I Never Want to Look at a Rope Again

After asking several people which route I should take with the rope, a cylinder, parallax occlusion mapping, tessellation and magic, I decided to test out the cylinder.

This was just a really quick bake to see if it would work out, and it held better than I thought it would.

The normals that I got from the bake were ridiculously messy.  I was guessing the waves were from the low poly mesh being too low compared to the high, and the artifacts were from where the cylinders intersect and the cage getting all jumbled up.

So what I decided to to was go in and hand paint over the messy bits.  This is what it look like at the minute, I'll add rope fiber normals after I have done the texture.

The texture was a bit hard to get right at first.  I had to go in and paint red lines over my cleaned up normals to see where the ropes sat.

Well, this is coming along nicely now, I just want it finished and to move onto other assets.  It has taken me a long time to get this far, it's always going to be harder when you first start as you just make so many mistakes.  I'm pretty certain the process will speed up a heck of a lot after I have got my first 3 assets entirely finished.

In regards to the bucket, I'll probably play around with the fibre normals a bit more, get some alpha planes in there for the rope fibers and fill the bucket with some water.  I have been playing around with fluid actors in UDK but I'll have to see if there are any better ways to do it.

Update

Okay so I haven't updated by blog for a while, I've been too busy trying to fix this damn rope!

Well seeing my last post and how my rope looked like a churro, I decided to remodel it.  This time I just made my own splines and path deformed two cylinders to them.  Already, it is looking a lot better, it's just a shame that it's almost 8,000 tris.

Oh and another thing that I had to go in and fix, I had to chamfer all the edges at the top of the bucket as when I brought it into UDK with a point light I could see the that the normals made it look too hard angled and nasty.  So the tri count has gone up a bit again...

Time for an Update!

Okay so I have been messing around with the diffuse on a couple assets, here's me trying to do minor adjustments with the bucket.

This is currently what the bucket looks like with a flat light in UDK.  I've yet to texture the rope. I'm unsure whether I have made the water leaks too dark.  But it's something I can really easily change down the line.

One thing that is annoying me about the bucket it the wood normals, they still look flat.  I think I am going to have to do some more tweaking.

Hmm... rope texturing has hit a wall.  I slapped a quick texture on there to see how the rope twisted.  I've noticed that there is no way this would hold up as a rope, the stands hardly intertwine.  This means I'll have to go back and change the rope geometry, most likely start from scratch!  Just as well I have a lot of incremental saves.

Texturing

Okay so I kind of finished the diffuse on the book, the only issue I'm having it is the saturation of the pages.  I won't be able to see if the colour is too strong until I get it into my lit scene.

I moved onto the bucket and I'm finding it hard to get the rivets looking good.  I might go into max and do a highpoly for them, would make it a whole lot easier.

I really need to get into my UDK scene and put up some lighting so I can check the textures, I will save me a lot of work in the long run!

Texturing

Okay so I have unwrapped my three assets that I plan to carry the whole way though, now I'm doing some texturing.  It's been quite a while!

I'm just playing around with the book at the minute, the image is in UDK with a simple light.  I've been working on it around 2-3hours, it's taking shape really nicely.

I Did it!

I spent an awful long time messing about with the rope...  a seriously awful time.  Having no luck with loft and many tutorials I went and spoke with a friend who game me a few ways that he would go about it.

I gave those a try and none of them were really turning out how I wanted.  They would only look nice when the tri count was through the roof.  I then had a little bright idea of how to model the strands...

I got it to work! And the tri count was... reasonably high.  But looking at games like Shadowfall coming out, and the fact that in my flythtough by camera is going to sweep right past this rope, it didn't bother me too much.

Okay, so here is what I did.

1) 6 sided cylinder
    44 segments
    553mm tall

2) Connect the edges
    -56 offset

3) Connect again
    Drag verts in

4) Twist modifier of 900 odd

5) Turbosmooth

6) Remove every other edge
    BINGO!

Put on a path deform and there we go, a nice organic looking rope.  I'll have to mess around with it a little but the basics are there.  This rope in the bucket is 1,840 tris... yeah, it's high! But again, it's fine as long as I can justify it.  I am going to look at it to see if I can knock it down some more.  I did try this method with lower sided cylinders but it didn't work.

Quite pleased with this though!  I've just got to figure out the unwrap and how to make it tileable, unwrapping it before twisting it may be the answer.

Update

It's been a while since I've posted anything up, so I thought I'd just put up what I'm doing right now.

I've been modelling a few assets on the desk, and I have moved onto doing three assets to a completed level.  So I can get an idea of what exactly I am going for.  I'm going to do this for the stool, a book and the bucket.

I've finished the low poly of the bucket, and I have been playing around with the rope and how to model knots, this may take some time to get right.  After I have this down I'll move onto the high poly.

