mediaHACK

Most of us think about finances in our nations fiat currency. In my case, USD is my currency I have dealt with all my life. So when I think I want to buy something I think: “how much will that cost me in USD?” In cryptocurrency we can exchange between currency pairs. For example LTC/BTC is a Litecoin to Bitcoin trade. A BTC/ETH is a Bitcoin to Ethereum pair.

Why is this important or useful? We can use a crypto pair to skip selling a bit of crypto for cash then purchasing the currency you want to buy.

Not every cryptocurrency exchange supports every crypto pairing. Coinbase only supports pairs for its four currencies: BTC, BCH, LTC and ETH. You’ll find that if you want to diversity your portfolio you’ll want to utilize a few different exchanges to get the portfolio and pricing that best fits your needs. For instance, I use four different exchanges (Binance, Poloniex, HitBTC, and Coinbase) and I foresee having to use more in the future.

USE CASE:

I wanted to add some more Stellar (XLM) into my portfolio. I don’t want spend more USD, I wanted to convert some LTC to XLM. Ideally I wanted to find an LTC/XLM pair. I searched a few exchanges I use and couldn’t find one. However, I did find that Binance has their own coin BNB. If you use their BNB then they cut the fee by 50%. So I traded LTC for BNB, then BNB for XLM.

This wasn’t as efficient as I had hoped, but by doing a little work I was able to save on the exchange. Speaking of saving, IF YOU USE COINBASE YOU SHOULD BE USING GDAX IF YOU WANT TO SELL OR WITHDRAW. It is cheaper and in the case of LTC, free; no fees. I’ll do a quick write up soon on that.

CRYPTOTIP

Want to leave me some crypto to support my writing?

Bitcoin: 1785aRMRVP8oZCNFTPcsGw2BoPa1BT7XVW

Ethereum: 0x96A6A14b1269254d5F7b9B9c206f019146AAd65b

As a small company I'm always looking for the best tools to get out our products with the right tools with the right development velocity. For our games (LD: The Woods or Towering Tower) we've looked at several engines and toolsets. As of now Unity 3D isn't our primary game development toolset. The direction of Unity 5 and soon to be available tools may make it a hard too beat choice.

Overall this was a great event. The Austin, TX event was hosted by Microsoft and the main presenter was Lead Unity Evangelist Carl Callewaert (@carlunity). The event began with Microsoft employee introductions and discussion on their developer investment in Unity and their (Microsoft) unified platform. Some hints about Windows 10 being very gamer friendly and upcoming announcements. They also promoted their Azure platform (which I've used and its pretty cool) and directed us indie game devs towards their BizSpark program; which I am a member of. It's a very cool program with lots of free apps and even monthly allowance towards their Azure platform. 

Microsoft is trying to reach out to Unity developers and has a Unity Developer Offer Program. This is mainly for devs that already have a game, but a good handout from them. 

Microsoft also mentioned they have a Game Dev Startup Accelerator where they host your game team in Redmond, WA for 6 months while you make a game. Neat. If I can find more information I'll post about it.

One other tool they talked about was Visual Studio being freely available. I live in OS X and Linux, but this is very good for Windows users. I know, I know, I could run parallels. I'll stick with MonoDevelop for now. For science, here is a link to VS 2015 CTP5.

Unity began their part of the event by introducing the Unity Austin team. These guys are made up from some of the local Austin game dev vets who had created Tsugi.

Unity loved Tsugi and acquired them. Smart move. Tsugi has become Unity Cloud. This suite of tools is amazing. It provides software configuration management (SCM) with continuous integration (CI) for game development workflows. Unity Cloud supports several version control software (VCS) platforms out of the box: Github, Bitbucket, Beanstalk even Perforce. It sounded like they cover the gamut of most common configurations, but the guys at Unity Austin were very excited and open to feedback and community involvement. If they don't support your platform, contact them and they'll see what they can do for you. 

For those that don't know, SCM platforms build, package and deploy software. They are usually connect directly to your software repos (VCS) like Git and SVN and use CI to, as it implies, continuously integrate changes and create new builds. Once your build is done the system will send you an email with link to the build. Now you have a web link to test the build which you can also share with others: privately through collaborators or publicly via a public link. Yes please! 

The Unity Austin team also talked about code hooks and build targeting. They explained you could place macro style blocks and Unity Cloud will detect them and build accordingly. For the build targeting, this is more useful for only building specific scenes and features. This is great for faster iterating. 

I asked Unity Austin what the cost is. They said they haven't made an announcement yet, but you will always be able to use it for free in some capacity.

Unity Cloud features and usage is a large topic and deserves a post all on its own.

Carl took over and introduced us to Unity Ads and Everplay. Both great tools for monetizing and user acquisition. If you've developed an app before you know its hard to get noticed and hard to make sales. Even as an end-user consumer I don't always buy. I look for free games, games with trials or games limited by content unlocks so I can see if I even like the game.

