The Grand Experiment Thread

Moderator: IskatuMesk

User avatar

Topic author
IskatuMesk
Xel'naga World Shaper
Xel'naga World Shaper
Posts: 8788
Joined: Sat Feb 07, 2009 1:40 pm
Location: M͈̙̞͍͞ͅE̹H̨͇̰͈͕͇̫Ì̩̳CO̼̩̤͖͘
Contact:
Status: Offline

Re: The Grand Experiment Thread

Postby IskatuMesk » Mon Feb 17, 2014 9:45 pm

me having artistic talent, that's a lol and a quarter
Image~[Gameproc]~Image
Warning: dialogue contains politically incorrect content. Viewer rearsore may occur.

User avatar

wibod
Zerg Creep Colony Landscaper
Zerg Creep Colony Landscaper
Posts: 433
Joined: Tue Oct 13, 2009 7:55 am
Status: Offline

Re: The Grand Experiment Thread

Postby wibod » Tue Feb 18, 2014 12:10 pm

mark_009_vn wrote:Explain to me why Mesk, why, why must you refuse to texture even though you got such artistic talent that could change the very faucet of the modern day?


Image

Awww shit nigga
Image

User avatar

Topic author
IskatuMesk
Xel'naga World Shaper
Xel'naga World Shaper
Posts: 8788
Joined: Sat Feb 07, 2009 1:40 pm
Location: M͈̙̞͍͞ͅE̹H̨͇̰͈͕͇̫Ì̩̳CO̼̩̤͖͘
Contact:
Status: Offline

Re: The Grand Experiment Thread

Postby IskatuMesk » Sun Mar 09, 2014 3:19 am

Some more filter forge sets, this time using a special master material I found on the internet. I don't actually know what the master material does, but it's significantly more complex than anything I've found in the stock files except for very high complexity demo shaders. I think it improves the specularity.

Image

A ground material with the spec parameter a bit too high.

Image

Another ground texture. Ones like this turn into a huge mess when at all distant from the camera.

Image

The opposite angle of that one. Try to ignore the ugly ass glowing shit.

Image

This one in particular looks pretty decent I think.

Image

Example of awful banding that happens to anything detailed at even a short distance.

Image

Some of the other materials I made largely by hand without the master material. Note the weird alphas on the one.
Image~[Gameproc]~Image
Warning: dialogue contains politically incorrect content. Viewer rearsore may occur.

User avatar

mark_009_vn
Zerg Creep Colony Landscaper
Zerg Creep Colony Landscaper
Posts: 440
Joined: Fri Nov 04, 2011 5:56 am
Location: [[THIS USER, HE IS A CHAR]]
Status: Offline

Re: The Grand Experiment Thread

Postby mark_009_vn » Sun Mar 09, 2014 9:44 pm

IskatuMesk wrote:
Image

Example of awful banding that happens to anything detailed at even a short distance.


You'll need to save the texture with mipmap on. Optimal resolution for texture for the particular object should be about 2048x2048 with intensive mipmapping.

Also you having artistic talent is NO lol and a quarter, I see an artist when I see one.
Image

"I'm begging you, let me work!" - Osamu Tezuka

User avatar

Topic author
IskatuMesk
Xel'naga World Shaper
Xel'naga World Shaper
Posts: 8788
Joined: Sat Feb 07, 2009 1:40 pm
Location: M͈̙̞͍͞ͅE̹H̨͇̰͈͕͇̫Ì̩̳CO̼̩̤͖͘
Contact:
Status: Offline

Re: The Grand Experiment Thread

Postby IskatuMesk » Mon Mar 10, 2014 1:58 am

Yeah mipmaps aren't working out too well.

I don't know how you think pushing a button in filter forge makes me an artist. You gotta get some standards, buddy.
Image~[Gameproc]~Image
Warning: dialogue contains politically incorrect content. Viewer rearsore may occur.

User avatar

mark_009_vn
Zerg Creep Colony Landscaper
Zerg Creep Colony Landscaper
Posts: 440
Joined: Fri Nov 04, 2011 5:56 am
Location: [[THIS USER, HE IS A CHAR]]
Status: Offline

Re: The Grand Experiment Thread

Postby mark_009_vn » Tue Mar 11, 2014 5:40 pm

IskatuMesk wrote:Yeah mipmaps aren't working out too well.

I don't know how you think pushing a button in filter forge makes me an artist. You gotta get some standards, buddy.


