Texture Space, Theory and Practice. - Page 3

Texture Space, Theory and Practice.

Share and discuss visual creations and creation practices like texturing, modelling and musing on the meaning of life.

Moderators: MR.D, Moderators

User avatar
Zpock
Posts: 1218
Joined: 16 Sep 2004, 23:20

Re: Texture Space, Theory and Practice.

Post by Zpock » 10 Jan 2008, 22:51

It's like ordering an XXL pizza and then eating one slice and throwing the rest in the bind. Then telling others this is a good idea.

Ok it might be good from a health perspective, why not just buy the smaller pizza in the first place?
0 x

User avatar
Argh
Posts: 10920
Joined: 21 Feb 2005, 03:38

Re: Texture Space, Theory and Practice.

Post by Argh » 10 Jan 2008, 22:58

Well, because sometimes you need a strip of that pizza that's exactly 512 long, for a giant area of a model, but you won't need space that large for the LOD elsewhere.

The only way to achieve that same slice of pizza with the next step down involves 2 hours of work, slicing up the model further, then having to do really tricky airbrush to hide the seams, like on that helicopter- screw that, I'm at spec, looks good, end of story...

That's the whole point of having a spec in the first place, guys. It's your upper limit. It doesn't mean you have to drive yourself crazy, or waste time.
0 x

User avatar
Wolf-In-Exile
Posts: 495
Joined: 21 Nov 2005, 13:40

Re: Texture Space, Theory and Practice.

Post by Wolf-In-Exile » 10 Jan 2008, 23:02

Argh, you could've said that you knew you could/should, but didn't choose to spend more effort on the UV map because you wanted to save time, and in turn saved yourself the trouble of making this thread. :P

Zpock: sorry for misinterpreting the tone of the sarcasm, please use smilies to avoid confusion. :wink:

I don't see how my position is extreme, in fact I phrased my statement very specifically. Its one thing to produce poorer quality models due to external factors like time, resources and so forth, but its quite another to denounce proper work practices (case in point, the 'polycount wars' we had not that long ago).

Yes, there are some people who give sound advice here, but if you don't know any better, its difficult to decide whose advice to follow, and whose to ignore.
0 x

User avatar
Argh
Posts: 10920
Joined: 21 Feb 2005, 03:38

Re: Texture Space, Theory and Practice.

Post by Argh » 10 Jan 2008, 23:12

The reasons I chose to have this thread:

A. I get sick of guys who can't skin like I do, even at the same rez, implying I'm "doin' it wrong". Whatever. Work is proof.

B. What I'm telling you guys is that, by being sloppy about levels of detail, you can often end up with too much space for minor crap that doesn't need it, and you often create more work for yourselves, hiding that, than is necessary. Why bother, in the name of "efficiency", when your spec was set well before you started your skin?

Think about it, people. Do you wanna rework an entire uvmap and re-spec, or do you want to get done and move on? I move on. Which is why I get nice stuff done, in one night, that seems to take most people a week or more :P
0 x

User avatar
Snipawolf
Posts: 4357
Joined: 12 Dec 2005, 01:49

Re: Texture Space, Theory and Practice.

Post by Snipawolf » 10 Jan 2008, 23:29

Well, I don't think you are doing it wrong, but doing it your way.

In my opinion, it is a horrible idea. I suck at both uvmapping and texturing, so any UV maps I make will eat space like no tomorrow.
0 x

User avatar
Peet
Malcontent
Posts: 4381
Joined: 27 Feb 2006, 22:04

Re: Texture Space, Theory and Practice.

Post by Peet » 11 Jan 2008, 00:06

Argh wrote:I get sick of guys who can't skin like I do, even at the same rez, implying I'm "doin' it wrong". Whatever. Work is proof.
Well your skinning is not the primary thing in question, it's uv mapping. You're basically saying you're a superior mechanic because you're a really good driver.
0 x

User avatar
smoth
Posts: 22297
Joined: 13 Jan 2005, 00:46

Re: Texture Space, Theory and Practice.

Post by smoth » 11 Jan 2008, 02:34

Important stuff:
Wolf-In-Exile wrote: Poly efficiency matters. The reason has been given many times already so I don't see the need to reiterate it here.

UV map efficiency matters. The less wasted space on a UV map, the better the quality of the texture, because you can put even more detail in it.

