My God, it's finally over... or Evaluation

So it's finally at an end. Now, I'd be lying if I said I was sorry to see it end, but it wasn't all bad in the end and I'll get to that in a bit. First however, I should really start by going back to my project brief and seeing whether I answered it in it's entirety. 

A rough overview of my brief was to produce a small to medium sized scene based around japanese alleyways. But that thee would be a greater focus on the VFX content and that this scene would be in many ways a backdrop to it. I was also to work on a secondary project to produce a VFX show reel that demonstrated various abilities and techniques.

I guess that my response to the brief is a game of two halves. On the one hand, I did manage to answer my brief correctly to produce a small alleyway scene. However, not only did I fail to manage to produce a full VFX reel, but I feel that the scene was the dominant focus. I suppose that this is more a case of me feeling like I have failed more than I actually have, although that would still be quite a bit.

If I view it a little more objectively then I feel a lot more confident about my outcome. Lets start with the scene. I am actually quite happy with how the scene turned out, there was a stage during the project where I felt it was horrifically dull and boring and that it would turn out a complete disaster. The overall scene has a very nice atmosphere and I, personally, really like the aesthetic and lighting that I've achieved. I think it was very important for me to have a strong colour palette so that it wasn't so dull and murky, it was important that I veered away from a grey/brown colour scheme. I also feel that my modelling and texturing skills have greatly improved whilst undertaking this project though there are still many areas for improvement and especially so with light mapping. Judging by the amount of light map issues and errors I've encountered I'd say that there's still a lot I don't understand. 

As for the visual FX side of the scene, I would say that I'm relatively happy with it. I am very pleased with how my rain shader turned out. I think that it has a high level of fidelity and that it works based upon the objects normals is great as it means that I can use the same material instance on each asset whilst cutting down on draw calls. Unfortunately though it is quite an expensive material so ideally I shouldn't have used it on almost everything. I think that it also works well in tandem with the rain particle system. The rain emitter actually gave me a lot of trouble as I found it quite difficult to get a cheap emitter to have a heavy fall to go with the scene, it was a lot of tweaking for that. Overall the other effects are okay I guess, not much to be said except for the cascading water. I feel that looking at the scene now that I should probably reign them in a bit as they're showing a much heavier fall of rain than is perhaps in the scene. Maybe switch them out for some just dripping water or a thin flow.

Before I proceed onto the other sections of my project or talk about the problems I faced and what I would do differently, here are a couple of my final renders and a short flythrough of the scene.

So as you can see it all in motion the rain shader really works it magic. But that wasn't the extent of the VFX that I worked on over the course of the project. As stated prior, I had tasked myself with producing a show reel. Unfortunately, I failed on this front as I didn't manage to produce one in time. I will definitely have one together for degree show so it's not a total bust, it just means that I failed to match one of my requirements for this project. 

I think the reason I wasn't able to produce one in time is that I underestimated how long it would take me to work on something as well as how much I would have to learn. Therefore my output was a lot lower than I initially thought it would be. But below are some of my renders of the projects that I was working on. The gas station explosion is the one that took up most of my time as there was so much that I had to rework or tweak or cut or add etc. It just took me a lot longer than I thought it would, but I think that it has the potential to be a major part of my show reel with a bit of love.

Unfortunately it's not more as there was so much that I had to learn for each project as well as working on the scene. New techniques or principles, new software and new pipelines all delayed my workflow. This was a major hindrance to my time management on this project. My complete underestimation of how much I would need to learn and how long things would take me to do were so costly. But there was also another big hindrance in that I found it troublesome to balance both sides of the project. By this I mean, one I was working on one side, be it the scene or an FX project, I would feel guilty for not working on the other and vice versa and this actually drastically slowed down work on either. I always felt my time would be better spent on the other thing. 

Because of this I feel that my scene suffered greatly as well as my various FX projects. Which leads me to how I would approach the project again. I'll split this into two; how I would approach the alleyway scene and how I would approach the overall project again. First off, the scene. I think that whilst it may look nice or what have you, it lacks interest. At it's core it is just a corridor. I think that it could do with some kind of expansion as well as a kind of focal point. I attempted to do this early on in the project by having that side alleyway, however, due to timing I really brought in the scope of it and ended up cutting it out. If I were to have dedicated my time fully to this project I would have kept that in as well as have the main street veer off round a corner or something to reveal either something like a refuse area or maybe something like having a police car parked with it's lights flashing would have given that focus to the scene. One idea that I did have was that I could have one of the signs either short out or be struck by lightning and fall the the floor as a tiggered cinematic in UDK that would block the player off though I deemed that idea unreasonable as it would have taken me far too long to research into destructive meshes in FX pipelines in the time limit I had. 