Laggy 3ds Max 2014

I bit the bullet and updated 3ds Max, and so far, I'm not too pleased with the changeover.  It's slowing me down as it is so laggy.  I've installed the most recent service pack but it's still not as smooth as it used to be.  Hopefully they will bring out another update soon.

I have not been working much on my FYP project for the past few days, two days of eyestrain coupled with other projects I haven't had the time for it.  I'm back on it now and I am in the process of modelling a small area of my scene, a table, pestle and mortar and a book.  I will fully complete these assets to get an idea of timescale and I will use them as a basis for all my other assets quality and texturing.

I will have to drop back into my schedule and change it, as I realised you want to be modelling and texturing together, to break up the monotony and so I don't get too sick of one or the other.

Research

My supervisor for this project told me to create a breakdown of the assets.  So have a sketch or image of the asset with textures, reference and a description.  I did this with three assets and stopped as I realised just how time consuming it was, I spent the entire day on three and I have seventy eight odd assets in my scene.  It was not a time efficient way of breaking my scene down.

So what I did was split my scene into textures, woods, metals, glass, fabrics, stone, plaster and plants.  This way I can get a general idea of what textures I am going for.  I also found when looking at textures that I will have to go take some pictures of my own, so doing this made it so I was able to move on with my work without getting too bogged down with the details too early on.

I will of course break down each asset as I am making them, here's a quick example at what I have been looking at doing;
Sketch/photo of the asset
Textures
Reference
Movie/Concept book reference
Description

This is by no means how I will present it, this is just messing with layouts.  Until I have decided on font and a theme I will just use plain text, black and white lines and backgrounds.

New Reference

I found some new videos of the interior of Radagast's house.  There is very sparse reference for what the environment is in the movie.  All I really have is a few seconds of footage.  On this new video I realised the ceiling is a lot lower than I had previously thought, so I went back and played around with it.

It took me a while to decide on a height, as the taller the room, the more mysterious feeling it has.  By this I mean if I have a flythough and in it you can't see the full ceiling you can only hazard a guess what might be up there, it makes you use your imagination.  On the other hand if the ceiling is lower, then it has a more cosy and compact feeling, which is what Radagast's house is kind of like.  I didn't want to have an extreme of either, I didn't want problems with getting my camera around if the environment was too compact, and I didn't want it too spaced out and not engaging.  So I went in between the two, this means I will have to mess around a little more with the position of the rafters but I would rather have realised this mistake now than further down the line.

Blockout Complete!

I think I can finally say that my block out is complete, I know where everything is going and I know where assets are instanced.  I've out a lot less emphasis on my back room as it is just a background to the main room.  There will be a lot of instanced assets in there, with only the sink, table, benches and shelves being anything dramatically new.

Completing the Ceiling

I have been working on getting the rest of the ceiling blocked out.  I had some issues with some of the rafters in the back room.  As you can see from the top image, the rafters look nicely spaced.  But as you can see on the bottom image, they look a bit odd.  I had to move these around a lot to get to this point, I decided that the camera angle in the top image is most likely what will be used in my flythrough.  So it was more important to get this looking right rather than the other camera angle.

Changes

I decided that the circular table on the left was not working, so I replaced it with a table with a drying/hanging rack.  I think this works better as it does not draw my eye as quickly as the circular table.

I also started putting some more whitebox assets into the back room.  I don't want anything too complex in here as this room is sort of a background piece, it's just to make the main room look good.

Cracked It!

 Argh!  It is surprisingly hard for me to get the ceiling right.  A few attempts later...

Much better, I now have the right number of rafters and they are in the right position.  All that I have to do now is tweak them by distressing and rotating them.  I might also make them a little chunkier, I'll have to experiment with dimensions.

Rafters

I went over to Photoshop and drew in some new rafters.  I then went and modelled them in to see what the proportions looked like.

I could instantly see that I had far too many rafters, and the gable wall leans too far into the centre of the building.  So I will go back into Photoshop make the needed changes and drop back into 3ds Max to remodel.

What a Mess

I decided to further block out my ceiling.  It didn't go all too well, I brought my model into UDK and the ceiling looked a terrible mess.  I might take a screenshot of my environment and bring it into Photoshop, then draw over the top where I think the beams should go.  Or maybe I will start with the main rafters and work my way to the smaller ones, after which I will put in the purlins.

Stick to the Schedule!

I spent a while putting together my schedule for the project in PS.  I wanted to jazz it up a bit so I can stick it on my wall.  I knew this wasn't wasted time as I will have to present it well when I hand my project in anyway.  I'll most likely go back and make some changes to the layout at the end of my project, but it's enough for now.

It's a very brief outline of what should be done when.  I didn't want to go into the specifics and end up getting all tangled and confused in it.  As long as I have these milestones then I have something to stick to.

Ceiling

I've been messing around with the ceiling and trying to get the structure correct.  I found it quite hard to wrap my head around how it would work.  I played around with the height and proportions for a surprisingly long time.

When I was happy with the general shape I then started putting in some beams to get a better idea of what works and what doesn't.