Unity's Ad network is pretty good from what I can tell. According to Carl, Unity paid out 20MM to developers for Ads last year. Now, I'm not sure how good a number that is in relation to the actual number of companies that profited from that. For all we know 1 company pulled in 20% of that revenue. Either way its still encouraging to see payouts.

One of the great features of Unity Ads is the game centric content. To paraphrase Carl: "No one wants to see an ad for a vacuum cleaner in your game".

Unity also has Ad customization to the level that you can block certain types of games from showing up to your users. If you're making an FPS, you can block other FPS game ads. 

If you've done any Ad campaigns then Unity Ads should be somewhat familiar to you. You can create your own Ads and set daily spend limits and cost per install (CPI). 

Via the Unity Ad SDK you can also put up reward videos. These are ad videos where you reward the player for watching a video. 

After I dig more into Unity Ads and the SDK I'll do another article.

Everplay is a free gameplay recording service. It can be added to your games via the Everplay SDK. According to the website you aren't limited to Unity3D. Everplay can be integrated with Cocos2D, iOS, Android or your homegrown engine. Gameplay replays are a good way to increase virality and overall increase your install base. 

Intro to Unity 5

At this point we broke up the presentation into 2 groups. Intro to Unity 5 and Windows game porting. I stayed for the Unity 5 intro lead by Carl.

Carl did a great job leading us from zero Unity to an FPS game with a level, lighting, weapons, an enemy and even a nav mesh with enemy path-finding. There is 4 hours worth of content. Carl has said he'll post the power point and a blog walk-thru of his tutorial. I'll cover some of the features I found great and not the tutorial, but I'll post a link here when I get the info.

For anyone new or wanting to know more about Unity, you could start with the manual The Unity Manual.

They have a landing page touting some of the updates coming in Unity 5.

The Editor UI. It seemed pretty close to what I remember. I haven't dove deep into Unity for a while; since probably version 4.3. Because this is the Pro version, I had the dark theme, which is pleasant. 

Assets loading seemed a little faster. Carl attributed it to the 64-bit editor.

Scripting with Unity is always a fun time. I'm always impressed with the code integration to the UI and how you can drag and drop assets and bind them to script variables and objects.

Physics and collisions seem to be really overhauled. We ran some tests with rigid bodies and box colliders. The team sitting next to me ran a test with was a 13" Macbook pro with a i5 processor, 8Gb RAM and Iris graphics. We had a pretty solid frame rate until we got over 1100 objects with physics in the scene. 

Unity 5 has updated shaders and new lighting. We were shown how to work with lighting a little, but what was really impressive was the Reflection Probes. They gave us reflective objects and blending without the constant and costly recalculating of reflections. 

The audio system in Unity 5 has undergone a great overhaul. We were shown how to create an audio mixer and create and bind individual sounds under a group. Now I'm not audio engineer, but I was very impressed and this is very powerful for our games. The groups can have child audio which then can have bindings themselves. We used this to create a sound ducking effect so that when a rocket was fired the background music was brought down and out of focus. This could be applied in many other game situations: talking to an NPC or creating game tension.

We were shown how to use a Navigation Mesh. I know what the nav mesh is and we used it for a prototype type at Night Owl Games, but the nav mesh is a Unity 4 Pro feature. I only have a Unity 3 Pro license. The nav mesh is a really cool feature. It allows you to select static objects and 'bake' them to create surfaces that objects can navigate across. 

As you can see above there are some highlighted meshes now which can be used for navigation path finding. Sweet!

Below is the Navigation Window. You can see there are some really neat features like the step height and max plane angle. You can configure how big your nav mesh agent (the object that will navigate your mesh) is. 

Like I said earlier Navigation Meshes have been a Pro only feature. I asked Carl if they were going to be available in 5 for free. He wasn't able to give a definitive answer. We'll see when Unity 5 releases.

UI seems to still have a problem when dealing with multiple resolutions and multiple platforms. During the demo another game dev asked about managing UI elements and multiple resolutions. There doesn't seem to be a clear solution for this, but Carl "feels the pain" of developers and said he's brought this to the attention of the Unity dev team. 

The downside. This was all done with unity Pro which carries a pretty heavy upfront price tag for independent developers. In addition to Pro licensing being $1,500, mobile platforms for iOS and Android are $1,500 each. That's $4,500. Ouch. Compared with Unreal 4's $20 + 5% is much cheaper with an industry powerhouse. If you make under $3,000 then you owe nothing more than the $20. If you make between $3,000 - $50,000 thats $2,500. Still cheaper. Now that doesn't mean I favor Unreal more. The roadshow's training and toolset demo has me very much wanting to move to Unity 5. 

VCS Who? CDN Wat?

Just in case you're not up to speed on these new-fangled-schmigity-smack-thingies, VCS and CDN I'll give you the galactic view (much higher than the proverbial 50,000FT level)

VCS stands for Version Control Software. Its what we use to track changes in our systems with the ability to go back and forwards in its history if necessary. They are great. Some you may have heard of include:

Concurrent Versioning System (CVS, old and eww)

Subversion (SVN)

Mercurial (Hg)

Git

A Content Delivery Network (CDN) is used to store assets on the web to maximize caching by providing a common and highly redundant server assembly. GitHub is going to be our CDN. It will give us the stability of GitHub's network and a VCS to manage our files.

