The End

I have made an animated gif of a walk through of my level for you internet users. It is on a loop and prompts you when it restarts.

Each frame is 4.5seconds long (may take a little while to load) Well.. I am finished.. finally finished! Its been a long journey but I think its been worth it, its also been one of the hardest things iv done and have dedicated a lot of my time towards it. I have learnt an incredible amount this year and understand UDK fairly well. Im please with what I have accomplished, I am within my limit of 80k triangles (measures as 1 asset), my assets comming to a total number of over 120 and with a few thousand more to spare! I have also been diligent with my Texture budget but all in all what I have worked to has been fine all along totally at 111mb / 120.

The point of this project was to give me a scope of everything in a 3d environment artists role, ranging from small props, human buildings, terrain creation, foliage and trees, animals, animation, day and night cycles, reflective rivers and their materials, UDK Material editing and also particle effects. Its been well worth the ride and have produced something pretty good. There is always the argument of 'I could have done more' which I would be lieing in saying I dont feel that way now but I think that is always the case. The point is what I have has been on purpose and have spent a lot of time particularly in the placement of my assets to make it visually interesting. I like to think most areas in my level are interesting to look at as each have their own values and characteristics. I am also glad I have done a night/day cycle to be honest. For the majority of my project I have had it set as midday just for something to work against but also for the ferns.. while miscellaneous most Ferns grow under the shade of trees during midday and I wanted to keep that and it makes it look pretty realistic.

I added spots of colour with my flowers to keep the eye busy, broken up areas with ivy and brambles. Spread my decals around and have quite an interesting layout going out.

If im honest the end result is NOTHING to what I had in mind, I had this image in my head before I started on production but I found it to change and evolve as I was doing it. I also found it incredibly hard to concept what was in my head also, which I did the concepting quite early in the project, which helped, even they were not what I had in mind. Anyway to be honest what i had in my head at that time was a lot smaller and probably not as interesting. While my fmp is quite big in its area compared to my peers - any less than that and I dont think it would have worked. Its a trickey one as occasionally I felt that I had too much spread around rather then small and compact, though on the otherhand having things compacted is probably not the best way to go about it.

So yes Visually im really pleased with my work, even the chickens I worked on. When I first imported them into UDK I was simply amazed at how realistic they moved! While small, crit from other people have been that they have noticed them straight away and have appreciated them. I textured my chicken and animals weeks and weeks ago, it was one of the first things I did in fact, and the textures helps also. In many games the chickens seem to be the general brown colour but I wanted something a bit different. You dont see that many black chickens in games and I thought it was an interesting touch. To be fair I could have made instances for different colours but I just ran out of time for something pretty trivial. I was working right until the end basically and some things I had to cut out - the eels for instant.

So yes onto my personal crit of the level now, It was ambitious and I was constantly concerned if there was enough foliage placed around, particularly around the market area. For weeks it looked bare to me and after studying it for around 8 weeks I quickly got used to it and became less impressed the more I seen it. I asked around and decided to leave it at this point.. I COULD have added more but it would have made it a bit more crowded.. too crowded and I wanted to distinguish between what people had been walking on and what not. Im still in 2 minds about that area but.. Its done now and what I have is alright. The other thing.. my MAJOR thing.. was the frame rate which unfortunatly wasn't brilliant. according to my meter I was getting an average of 20fps and made some areas jerky to walk. Particulary the start where you turn to the left where it paused for a few miliseconds and everytime that position was looked upon it happened. As I mentioned I made LODS of all my foliage as they got into the distance and also edited the materials but nothing was happening.. I also looked on forums and I couldnt find anything I could do about it.. It was quite a lost cause and its my MAJOR concern.. In fact I handed my work in today but I felt so wound up about it I looked to see if I could do something about it. 1 theory was to make an auto executable file for my level which plays on its own.. At first yes I noticed a difference but not enough.. there were a few values I could play with but these were core UDK technical things which I dont fully understand. (and I think my role as an artist I cant really know) im still a novice at UDK and all things level design. I can only say I understand the fundamentals which I have practiced throughout this whole project but im still inexperienced in the technical field. Anyway apart from the frame rate issues Im pretty happy with what I have accomplished. I have some nice views all around. I did make an executable file (after I handed in however..) and after tinkering which I have done the rest of the day managed to get the frame rate a little higher which Ill defiantly use for the degree show. I have not changed my level at all but the actual UDK engine itself! Im.. pretty annoyed at myself for this but I got so stressed the last week being paranoid about 'does it look real, is this alright?' that I was concentrating on that. The visual elements were the key to this level and I did have frame rate in the back of my mind for quite a while.. but I have to remember my project was pretty big, the area was big and it needed filled. I cant say I would change anything.. as for 1 I physically cant due to deadlines. The Frame rate was my cause for concern but It wasnt really in my expertise. While I still feel frustrated by that there isnt really much I can do now. I will be presenting to the tutors later on today and discuss the issues I have had.. Anyway yes so in conclusion this project has been long and I have put a lot of work into it, I have learnt a lot of new skills in taking on this challenge, I feel more confident in what I can achieve now, I have produced something that im quite pleased with, but my main crit was the frame rate.. Its been a good project all in all though and im glad I did it, doing something else I think I would have gotten bored very very quickly. If I hadnt taken that trip to Norfolk Im sure it would be a very different project sitting here - likely one which I wouldnt have been so passionate about. So there you have it, my Riverside Settlement FMP.. Its been emotional..

