This post will be a bit of a rant. I recently got a Dell Precision M4800, a professional mobile workstation.
It comes with software preloaded called Waves MaxxAudio, it provides some of the same features other notebooks include that can “make audio sound better”.
I’ve been comparing the sound output with MaxxAudio enabled, and MaxxAudio disabled. And while it does sound somewhat better while on. The audio still sounds terrible for some strange reason.
I turned the audio enhancement off, and the audio that came out sounded like a megaphone. I instantly knew something was up. So I stuck the audio from the headphone jack straight into a calibrated audio spectrum analyzer and found that:
1. When MaxxAudio is ‘off’, It isn’t actually off. The mids are boosted, and bass/treble cut off to make the audio sound much worse than it should be
2. When MaxxAudio is ‘on’ It boosts the bass and treble to levels where music begins to sound nothing like what it was intended.
So what I did next was uninstall the realtek driver, and rebooted my PC. Music sounds much better now, not distorted, and full of adequate sub bass that i’m used to hearing from a semi-decent DAC.
Now, I really expected better from Dell. I find this completely unacceptable in a professional notebook. Audio coming out of the speakers, when audio enhancements are “off” should not be tampered with at all.
My advice is, UNINSTALL THIS CRAPWARE the moment you get it. Use the stock driver provided by microsoft. Wave audio’s enhancement software is nothing but a scam and a battery/CPU hog.
This is the only way to make the speakers and headphone jack sound normal again.
One of my new projects is figuring out how to automate my Pioneer VSX-822-K AVR. There have been posts before on how to do this with telnet for higher end AVRs (VSX-1022). But there doesn’t seem to be any documentation on mine.
My plan is to make a PHP web interface for controlling it because:
1. The Pioneer Control App for Android is laggy and crappy
2. I can control it from a web browser as well
The VSX-822-K uses port 8023 for telnet commands. Only some of the commands that worked with the VSX-1022 worked with the 822. Many of the function commands are changed. I went through every possible FN combination below. This information doesn’t seem to be available anywhere else but here.
10FN Video H
15FN DVR/BDR H
25FN BD H
44FN Media Server
49FN Game H
TO get the rest of the commands like tuner preset+/-, I installed Shark for Root on Android. This application is like WireShark in that it captures packets to and from the device. I then opened the AVR application and made sure I pressed every button that was available.
But then I remembered that the app downloaded device specific data when I first opened it. I went on a hunt inside ESFileExplorer for the related application files. I found them in /data/data/jp/pioneer.avsoft.android.controlapp I zipped that up for inspection. And then I found the jackpot
Every command the AVR uses, implemented in HTML and JS, by japanese programmers at Pioneer Electronics. I can’t blame them for using HTML, It’s easy, but it’s not responsive at all. No wonder the app was laggy.
After sifting through the code, here’s what I got out
PWR0 Power on
PWR1 Cold standby
PWR2 Network standby
MUT0 Mute on
MUT1 Mute off
?GAP Prints OSD info
FU: Function up
FD: Functon down
10FN Video H
15FN DVR/BDR H
25FN BD H
44FN Media Server
47FN DMR (doesn't do anything?)
49FN Game H
PO Power On
PF Power Off
VU Volume Up
VD Volume Down
MO Mute On
MF Mute Off
MZ Mute toggle (doesn't work?)
##PR (30 avail)
0100SR Advanced Surround
12PB Skip Reverse
13PB Skip Forward
40PB iPod Control30
Some other things to note:
The AVR only allows one telnet session at a time, otherwise it will refuse the initial connection.
The remote is a huge pain in the ass to use, if you want to change settings you have to go into the internet radio function menu, just to get the 80’s style menu to open.
What’s worse is the menu goes from digital in the CPU, to analog in the DAC (so it can be outputted to composite video), THEN it gets converted back to digital for use in HDMI. It looks really ugly to say the least.
When you are using SDR# you are looking at a chunk of radio spectrum. Standard radios only have the ability to process one signal at a time, you now have the ability to not only hear, but also see the signals.
