Week 13
Screenshots of Final ProjectToday we had our presentation. Overall we felt it went really well, We presented it in a virtual cave that Tim had made in conjunction with my studio 4 group and was planning on using it for both this and his studio IV temple project (http://temple-experience.blogspot.com/).
Erik, Stuart and the Tutors seemed impressed by the final product and gave us very helpful feed back that we shall try to impliment for the exhibition tomorrow. These included fixing up the ground texture to make it look more real, putting in roads, and fixing up some of the trees and surroundings etc. Dr Prangnell is coming out tomorrow also to see the final product. There may also be others from the Archaeology faculty coming as well.
We both really enjoyed working on the Paradise project. We found it a very challenging yet extremely intersting project to work on. Dr Prangnell was great to have as a client, he was very helpful and provided us with all the resources we needed to complete the project. We both also worked extremely well together and felt that we made a really productive and effective team.
Weeks 11 & 12
Textured scene with working triggers and information pointsOver the last two weeks we worked a lot on finishing off our project.
Ted finally got the trigger system worked out and Dr Prangnell also supplied us with the information for the information points, so a lot oflast week Ted spent setting up the triggers and information points.
In both week 11 and 12 we met with Erik to show him our progress as well and get a critique on our work so far. He gave us a lot of good ideas on places we could take it before the final presentation in week 13. He also supplied us with a lot of information from the Quest forums and online help to help us overcome some of the problems that we have been encountering.
Tim has been continuing working on the model and texturing side of things, He now has all of high quality models in quest and working. He spent many many hours sorting through all of them and flipping the normals back to the way they should be.
He has now also finished re-applying all the textures and adjusting them appropriately, and we are now just applying the finishing touches to the scene. Ted has also added background bush sounds to the scene, making the user experience that much greater and effective.
We are very pleased with how the project is looking and we look forward to presenting it both on Monday and also in the exhibition. Dr Prangnell is coming out for the exhibition and said he can’t wait to see the final product.
Texturing in Progress
Week 10
This week hailed some good and some bad news for us.
The good news is Ted has worked out how to use the nature painter. At first he said it was a bit tricky. It took him a while to test out the logic flow of using nature painter. But after that, it was just as easy as “painting” an entire area with the tree you modelled or in the library by selecting and mouse-clicking.
The bad news is that, as we feared, none of the textures were kept during the transfer process, and unfortunately everything needs to be retextured within quest. Fortunatly though, the models did keep the UV coordinates, so that should make retexturing a lot easier. Tim is taking charge of texturing from this point on. So hopefully, by next week we will have detail-textured buildings and environmental surroundings ready together and then Ted shall proceed to work on sounds.
We are still trying to sort out the trigger system. We have a basic one working and are getting very close to having the system that we want, however there are still a few bugs to work out.
We also went and met with Dr Prangnell this week and showed him our prototype. He was extremely impressed with it and was amazed how he was able to virtually walk in and around the township. He told us that he had had no idea of how powerful this would be, and he said he can’t wait to see the finished product. He also apologised greatly for not returning our emails and phone calls. He said he had been away on work and then had come down really sick, so he hadn’t been into his office for the past few weeks and only just got our messages last week.
Week 9
This week had our prototype presentation. The actual presentation didn’t actually go that badly, we did, however run into some major problems in the days leading up to it.
The first major problem that we ran into was when we when we first tried importing our models from cinema 4D into Quest3D. Apparently Quest is able to import a number of different 3D file formats including 3DS, LWO, DXF (strangely it does not import OBJ files) and many more, however we discovered that this is not the case. After many attempts at trying to import our models into Quest as a 3DS file, without success. After looking at some forums we realised that we needed to use quest’s native file format (.X). Unfortunately, even though there is a .X exporter available for Cinema 4D, every time we tried exporting from cinema 4D, the program would continually crash. Finally we discovered that the only way to effectively export from Cinema 4D to Quest3D was to export as an OBJ and then import that into 3D Studio Max (3DS files don’t seem to export properly from C4D either). We then exported the model again from 3D Studio Max as a .X into Quest.
The second big problem that we ran into was that during the process of exporting from Cinema 4D into 3D Studio Max, many of the normals were flipped in random places. This caused great time consuming problems as we were exporting large scenes at a time and we had to manually search through the whole scene to find these flipped normals and manually flip them ourselves. To start with we did this in 3D Studio Max, but later we discovered that you could actually do this in Quest itself, and this made it a lot easier.