Actually, by industrial standards that's how you do it... The only thing you need to texture is to copy/paste shits. Believe me, people write books about it.
Image

"I'm begging you, let me work!" - Osamu Tezuka

User avatar

Topic author
IskatuMesk
Xel'naga World Shaper
Xel'naga World Shaper
Posts: 8788
Joined: Sat Feb 07, 2009 1:40 pm
Location: M͈̙̞͍͞ͅE̹H̨͇̰͈͕͇̫Ì̩̳CO̼̩̤͖͘
Contact:
Status: Offline

Re: The Grand Experiment Thread

Postby IskatuMesk » Sat Mar 15, 2014 5:02 pm

As Retribution is considered dead now, I will talk a bit about the third scene that I was working on.

The third scene was the Anahn Industrial Foundry. It was not only the largest and most complicated of the scenes I had created thus far, it would serve as a basis for building my props scene to be used to create more scenes later down the line.

The Foundry is one of the few things I have attempted to model based on concept art. I used a concept art to establish the basic camera angle and thematic I would go for. I attempted to find more concept art related to industrial scenes, but almost everything I found was too far beyond implied design to apply to my work, or was far too low resolution to be of any use. Modeling off of concept art is a nearly impossible task because I find it hard to bring ideas into fabrication. I work best by just swinging my dick around and yolo'ing without any foresight. I did have an established goal I wanted to do, though.

First, I had a foreground to make. This would be super high detail, because it would be where the camera was confined to.

Then I needed a background. This would be built out of largely modular pieces that could be recycled for filler scenes and other areas later on in the project.

Then I needed a skybox. The skybox was to be built out of three layers. The first two layers would be meshes and the third would be a series of composite mesh/texture hybrids using some experimental materials I was making.

Image

This is as far as I got before the concept art was useless to me.

Image

At this stage I am just flailing around. The camera bounds area is at ground level, where the first bridges and magma would be. That small tower with the three levels exposed is the control tower and would technically have 5 or so levels to it. I initially planned to model this out but then decided against it since the camera would not be inside it. However, I was never able to find a good material to mask the windows with.

Image

Still very early in. Trying out different detail ideas. I built many extrusions with the intention of splitting them so I could apply different materials to them, since I don't know how to UV or texture.

Image

It was suggested I tried to Loft a curve for the skyrails. This was too difficult for me to do, so I made a pipe and just extruded it. This entire scene is built out of extrusions and inset. That's all I do - make a block, extrude the shit out of it. The lower level begins to take shape - this is where the camera would originate from in the animation in my head.

Image

Minor detailing. I originally wanted to model/animate the trains/railed devices that would use those rails, but quickly scrapped the idea because they would be far too hard to model. I added a box to scale - the box is the scale of the average Anahn (8 feet) and represents the character scale in the UDK.

Image

Some more minor detailing and a few props I may instance later, within the UDK, for detailing. The magma takes on a cheesy complexation. I don't know how I'll make the lavafalls yet - they remain the toughest thing to figure out for the scene.

At around this stage I considered the project dead, but opted to continue work in a small bid to improve my fledgeling UDK knowledge to what extent I could. Struggles with the cat and severe depression stagnated work on the scene severely.

Image

Image

Around this time was maybe a day's worth of work total, since I worked off and on with the scene very sparingly, and most of the time was spent being confused and not knowing what to do.

Image

In particular, I didn't know how to detail the scene, and opted to ignore detailing entirely for the duration of this miniproject.

Image

I originally planned the ground level skyway to enter hangars, but didn't know how to model the hangar interiors, so I left them entirely untouched.

Image

The scene at a glance. I wanted the slag pits to be viewable from ground level but failed to measure the heights appropriately, so they were actually out of view.

Image

I learned a new technique to detail generic scaffolding. Inset a square, extrude, then inset again, collapse the new inset, inset the four pieces and extrude inwards. I applied this technique liberally throughout the scene from hereon, since it was the only way I could detail the surfaces.

Image

I never could figure out how to detail this interior.

Image

So I opted to skip interiors and focus on blocking out basic geometry for the exterior where it was within my ability to yolo my way through.

Image

The ground level bridge confused me for a long time before I decided to extrude it "off camera" and sweep it under the rug.

Image

Blocking out the area immediately surrounding ground level. This area absolutely must be high detail for the scene to have any chance of succeeding. I didn't know how to detail it, so I hoped I could make props later on to mass copy paste once I had the components set out.