FACT
Most of the rest of this thread is crap.
0 x

User avatar
Argh
Posts: 10920
Joined: 21 Feb 2005, 03:38

Re: Texture Space, Theory and Practice.

Post by Argh » 11 Jan 2008, 02:48

@Peet:

No, I'm saying that mechanics aren't quite what everybody insists they are. That there is more to making a good skin, fast, than just packing your uvmap and then struggling with crappy coordinate spaces. Aligning stuff, so that all lines are straight as painted, and is easy to get light-sourcing on the bevels right, for example, has huge payoff, and only minor drawbacks- one of them being that it's "inefficient" on texture space usage.

I mean... look how damn long it takes most of these guys, before they can show off a piece? Is that a model you really want to emulate, so that you can pat yourself on the back about how efficient you are?

Or, do you want to pay attention to the fact that mirroring, for example, allowed me to stretch the Bomber's upper and lower halves across most of a 512, use big space for the engines, and still "waste" a lot of map? What, should I have carefully angled the halves of the Bomber to nest, screwing up my nice neat lines, just to get more efficiency? :roll:

There is efficiency, then there's being anal-retentive, for the sake of being anal-retentive. The bomber's uvmap is efficient- all detail levels are pretty much the same:

Image

See? Oooh, 20% of the map is "wasted". But... wasted, in what sense? The engines could have been "more detailed"? Whatever. They are teeny-tiny, in-game- players never, ever see them in full detail anyhow. If anything, they could be smaller and use less space. Wait, I could've made the top surfaces bigger, and more detailed, then squeezed the other stuff smaller, right? Sure, but meh, we're talking maybe 5% on total size, here. Would 5% add a lot more coolness? NO. So, who gives a fig?

Moreover, when guys like the DoW skinners wasted the better part of a 512, simply because they didn't feel like mirroring, and used multiple textures on that one model, I just have to laugh at this vein of argument, frankly. They weren't worried about saving space, there- they wasted it, in giant amounts, on something that could've been mirrored quite safely, and players would've never been the wiser.

@Smoth:

So sayeth ye, who didn't bother telling Fanger he was wasting massive amounts of his uvmaps on the underside of tank turrets, which is why this whole thing got started in the first place :roll:
0 x

User avatar
Peet
Malcontent
Posts: 4381
Joined: 27 Feb 2006, 22:04

Re: Texture Space, Theory and Practice.

Post by Peet » 11 Jan 2008, 03:02

Argh wrote:So sayeth ye, who didn't bother telling Fanger he was wasting massive amounts of his uvmaps on the underside of tank turrets, which is why this whole thing got started in the first place :roll:
Uh..wut. You created an epeen-brandishing fest on here because smoth neglected to notice a few hidden faces on another person's WIP model?
0 x

User avatar
Argh
Posts: 10920
Joined: 21 Feb 2005, 03:38

Re: Texture Space, Theory and Practice.

Post by Argh » 11 Jan 2008, 03:15

Hidden faces?

OMG, LOL!!!

You obviously didn't pay a lick of attention, to what that shot showed, or you didn't understand what it meant.

Here, let's see what started all of this, so that everybody can judge fairly. Smoth, as a teacher, who supposedly knows so much about uvmapping that he's labeled me full of crap, let this piece of junk go without comment:

Image

The red stuff, btw, is my markup, showing all of the faces on the undersides of the turrets and the bottom of the tank.

The green stuff is the stuff I can see should be mirrored, just from looking at the map, unless the columns need to have different ends.

This is efficiency? Give me a break. I'm the lousy teacher? Sheesh. I must be terrible...
0 x

User avatar
smoth
Posts: 22297
Joined: 13 Jan 2005, 00:46

Re: Texture Space, Theory and Practice.

Post by smoth » 11 Jan 2008, 03:20

Argh wrote:Here, let's see what started all of this, so that everybody can judge fairly. Smoth, as a teacher, who supposedly knows so much about uvmapping that he's labeled me full of crap, let this piece of junk go without comment:

Image
Easy, fang is a noob, and he was just playing. You project yourself as lord on high of modeling. HENCE why your uvs are under scrutiny.
Argh wrote:The green stuff is the stuff I can see should be mirrored, just from looking at the map, unless the columns need to have different ends.
Good that you can spot it... congrats..
Argh wrote:This is efficiency? Give me a break. I'm the lousy teacher? Sheesh. I must be terrible...
That is not my uv... LOL.
0 x

User avatar
Argh
Posts: 10920
Joined: 21 Feb 2005, 03:38

Re: Texture Space, Theory and Practice.

Post by Argh » 11 Jan 2008, 03:23

I know he's just playing. I know he's still figuring stuff out.

I just came there, saw what he had posted, and wanted to help. That was the biggest single issue I saw, and I said so. Then everybody jumps on me, for the better part of 48 hours now :P

It's kind've retarded, people. I was just going to post one critique, and when Fang said, "hey, I'm not even slightly done yet", I said, "fine, I'm moving on", and I thought that was the end of it. But, oh noooo. Peet hounds me via PMs, Forb decides to have little jokes on me, and I'm put in the invidious position of having to explain something that I don't even have a really coherent theory about, but know is working.

I just wanted to help Fanger out. I could see he's finally getting somewhere, and I was happy for him, and wanted to help.

You guys buried that help, under this waste of time, over stuff that's my personal choice, that I've never bothered explaining to anybody up until now. All I really know is that it's working for me, and I think that you guys who're obsessed with packing the maps are wasting valuable time you could be spending painting... or lower your spec so that it becomes an exercise with a real point. Paint everything in Gundam with 256s, Smoth, and make it look better than PURE. I'm sure it can be done- a year from now. I don't have a year.
Last edited by Argh on 11 Jan 2008, 03:33, edited 1 time in total.
0 x

User avatar
Peet
Malcontent
Posts: 4381
Joined: 27 Feb 2006, 22:04

Re: Texture Space, Theory and Practice.

Post by Peet » 11 Jan 2008, 03:24

Argh wrote:I just came there, saw what he had posted, and wanted to help. That was the biggest single issue I saw, and I said so. Then everybody jumps on me, for the better part of 48 hours now :P
Because your "teaching" consists largely of arrogant and demeaning blather directed toward the "student" and toward others.
piece of junk
etc.
0 x

User avatar
smoth
Posts: 22297
Joined: 13 Jan 2005, 00:46

Re: Texture Space, Theory and Practice.

Post by smoth » 11 Jan 2008, 03:27

Argh wrote: @Smoth:

So sayeth ye, who didn't bother telling Fanger he was wasting massive amounts of his uvmaps on the underside of tank turrets, which is why this whole thing got started in the first place :roll:

Missed this one...

ah, that would be because fang does not act like he is the expert and guru of modeling. Fang was just playing around and trying the uvmap thing out. He did not say it was done or even efficient he was just saying, hey look I MAED A UV... that was all.

Now, you can rage all you want but frankly, the fact is that my uvs are tighter then yours and I use my space better.

Fang was just playing around

you claim that you are a sort of expert.

HUGE DIFFERENCE.
0 x

User avatar
Maelstrom
Posts: 1950
Joined: 23 Jul 2005, 14:52

Re: Texture Space, Theory and Practice.

Post by Maelstrom » 11 Jan 2008, 03:36

Your bomber skin is a pretty efficient skin, I agree. With a minor effort, it could be made more efficient, but in this case it is not really worth it.

However, the map you posted at the start of the thread? HORRIBLE. While the unit itself does look good, yes, it could look better with a VERY minimal effort. Seriously, it is not difficult to rearrange a UV map to be more efficient than that.

I hear you crying that its not worth the effort? Bull. It might take a little more time, but the time you spend making it more efficient pays off in the end. You can make the units look even better than they already are.

You can still have an efficient map with alligned edges. I agree, having edges that are not straight are not good. But efficiency does not have to mean misalligned edges at all. That claim is complete CRAP. Im sorry, but it is. Having properly alligned edges does not mean you cant be efficitent. Take the map that Zpock posted for instance. That is a nice map, hardly any wasted space, and alligned pieces. A quick glance at the map reveals a few sections that could possibly have been mirrored, although I couldnt comment unless i saw the model, but that is not the point I'm trying to make. It is a map with very little wasted space, and aligned edges.

Argh, you are seriously full of crap. There is no nicer way of putting it.
0 x

User avatar
smoth
Posts: 22297
Joined: 13 Jan 2005, 00:46

Re: Texture Space, Theory and Practice.

