Topic: Poly Count (Read 3534 times)

starforce2 · « **Reply #20 on:** March 17, 2004, 02:12:55 am »

Quote:

Most modelers are well aware of the poly counts of their ships. Of course, most modelers also build ships for themselves, and not others, so they will add as much detail that they want and what their own computers can handle. This is more of a hobby than a community service, so we tend to indulge ourselves first, then worry about what the rest of the world wants.

excactly...those of us who are willing to spend the money on computer hardware want to enjoy it. You wouldn't buy a ferarri to drive 20mph....

Atrahasis · « **Reply #23 on:** March 17, 2004, 03:44:16 am »

I'll need to bring some facts to light to discuss this:

1) For the vast majority of the playing time, the most important model that is used is LOD 2, not LOD 1. This is because LOD 1 seems to kick in at a range of only 3 (or less) or so away from you, and LOD 2 kicks in at ranges beyond that out until when LOD 3 is used (which i don't know the exact number for, but it's pretty far). Ergo for most of the ranges where the shooting occurs and starships spend most of their time, it's LOD 2. This means:

A) You can make LOD 1 rather high-poly if you like, but I try to stay below 6000. Granted things may get tricky if you have saucers and cyclindrical warp engines to deal with.

B) LOD 2 is the model that you should really try to engineer to lower polies on. It's the model where you should be pulling all the tricks you know yet at the same time not sacrificing discernable quality. But since range 3 is actually somewhat far away from you, especially when th screen is at normal resolution, you don't need to make it that detailed....just sufficient. So for a LOD 1 model at 6000 polies, you should try to shave as much of that of as you can for LOD 2, say down to 3000 or so.

If at all possible, I believe if you can make a model at 1500 polies or less, you can make LOD 1 and LOD 2 the same model if you like.

Atrahasis · « **Reply #24 on:** March 17, 2004, 03:53:15 am »

Also, I might add that people should experiment more with the "optimize" function in Max with their meshes, to make the lower-poly meshes, but it's not always a viable solution to every problem. For example, I used it to lower the poly count on the TOP of a Fed saucer by 400 polies, but I couldn't use it on the BOTTOM of the saucer because it created mesh warping that made the textures look odd, because the bottom of the saucer has a steeper angle. However, if you consider that the BOTTOM of a ship model in the game is not seen often (I don't think the camera even pans to the bottom angles) then that means for LOD 2 you can get away with bastardizing the bottom areas of ships that are never seen, as much as you like, even if it causes warping or what not. LOD 1 people may want to use for renders outside of the game, but nobody cares about LOD 2.

ModelsPlease · « **Reply #25 on:** March 17, 2004, 04:24:59 am »

I will go on record as stating this I wish I could build ships for this game or even learn to texture.Alas my choice in being in the military and deployed doesn't truly allow me too take classes for it.But I appreciate everything each and every one of you modelers do.It was this passion that you all have for Star Trek and gaming that even made me join this forum and join in as a voice in the chorus.So thank you all and with that being said.............
GIMME MORE ModelsPlease

-MP

red_green · « **Reply #26 on:** March 17, 2004, 05:01:35 am »

Yea I can see where being deployed could put a crimp on things. Keep in mind that many of us just putz around in this. This forum has been kind of a class for me. I read stuff, it makes no sense. Then a month later while I am doing a model I remember reading something about the problem I have and it makes sense then. Like Atrahasis talking about LOD's . I really don't know how to do them yet but from reading here I probably could try them and struggle through it. Maybe asking a question or two. This stuff does take a while to pick up. I just now can make a so so model after about 16 months off and on. Fortunatly for me I am building for SFC3 which has dynamic LOD's. I started out by retexturing other peoples stuff. Most of it wasn't an improvement either. I just changed color schemes using a free editor. Then I bought milkshape for 20 bucks and its been a lot of fun but also a lot of frustration at times. I have to re-evaluate what it is I want out of this hobby.
I mean I am spread out all over amongst SFC1, 2 and 3. If I focused on 1 I could have a decent mod out by now.

