))
+low/no difficulty curve, you slap a premade model in front of someone, an easy to use vertex painting editor (ingame would be cool) and a 5 year old could design armor for a quest.
my only major concern is if there is an easy way to store vertex painted information outside of a model file? Otherwise most armor pieces would have to be seperate meshes...
There are some odd choices that you may be making in order to create something unique and original, but if you're looking to create a stylized humanoid with a large upper body and long arms and short feet then I recommend taking a closer look at actual cartoon characters that are similar to what you have in mind and see how they approach solving the same problems. You'll probably find that most adhere to certain conventions regardless of how extreme they are simply because you still want to remain in the bounds of believability.
This does demonstrate that technically you are capable enough to bring an asset like this through to near completion, but I don't think going the whole normal mapped route is the answer with the restrictions you're working under. I don't think vertex colors are it either... if you take a look at even most mobile games, you'll find that these days they often look more modern than that.
You shouldn't use another texture for the face if you can help it. In many games where you have to swap out pieces for customization, a technique that is used is: http://en.wikipedia.org/wiki/Bit_blit
To be honest, I think character customization shouldn't be your priority though. Maybe it's good to have one or two nice looking characters to show off, but you may be better served just building simple temporary models and just get the game working. Maybe if you can make a decent build of the game available then you'll be able to attract people to build assets for it.
As for the WoW comparison, I think it's common for some people to take a look at the game and dismiss it for it's low res nature, but if you examine a lot of the stuff in that game it's pretty sound work. Aesthetically WoW remains believable and appealing where your current model is coming across as not thoroughly explored. Giving your model all the detail and tech in the world won't matter if it doesn't appeal to players in some sense, and that's the part about art that is probably the most difficult to develop.
To be honest, I think character customization shouldn't be your priority though. Maybe it's good to have one or two nice looking characters to show off, but you may be better served just building simple temporary models and just get the game working.
this
it's not the right thread but since it's been brought up ...
if you're still serious about this project as a game, i wouldn't even think about a single solitary piece of tech until you've done the following :
render a plane
put a box on top of the plane
make the box move through player input
put a sphere somewhere on the plane
move the box to the sphere, and through player input make the box flash three times
after three times, make the sphere flash
print 'one' on the screen
repeat this
print 'one two'
render 6 boxes
control the 6 boxes through 6 different PCs on a LAN
make them all flash three times, and when they have make the sphere flash
print 'one two three'
control the 6 boxes through 6 different PCs over the internet
make them all flash three times, and when they have make the sphere flash
print 'one two three four'
crack a bottle of champagne, and then get back to your characters and bits of tech and class descriptions and backstory and all that stuff, cos you might, just might, get something out of it.
yeah, that might be good, i'm not sure though. I'll probably end up using some textures for the environment at least, probably going to aim for something simple, windwake esque.
I've messed with realmcrafter in the past, but I want to use unity because it just seems more customizable and is also much easier to convert a game to a different platform if i want to at some point.
Replies
the bake is done in xnormal with the 4 mil poly model shown above, i have a sphere cut in half on the bottom for bouncing light, any suggestions on making the bake better? if i put geo above the model will it block the light, or just bounce and make the top of the model less harshly lit?
Even if you pursue and develop this design (which I'd recommend against), you're going to have a tough time animating that mesh with the topology you've got right now. As it is, this looks like 90% of the sculpts we see from beginning artists, with too much focus on surface details like veins and not enough thought put into how muscles actually work, or any thought as to realistic proportions.
Look at the best examples of fantasy characters - even though they don't exist, they look like they could, because of a believable underlying musculature structure and proportions, no matter how stylized. I'd normally suggest studying anatomy to gain an understanding of how muscles overlay the skeleton to make your concept more believable, but something tells me this won't happen.
Fact is, if you're dead-set on making this ridiculously-over ambitious project, people aren't going to work on something with this level of art. I think you'd be better off actually studying and practising on improving your art skills before worrying about trying to make a 'name' for yourself as a designer, but you were told this ad nauseum in your other thread, and it fell on deaf ears.
/shrug
As far as my muscles, and bone structure go, i really thought they were believable, i have been using reference images a lot as i sculpt, but exaggerating at my discretion. I don't want to make the same troll and orc as everybody else, but i also want it to look fairly good. Personally I feel i've at least hit the WoW level of quality on this character.