Our final problem occurred Monday morning just before the presentation, when we opened up the project after working on it all weekend, only to discover that during the save process the night before, that the file had corrupted, and many of the models were missing from the scene. Luckily we still had an older version of it saved, and we were able to make some modifications quickly to that and show that for the presentation, although it was still not quite as good as the one we had intended to show. This was unfortunate, but we still felt the presentation went alright.
Screenshot of original prototype before it was corrupted
Week 7 & 8
Well we are finally getting the hang of Quest. Although the process of learning the program is a constant journey(it was quite unintuitive when we first attempted quest), we seem to actually be getting places with it now. We should have it fairly well worked out before the prototype. It seems to involve a lot of knowledge of scripting so we may have to brush up on our coding skills since we havn’t been coding for 2 years now.
Ted also went through a number of tutorials on the web to get the basic idea of how to trigger an event in Quest3D. For our project, we need to have trigger points in front of certain buildings to enable the text box to pop up. Ted realized that, to do this, we need to have the most up-dated position values to be calculated in order to determine the distance between the trigger point and the walkthrough camera. However, instead of setting the logical flow to calculate the current position there is collision template already available to use. We just need to learn how to modify this and add new logical arguments of our own into the flow chart .
Meanwhile, Tim is still modifying 3D models so they are easier use within quest. Initially he had created each element of the models as separate objects arranged hierarchically under the name of the model (e.g Courthouse or School of Arts), however during the transfer process the hierarchy system was lost and we ended with a massive amount of really disorganised objects. In order to solve this problem, I had to go back through all of the models that I had created and find any objects that could be grouped together and merge them together into 1 object. (e.g. Stairs, Pylons, Frames etc.) This is a very lengthy process and it may take tim even the whole next couple of weeks to complete it. In the meantime he has made some low poly quality models for Ted to use within quest to create the scene. Then it should just be a simple process of then replacing the low quality models with the high quality ones. We are hoping that Quest will keep the textures when we transfer the models into it, however I have a feeling that it will. I'm pretty sure that even if it does lose the textures during the transfer, It should at least keep the UV mapping.
We still have lots to do but things seem to be coming together and we are feeling confident that we will have it all sorted for the prototype.
We have been trying to get hold of Dr Prangnell, but to no avail. Tim had tried calling him and emailing him on numerous occasions, but he hasn't gotten back to us. Its been about 2 weeks since we last spoke to him and although he did say he would be away for a week (during week 6), he should be back by now (week 8). We'll try him again on monday and if we cant get through to him, we shall call the front office and see whats going on.
Week 6
As I said last week, Ted went to Sydney to take some exams for an external course he was taking. Because of the difficulties that we had run into last week with Quest 3D, while he was away I decided to investigate other engines. I played around with a number of different engines Including the Unreal 2 engine, the Torque engine as well as many others. After testing all the engines for ease of usability, I decided we should use the Unreal 2 engine as our secondary engine. I would however still like to go with the Quest engine if at all possible.
This week we also had our informal presentations for the project to check our progress. Unfortunatly Ted was still in Sydney, So I had to do the presentation myself. I actually didn't realise that there were presentations on, and only found out when I showed up for the ordinary monday morning lecture. I explained the issues that we had run into with Quest and that we were looking for a possible other alternative. Erik and Stewart mentioned a number of other alternative for us, including Director, Active Worlds, Ogre, Torque as well as Unreal 2. I explained that I had looked into a few of these and had chosen Unreal as our second choice.
After Ted got back from Sydney, he told me that he was getting the hang of Quest and that it would be fairly easy to use it to do what we want. So for the moment we are still going to use Quest3D as the realtime rendering engine for the project.
Other then that we have a little bit of catch up to do, but we are both comfortable that we are able to get everything done in time for the prototype.
Weeks 4 & 5
The project work went on hold for a bit during week 4. I came down quite sick at the end of week three and was ill until the end of week 5. Ted had study for exams during this week as well so this meant he wasnt able to do much this week either. I did however, while I was sick, do some modeling work from home, just fixing up any of the area's that Dr Prangnell thought needed changes. I also started working on importing the poser figures into the mesh in the appropriate places.
On tuesday of week 5 I met up with Dr Prangnell again briefly, as he was doing a presentation on the Paradise township up north and he wanted to use some images and video flythroughs from our work from last semester, to include in the presentation. Unfortuantly Ted was unable to make it to this meeting, but I still went in to St Lucia to meet with him and give him a copy of the work. I also spoke to him about our progress and showed him some updated work and he was very interested in how things were going and asked if we can give a presentation on it to the UQASU at the end of this semester. He said that overall it was coming along really nicely and he can't wait to see the finished product.
