First podcast

Had a new experience this week. On Tuesday, 4th of August, myself, Jack and Callan all traveled to uni to be part of 8 and a half bit gaming podcast. This was the first time that we have ever been part of a podcast, and it was really fun.

Creating Dyadic has been a great experience as it has allowed me to get involved in things that I have never done before such as marketing, creating greenlight campaigns and even podcasts. We were all a little nervous to begin with as we didn’t know James or Paul very well. However, that all changed within a few minutes of talking to them. They had a way of making us feel a lot more comfortable.

We were asked questions about Dyadic, we answered them and then it would usually go off track in some way. It was really fun. I’m glad that we participated in this podcast as if I want to be serious about games development, I’m going to need to be able to talk to people about my games. And this experience gave me some more confidence.

I won’t ruin too much, but we talked about our greenlight plans, our futures and even a love story! I will make another blog post when it’s available to listen to.

Advertisements

Handsome Dragon Games website

I worked on yet another website this trimester. I’m starting to get a name for myself…

At the beginning of this trimester, after deciding what name we were going to use as a team (Handsome Dragon Games), we purchases the domain name and hosting. Then I was in charge of making the website. This task basically fell to me as I was the only one with web experience and we needed it done as soon as possible so we could start promoting ourselves.

At the beginning of each website I develop, I forget how much I hate the whole process. Websites was how I first got started with code and which spiked my interest in making games, but boy do I hate making them nowadays. There is just so much bullshit. I think I have identified why I hate web development, and found ways to make it better.

My main problem is I know PHP. So I made the website with PHP. I just don’t enjoy writing that at all, so it’s miserable. But recently I got involved with Node and I am enjoying that so much more. Next time I am on for making a website, I will use Node instead and see how that goes.

Another problem I have with web development is making a sytle. I planned out the style, created it from scratch with CSS (though I have discovered Less since then and will use that next time to make this a little less painful) and then found out that the colours didn’t work. So we spent about 4 hours as a team just changing the colours and seeing if they work. However, css doesn’t have variables (again, will use Less next time!), so it was super annoying having to change the colours. Sure, you can use Ctrl+F to find the colours, but then you got to edit them everywhere they are used. Eventually, we settled on colours even though no one is truly happy with them. I think the problem is the theme isn’t very good.  What I did was look at a few themes and tried taking the best things and putting them together. I don’t think that worked very well.

Now that I have some experience with Node and Less, i’m once again looking forward to making a website. I really want to see if this makes the process a lot better or there is something else that I am overlooking at the moment that makes me unhappy to make websites.

Dyadic’s Greenlight Debut

It’s finally time… time to put Dyadic on Greenlight. We have all been working extremely hard over the last two weeks to make sure everything is up to scratch to make sure our Greenlight campaign goes well. Admittedly, I have probably put a bit too much time into this over Studio, but this was really important to me and my team so I made that sacrifice.

We have been doing many things outside of making the game to make sure we are ready. Things such as making promotional material (Posters, Business cards and Website), creating the videos needed for the Greenlight page and lots of writing. It has been a really interesting experience as these are things I haven’t really been apart of when making a game. All other projects I have worked on have never come this far, and it’s exciting being in this position.

It’s a little scary thinking that we will be showing off our game and people will be able to leave all sorts of feedback. I’m looking forward to reading all that feedback (negative or positive) and hopefully being able to shape the game that we want to make and that people want to play.

Everything was ready in time and the game was placed on Greenlight successfully. We didn’t miss anything vital, so that’s a good start. Now we need to continue marketing the game and getting lots of people onto that page voting for us. Within the first day we had a massive hit due to all the followers we have acquired through social media and lots of people sharing our page and post about greenlight. It was really motivating scrolling down facebook and seeing how many people actually cared enough about us and our project to share us. Let’s hope this all works out for the best.

OpenGL life

I got stuck into learning how opengl works over the last week. Thankfully, I wasn’t alone on this endeavour. Pat managed to help me a few times when I got confused for stuck. So thanks for that Pat!

I followed along the tutorials on open.gl. I decided to use SDL for creating the window and opengl context and glew to obtain all the function pointers needed for opengl. Unfortunately, I had more trouble setting up SDL and glew than I would like to admit. I downloaded SDL and set up the lib and include folder. I linked them within Visual Studios and I was getting linking errors. After a little looking through the lib folder, I realised I set the additional dependencies to SDL.lib and SDLmain.lib when they actually needed to be SDL2.lib and SDL2main.lib. No biggie, I found the problem fast enough. However, the next problem stumbled me a lot more. I knew I needed to move the dll files into the project directory, but for some reason, it didn’t work like it normally does. Normally I can place the dlls next to the .vcxproj file and Visual Studio will have no issue finding it. However, this time to run it in VS I had to add them to both the Release and Debug folders. I placed these files around everywhere and eventually it worked. Now I know for future reference how to solve this issue and where to put the files.