Your sample rate is your bandwidth. If you only want to see 1MHz of spectrum, choose 1.0MSPS. Higher sample rates above 2.4MSPS are not recommended
Interference from very strong signals will show up as mirrors that run in the opposite direction of the actual signal or only appear at the edge of the screen.
Interference from very strong signals will show up at intervals corresponding to your MSPS rate
If you have a signal at 100MHz, and your sample rate is 1MSPS
You will see a duplicated signal at 101MHz, 99MHz, 98MHz, and 102MHz.
If you have a signal at 100MHz, and your sample rate is 2MSPS
You will see a duplicated signal at 98MHz, 96MHz, 102MHz, and 104MHz.
The two display methods of SDR#
What is the FFT and Waterfall?
FFT: Top half of the screen
Waterfall: Bottom half of the screen
How to enter frequencies into SDR#
Type it into the VFO
Click on the bottom or top half of a number
Drag the waterfall or FFT
Use your scroll wheel
Eliminating the center signal spike
If you want to get rid of that DC spike in the center, enable Correct IQ
Decoding digital tips
For decoding digital signals
max out the volume, use WFM, change frequency step to 6.25KHz, and change the bandwidth to fit perfectly around the signal’s envelope.
Set your sample rate to .900000001 MSPS (I personally have had the best results with this)
Also uncheck the filter audio box, chances are, your application will want a clean signal coming from SDR# to decode digital data properly.
If you are decoding time sensitive information (Trunked radio decoding), turn down your audio latency so it is low enough to where the audio doesn’t pop from buffer underruns, but is fast enough so that the beginning of conversations aren’t cut off.
Keeping things organized
If you are going to be using your SDR for multiple uses, like HF, VHF, and Digital. Duplicate your SDR# folder for each instance. This will save configuration time, so you won’t have to change modulation, volume, or Gain every time you want to listen to different modes.
When to use AM
If you do not have an upconverter, there are only very few times you will use AM,SSB,USB,LSB,DSB,CW,or RAW. These include
Listening to Airband (120MHz)
Listening to CB Radio (26-27MHz)
12 and 10meter ham bands (24-30MHz)
Low tech gadgets that run on 27 and 49MHz (RC Cars)
Remote controls that run on ~400MHz (Car door unlocks)
some low tech ISM devices 902-928MHz (home automation, weather sensors)
Otherwise use NFM or WFM
Common interference in the home
If you use Wi-Fi or Ethernet in your home, you may see lots of interference in the 140-160MHz range, when browsing these ranges don’t use Tuner-AGC unless you have already locked onto a signal you want to listen to. If you are just browsing, keep the gain somewhat low to avoid overloading the device.
When to use AGC
Use AGC when you are aware that checking the box may lead to visible interference, only use AGC when you are sure the frequency you are on, is the only signal you want to listen to.
Tuner-AGC seems to work better than RTL-AGC at picking signals out from noise
Generally RTL sticks need calibration. If you are only listening to FM Radio, calibration probably isn’t needed, but for narrow and more precise signals, it is required.
It is recommended to wait several minutes before calibrating, when the temperature of the dongle warms up, the frequency could shift 4+ppm.
Calibration is calculated in ppm (parts per million) AKA Hertz per MHz
so, if your correction is 50ppm, a signal at 1,000,000Hz(1MHz), will be 1,000,050Hz(1.00005MHz)
Remember that this scales the higher in frequency you go. So at 900MHz, the correction will not be 900,000,000/900,000,050
It will be 900,045,000. At 900MHz, a difference of 45 KHz is huge, especially for narrow signals like voice.
When calibrating your dongle, always opt for using a higher frequency reference point.
Calibrating to a local NWS station (162MHz) is good, but calibrating to a local GSM tower, or trunked radio control channel is much preferred.
Getting a clearer picture
If you are listening to narrowband signals, it may be helpful to get a clearer view of the signals you are looking at.