And if I were to approach the whole project again I would do so very much differently. Most importantly is that I would drop the scene altogether. I felt that having my time be divided like this ended up affecting both parts of my project and that it detrimentally affect my output. If I had focused solely on producing good FX as well as a couple of larger FX based sequences in matinee it would have benefitted me tremendously. 

I feel like now I am having to play catch up for the time that I either messed up or missed out on for my FX over the next couple of months in order to build a strong enough portfolio/show reel to get my foot in the door. Although I wouldn't describe it as a complete bust as having the time to look back over the project I can definitely see how far I have come and how much I have learnt about the practices and principles of being a good FX artist and that with this project under my belt I feel confident in my ability to secure myself either an FX or junior FX artist position.   

Flythrough

My first run at my flythrough. Just seeing the timing and speed of it all really so that once I'm happy enough with it I'll build the lighting at production and then render it out at a higher quality.

Just some really quick updates whilst I work away on the flythrough and getting my final renders in order. This is just a quick render of how it all looks finished properly with all the correct post processing and stuff. I know it's not dramatically different from the previous render but it's as it stands now I guess.

And now for something completely different. Been going back to my other scenes for a bit. Recently been trying to get the plume of fire and smoke after the second explosion to look better. The one on the left is the original and the one on the right is the reworked one. It's a little on the expensive side so I'll definitely have to strip it back. It's a shame because I really like how it turned out using both faked lit particles and macro UVs. Unfortunately what pushes it completely over the top is the use of UV distortion so that'll have to be the first thing to come out. But it's fun to experiment with these things.

And then here's how it looks with the reworked plume. This still needs so much work. When I'm done with this I think I'm going to have to really practice my fire and explosion effects a lot. They're just so dull and lack the visual impact that they really should, especially my explosions. But the next couple of days I'm going to be working on some new flipbooks for the fire to hopefully break up the repetition. And other than really working and reworking the explosions, I'll be adding in a glass shattering effect from the store after the initial explosion. Just to try and tie it into the environment a bit more.

Final Render Ideas

So as promised, just took a quick look around my scene to see if I could find any interesting angles. Unfortunately, whilst a simple street scene is great for it's ease to produce it becomes very difficult to find interesting angles for renders. Below are just some of the angles I was looking at. But first, just to show the difference between post processing and the use of colour look up tables. Such a useful feature.

The latter is with the colour look up table applied. I really tried to sap some of the blue out of the scene as I was gradually going more and more overboard with it to the point where it had completely washed the whole scene out. Below is a selection of angles I was looking at.

Some are obviously stronger than others. But as I've mentioned prior, being such a simple scene I'm almost restricted to simple single point perspective shots. I really like the portrait shot, but if I'm to commit to it I'll have to play around with the field of view angle and drastically reign in the bloom as it completely overpowers the whole shot. Other than that I really like the last one and the couple in the middle. Once I've decided on which I like most and which I think are strongest it'll then just be a case of finesseing it before taking my final renders.

More Rain

Just a quick update. I've fixed a few things that I'd forgotten about but most importantly I finally found a good method for making better rain. So it looks much more coherent now. Unfortunately my PC is lagging the level a fair old bit at the moment so it wasn't until I uploaded this video that I realised that the misty elements I have in the new emitter are moving way too fast, but that's such an easy fix it doesn't bare talking about.

Now I'm onto the last little bits. Most of my day will be tweaking the post processing and trying to find a nice angle or two for the final screenshots. So you should probably expect an update either later today or tomorrow of several different angles whilst I deliberate over which looks best.

Nearly There

So I'm almost at the end of this project. Finally, it's been such a long project. But I've added in the cloth and lantern animations, and whilst they're not perfect by any means, but they do add a lot to the scene by making it feel that bit more viable. So I thought I'd give you a look at how everything is in level now.

Like I said, I'm almost there. The only things left to do though are more tweaking things and doing some post processing. But it's so close to being done :)

Updates on my other projects

So in between working on my FMP I've also been working on some side projects as you probably know, so I thought I'd just give a quick update on where they're at. 