Smiley · « **Reply #27 on:** March 17, 2004, 06:20:53 am »

Polies.....well, I try to make models for more than just one game, and recently I have been concentrating more on Armada 2 - because, well, I started out modding in Armada and I feel like I'd like to give something back, plus - even though it's and old game there are still a lot of models that haven't been made - the Jem'Hadar for example, even though they are a popular high profile race, they have not many built ships!....Ra....sorry went off topic there - but my point was that I make low poly models for Armada, I recently made a 827 poly Jem'Hadar Battleship for MP2 and I am very pleased with it, but it does use 2 1024x1024 TGAs and also uses lights maps, so that's like 8mb of texture memory already, of course the beauty of that it mip-mapping - you create copies of the same textures at lowering resolutions and the computer selects the appropriate one for the level of detail required, SFC can use the same thing....I think?
Plus there is a point I'd like to make that no one else have made - SFC3 using dynamic LODs? Does it not?
It all depends on the mind of the modeller and his target audience....whether the audience be his/herself, SFC1, SFC2x, SFC3, BC, Armada 1, Armada 2, Homeworld 1, Homeworld 2 etc.
I would like to go on record to say that all modellers should do their best to make the best models that they can at the lowest possible poly count - that's the mark of really good modellers, plus it's more efficient (green for the economy, Borg stylie) - but once again, depedant on the target destination of said model. I'm a newish modeller, I know that I could do better, but when I started I just actually wanted to get the right shape! lol

G_Kane · « **Reply #31 on:** March 17, 2004, 10:57:49 am »

Its interesting to read all the posts.

I have a good idea how some of you feel.

I started making ships because of this forum.

I wish I could make ships as good as what I've seen here,but it will take time.

now I try to make ships with a poly count of 2000 or less.

I don't want you guys to change because of me. Keep up the good work.

I love art and the work I've seen here is art.

Atrahasis · « **Reply #32 on:** March 17, 2004, 05:05:51 pm »

I don't think people should "keep their mouths shut" when it comes to constructive criticism. When I first started I got plenty of criticsm and I welcomed it, because I really wanted to know how to improve. You might be acquainted with the phrase "a trained eye"....it doesn't just evolve on its own, it needs input. I see a lot of work that could probably be improved but I don't notice anyone saying anything about it, and so the authors keep going about their work in the same style, not that that's necessarily bad, but it often means he's not improving...as in adding to his skills. Expanding his horizons. Whatever you wanna call it...

red_green · « **Reply #33 on:** March 18, 2004, 03:24:36 am »

No one was asking anyone to keep there mouth shut. Criticism is good. Unfortunately alot of criticism is too vague to be helpful. Responses like you destroyed that model or that sucks do nothing to guide a modeler to a better end product. I see responses like that and it doesn't sound like criticism, it sounds like an insult.

Other criticism usually involves personal taste. Someone might think the buzzards are the wrong color but it might be that the modeler prefers them that way. Criticism is best when a solution is provided. Like if you notice that a mesh is distorted instead of just saying the mesh looks distorted, it would help to explain how to correct it. Its very likely that the modder knows its distorted but doesn't know how to go about fixing it other wise they would have done it.

How criticism is presented is important. Just keep in mind that the model may have as much as 50 hours of work in it. It may be one of the 1st ones a modeler is doing. They may not know how to fix it.

I don't think any criticism in this thread was presented in the wrong way. But I have seen criticism go awry before.

Atrahasis · « **Reply #34 on:** March 18, 2004, 03:48:20 am »

Yeah, I agree with everything you said, a "fix" should be suggested when appropriate, but at times a plain simple "it just doesn't look right" can also be helpful. I've recieved some of those in the past, and in retrospect they were helpful comments because some things are hard to nail down with words.

MajorRacal · « **Reply #35 on:** March 18, 2004, 07:57:06 am »

Just to chip in, I agree with you there... Apart from the idea that other people might like to make use of my creations in time, feedback, good or ill is paramount to development, and it's one of the main reasons I like to post here. I've had some interesting feedback over the past 10 months, some of which I've taken onboard, and some I've ignored, but it's all been useful (and I thank you all for it!)

As for the poly count, I'm guilty of building higher poly models than perhaps is wise (now I'm working to claw the counts down without spoiling them)... The Liberator (oddly) is probably my worst offender with over 30000. Some of my earlier attempts breach the 10-15K level, but I'm trying to strike a balance of 4-6K wherever possible.

MajorRacal

G_Kane

Re: Poly Count

Captain Ron

Re: Poly Count

starforce2

Re: Poly Count

Chrystoff

Re: Poly Count

Captain Ron

Re: Poly Count

News: