DSD Windows Port

[racingfan360]

From the DSD readme file:

"e:" is the beginning of the errorbars display. Each "=" indicates a
detected error within the voice data. "R" and "M" indicat that a voice
frame was repeated or muted due to excessive errors.

HTH,

Jim

 

[sealltech]

This is a great port over to windows noobs download v2 of the dsd port works without any headache. decided to take it a step farther and make an all in one unit... Works very well.

Mototrbo scanner v2 build - YouTube

 

[SCPD]

Sealtech. Like the setup you have for DSD. Where did you get the monitor for the front of the tower?

 

[sealltech]

it was a 7 in monitor that I got from a local car audio shop in town just had it laying around. Took it and molded it in. But now on tigerdirect they have a 7 inch touch screen monitor that uses usb for power and video for around a hundred bucks. those 7 inch monitors with very little work fit very well in the cd tray slots. Then you can set it as the primary monitor and setup dsd to autostart with windows then bingo you have a scanner that requires no keyboard mouse or external monitor.

 

[SCPD]

Sealltech: Thanks for the info. Can you link me to the td website with the monitor?

 

[w9ran]

I am decoding P25 using a Realtek 2832u RTL-SDR USB dongle, the supplied antenna, SDR# (SDR sharp) and DSD.

Great to know it can be done! I too have been having a blast with the DVB-T dongle, and am hoping to get P25 decoding working next. I can get audio into the VAC but am not sure exactly how to tell DSD to use the output of VAC as its audio source. The default is /dev/audio - but how do you set that in the Windows world? This is on a new Win7 machine that I'm less familiar with than XP...

Am I correct in assuming that the output from DSD will go to the speakers regardless of VAC? Too much new software at the same time ;-)

Let me recommend the SDSSharp group: SDRSharp@yahoogroups.co.uk and especially the cool things Bob Rich is doing to add automatic tuning (just like a scanner), logging, and more. Check out his video demos on Youtube at "jcims" channel.

Thanks and especially thanks for the Window port of DSD.

73, Bob W9RAN

 

[DaveNF2G]

According to the instructions for VAC, if you want to hear the audio that is going through the VAC, then you also have to designate an electronic path to your soundcard.

 

[w9ran]

Looking for help with VAC and DSD under WIn7, running on a 3.4 Ghz dual core Pentium D machine.

I can configure my SDR to put its output on Virtual Audio Cable 1 and I can select the Wav Out device in VAC Audio Repeater to "Speakers (Hi Def. Audio)" and hear raw P25 audio in the speakers. But I am not sure how to correctly set the Wav Out to send audio to DSD, and moreover DSD does not seem to be working right no matter what I try (see error messages below)

Here is my litany of questions:

VAC Audio Repeater:
1. What should Wav Out be set to to send audio to DSD? The options are: None, Microsoft Sound Mapper, Line 1 VAC, Speakers (Hi Def Audio), or Line 2 VAC.
2. What else needs to be configured in VAC? I'm assuming only one audio cable is required. I can start and stop the sound with Audio Repeater so I think that much is working at least.


Windows 7 Sound configuration:
1. What should the default Playback device be?
2. What should the default Record device be?
3. What is Microsoft Sound Mapper and how do I configure it (this seems to be a point of confusion).
4. Since DSD is looking for audio on /dev/audo - How do I configure what that is?

DSD ERRORS: I have had no success getting DSD to work using the following three tests:

1. When I run (from command line) DSD (no options) I get the following and nothing else (i.e. no errorbars display or anything)
Digital Speech Decoder 1.4.1
mbelib version 1.2.3
Error, couldn't open /dev/audio
ioctl reset error
ioctl speed error
ioctl stereo error
ioctl setfmt error
Audio In/Out Device: /dev/audio (then nothing until I use CTRL-C to exit the DOS shell)

2. When I run (from command line with the sample file in the same directory) DSD -r p25sample1 - I get the following and nothing else:
Digital Speech Decoder 1.4.1
mbelib version 1.2.3
Error, couldn't open /dev/audio
ioctl reset error
ioctl speed error
ioctl stereo error
ioctl setfmt error
Audio Out Device: /dev/audio
Error: could not open p25sample1
1 [main] dsd 192 exception::handle: Exception: STATUS_ACCESS_VIOLATION
581[main] dsd 192 open_stackdumbfile: Dumping stack trace to dsd.exe.stackdump

(I can send the stackdump file if anyone wants to see it)

3. When I run: DSD -r p25sample1.wav (adding the .wav extension) - I get the following and nothing else (no audio)
Digital Speech Decoder 1.4.1
mbelib version 1.2.3
Error, couldn't open /dev/audio
ioctl reset error
ioctl speed error
ioctl stereo error
ioctl setfmt error
Audio Out Device: /dev/audio
Error: unrecognized file type
Playing p25sample1.wav (but no audio is heard)

I am at a loss to explain why I can't even get DSD to play the sample file, but I get the exactly the same errors when I repeated the above tests on my laptop running XP. Can't help but think I'm (consistently) missing something obvious.

Clearly there are many possibilities of things that are incorrect, and I'm hoping that someone who has more familiarity with the DSD and VAC settings can help me eliminate some of the possibilities.

Thanks and 73,
Bob W9RAN

 

[ilgrant]

Due to the highly technical nature of the USB Dongle and it's attempted integration with DSD is it possible to create a separate thread?

 

[w9ran]

I'll be happy to do that and I'll also volunteer to write up a detailed FAQ document for others who would like to take advantage of this very inexpensive way to listen to digital signals after I figure it out myself.

As per my previous post however, I am unable to get DSD (dsdfor win) to run at all on either of three PCs tried (two XP and one Win7).

I would really appreciate it if someone who knows the DSD.EXE code could tell me what the above errors are, and what to do about them. I've searched extensively but everything points back to this forum, and this or the original DSD for Linux thread.

It appears that DSD has a problem with my hardware. All three of my PCs are DELLs - has anyone encountered these errors with DELL machines, or when using the DELL integrated audio system to drive DSD?

Bob

 

[DaveNF2G]

dev/audio should be replaced with dev/dsp

I am using a Dell Inspiron laptop under Windows XP.

 

[sealltech]

Xeno194 This would probably work for that purpose. Hope that helps

http://www.amazon.com/iMo-eye9-9-inch-Touchscreen-Monitor/dp/B004FG6XEC

 

[w9ran]

Dave - NF2G - thanks for that suggestion but it still won't work that way for me. At least when I enter "dsd -i dev/dsp" from the command line, I get the same series of error messages along with an error about dev/dsp tacked on. But I found and tried the "version 2" DSD code on one of the DSD-related threads, and that version works for me on all three PCs, both XP and Win7. The error is definitely related to the input switch but thus far no responses from those in the know.

73 Bob W9RAN

 

[mtindor]

Dave - NF2G - thanks for that suggestion but it still won't work that way for me. At least when I enter "dsd -i dev/dsp" from the command line, I get the same series of error messages along with an error about dev/dsp tacked on. But I found and tried the "version 2" DSD code on one of the DSD-related threads, and that version works for me on all three PCs, both XP and Win7. The error is definitely related to the input switch but thus far no responses from those in the know.

73 Bob W9RAN

1. Default Playback should be set to your normal soundcard speaker output
2. Default Recording Device should be set to Virtual Audio Cable
3. DSD for Win needs to be run with:

-i /dev/dsp -o /dev/dsp

Specifically, you don't get to choice which soundcard you are using via a commandline switch in DSD for Win. To overcome that, you set the Default Recording Device and Default Playback Device to be Virtual Audio Cable and your normal soundcard speaker output respectively.

For the purposes of DSD for Win, interpret "-i /dev/dsp" as mapping to your Default Recording Device and "-o /dev/dsp" as mapping to your Default Playback Device

Normally I have:

Default Playback Device: my laptop speaker audio device
Default Recording Device: Virtual Audio Cable

and in my SDR software I have:

input: Funcube Dongle (insert your SDR devie here)
output: Virtual Audio Cable

And if I want to hear the raw output from the SDR software through my speakers as well, I'll fire up Virtual Audio Cable's Audio Repeater (MME) and set:

Wave In: Virtual Audio Cable
Wave Out: Speakers / Headphones (Sigmatel)

Mike

 

[mtindor]

Looking for help with VAC and DSD under WIn7, running on a 3.4 Ghz dual core Pentium D machine.

Bob,

To further clarify my previous post...

* I use a Funcube Dongle
* I run Windows Vista
* I use the built-in (Sigmatel) audio device for output

1. I plug my Funcube Dongle in so that the audio device exists in Windows Vista

2. I set Default Recording Device to Line 1 Virtual Audio Cable

I usually set it's audio level to "2" -- but I think that is specific to me using a Funcube Dongle, which doesn't work well if it's audio level is too high when it's fed into the SDR software.

3. I set Default Playback Device to Speakers Headphones (Sigmatel High Definition Audio Codec)

4. I fire up my preferred SDR software (SDR-Radio) and configure it's input to be Microphone (Funcube Dongle V0.0) and it's output to be Line 1 (Virtual Audio Cable)

* see attached images at bottom of post

5. My audio on my laptop doesn't have an option for "What You Hear", so I can't simply listen to what's being input from the SDR Software into DSD on my speakers without using VAC. I rarely want to hear what's being output from my SDR software when I'm running DSD. If I'm not running DSD/DMRDecode and I simply want to listen to standard FM Audio output from the SDR on my speakers, I just change the output in the SDR software to be my speakers instead of Line 1 Virtual Audio Cable.

If I want to listen to the output from the SDR while I'm running DSD / DMRDecode, I will fire up Virtual Audio Cable's Audio Repeater (MME) and set it up like this:

6. I use multiple versions of DSD for Win. I use the original found on these forums and then I use the version modified by Woodpecker to add a filter [which helps increase decode quality in DSD for TRBO (and supposedly P25 and maybe NXDN). I often monitor weak signal TRBO, and the version put out by Woodpecker [with the filter in it] works much better for me.

If you're looking to try Woodpecker's version, I have a copy [which may or may not be the latest] here:

Index of /dsd/dsd_with_filter_by_woodpecker

NOTE: Woodpecker's DSD is for Windows and there is no source code.

If I'm running vanilla DSD for Win, I use this commandline:

dsd -i /dev/dsp -o /dev/dsp

If I'm running Woodpecker's DSD for Win [with filtering], I use the following commandline:

dsd_filter -ff -i /dev/dsp -o /dev/dsp

* -ff is a switch to enable filtering
* dsd_filter is just what I renamed dsd.exe so it could be run from same directory

NOTE: Because o Cygwin limitations, you can't reference specific audio devices -- only the defaults. So, for DSD for Win, -i /dev/dsp always refers to the Default Recording Device and -o /dev/dsp always refers to the Default Playback Device.

Because of the above, this means that you can also abbreviate commandline since DSD for Win defaults to using /dev/dsp for input and output.

dsd [plus commandline switches] - ex: dsd -mg , dsd -f1, etc.
I hope this is helpful to you.

Mike
----------------------------------------

 

[w9ran]

Mike - Thanks VERY much for taking the time to post this detailed and specific setup information. I'm sure it's going to be appreciated by others like me who come to this thread looking for help.

I wonder if you might be able to do two more things that might help me improve the quality of P25 decodes. I've tried Woodpeckers filtered version, but like the original I am still getting a lot of errors in the errorband display along with enough audio dropouts that it is hard to understand most transmissions. The quality of the voice decoding does not improve much by varying inlvl - in fact the quality (or lack thereof) is fairly consistent from "0%" to 60% or more, or about 90% of the SDR# audio gain slider. At the "best" audio gain setting I'm still getting a lot of errors. To try to eliminate possible causes, it would be helpful to know the following:

1. The filter type, bandwidth, and order settings from SDR# (and any other settings that might be germane)

2. A screen capture of the Virtual Audio Cable Control Panel showing all parameters for a known working VAC setup.

Any information as to the required RF signal level needed for error-free P25 decoding would also be appreciated.

This is amazing software and it's been a great learning experience - many thanks to those who have shared their wisdom and expertise!

Bob

 

[mtindor]

Do I have to install Cygwin in order to run this on my windows machine? Guess looking for a run down of the steps? Thank you for this.

No, you shouldn't need to. You should only need to have Cygwin installed if you are going to compile the DSD source.

Read the thread - the whole thing is basically a rundown of steps.

Mike

 

[mtindor]

I wonder if you might be able to do two more things that might help me improve the quality of P25 decodes.

Bob,

1. A screenshot of my VAC Control Panel is at the bottom of this post. However, if you use my screenshot I'm sure you are prone to more problems than you already have. Depending upon what the input audio device supports and your sound card supports, your settings might have to be different than mine. I run my VAC across the board as 96000 / 16.

2. I set the audio level of my "Funcube Dongle" Recording Device (in the sound settings) so that it's audio level is extremely low -- like "2". I do this because too much gain into the SDR software from the audio recording device [Funcube Dongle] will result in a really bad S/N ratio. I don't know if that will happen with your audio device. In general, for the SDR component's "recording device" I would suggest setting the audio level very low in your Windows sound settings for that device in order to make sure that you aren't feeding more noise than signal into the whole setup from the start.

3. I adjust my AF Output (in SDR-Radio) so that my INLVL in DSD is about 20-25% for the stuff I monitor, mostly TRBO. Remember, the SDR software is feeding it's output into VAC [for use by DSD / DMRDecode / Unitruner]. So I just use the output slider in SDR-Radio to manage the INLVL in DSD.

4. As far as filter type, bandwidth, and order settings -- I'm not sure what you mean by "order" settings, and I'm not aware that I can set the "type" of filter. I'll have to try and find out what type of filter is being used, if any at all.

The bandwidth -- well, I guess I'd have to show you a screenshot of what SDR-Radio looks like when I'm decoding. I adjust the bandwidth for every signal basically. And for the next day or two I'm not using an outside antenna, so I can't monitor any TRBO stuff. I can't monitor any P25 stuff either to give you a screenshot because of that. I'll try to get you a screenshot of that in the next couple days.

Mike

 

[w9ran]

Mike, thanks and understood. I was mistakenly thinking you were using SDR# (SDRSharp), but maybe someone who is getting decent P25 decodes with it could pass along what Filter settings they're using (and their VAC settings).

Reducing the FCD audio level makes sense as you describe. The difference is the DVB-T dongle provides I/Q data at 2 megasamples/sec directly via USB rather than going through the soundcard.

SDR# fixes its audio sample rate at 48000 and provides a volume slider, but I've not been able to eliminate the DSD errors at any volume level when using VAC to pipe this signal into DSD. So that's why I'm trying to see if VAC is configured correctly, or if Windows is doing any kind of resampling behind the curtain. When I locked VAC to 48000 (just as you've done at 96000), SDR# produces the "Pa UnanticipatedHostError" and won't run, but that's clearly ans SDR# issue.

73, Bob W9RAN

 

[mtindor]

Bob,

Just out of curiosity, what is your CPU specs? I run a Core 2 Duo T5750 Mobile CPU. 2 cores, no hyperthreading. This thing doesn't work well with either DSD or DMRDecode. On most decodes [with the SDR software running and already using 20-30% CPU] its choppy with errors because the CPU is maxing out.

I imagine you're running something with higher specs than this. but i you aren't, that's going to be a big part of your problem.

Mike