This is the progress so far on my explosion scene. I've just block out the second explosion kind of just roughing out it's timing about and the scale. Like I said it is just roughing it out so there are some timing issues and the biggest problem right now is trying to figure out how to blend the fire with the explosion without it looking like it's just sitting there unaffected.

And then this is that magic spell thingy that I was working on a while back that I finished a week or two ago but just forgot to upload here I guess.

 

Lantern Animation Test

I think I've now entered the stage where I'm doing final tweaks and adding in the little details now. Yesterday I made a couple of decals for like grunge and splatter for the base of the walls and roll up doors to emulate the build up from people walking around in a damp wet alleyway. I also created some leaking damp decals to put around things like air vent and air conditioner units to replicate the build up underneath such assets. 

But a couple of days ago I made a quick animation for the lanterns. It's only a quick test more than anything to see how they would look in UDK. However I'm having some trouble getting it to work. I don't know what I'm doing wrong as I've never imported an animation before so I'm sure my next couple of days will be looking up tutorials just to get it in, to which I'll probably find I'll need to redo it anyway.

The Actual Level

So I realised that I've put up a lot of progress shots but I've not actually put up anything of it actually in real time yet. So you've not been able to see it all working properly. This is still a WIP so obviously there's stuff that needs tweaking and changing. But here it is anyway.

Watching it back I see that I need to work on the splashes, they pop too much and come up to high and are much too opaque so they're kind of drawing the attention way. I know I've been harping on about it for a long time but there's still not enough rain. The shader and the water running for the awning give off a nice atmosphere but they imply a far heavier rain fall than the particle emitters are showing currently. So after so much tweaking I'm going to have to go back to it because it's just not enough.

Updates and Paintovers

So not updated in a while so I thought I'd post up what I'm working on. Unfortunately there's not a massive change since the last post as it's mostly just tweaks at the moment. But I guess what I've been working on the most is just tweaking the amount of moisture and fog in the air. I also tweaked the steam vents finally.

I just did a quick paintover to of some more things that need to put in to just sell it a bit more. Basically just adding in some water cascading off of edges and make some grungey splashy decals for the walls and stuff. I also need to really get round to making some procedural animations for the hanging cloth and lanterns to make them look like they're blowing in the wind.

And now for something completely different. This is just a progress shot of how I'm getting on with my explosion stuff. The major things I need to fix is basically to make some more flipbooks, even if it's just like two then I can get a better sense of variety to make it look more natural. Other than that are the things that I need to add, like some large embers falling from the roof, and I'll make some glass shattering particle systems that follow the shockwave. And then I need to make the second explosion, which will be a lot of my focus at the moment.

So what's next? I need to get this explosion at least to a stage where all the elements are in so that I can ask for crit. Then I need to balance my time between the scene and other various fx to get the practice.

Rain and Stuff

So I know it's not much but I thought I'd just do a quick update has I've got everything finally in with lightmaps and collisions so I feel like it's going well more than anything. So bits that I've tweaked since my last update; I've worked on the rain emitter a little more so that it has more directionality to it though it still needs more work as I keep going back and forth on the opacity, it's either too distracting or not noticeable enough.

Other stuff is that I've made the fog more dense so that it feels like there's more moisture in the air. Though looking at it now it feels too blue, I should maybe tone down the colour on it. And a very slight thing is that I changed the fabric signs to red as the blue was blending far too much with the sense. Too much blue everywhere. It's only a small thing but I feel it actually does quite a bit. 

So this weekend I'm going to try my hand at some procedural animation for the fabric signs and the lanterns so they blow a little in the wind, give it a more natural feel. That is unless I get distracted by working on fun fx.

More Bits and Pieces

Right, so I spent my morning trying to fix some rendering errors with my ground attack thing to only discover that I couldn't get it to go away so I'm sure it'll be something that I'll have to look into later for when I inevitably have to rerender this out. No idea why it spawns two emitters on top of each other a fraction of a second apart but oh well. 

So this is the AoE type ground attack that I've been working on that I was talking about yesterday. In the first video it's just the normal one but in the second video I was experimenting with staggering the timing for a greater sense of impact. It's supposed to be that there's an initial hit then a brief pause before the main impact explodes outward. I think the idea is sound and would give the effect greater presence but I think I'm off with the timing as it isn't gelling together properly. 

So improvements. Obviously I've mentioned the timing. But I think the overall effect lacks that energy and impact you'd expect from an effect like this. I'll have to readdress the textures and see if I can't get some fast panners in there to maybe try and give it that energy. The other major problem is that I don't like the rocks that shot out. It's supposed to be the displacement of earth from the force of the impact but I think I've over done it. I think it would work a lot better if I tried something more subtle like a large indented crack where just some of the chunks of earth round the outside are sticking out.

Some Quick Updates

So I've been working on a few things recently just between projects. One is like an AoE ground attack and the other is this kind of ice spell. The only thing is that the videos I rendered out for the AoE seem to be having some troubles with spawning emitters on top of each other so I'll have to retry rendering it out tonight and try and fix the problem.

But on with the show I guess. I've had this idea rolling around in my head for a while, and especially after playing DotA I admit. But it was a good opportunity to learn about how ribbons work and how to implement them. So I worked up a really quick concept to just kind of help me visualise it. Below is my concept and a quick render of where it's out.

I apologise for the poor quality of the video but I can't seem to get ribbons to render in UDK, I had the same trouble with the previous gas station project. I'm only really just started with it so it's got a long way to go. A real big problem I see is that it lacks luster, it's not got that great sense of energy that something like this should have. But some things that I've seen that need changing is the centre point. It needs more too it to convey that sense of energy but also I think it's a little too long. The sparkles last too long around the centre, I want them to live longer as they leave a trail but they need to die really quickly at the centre. And the other thing is that I need to change the whispy ribbons that come from the centre, I think it's too expensive to have so many ribbons so I'm thinking I'll make some kind of whispy mesh for those parts.

Some Other Projects

So I've not updated this blog with other projects that I've been working on other than the environment. The idea of this is to have a gas station that gets by a RPG causing an explosion then burning for a bit before it triggers an even larger explosion.

I had to render the second video as the actual render from UDK didn't render out the ribbon data on the RPG so that's why I had to do a screen capture so that's why it's a little worse quality.

Initially I was having a lot of trouble understanding how the flame would expand and grow over time so I just whacked out a quick sketch just to clarify things for me so I'm not so stuck in my head with things.

But I've got some really great feedback about this and now I've got a better idea of what to work on and where to take it. So it's all good :)

Brief Update

So just a brief little update of what I did yesterday. I've put in some rain but I think it's barely noticeable so I need to make it more visible and increase the quantity. I just think that it doesn't gel with the rain shader, the shader shows a much heavier rainfall than the particle emitter.

I also put in some fog to add some atmosphere into the level. I think it compliments the steam and the wet damp feeling well. But I think it might be a bit overpowering right now so I'll probably go in and tweak it. 

The next thing that I feel I need to address is the lighting and colour palette, it's just not popping right now. So I think that I'll take it into photoshop and play around with the contrast and hue/saturation and try and see where I want to take it.

Bit Of An Update

So I realised that I've not updated my blog in a while. Mostly that's because I've been working on other projects the past couple of weeks (that I'll make an update for soon, probably over the weekend) but even so I've made some minor improvements. 

Since you've last seen it, I've got the rain shader up on almost everything (although I'm still working on generating some of the textures for it and then I'll need to finesse it). And I've started to reimport things with their light maps so hopefully it'll start to look a little nicer and then I'll get to work and polishing out the lighting.

So I still have ways to go. I've still got to finish off making light maps for everything and then finish off the rain shader completely and then hopefully I'll have some time to just dive in and make some fx to really help sell the environment.

Pipes Galore

Ah so I've finally got all the pipes into my scene. It took me a lot longer than I had expected, probably due to laziness and apathy more than anything, I just couldn't look at anymore pipes. So here are just a couple of shots showing all the pipes in level.

And this last one is what they look like from ground level. You can't see them all that well to be honest but I think they give the scene this sense of compactness, which is what I liked about these types of alleyways in the first place. And once I've got random bits of steam and running water off them I think they'll add a great deal of atmosphere. 

But the most important thing is that with this comes the last of the assets that need to go in. Now obviously I've only populated the alleyway but the areas that I'll block off will simply be done by me duplicating sections of the level so it won't take me long to fill those out later on in the project. So my task for today is to get the Up Vector working on my rain shader and hopefully I can generate the flipbook ripples for the ground textures and get that in and that should be the rain shader almost finished. 