There's nothing wrong with trying to develop a more unusual design to differentiate your work from the thousands of other orcs and fantasy creatures, but without a believable, functional anatomical base, it won't look fresh or interesting, it'll just look wrong.
How is this orc even supposed to maintain a standing position, with those massive arms, with such tiny legs? If you're using apes as a starting point, notice that although the legs are shorter, they're not so short as to be little more than appendages.
Also, notice the width of the pelvis in the image you posted - this gives the ape a solid core to achieve and maintain balance. We know that apes walk on all fours, but nothing in your design suggests this - it just looks wrong. Also, even if your orc did walk on all fours, how are those tiny legs supposed to propel it forward? It'd have to take about ten tiny steps for every arm motion.
Speaking of arms, look at them. Why are the forearms more muscular than the biceps? Why are they noticeably longer than the upper arms? You say you've studied anatomy closely while working on this, but I'm sorry, I'm just not seeing that.
You're kidding, right? I mean, not to sound like an asshole, but you have to be kidding or trolling.
http://dl.dropbox.com/u/1310775/krugerjdunningd-unskilled.pdf
If you're deadset on this MMO idea you don't really have the luxury of lovingly crafting every asset. There's a saying that applies: Good, Cheap, Fast; choose two. You've got no budget and you want to finish this in your lifetime so cheap and fast means you don't have the time to make it look good.
Ouch, seems a bit harsh. I mean I understand that it is important to have a good underlying structure to your characters, but if you can't create something completely fictional within the game world then we as designers are going to get stuck in a rut of cookie-cutter art our whole career.
Now don't get me wrong, I do see some obvious flaws within the design, but nothing that would suggest that he scrap what he has. I would suggest going back into Zbrush and tweaking some of your forms to create a more stable character, or give him something to ride that would eliminate his use for legs and walking all together.
I see too often cool designs thrown out the window because they don't look like they would function in the real world, when the fact of the matter is that games are not the real world. You should be able to go balls to the wall and create things you could never imagine seeing anywhere else, creativity essentially has to limits when it comes to video games.
Sorry for the rant, and I don't want to sound like I'm negating everything you said danshewan, you did bring up some great points. Also I will have to agree that this is very far off from WoW style, so if thats what you were going for achillesian then you may need to rethink your approach a bit.
All in all, I like the design, I'm eager to see where you go with the texture on this one. Keep it up
Maybe I can make the legs a little longer, its really a big part of the style and ties into how these races evolved. Forearms determine grip and hand power, so maybe these trolls mold rock into weapons and homes with bare hands. Maybe they live in a very small area, and never travel much, resulting in underdeveloped legs. I'm just tired of the uniform humanoid sillhoutte for every mmorpg race, and this is my attempting in changing it up. But maybe it is over exagerated.
@tRens
Maybe I do think I'm better than I am, according to those graphs almost everybody either overestimates or underestimates their abilities though.
But seriously, given that my model and wow player models are within 200 tris, and very similar texture sizes, are mine that much worse? Maybe I'm just insane, but they don't seem miles ahead of my wip.
@Jungsik
Thanks, most mmo's these days don't have ugly or fat playable characters, and I think its important to realize another world is full of different and unique body types, and even if they look monsters it doesn't mean they are. One of the ongoing themes of the game, and the fact that you can pick any faction as any race, is just going by someone's appearance doesn't always verify friend or foe, as is so often the case in real life.
@Justin
The separate head texture is because of multiple head choices. Armor is going to be separate meshes so I didn't have to break the naked troll texture up into 10 different pieces like WoW does.
@Mike
Thanks alot for your words Mike, ultimately I feel you and danshewan both have valid points. And ideally I can pick something in the middle.
Style wise I agree this is a different style from wow, I'm aiming more for their quality bar, not their style.
No offence, but honestly yes it's way off the WoW model, but I'm interested to see where this goes and wish you luck with it.
So you Z Sculted all this, baked it down, and THEN asked for crits?
Shouldn't you have asked for crits at the Zbrush stage and then after getting those used what you got and then baked it down?
the texturing looks nice except on the small legs (to much dark there)
if you take the postet picture of the skeleton as a reference you could try to shorten the arms to the knees . At the moment if they would hang down he would always contact the ground.