When you first start using SDR# turn your contrast down to a level where:
no signal = blue
weak signal = light blue
med signal = orange
strong signal = red
Note, while gain controls how the signal is heard and viewed, contrast is only visual.
This mod allows your GPU to disable its fans when the temperature is cool enough. Flash at your own risk, I use this myself, but I can’t take responsibility for fried or bricked cards. Use nvflash to update. Consider using a piece of software to throttle the fans according to temp levels (EVGA Precision X works)
I oftentimes see people asking why there is so much interference on their SDRs. Most visible interference is in fact not caused by real signals, but caused by the RTL dongle overloading.
Eliminating visible interference
The first to troubleshooting interference is turning off any type of AGC. AGC maxes out the gain slider which makes nearby strong signals to appear where they shouldn’t be.
37.2 dB is a good level
Turning on and off AGC, Manual Gain 37dB
Change your sample rate. Understanding the sample rate is quite easy. 0.25 MSPS will only show 250khz of spectrum on your waterfall/FFT screen.
It isn’t recommended to use 0.25 MSPS because this can cause interference on frequencies every 250KHz. Take for example, a frequency on 850MHz and 850.250MHz. If you set your sample rate to 0.25MSPS, and either the signal is too close, or your gain is too high. Both 850MHz and 850.25 will mix causing bad decodes. The same can be found up to 10 iterations of 250KHz past the original signal. To see this, find a strong signal, set your gain high. Center it on your waterfall/FFT display, and increment the frequency by 1MHz. You will see the same signal every 1MHz, slowly decreasing in strength as you increase the frequency. Now use 2.048MSPS, you will notice mirrored interference every 2.048MHz
250KHz interference, 929.6125 MHz signal
.9000001 MSPS is recommended for digital modes because it provides a small enough chunk of spectrum for visually inspecting the signal, and it doesn’t cause even overlaps. So instead of 849MHz, 849.5 849.75 interfering with 850MHz, interference will only occur every 900KHz, which has a lesser chance of overlapping interference.
How to eliminate real interference
Real interference is caused by any type of RFI near your SDR receiver. Simply moving the dongle and antenna away from the source of noise (computer, TV. AVR) will lower interference.
To go even further, you can add a ferrite core, or an RF shielded case around your SDR to block unwanted signals.
A few months ago I wrote a simple script in batch that allowed me to record each call in MP3 format with the TGID, RID, date recorded, into a folder that contained the proper year/month/date.
It’s still a little buggy, there are much better ways of doing this.
But what I recently did was make a web front end to read the folders and files, then spit out a table with all the data.
Right now it’s not very efficient, but I plan on fixing that by migrating from, a loop that scans every file from a directory, to a database application. I’ll be using a program called Everything Search by Void Tools. I chose this because it has an extremely fast indexing engine and a command line interface.
This should speed up my page load times by a lot. With my current setup a query takes about 5 seconds, with a directory that has a month of calls, 26,000 files, and almost 1GB of file space taken. If I don’t make this change, page load times will only increase.
I hope to be done with this migration by the end of the week.
This is the method I used. I didn’t feel like paying for Clockworkmod tether, so I found a solution that is almost identical, with alternate software.
No root needed
I’ve tested this with my Nexus 5 running 4.4.2
The problem: Using the stock tethering method, only redirects you to the T-Mobile page where they advertise their tethering option. By default you can’t access any other internet site.
How they do this: T-Mobile counts the hops to their servers using the TTL function, if an extra hop appears, T-Mobile will assume you are tethering without permission.