Image

The prime_structure_skybridge begins to take shape. This piece is largely ignored for the remainder of the production because I don't know how to detail it. In particular I didn't know how to model cables without spending extraordinary amounts of time extruding individual curves by the thousands, so I opted to ignore all detailing that required cables, which was most of the scene.

Image

Areas that require massive detailing to succeed, as these would probably be viewable during the take.

Image

The secondary hangar begins to take shape, and would be largely ignored from hereon.

Image

I intended to instance the secondary hangar structure for background purposes but got lazy.

Image

I block out low-detail components on one of the prime towers. This tower was originally intended to be used for a railed elevator, but then the design was changed to a crane. I couldn't model a crane, so then I later simplified to be a simple support for one of the upper structures.

Image

During modeling the tower I was in an extremely dark period, so progress was extremely slow, taking many days to progress in spurts of 4-5 minutes of work an hour at most.

Image
Image

Once the tower was satisfactory I slapped together a quick skyway for it to connect to.

Image

During this period work sped up considerably.

Image

Quick and dirty pipes. I didn't care to detail them. I knew I couldn't, so I moved on ASAP. The anal-bovine mesh, aka the first hangar, got only a tiny amount of its required detailing done.

Image

The second upper level was very hastily cobbled together in a few minutes and then ignored for the remainder of my time with the project.

Image

I tried to model external plating for the pipes akin to a concept I had seen, but they still look like shit. They need more attennae and cabling and such to look any good.

Image

A look downwards.

Image

I tried to block out components for the ground level but found it impossible to detail at the level necessary. I had hoped props could fill it out; fences, lights, etc.

Image
Image

Since iteration is so easy in the UDK, I opted to consider the scene a prototype and move on to making props and putting the scene into the UDK. No further work was deemed necessary because the project is unlikely to ever move again.

Image
Image

I unified old meshes, ones used for my second scene, and some new ones, into a single scene. Only a few of these are destined for the Foundry, others were for other scenes.

Image
Image

Putting the scene into the UDK took over a day because I somehow fucked up the scene file and almost lost all of my work on the foundry, and many hours were spent trying to recover it. I had to manually rig the tiling rate and smoothing on all 200 of the Foundry's objects, plus the props, then get them working in the UDK. These are my very first screenshots I took of it.

Image

At this stage I began researching materials, and quickly realized that lava materials were extremely complicated and would require considerable time to get working right. I stopped my work with the scene and concluded my time with the UDK, since any further time spent now would exponentially increase for no gain. Total real-time work spent on this would have been like a week, if you include all the fucking around in the UDK and busting the scene and such. Scenes are easy to make, and this thing is super low poly still. However, getting it looking good is no trivial matter, and not worth my time.

I will now close this thread as no further additions will be made to it. Should work ever continue, I will re-open it.
Image~[Gameproc]~Image
Warning: dialogue contains politically incorrect content. Viewer rearsore may occur.

User avatar

Topic author
IskatuMesk
Xel'naga World Shaper
Xel'naga World Shaper
Posts: 8788
Joined: Sat Feb 07, 2009 1:40 pm
Location: M͈̙̞͍͞ͅE̹H̨͇̰͈͕͇̫Ì̩̳CO̼̩̤͖͘
Contact:
Status: Offline

Re: The Grand Experiment Thread

Postby IskatuMesk » Mon Apr 21, 2014 9:08 pm

I am unlocking this thread so it may follow my Starcraft 2 experiments.

For start, here's a Lineage 2 voidhound with a test pet fireball. I made the spec/normal with MindTex.

Image

Tera character model. I did her arrow mostly by hand, only the mesh is from the game. I used it to learn GL animations, animated opacity, animated visibility, glow, and other such shit. It has a projectile and hit effect ingame as well; if I get far enough to make a proof of concept demo, I'll show them.

The problem with tera models lays in that 90% of their animations move the mesh to take advantage of the Unreal skeletal mesh actor system. Can't do that in sc2. You could freeze them in place but the anims will look really weird. So I'll need stuff from other games, or somehow need to re-animate those animations, to get places.

Image

The hair's got crazy moire going on from the high detail. I'm an idiot, so I don't know how to simplify the texture without destroying it. I did have to recolor the armor to match the bow, though. That was surprisingly difficult.
Image~[Gameproc]~Image
Warning: dialogue contains politically incorrect content. Viewer rearsore may occur.


Return to “The Trampletorium”

Who is online

Users browsing this forum: No registered users and 1 guest