Continued

Once again.. Have neglected blogging for.. 2 weeks this time! ehem.. Right These are the models I have completed since then and make the final elements to my asset list

Yes I had a lot of issues with some of my more complex animations, the common error being that it was rigged properly in max and worked perfectly fine.. but exported as an FBX was where the issues arose where in UDK most of the rigs had gone a bit mental. For example..

The above showing how it displayed in max and below how it imported into UDK

And this was quite a mild example, I did find online that it was something to do with the heirachy within 3dsmax.. though were set up correctly. To be perfectly honest I dont really understand how I managed to fix some of them, playing around with values of the bone references and tickboxes.. This simply wasnt documented anywhere so was at a loss. Eventually I got them to work though but for example the Courmorant, the issue was so bad I eventually had to delete 1 wing and have it a seperate object.. It was crazy and I still dont understand it even though iv worked on it. Yes I got my pub textures from other textures which I had created (as my limit was rapidly getting nearer and nearer) I also made a few more decals which i placed in the level quite a lot, mainly due to the fact that the UDK terrain I found had a glitch also. I would be painting the textures onto it normally when all of a sudden a huge rainbow effect appeared as seen here.

Once again I dived to Google trying to troubleshoot and apparently it seems to happen when too many 'complex' textures are in place. I tried many ways to get it to work properly but it simply was not having it so I ended up making a few decals and spreading them around. These were the decals I used:

Among these things the rest of what I had to do involved foliage placement and LOD. I found a few framerate issues in the level so decided to place quite a few more LODS around and also made instances of a few models without the foliage movement effect. Though better still the Frames per second remained a little off and I havent really been able to solve it since. I think I did the right thing with them LOD's and material instances but this is quite a technical problem - something I dont really understand. Anyway after a lot of tinkering, placing of animations into the level, playing with the lighting and fog volumes my project came to a close And to be honest Im quite happy with it. So yes I was working late into nights the past few days, it was a lot of tinkering around and dealing with issues. Eventually though it came to the point where I didnt think I could do anymore, my frame rate still seems a bit shifty.

Major Update

Right... Apologies for my lack of blog entries recently as I have been extreamly busy with my Fmp trying to get things done..
My last blog entry was my working on high poly rock creation (that seems like weeks ago now!) SUfficed to say I figured out a way to make a quick low poly version. So much though it was laughable. As I was using the turbosmooth to get the distortions in the rock face.. Simply setting their values to 0 made a lower version of about 300 triangles, As I wanted the shape I left it at that but I dare say I could lower even more.
Then after that it was whittling down my list of 124 odd assets to be modeled and textured.
Here are all my models; Counting on screen it wont add up to 124 as Iv not taken into account my animals (which ill have to export with animations later to not confuse UDK.. and myself..) and a few other misc assets which need to be reexported which will be done later.




As im right in my Fmp now I didnt document everything which I did, basically the foliage used the same techniques as previous with the high to low bake onto alphas. The rocks were all done the same way with some rock textures I taken from Bradgate. Other models were done from my reference of Norfolk. The Ivy (well the leaves) were referenced on a walk around my area. The piers were based on assassins creed, textures reference from my Norfolk snaps.

I used exactly the same techniques used that I have mentioned previous in my blog and this is them.

I finished these assets a week or so ago and since then have been reshaping my terrain quite a lot and placing a few of my assets around, this is not in anyway finished yet but these last couple of weeks im confident I can get it done.