...neither of these things have anything to do with quality or execution of a piece.
a lot of the sculpting is very confusing. what are the rows that you put on all the muscles on the upper arms and torso? and what are the protrusions on the arch of the chest? the sharp Vs on the belly are distracting and unnatural.
is there a concept for this guy? it might clear up a lot of the anatomy issues people are having if we can see what you're shooting for.
tried to listen to some of your guys suggestions, legs are now longer, tried to make some of the torso areas more realistic, gave him a diaper but kept the asscrack, the boots look like shit, but the LP has so little detail there it wont matter, and i'll just poly push to get the right shape i want there. Also ignore this head.
yeah, i tend to get better forms with LP, once i bake i push and pull until it looks good. A lot of people have problems with HP looking blobby, myself included in some areas.
@cholden
Thanks for your words chris!
---
For those of you who read my big post about the project itself, you know one of the inherent problems with the project is the collection of art styles, and art qualities. At first i thought i could just moderate and throw out what i don't like, but because this project won't likely have many helpers at the beginning if ever, i thought it might be easier if i streamlined the art style, and lowered the difficulty curve. This makes it easier to accept anyone's help, if they are willing to give it.
and the answer is: vertex painting
although they won't serve well as portfolio pieces (but neither does my normal shitty artwork) doing the game entirely vertex painted might kill several birds with one poorly colored stone.
+lower system requirements (ability to run on mobile platforms (probably make lod models or something for that))
+low/no difficulty curve, you slap a premade model in front of someone, an easy to use vertex painting editor (ingame would be cool) and a 5 year old could design armor for a quest.
my only major concern is if there is an easy way to store vertex painted information outside of a model file? Otherwise most armor pieces would have to be seperate meshes...
This does demonstrate that technically you are capable enough to bring an asset like this through to near completion, but I don't think going the whole normal mapped route is the answer with the restrictions you're working under. I don't think vertex colors are it either... if you take a look at even most mobile games, you'll find that these days they often look more modern than that.
You shouldn't use another texture for the face if you can help it. In many games where you have to swap out pieces for customization, a technique that is used is:
http://en.wikipedia.org/wiki/Bit_blit
To be honest, I think character customization shouldn't be your priority though. Maybe it's good to have one or two nice looking characters to show off, but you may be better served just building simple temporary models and just get the game working. Maybe if you can make a decent build of the game available then you'll be able to attract people to build assets for it.
As for the WoW comparison, I think it's common for some people to take a look at the game and dismiss it for it's low res nature, but if you examine a lot of the stuff in that game it's pretty sound work. Aesthetically WoW remains believable and appealing where your current model is coming across as not thoroughly explored. Giving your model all the detail and tech in the world won't matter if it doesn't appeal to players in some sense, and that's the part about art that is probably the most difficult to develop.
this
it's not the right thread but since it's been brought up ...
if you're still serious about this project as a game, i wouldn't even think about a single solitary piece of tech until you've done the following :
render a plane
put a box on top of the plane
make the box move through player input
put a sphere somewhere on the plane
move the box to the sphere, and through player input make the box flash three times
after three times, make the sphere flash
print 'one' on the screen
repeat this
print 'one two'
render 6 boxes
control the 6 boxes through 6 different PCs on a LAN
make them all flash three times, and when they have make the sphere flash
print 'one two three'
control the 6 boxes through 6 different PCs over the internet
make them all flash three times, and when they have make the sphere flash
print 'one two three four'
crack a bottle of champagne, and then get back to your characters and bits of tech and class descriptions and backstory and all that stuff, cos you might, just might, get something out of it.
http://www.thegamecreators.com/?m=view_product&id=2130
thumbs up to this, I think this is actually a good choice, it'll save some crazy amounts of time and let you focus on getting stuff done.
(now pick up unity)
yeah, that might be good, i'm not sure though. I'll probably end up using some textures for the environment at least, probably going to aim for something simple, windwake esque.
@danr
good tips danr, i can do about half of this list, i'm working in unity, that alone cuts several corners.
@XenoKratios
I've messed with realmcrafter in the past, but I want to use unity because it just seems more customizable and is also much easier to convert a game to a different platform if i want to at some point.
@eld
thanks for your vote, yeah the whole idea really solves a lot of problems the project had.
---
My newest problem however is getting vertex painting into unity 3, which i have had 0 luck with.
simple, easy to understand, easy to edit, kinda cute