However I think this is the last I'll look at it for a while as I've been feeling really stagnated in my work. So now that I have everything in level I can afford myself a break and make some big explosions cause at least that's fun :)

Some Little Bits

Just thought I'd shove up some little progress shots. So yesterday I just spent the day doing a few tweaks here and there. I created some new diffuses for the role up doors so that now it doesn't look like the same one repeated over and over. I also made some emissives instead of simplying using the diffuse again, what it was just to see roughly how they'd look. But at least now my bollards don't glow. And I also generated a cube map, I know it's a little early to be doing it but my rain shader looked awful with a sunny sky cube map I'd borrowed from UDK just to see if it was working. 

I also remembered that I'd not shown that I had put all the signs in my scene yet. It's only a sort of temporary placement as they're a little too statically placed. I'll probably move them around and the like today when I finally get round to putting all the pipes in level.

So that's my major task for today, to try and finally get some of the pipes in level. First assets to be finished last to go in. I also might get bored and tweak the steam as it's very WIP at the moment. And who knows, I might even get bored enough that I build the rain particle.

A Little Progress

So I've started working on the lighting a little bit to try and get it closer to what I had envisioned, it's obviously not there yet, not even close. But at least it's a step in the right direction. I've got the overall colour scheme better and I've put in lights to back up emissive light sources but now I feel that they're a little too bright and kind of wash out the scene so I'll have to go back and finesse them. 

I have also made a start on the rain shader. It's only really a base right now but it just helps me to know where I've got to take my scene. I've still got to generate a tiling flipbook normal for the ripples and once I've done that I can then work on the normals more as they seem a little intense. Once I've done all that I can then go and do the work with the up vector stuff so that the shader will show either ripples or running rain water depending on the direction of the normals.

So stuff that I've got to do still; I've still got some assets to place in scene like a lot of the pipe work. I also need to start on some of the FX like steam, I think that will bring in a lot more of the mood and the like. Plus it's fun.

Revisions

So I took the screenshot and did some paintovers to get a better sense atmosphere and mood. It's only a quick thing but it helps me to know where to take it. But I was still having some trouble with the lighting and colour palette.

I went back to my reference to find good images of mood and lighting to help inform my decisions on lighting. As you can see below I've been just using some overlays to get a quick sense of how it will look.

Below is an screenshot taken from Bioshock because I really liked the colour palette and mood. Even though it's not accurate to the reference I've collected I think it's a much nicer tone for my level. 

 I think the lighting and tone in these last two is much nicer than the ones that try to adhere to the reference. I've found this very useful to help understand where I need to take my level now. Hopefully it can get me on the right path now.

Update

I know I've not updated in a while so I thought I'd take some time to just show where the projects at at the minute. I've started to put my assets into the scene now. I think it's definitely been improved by taking a step back and redoing it but I have cut it down by half so that it's not so noticeable that I'm reusing a lot of my assets. Saying that I do need to make some new textures for some of the so it's not so noticeable, especially the roll up doors.

So here it is as of now. I'm not too hot on the lighting and the colour scheme so I have taken a screenshot with it having neutral lighting and I'll do some paintovers playing around with the colour and atmosphere.

New Directions

So I started on placing the assets in level when I realised that as I had been building the level earlier on in the project I had not been thinking about how it would all go together properly. As you can see below the girders that I supposed to line up between shops cut through doors and shop fronts. 

So this means that I'll have to rebuild it from scratch remembering to pay close attention to how I'm actually building it. I might take this chance to decrease the size of it so that it all feels more unique rather than the recycled assets. Hopefully it'll turn out as a better finished piece because of this so I'm okay to redo it.

Assets Finally Done... Hopefully

So I might finally be done with the assets. It's taken me long enough to get through them. Now they're probably not the finished assets that'll be in the final renders as I'm sure there are plenty of errors or things that I'll decide to change later on. But for now here it is. I've just spent the last couple of days finishing off the last couple of assets; signs, lots of signs. (And an air conditioner)

And here is all my assets together. These will all be either modular or repeatable and be used a lot to build up my level. Some assets will have either different diffuses or alter the colours directly in UDK like the roll up door as it is used so often.

Little Update

So I've not updated in a while so I'll just put up some of the stuff I've been working on. It's basically just a few assets I've been working on, that's really all I've been doing really. Below is the stuff I've done this past week (with exception of the awnings, they're all just normals at the moment but they'll be done by the end of the day)

But here is a shot of all the assets I've done so far. Soon I'll be finished with asset production, soon. And once I'm finished with this I'll then start putting everything together in level like an elaborate puzzle. Fun times...