Throughout week 5 Ted & I also did some testing in Quest3D, however we found it a very difficult and complex program to work with. It doesnt seem to "think" the same as many other 3D programs, and is different to anything else I have ever used. We are starting to think that maybe it is not the best real time renderer for the project. Ted is going to continue playing around with it and see if he can figure it out before he goes to sydney for the weekend. If not, we will have to move to something else. We are seriously considering using the Unreal 2 engine if Quest3D doesnt work out. We have even concidered shockwave 3d or VRML as last resorts, but I dont think it will come to that.
Week 3
Week 3
ProposalThis week was our proposal presentation. We had spent quite a bit of time preparing for this proposal, but unfortunately I think we slightly headed down the wrong track. We spent too much time discussing what we did last semester and although we did discuss how we were going to use our previous work in implementing our plans for this semester, I don’t think we covered this deeply enough. One mistake that we seem to have made on a number of occasions and really need to take care that we don’t do again, is that we were just too broad and didn’t go into enough detail of our proposed plans for the semester.
We did however provide numerous illustrations and a brief story board so as to better get across our ideas, and I did feel that we were able to be successful with proposing our semesters plans, it just would have gone a lot better if we had gone into more detail about what we were going to do THIS semester and how we were going to go about doing it.
I also met with Dr. Prangnell that Monday afternoon and showed him the proposal that I had presented. He was very impressed with it and the work that we had gotten done so far. I discussed with him an idea that Ted and I had had for the project in order to make it more interactive and educational. I had mentioned this idea to him in our first meeting in week 1, however we went into more details with him this time and explained how we were going to implement it. (see week 2 blog) As explained in week 2’s blog, we decided to use poser figures as information points (see Layout Plan Below) to give the user information about certain buildings and areas within the paradise township. Dr. Prangnell said that he would write up some brief information about each of the buildings or areas that each info point (Poser figure) relates to.
Week 2
Group RolesThis week ted and I got together and discussed our roles for this semester. Due to the fact that we no longer had a group of 4 for the project and the amount of work that we had to do was even bigger then before (we now have a new program to learn, this being Quest 3D), we decided to split the roles as such. I would continue mainly consult with the client on a fortnightly basis, although ted would often come along as well. I would also continue working on additional modeling and detailing existing models & doing all the texturing. As well as this I shall be conducting further historical research and documenting our work as we go along.
Ted's role was to deal with the more technical side of things, such as learning Quest 3D and then importing the models into this realtime engine as well as doing any of the required scripting. As well as this he is also continuing to work on our website, which will be hopefully be soon hosted by the UQ Archaelogical Sevices Unit. Finally He is also conducting further research into the capabilities of Quest 3D and possible final output methods e.g. VR Cave etc. (see below)
New IdeasAs mentioned last semester, this semesters plan is to take a 3D reconstruction of the 1890's Mining Town of Paradise from central QLD and make an interactive walkthrough and virtual tour of the township as it was in 1890 using the Quest 3D realtime Engine.
This week we also came up with an idea that we could make the project more interactive and educational for the users by using poser figures as information points to give the user information about certain buildings and areas within the paradise township. The reason that we decided to use poser figures as the information points instead of just, for example, sign posts, is that studys have shown that users migrate towards human figures within a virtual environment and will generally try to interact with them. We decided that when a user clicked on a figure, that a screen would come up displaying information relating to the surrounding structures and area.
We had previously looked at having figures within the township, but never thought of using them in this way.
Week 1
MMDS 3802 - Semester 2 Week 1
This week we first met together as a team again. Unfortunately two members of our team, Thara Amerin and Simon Legrand, decided to leave the project. Simon actually decided to leave uni and Thara decided she no longer wanted to undertake a project of this nature as it wasn’t the field that she wanted to end up working in. This means there is only two of us working on the project now, being myself - Tim Davidson, and Ted Lai.
This week we also met with Dr Jon Prangnell, our client, to discuss our plans for the semester, as well as find out the direction that he wants us to take the project.
We asked him to give us a list of his priorities for the project. These included:
1. Keeps the Historical Accuracy
2. As realistic a possible.
3. Can be used for educational and research purposes for schools, universities and museums as well as exhibited in archaeological exhibitions.
4. That we complete the project in the required time frame.
He was very excited to see where the project would take us this semester.
Screenshots of Final Project