On your phone: Get Proxoid Go into your settings>More>Tethering & Portable hotstpot, check the box that says “Portable Wi-Fi Hotspot” Take note of the Wi-Fi password Open Proxoid and check the ‘Start’ box
On your computer: Connect to the Wi-Fi Network, and enter the password noted previously Open a command prompt and type ipconfig Look for the default gateway for your current Wi-Fi adapter (usually 192.168.43.1) Take note of this IP address
Checking if the connection works: In command prompt type ping (ip address) e.g. ‘ping 192.168.43.1’ The response should not time out Close the window
For firefox: Go into Settings>Network under connections, click “settings” Select manual proxy configuration In HTTP proxy, Address: (IP noted earlier) Port: 8080 Check the box that says “Use this proxy server for all protocols” Click OK
This article is meant to go with my video, but the video I have is older than the steps I show here. Follow this tutorial, but use my video if you want to know how specific programs work. (don’t follow the video, just watch it)
Note: Lots of things are changing rapidly in the areas of SDR, especially the $9 RTL-SDR’s So What I say today, could change tomorrow. That said, I will provide as much detail as I can, and will list the versions of software I use.
Intel 2nd generation Core i5 or better (AMD processors have been untested)
You need to at least somewhat know how to work SDR# and UniTrunker. I suggest you play with some analog stuff and get familiar with the UI before getting into advanced things, or else you’ll tear your hair out.
Rick (Developer of UniTrunker) has implemented RTL-SDR INTO his program, so we don’t have to use this setup in theory, The latest preview build of Unitrunker has RTLSDR tuning support built in (no need for SDR#) but audio decoding isn’t too great, I still suggest using the tutorial below.
Virtual Audio Cable uses lots of CPU resources, even if it doesn’t seem like it. It seems VAC works well on some machines, and bad on others. My latest videos show decoding being done on a desktop with a core i5 4670K. I know this setup works with my hardware, and you can replicate my build for ~$300(used). Decoding varies a lot from PC to PC, and is generally not recommended on PC’s older than 2008.
Lets go through how this setup works. Below is a picture to familiarize yourself with what’s happening.
You have 2 USB Dongles with a (special)driver loaded that allows you to communicate directly with the chip inside the dongle. The chip controls the frequency at which you listen to. If you tune it to 100MHz for example, you’ll see the end of the FM broadcast band (87-108MHz)
SDRSharp and UniTrunker can utilize this driver to feed audio into their programs. When you listen to trunked audio, you have to have at least 2 receivers. One receiver is dedicated to listening to the control channel, the other listens to the voice channel. In this setup UniTrunker will listen to the control channel from SDR#1. Control Channels send data like, who’s talking, who joined the network, what the frequencies of the channels are, and how many channels there are. Control channels usually transmit 24/7, so they are easy to pick out in a waterfall display. Unitrunker reads this information into a nice window like this.
This window will only pop up when UniTrunker has detected a compatible radio system
Now lets talk about DSD. All DSD does is listen to Virtual Audio Cable for audio it can decode. Think of Virtual Audio Cable as the pipes that send audio from one program to another.
SDRSharp gets its orders from a plugin “Zefie’s Mod”
Zefie’s Mod is a 2 part plugin, it requires the installation of VC++ REDIST. Part one emulates a radio interface in UniTrunker. This is shown here
Part 1 is also known as remote.dll
Part 2 is a plugin for SDRSharp, It takes orders from remote.dll via a file called sdrsharptrunking.log. Sdrsharptrunking.log contains the current channel you should be listening to.
This is what the plugin looks like
Unitrunker tells the debug receiver to listen to a channel, the debug receiver writes the file called sdrsharptrunking.log, Zefie’s Mod SDRSharp plugin listens for that file, and tells SDRSharp to change its frequency based on the file.
This is currently the only way to listen to the voice channel. In analog trunked systems, you can directly listen to the output audio without the use of DSD. But if your system uses P25 CAI you will need to decode the audio, otherwise you’ll just hear weird noises any time someone talks.
DSD is a program that decodes a number of digital voice protocols. It’s still experimental and has bugs. As of version 160 it does not support P25 Phase II, but support is currently being worked on here. By default, DSD is not very good at decoding audio, It tries to guess what type of signal it is receiving and sometimes its guesses cause the audio to break up. Command line switches are used to specify what type of signal it should be listening to. This will be discussed further, later on. DSD requires some DLL’s to work, and can only utilize the “default” audio in and output for Windows. The audio you hear out of DSD should be words that we can understand.
