Add documentation on log viewer and hard drive performance.

README.md

@@ -37,6 +37,8 @@ There are two ways you can generate training data from AirSim for deep learning.
 
 ![record screenshot](docs/images/record_data3.png)
 
+There is a [Log Viewer](docs/log_viewer.md) provided that can visualize the streams of data that are published by the simulator (and your drone).
+
 If you would like more data logging capabilities and other features, [file a feature request](https://github.com/Microsoft/AirSim/issues) or contribute changes. The data logging code is pretty simple and you can modify it to your heart's desire.
 
 A more complex way to generate training data is by writing client code that uses our APIs. This allows you to be in full control of how, what, where and when you want to log data. See the next section for more details.

docs/hard_drive.md

@@ -0,0 +1,52 @@
+# Busy Hard Drive
+
+There is a reason we recommend running your Unreal Environment on a Solid State Drive (SSD).  Between debugging, logging, and Unreal asset loading
+the hard drive can become your bottle neck.  It is normal that your hard drive will be slammed while Unreal is loading
+the environment, but if your hard drive performance looks like this while the Unreal game is running
+then you will probably not get a good flying experience.  
+
+![Busy Hard Drive](images/busy_hard_drive.png)
+
+In fact, if the hard drive is this busy, chances are the drone will not fly properly at all.
+For some unknown reason this I/O bottle neck also interferes with the drone control loop and if that
+loop doesn't run at a high rate (300-500 Hz) then the drone will not fly.  Not surprizing, the control
+loop inside the PX4 firmware that runs on a Pixhawk flight controller runs at 1000 Hz.
+
+### Reducing I/O
+
+If you can't whip off to Fry's Electronics and pick up an overpriced super fast SSD this weekend, then the 
+following steps can be taken to reduce the hard drive I/O:
+
+1. First run the Unreal Environment using Cooked content outside of the UE Editor or any debugging environment,
+and package the content to your fastest SSD drive.  You can do that using this menu option:
+
+![Package Unreal Project](images/package_unreal.png)
+
+2. If you must use the UE editor (because you are actively modifying game assets), then at least
+don't run that in a debugger.  If you are using Visual Studio use start without debugging.
+3. If you must debug the app, and you are using Visual Studio debugger, stop then Visual Studio from 
+logging Intellitrace information.  
+Go to Tools/Options/Debugging/Intellitrace, and turn off the main checkbox.
+4. Turn off any [Unreal Analytics](https://docs.unrealengine.com/latest/INT/Gameplay/Analytics/index.html)
+that your environment may have enabled, especially any file logging.
+
+### I/O from Page Faults
+
+If your system is running out of RAM it may start paging memory to disk.  If your operating system has
+enabled paging to disk, make sure it is paging to your fastest SSD.  Or if you have enough RAM disable
+paging all together.  In fact, if you disable paging and the game stops working you will know for sure
+you are running out of RAM.
+
+Obviously, shutting down any other unnecessary apps should also free up memory so you don't run out.
+
+### Ideal Runtime performance
+
+This is what my slow hard drive looks like when flying from UE editor.  You can see it's very busy,
+but the drone still flies ok:
+
+![Package Unreal Project](images/ue_hard_drive.png)
+
+This is what my fast  SSD looks like when the drone is flying in an Unreal Cooked app 
+(no UE editor, no debugger).  Not surprizingly it is flying perfectly in this case:
+
+![Package Unreal Project](images/cooked_ssd.png)

docs/log_viewer.md

@@ -0,0 +1,29 @@
+# Log Viewer
+
+The LogViewer is a Windows WPF app that presents the MavLink streams that it is getting from the
+Unreal Simulator.  You can use this to monitor what is happening on the drone while it is flying.
+For example, the picture below shows  a real time graph of the x, y an z gyro sensor information being generated by the 
+simulator.
+
+### Usage
+
+To use this LogViewer, connect the simulator `before` you run the simulation.  Simply press the blue connector
+button on the top right corner of the window, select the Socket `tab`, enter the port number 14388, and
+your `localhost` network.  Then press the record button (triange on the right hand side of the toolbar).
+Now start the simulator, pick some mavlink items to graph,  you  should see something like this:
+
+![Log Viewer](images/log_viewer.png)
+
+The drone view here is the actual estimated position coming from the PX4, so that is a great way to check
+whether the PX4 is in sync with the simulator.  Sometimes you can see some drift here as the attitude
+estimation catches up with reality, this is more visible after a bad crash.
+
+### Installation
+
+If you can't build the LogViewer.sln, there is also a 
+[click once installer](http://www.lovettsoftware.com/LovettSoftware/Downloads/Px4LogViewer/Px4LogViewer.application).
+
+### Configuration
+
+The magic port number 14388 can be configured in the simulator by editing the [settings.json file](settings.md).
+

docs/prereq.md

@@ -4,7 +4,7 @@
 ### Computer
 Intel Core i7 or equivalent PC with 32GB of RAM and a SSD and GPU with at least 4GB of RAM. Large environments in Unreal requires a lot of RAM and a beefy GPU card. 
 The typical computer hardware we use for development purposes usually have 6 to 12 cores and graphics card such as NVidia TitanX or NVidia GTX 1080 with 32GB to 64GB RAM
-and fast SSD.
+and fast SSD.  See [why fast SSD](hard_drive.md).
 
 ### RC Transmitter and Receiver
 This is the remote control that you usually use for real RC vehicles such as quadrotors. Our favorite is [FrSky Taranis X9D Plus](https://hobbyking.com/en_us/frsky-2-4ghz-accst-taranis-x9d-plus-and-x8r-combo-digital-telemetry-radio-system-mode-2.html) because it has built-in USB ports.