GLEW was also a bit of an issue. Not as big of an issue, but I still ran into one problem. When I set all this up, the glew website was down for maintenance. Luckily, I already had the files downloaded from when I used them a couple of months ago. So I put all the files needed into the include and lib directors, move the dlls into the Release and Debug folder and set the Dependencies of glew32.lib. However, it was still having issues. I spent a few minutes looking through all my settings and checking the files were in the correct folders and then I remembered I needed to add the dependencies opengl32.lib. Once I added that, it finally built and I had a black window that stayed open for 1 whole second and then closed. Progress was made.

I followed along and I learnt how to make an opengl context, initial glew to get opengl function pointers, and create the Vertex Buffer Object. I even understood a large proportion of how the rendering pipeline works. Then I got to shaders… The tutorial is a little vague here, as they never talk about the most popular ways of writing shaders, and they didn’t really explain you have to parse the shader. I decided I want to have .shader files (because they sound cool) with both the vertex and fragment shader in one. The way I establish where the vertex shader and fragment shader start is with #vert and #frag at the beginning of the shader.

My parser went through a few iterations, starting with it opening the file, looping through it line by line, checking if the line contains #vert or #frag and then adding the lines to the specified shader source. However, this wasn’t really a good way of doing it as it was potentially doing two finds each line of the file. The final iteration is it’s own class that has a shaderId which is just the program GLuint. It loads the whole file into memory then does 2 finds to find the position of #vert and #frag. Once they have been found, the source is created by getting a substring between the opening tag and the next tag (or the end of the file). This way there only needs to be 2 finds instead of a potential of 2 each line. In my shader class I made a large mistake which actually effected the end result not rendering. But more on that a bit further down.

Next I learnt about setting attributes for the shader and uniforms. These were simple enough as I understood the concepts from using shaders within Unity before. This is where I ran into one of the large problems. I was creating the Vertex Array Object but it was crashing everytime. I tried moving the code around and still nothing. It turned out I had to do glewExperimental = true before I initiated glew. It appears that not everything is set up unless you have that line. Once that was done, I thought it was all good, I ran it and … nothing. To be expected. Now I will outline the other issues I had.

In my shader class, for the fragment shader, I forgot to change it glCreateShader(GL_FRAGMENT_SHADER); I left it as GL_VERTEX_SHADER for the fragment. The next issue was a silly one. When I was drawing the rectangle, I wrote the code on the wrong line and it was just outside of the core loop. Oopss. I picked up on that one fast luckily. The last major issue was I forgot to enable the program. I knew I had to do it too because I created functions in the Shader class to enable and disable. I just forgot to call it. Once that was all done, it finally drew a triangle. I was so happy because in totally this took me about 8 hours to write and understand what was going on. I didn’t want to just copy paste code, I wanted to actually understand. And I can say, I understand a lot more than I did before. Obviously it didn’t ALL stick the first time, but continuous use will help with that. I also changed the color uniform to be a random colour each update… so that looks cool. Anyway, here are my precious pictures:

HappiestDayOfMyLifeImCreativeISwear

R vs Processing

We have been tasked with showing visual data from a flocking simulation so that someone can tweak variables that affect the way the simulation runs with some insight. Suggestions were things such as spitting out a CSV and making graphs in Excel and Heatmaps. I will be completing Heatmaps, but instead of graphing in Excel, I decided I would look into R and Processing.

After doing a bit of research on both, I found a useful video on how to make a graph in R, so I took that option first. I downloaded and installed R, got myself an IDE (R studio) and started playing around with R. The way you run R in the IDE is different to anything I have done before. It only runs lines that are highlighted, mean you can run a single line in your program or you can highlight it all and run the whole application.

I downloaded a library for R called ggplot2. This library was created to easily plot data from a table. To create the table, I loaded in a CSV. Reading CSVs in R is so simple. They have a function specifically for reading them in, and once you have the table, to get data from a specific cell you just have to write the variable name then the heading or index next: table$heading1.

After creating a bar graph, I ran into my first problem. This was running in the IDE and was only showing one Graph at a time. So I decided to try and export the program to see what it does as a .exe…. however, that isn’t an option (as far as I could tell). All solutions I found said you could write another program to launch your R code as it’s an interpretive language, but you also need R installed on the deployment machine. This was the biggest killer as I plan on running this on Uni Computers where R isn’t installed. There is also the fact that I would never want to ask people to install R just to see some graphs that could have been done (and probably nice looking) in Excel. That leaves Processing.

Before even diving into finding libraries for processing that make graphs, I determined if you could export it as an .exe. And you can! So this is off to a good start. Another really good thing about processing is you don’t need to install it, so it can be ran anywhere. You just need the .exe you download and you are good to go.

After a bit more research through all the libraries, I found once called giCenterUtils. It seemed to do exactly what I wanted, so I pulled the library into processing, and tried it out. I parsed my CSV file and passed all the lifetimes of the prey to a barchart. Then I specified where I wanted it and voila, there is a nice bar graph on my screen. The following picture shows the lifetimes of the prey in a bar graph form then displayed 4 times. That way once I have recorded more data, I will be able to show various plots on the screen.

Processing

My plan from here is to add titles to the graphs then have arrows on the sides of the screen so you can go between the graphs, allowing for more than 4 graphs to be shown.

Network Draw Application

I have really enjoyed this task a lot.