Look for a device called “Bulk-In, Interface (Interface 0)” There may be two
Select WinUSB as the target, and click “Install Driver“
If the driver install fails (usually on Windows 8), Select “Edit Name“, Rename the device to “Interface” and try the install again.
Driver installation should be successful, Check the second Bulk-In and make sure the second device has its drivers installed too.
You may close zadig now.
Installing Virtual Audio Cable
You can purchase, or find an alternative to “Virtual Audio Cable”
Go through the setup for VAC and open the “Control Panel” (blue icon with yellow cables)
Under “Cable Parameters” set “SR” to “22050 … 192000” and “NC” to “1 … 1“
We are lowering the fidelity of the cable to save CPU usage.
Click the “set” button
You may close Control Panel now.
Crossed out steps above have been found to be un-necessary.
Configuring Windows Sound
Right click on the speaker icon in the system tray
Click on “Playback Devices“
click on the “Recording” tab
Double click on “Virtual Audio Cable”
In the “Advanced” tab, set the fidelity to “48,000hz , mono“
Set the default recording device to “Virtual Audio Cable”
Click on the “Playback” tab
Double click on “Virtual Audio Cable”
In the “Advanced” tab, set the fidelity to “48,000hz , mono”
Using SDR# for the first time
RTL-SDR Dongles have a warm up period of 5-10 minutes. During this time there may be a frequency shift of 5PPM or more.
Please open SDR# and select RTL-SDR/USB
Press “Play” in the main window
Feel free to play with SDR# to become accustomed to the user interface.
You can change frequency by dragging on the FFT/Waterfall, by clicking on the top or bottom of numbers, or by putting your mouse cursor over the numbers, and typing a frequency in. Look for FM radio stations, or Weather radio stations nearby and see if you can listen to them, set your mode accordingly WFM for FM radio, NFM for everything else.
If you feel like you might mess up the software, feel free to make a copy and run that one instead.
Once you’re back- Calibration can be set by going into “Configure“
Take note of how some settings may be locked, this is because you are currently running
Press “Stop” in the main window
“Device” selects which RTL dongle you will use
“Sample Rate” selects the amount of spectrum you will be able to see. Note that 2.8+ MSPS is not recommended due to audio cutouts, 0.25 is not recommended either due to interference.
Set this value to 0.900001 for optimal performance during trunking. Doing so will show 0.9MHz per fully zoomed out screen. Take note of how this correlates to the MSPS.
“Tuner AGC” can help in situations where you need the best signal possible for long range listening, however it is not recommended for nearby systems due to the possibility of overloading the SDR and causing harmonics in the FFT/Waterfall.
“RF Gain“should be set to a maximum of 32.8dB to avoid overloading.
Under “Frequency Correction” you can set the amount of correction needed. If you don’t set this properly, your decodes will not sound as good as they should, or not come in at all.
The PPM varies by unit. Generally around 55-65 in R820T based dongles
To calibrate your tuner you can find a control channel on radioreference, and tune to it inside SDR#. (you can also calibrate using a known NOAA weather frequency)
Notice how changing the PPM slightly changes the frequency, zoom in to get a better look.
Do this for both SDR’s, switch using the “device” drop down menu
Be sure to remember your number, It will be needed for UniTrunker.
Note: In practice, UniTrunker must always be opened before opening SDRSharp or the RTL device will not show up. If you currently have SDRSharp running, please close both UniTrunker, and SDRSharp, then open UniTrunker only.
Double clicking on the receiver will open the setting panel for the corresponding device.
In the “info” tab set the “RTL Device” to the second device on the list.
Set your “correction” to the PPM you had set from SDR# (e.g. 57.00000)
set “gain” to 300
checkmark “Auto Gain“,
Set your park frequency to the control channel frequency (e.g. 854.16250)
Start the receiver
If your dongle successfully found a control channel, a new window should pop up.