Step 1 - GitTFO

This isn't a Git usage tutorial I am going to assume you have the following and have git working on your system. On Windows you're going to have things like the SSH keys all configured as I am not going to cover that. 

Install Git

Get a GitHub account

Step 2 - Create a Repo

Create a repo for your CDN. If you're new to GitHub then click your username in the top right menu to go to your profile page. From there you should see a button above your repositories list that says "New Repository".  

Once here enter a Repo name. Its all relative to your account so I chose to create a repo called cdn. 

Also you can optionally enter a description. I entered: "Using Github as a CDN"

Leave the repos' visibility to public as you'll want anyone to be able to get to the resources.

Click "Create Repository"

Step 3 - Enable Pages

After you've created your repo we need to enable GitHub Pages. At this point your repo's GitHub page might empty and have the standard boilerplate "Setup, Next Steps, [...]" details.

This is OK. What we want to do is access the repos' admin section. Click the "Admin" button up near the top right of the screen under the GitHub main toolbar; the one with the wrench and says Admin.

The page content has different sections blocked off: Settings, Features, GitHub Pages, DangerZone. We want to click the "Automatic Page Generator" button inside the GitHub Pages section.

Go to the bottom of the page and click the "Continue to Layouts" button. None of this matters as we'll be wiping it out with our own assets and page(s).

Next we are at the layout selector. There is a section at to the top with themes, then a small toolbar for Hide, Edit and Publish. Click Publish.

Now we wait. GitHub says it can take about 10 minutes while your pages activate. I found this to be true. After its done we can add our content and overwrite the stuff they setup.

Step 4 - Local prep

If this is your first repository and your'e on a *nix or OS X machine you may want to follow the instructions they have for the global settings:

Code

git config --global user.name "mediahack" git config --global user.email [email protected]

Based on the recommendation of this Gist: https://gist.github.com/1086274 we're going skip to most of the details in new repo page tells us to do. The author is using the name Alice for their project. You can replace it with your own project name. Here we go!

You can proceed with command-line or with a GUI.

Command Line

On your machine create a dir for the repo: mkdir <YOUR_GH_USERNAME>.github.com

cd <YOUR_GH_USERNAME>.github.com

git init

echo "<h1>Hello Alice</h1>" > index.html

git add index.html

git commit -m "initial commit"

git remote add origin [email protected]:<YOUR_GH_USERNAME>/<USERNAME_AGAIN>.github.com.git

git push -u origin gh-pages

git branch -m master gh-pages <- renames the "master" branch to "gh-pages"

Get a GUI

Some people don't like the command-line; I love it. I must admit I do use a GUI for my Git tool. There are tools like GitHub for Windows/Mac, Tower for Mac and SourceTree for Mac. I'm sure there are others.  Personally I am using SourceTree and I am very happy with it.

Step 5 - Add Content

Lets add files to our repo and can use them like a CDN! At this point our local files can be modified, committed and pushed with our tools and it will update our assets. Love it.

Step 6 - Trust but Verify

Check your GitHub Pages url to see if your files are being served up. The url format is: 

http://<GITHUB_USERNAME>.github.com/<REPO_NAME>

After that the assets are in the directories and folders based on where you added them to the repo.

First!

Not really. This is not new. I've just started doing it and I'm documenting for anyone else interested :D

Other references

https://github.com/viewsource/viewsource.github.com

I wasn't able to immediately find an example or solution, click here for the unofficial but referenced docs, but eventually I did.  Sublime Text 2 (ST2) Build Systems are JSON objects. The cmd property executes commands :P However, most documents I saw used the property as an array. The internal cmd tool doesn't seem to like running multiple commands. So to resolve this I first set the shell property to true then set the cmd property as a string. It was also necessary to escape the quotes required in the command to run.

This first pass ST2 Build System runs a specific Ant file then launches my currently active file in Google Chrome.

{   "shell": true,   "cmd": "ant -f \"$project_path/build/build.xml\" && open -a \"Google Chrome\" \"$project_path/dist/debug/$file_name\"" }

For anyone not familiar with the syntax above, the command says, "run the ant command witth -f flag for the file located at MY_PROJECT_PATH/build/build.xml AND (&&) if this is successfully completed open the application (-a) Google Chrome with the copy of the file I am currently working on but it is located in MY_PROJECT_PATH/dist/debug/. 

If you copy this exactly it will probably fail for you. I have a specific project directory structure and a specific output directory. However, this does NOT change the main focus which is running multiple commands via the ST2 build system.

Working example of my directory structure:

/build

build.xml

yuicompressor-2.4.7.jar

/dist

debug

/js

my.html

/src

assets

js

my.html

MyProject.sublime-project

MyProject.sublime-workspace