At the very end of week 1, we started talking about networking in class. It was all new stuff to me, so I found it extremely interesting. I have been interested in getting started with this stuff for a little. I did dip my toes into it back in Make-a-Thing last year, but I had some help from Pat.

The task we were given was to create a drawing application that would connect to a server (provided) and would allow multiple people to all draw together. We are able to send circles, squares, lines and single pixels to the server. I mainly focused on pixels as I wanted players to have that freehand drawing functionality.

I grabbed SFML 2.3 and linked it with a new project (I’m getting pretty good at this…) and got started. Then I got started. One of my goals for this project was to try and keep my project tidy and code in relevant files. I feel like I have achieved that well as all my networking code has gone into it own file, UI stuff into another etc. You can view the files on github.

My original idea for sending packets would be I have two arrays the size of SCREENSIZE * SCREENSIZE. Each one would represent a single pixel. Then, every interval, I would check the difference between them and send the pixel packets. I thought that would work well and I implemented it…overlooking a crucial thing. You can never draw over pixels. So…that was a problem. Also, it was difficult to send multiple pixels at once without being in-efficient.

After learning that it was possible to send multiple pixels in a single packet (and advised due to the overhead of each packet being so large for such a small amount of data), I changed the system. Instead, I created an array that would hold all the pixels to send each interval and when it was time to send the pixel packets, I would send the array instead, making that I would need to send way less packets, and also i’m not sending tiny amounts each time. This was much more efficient and I found on average I was sending 10x less packets.

I also worked on a few UI elements. So I created Buttons, Text Input Fields and Sliders. I thought that it would be worth investing time into these as I can use them in the future. The one I had most difficulty with was the text input field. The reason being was I didn’t understand how strings worked well enough. There was also an issue here with unicodes. Eventually, I realised strings are just an array of characters, and I worked through with the unicode characters. After much tinkering, I finally got a field where when you click on it, you can change the text, add more text and if you click out or press enter, it no longer has focus. I ended up using that so the user can type the IP and Port then having a connect button. The sliders were used for RBG.

The final task was to make a heatmap. I decided to map the user’s mouse position rather than when they click. To do this, I had a heatmap class that had an array to keep track of previous mouse positions. Similarly to the sending of packets, ever x interval, I would loop through the array and modify the image. I ran into a silly problem with this. It wasn’t drawing the pixels correctly and it was creating lots of images really fast. What was happening was the image was being saved and opened too fast, causing issues. This happened because I forgot to reset the interval. It was a really silly mistake and I laughed when I found it.

In the end, I was really happy with my draw client. I am really looking forward to doing so much more with it. There is a part two to this task that got completed over the next week. My understanding of networking is a lot stronger, and I believe this is an area in programming that I would like to focus on, alongside tools.

Link to github: https://github.com/Silcoish/studio3drawclient

Making a Raytracer Multi-Threaded

Recently we were given an un-optimized raytracer so we could work on making it multi-Threaded and some optimization techniques. During my time working with this project, I spent a lot of time trying to understand what multithreading is and how it works.

During the classes I asked lot of questions to help further my understanding. I feel like I have a good grasp of the concept now, allowing me to implement it into the raytracer and actually understand what is happening.

I used openmp for multithreading for the project. When I first got the project, it took ~64 seconds to render. By the end, I got it down to ~11. I would like to continue further with this in the future, but at the moment, I am focusing on other projects.

One problem I ran into while working was that you had to enable openmp. I was implementing it without seeing any effects. This signalled a red flag to me, so I researched it a bit and found out that you had to enable it within visual studios. Once that was done, the program was running much much faster.

What I did was use openmp to determine how many processors the computer had and based everything off that. To do that, it was one simple function:

unsigned int nProcessors = omp_get_max_threads();

Once that was complete, I was able to tell openmp to create threads equal to 4 times the amount of cores the computer had. I did that with:

omp_set_num_threads(nProcessors * 4);

I came to the number of threads by trial. I found that it was much faster to have double the amount of processors rather than the same amount. Then I kept increasing it, with it peaking at around 4 times the amount. Anything more and it was slowing down due to the excess time required to keep swapping threads.

The next process was the utilise multithreading. To do so, at loops where the program can run in parallel, I used:

#pragma omp parallel for

This broke the for loop down and assigned work to each different thread. An example of that is if there was a for loop like so:

for(int i = 0; i < 100; i++)

and there is four threads, it will assign the work load equally. The first thread will compute from 0 -24, the second thread will do 25-49 and so on. Then they can all run parallel making the program run much faster. At this point, the program was running at ~16 seconds.

In the way of optimisation, I didn’t really do too much. The main thing I did was think about how shadows were handled. I realised that when doing intersections to determine where the shadows were, if it intersected with one thing, then we knew we would have to cast shadows. Therefore, rather than loop through all the rest of the checks, I just returned form the function. Upon comparison of the old and new versions of the rendered images, there were no differences.

Overall, I feel like with this project, I have achieved a much better understanding of how threads work. I managed to do a little optimization appart from the threads and got it to render in ~11 seconds rather than the original ~64 seconds.

Link to raytacer: https://github.com/Silcoish/studio3raytracer