Troubleshooting the control channel (no video)
If you do not see a system pop up, follow these instructions
Go into the scope tab and look at the signal coming in.
If it is random and not uniform, you have not found a control channel
P25 Control channel example scope image
Static, no signal scope image (seen at 21:35)
A way you can detect if you found a control channel is in the info tab
This indicates a P25 signal, with a health of 99/100 was found, and is being decoded.
If you still see no systems, go into radioreference and try typing alternate control channels into the park frequency text box. See pic
If you still can not find a control channel, try a different system, try a better antenna, or go close UniTrunker, Open SDR#, and manually browse for a control channel visually.
This is what a control channel looks like, move your cursor over that frequency and try that the next time you open UniTrunker
Now that you have detected a system in UniTrunker you can begin filling out, or downloading radioreference data.
This is the download button, if you are a premium radioreference subscriber, you can enter your username and password, and UniTrunker will fetch the latest system info, system name, talkgroup names, etc.
Your frequency list should populate with channels like this.
If it does not, stop the R820T receiver, close the system window. and Start the receiver again.
UniTrunker can decode the control channel, we must now add a second receiver that will allow us to follow voice channels.
In the main UniTrunker Window, Add a Debug receiver (the video shows a bit differently, follow this step instead)
Check the box which corresponds with the type of audio you want to listen to (P25/VSELP/ProVoice)
Now for some math. Take the average of your lowest and highest frequency channel. This is done to ensure that the time it takes for the SDR to switch channels is even. SDRs do take a few ms to change their frequency, and you want to have the lowest latency possible. This will vary by radio system, and needs to be changed when you want to listen to a new radio system.
For example: My highest frequency is 856.16250, and lowest is 854.1375. The average is 855.15. So I would put 855.15 into the box.
In the main window under options enable listen
Press play on the debug receiver.
A new file should be created in ./SDR/UniTrunker called sdrsharptrunking.log
This is the file SDRSharp will read and follow, so we can get voice out of the program.
If you do not see sdrsharptrunking.log, you have not properly installed Visual C++ redist.
Following the Voice channel in SDR#(no video)
Now open SDRSharp in the SDR folder named VOICE
Press Start and duplicate my settings
Now we can configure the plugin
On the bottom of the left settings panel in SDRSharp there is a plugin called “Trunker (Plugin)“
Click set and find the UniTrunker install directory.
Also set Single log file output to that same directory.
Click on show options and in the log style box, type %t %s
Set the Parked Str as NOCALL
Check both “enable” boxes
Your VFO should match the parked frequency (if there is no call)
Now, when a call appears, UniTrunker will direct the plugin to write sdrsharptrunking.log, which will get read by SDRSharp, and the audio coming out of SDRSharp can be listened to (on analog systems), or for digital, there will be just one more step.
This is the final mandatory step in decoding P25 voice.
Download this batch file, and put it into the same folder with dsd.exe and cygwin1.dll
It will start DSD without having to open command prompt
Make sure your default audio output is still your sound card, and your input is the virtual audio cable
In most cases, this should work without problem
If you want to see what options are available in dsd.exe (to get better decodes) open a command prompt, navigate to the dsd folder, and type in ‘dsd.exe –help‘
Personally I use ‘dsd.exe -l -f1 -mc -u 9‘. But it varies by system.
That’s it! You should be hearing audio coming from DSD.
Don’t forget you can set call priority and lockout in UniTrunker
Here are a couple extras
Programs for Streaming
I advise you to make only unlisted streams. I know I can’t stop you from making public streams, but If you like what you hear, and don’t want to lose it. Stay far away from streaming services that allow criminals to tune in. This is the number one reason why cities are going encrypted.
Programs for recording
Audacity – has a voice activated recording mode to save silence
Log Recorder – based on SOX, written in batch. Puts each conversation into a folder based upon the date, year, in chronological order. Outputs mp3’s named by their TG and RID. Highly configurable