Post by smoth » 11 Jan 2008, 03:42

I see what you did thar....
Argh wrote:You guys buried that help, under this waste of time, over stuff that's my personal choice, that I've never bothered explaining to anybody up until now.
I didn't burry shit. I said hey, fang want help with your uvs... he said nope and I said ok. YOU made the mistake of thinking he wanted a critique when he did not ask for one.
Argh wrote: All I really know is that it's working for me, and I think that you guys who're obsessed with packing the maps are wasting valuable time you could be spending painting... or lower your spec so that it becomes an exercise with a real point. Paint everything in Gundam with 256s, Smoth, and make it look better than PURE. I'm sure it can be done- a year from now. I don't have a year.
Very simply put: Gundam is something I ACTUALLY CARE ABOUT and I will make the textures very detailed and devote good time on them because I am doing a project I really care about. Pure might be a throw away project for you but gundam is a project that I take very seriously. NOONE GROCS GUNDAM LIKE I DO.
0 x

User avatar
Argh
Posts: 10920
Joined: 21 Feb 2005, 03:38

Re: Texture Space, Theory and Practice.

Post by Argh » 11 Jan 2008, 03:49

@Smoth: Fang is not "playing around", in that sense, Smoth. He's supposed to be learning how to uvmap and skin. Nor did I attack Fang, in any way. You guys cannot, apparently, just let critique be critique. Oh, noes, if it's not totally positive, it must be an attack. Whatever.

And what, PURE is a throwaway? I don't care about it? Not making sense there. Time invested has a curved relationship with quality. And we're not building ships in a bottle here. There is a shelf-life on the human effort involved.
I hear you crying that its not worth the effort? Bull. It might take a little more time, but the time you spend making it more efficient pays off in the end.
Oh... and where is this supposed payoff, Mael? Over the rainbow, where I get hours of my life back? Where's the payoff? Why do you kid yourself? :roll:

If you guys really think that "quality" somehow erupts magically out've "efficiency", you're silly. If you think that it's worth an extra day of my life, just to be a "little more efficient", when I can achieve a level of quality that's acceptable, then you're wrong. An extra day is a whole 'nother unit for PURE that isn't done. This argument has cost me two days.

I'm going back to getting things done now, argue with straw men all you want.
0 x

User avatar
smoth
Posts: 22297
Joined: 13 Jan 2005, 00:46

Re: Texture Space, Theory and Practice.

Post by smoth » 11 Jan 2008, 03:54

LOL, you didn't read my post.

Truth: I care about gundam

Truth: I have tight uvs and you cannot argue that.

Truth: you put on the pretense that you are some sort of knowledgeable person in modeling.

Truth: you are not.

Truth: I spend time giving as much detail to gundam because I care.

Truth: you rush your pure texturing.

Truth: better usage of space = more space for details.

Truth: you do not shadow any part of your pure models.
0 x

User avatar
rattle
Damned Developer
Posts: 8278
Joined: 01 Jun 2006, 13:15

Re: Texture Space, Theory and Practice.

Post by rattle » 11 Jan 2008, 04:36

I improved my texturing skills while you guys are arguing over crap, I guess that makes me a winner.
0 x

User avatar
Fanger
Expand & Exterminate Developer
Posts: 1509
Joined: 22 Nov 2005, 22:58

Re: Texture Space, Theory and Practice.

Post by Fanger » 11 Jan 2008, 04:42

JEEZE im fucking sorry I posted a picture of a s3o model.. It wasnt meant to start a shit storm.. GOD how do I constantly start shitstorms just by making something, and posting it somewhere..

That Uvmap could possibly be mirrored in a couple of spots, and I suppose the bottom of the turrets and tank prolly dont need to be the same size.. Im not going to dispute that.. Mainly I was confused by your post, because basically you pointed out a bunch of stuff that I hadnt actually done yet.

I probably should have made a note somewhere, but you never seem to comment on my stuff so I didnt expect anything, and actually thought no one had noticed that post, since it lay dormant for like 2-3 days before someone replied..

My uvmaps are shit (I KNOW THIS ALREADY) My textures could use alot of work, and in most cases arent finished (Again already known)..

I apologize for posting a picture.. side note this sort of stuff is the exact reason I havent posted a thread about EE here.. I must have epic shitstorm starting abilities...
0 x

Post Reply

Return to “Art & Modelling”