NOTE: THE SKYDOME SEEN HERE IS NOT MINE AND IS ONLY PLACED FOR DEMONSTRATION PURPOSES
NOTE2: THE LIGHTING WILL BE A NIGHT/DAY CYCLE AND WHAT THE ACTUAL DIRECTION IT IS LIKELY TO CHANGE





























OOOOOOOOOOOOOOOOOKAY So there you have it, quite a lot gone on since my last post and again as I have to now finish working on it there may not be another post for a few days..

Some areas seen above are to be finished and placed properly,
Now ill make a list of what else I think needs doing;

-Sort out foliage near start of level,
-place some ivy at start of level
-Make Clumps of foliage around fence posts,
-Make a few crack and indenting decals,
-On the street section dirty and place more assets around, barrels, waterbuts outside, pile of logs
-Make some shorter grass to scruff up the market a little more
-Have some ivy climbing more of the beams, and cover harsh edges of things with grass and other foliage
-Grass Undercarts
-Straw Decals
-Twig Decals
-Ruts in the ground for cart decals,
-Make an instance of the market mesh and adjust colours a bit to make each more unique
-Make instances of other materials and add a more earthy element to it with Gradiants/Multiply Nodes
-Grass around edges of paths
-Make some bramble bushes and spread around level
-Make a culdren tripod for my little hidden area
-Make the wheat more golden looking
-Place rocks underthe water sections
-Muddy up the boat area
-Finish terrain repainting
-Adjust pivot poitns for ropearoundwood
-Lighten the shadows as they are too harsh
-Make my own skydome (went to Beacon Hill for another panorama which ill have to edit
-and make a few material instances of the tree leaves to vary slightly.

SO.. Lots to do but Im confident Ill be able to get this done... Until Next time..

Rocks P1 Mesh Construction.

Right well I did have a look at rocks a few days ago but that was just before the presentation and I thought it would take a long time. I did some scouting around and found quite a quick and easy tutorial of rock producing albeit high poly but I thought I could use that and then use render to texture to get the normals from it, then can overlay my own textures which Ill get from Bradgate Park.

The tutorial was from a guy named Sasha Henrich http://saschahenrichs.blogspot.co.uk/ under the heading of Procedural rock creation.

I also found a few other similar tutorials on youtube doing this same work flow, so I will just do a quick test now to see how easy it is to do.

SO;
Create a box with a few segments,

Add a spherify modifier..

Add a turbosmooth modifier with a few iterations..


Next making a couple of displacement maps, the first using the 'cellular' built in texture, dragged to the material editor..

and doing a similar thing but with the built in 'noise' texture.

A couple of Noise modifiers to adjust how rough the surface is


Then a FFD 2x2x2 modifier to adjust the shape.


It really is that quick and easy to do and with this technique it would be quite easy to churn out a few rock models for my level. Now I do need reference for stone textures.. In a few days ill be going to Bradgate Park to get a few more snaps of these as well as more textures for my terrain so Ill leave it at this for the moment.
There will be a time where Ill use Mudbox which I have never used before to create seamless textures, I bought the education suite in the first year and its never been used. Having a quick look through Sasha's tutorial it seems fairly easy to use and im kicking myself for have never used it before.. Oh well!
Seems straight forward enough and I should be able to get some decent models produced - what I have done so far is pretty high poly stuff but then I would be making a lower poly mesh of my high poly and render-to-texture a few maps projecting onto the low poly. Im quite looking forward to this but without suitable textures it would be a bit pointless to continue until im at that stage.

Post Presentation

Alright so a couple of days ago we had a presentation of our work and mine went pretty well, what was said I had an inkling about anyway but it was good to hear the feedback first hand as this made me realize I was on the right sort of waveline.

First of all Ill post up the images I shown in my presentation here which was basically a run through of my level. The lighting went a little too orange though as I was building the night previously but this shows my most up to date version of my level.





















So yeah That was it, so the conclusion to this presentation was that what i had was good and coming along but the main concern was with the paths looking monotonous and that more effort needed to be made to the terrain - made less hilly and harsh but also to have a few rocks placed around, have more older looking trees in the area as one of the crits was that a lot of the trees looked the same age.
And yes to finish off all my other planned assets, get my day night cycle working, get my animals placed and animating, learn how to do fire for my lanterns for the night section.

A lot to do but im confident ill get it done during these last few weeks.