Quote from Old Forum on 7. 11. 2018, 10:22

admin wrote:

Have you tried direct RDS decoding with RDS Spy via 192 kHz sound card?
Other users will be pleased if you provide a simple information about your sound card brand name/model type, operating system and add your experience with that hardware.

Quote from Old Forum on 7. 11. 2018, 10:22

admin wrote:

The direct RDS decoding facility has been originally developed and successfully tested with Creative E-MU 0204 USB under Win XP and 7. I can recommend this sound card.

In Windows 7 the E-MU 0204 input registers itself as "microphone". This is not optimal as Windows 7 automatically assign some stupid digital preamplifier to this sound card causing the audio data getting overloaded although the input signal is about 20 dB below A/D converter's saturation. The "microphone" sensitivity must be adjusted in the sound card's system settings.

Quote from Old Forum on 7. 11. 2018, 10:23

Jens_Mander wrote:

It works perfectly with E-MU 0404 PCI. It is necessary to add a new profile in the "PatchMix DSP"-Software to switch to 192 kHz.

Quote from Old Forum on 7. 11. 2018, 10:23

benko5009 wrote:

Hi All,

At the moment I'm using the following solution:
Tuner -> MPX out -> external RDS Decoder(TDA7330B)->Data, Clock, Ground -> PC Soundcard line-in -> RDS Spy

Using Tuner -> MPX out -> PC Soundcard(192MHz) -> RDS Spy will it be better considering speed and sensitivity ?

BR,
Balazs

Quote from Old Forum on 7. 11. 2018, 10:23

admin wrote:

Thank you for this question. I cannot guarantee it for all cases but in my case the 192kHz sound card solution is considerably more sensitive than the solution based on Conrad RDS Manager and Clock & Data. It is probably due to DSP processing which provides excellent spurious rejection, maximum precision and insensitivity to input level adjust.
When BER is 75% with Conrad, I get 40% with 192kHz sound card.
The Conrad's analogue part is very well designed, it contains extra tuned LC bandpass filter. This is usually not a part of common TDA7330 based decoders so I suppose that difference would be really significant in your case.

Quote from Old Forum on 7. 11. 2018, 10:23

benko5009 wrote:

THank you for quick answer !

Next week I'll receive a Creative E-MU 0404 PCI card, and can make some tests.

BR,
Balazs

Quote from Old Forum on 7. 11. 2018, 10:23

admin wrote:

Let us know your results. Thanks!

Quote from Old Forum on 7. 11. 2018, 10:23

Quote from Old forum on 7. 11. 2018, 10:23

Jens_Mander wrote:

It works perfectly with E-MU 0404 PCI. It is necessary to add a new profile in the "PatchMix DSP"-Software to switch to 192 kHz.

benko5009 wrote:

I also have E-MU 0404 PCI. When I want to add a new profile there are no possibility to switch to 192 kHz. I see only two possibilities to choose from: 44k/48k and 88k/96k 

Can somebody help ?

Update: Sorry, I forgot to install an update Now, 192 kHz is available.

Thank you,
Balazs

Quote from Old Forum on 7. 11. 2018, 10:24

benko5009 wrote:

Here are my results of compare the two solution:

1.)
Tuner -> MPX out -> external RDS Decoder(TDA7330B)->Data, Clock, Ground -> PC Soundcard line-in -> RDS Spy

2.)
Tuner -> MPX out -> PC Soundcard(192MHz) -> RDS Spy

a.)
Not easy to compare, and not a significant difference, but in my case the BER value in the second case(direct MPX) seemed to be better(lower) by 10%

b.)
I tried to get at least the PI code of a week station since few days using the first solution (TDA7330) but without success.
Today, as I started to use the second solution(direct MPX), RDS Spy managed to get the PI code !

So, I can confirm that Creative E-MU 0404 PCI with Windows XP SP3 works perfectly !

Here is a picture on RDS Spy "Status Information" using direct MPX:

Best regards,
Balazs

Quote from Old Forum on 7. 11. 2018, 10:24

admin wrote:

For your information, the first waveform in the Status window shows the signal after synchronous demodulation and filtering. Effectively it is a 1187.5 Hz sine wave that inverts its phase on each occurrence of log. 1 in original RDS data.
The second waveform shows the original data, i.e. this signal is equivalent to standard Data signal as present on output of the common RDS demodulator IC.

Quote from Old Forum on 7. 11. 2018, 10:24

admin wrote:

Beware of Chinese low cost "HiFi Super Pro 192 kHz Sound Card" products offered mainly on eBay (DAC68 etc.). None of these toys can record using 192 kHz sampling rate. Always check for the "192 kHz recording" explicitly mentioned!

Based on the experience, I recommend never to buy any low cost Chinese equipment if you are not 100% sure that you know what you're doing!

Quote from Old Forum on 7. 11. 2018, 10:24

benko5009 wrote:

Hi All,

My laptop also works well with Direct RDS/MPX (192 kHz)
HP 620 WT261EA#AKC
OS: WIN7

Sound- video- and game devices:
High Definition Audio sound device
Intel(R) High Definition Audio HDMI

You can set:
2 channels, 24 bit 192 kHz

I think some other laptops/notebooks also usable installing the appropriate audio driver

BR,
Balazs

Quote from Old Forum on 7. 11. 2018, 10:24

DaveB wrote:

How can I tell for sure that the audio card in my laptop CAN actually handle 192kHz audio input sampling? Please note the following points.

Laptop is a Packard Bell EasyNote LM86
Realtek HD Audio – audio codec ALC272X
Latest drivers downloaded from Realtek which allow me to set the mic input to 16bit 192000Hz (Studio Quality). There is a recording device that allows you to route the output from the line out to recording software / RDSSpy which can be set to 16bit 192000Hz.
Relatek websitr says the ALC272 codec supports 192kHz input sampling – but codec AL272X was not listed – even though the drivers were sucesfully updated.
But I cannot get RDSSpy to decode in direct mode – I can certainly prove that I am feeding the speaker output from SDR# to RDSSpy via StereoMix – and the internal RDS decoder in SDR# is giving full decoding. I copied a link to a picture of the status information window. The top waveform doesn't look right – its not a nice sine wave as shown in the screenshot posted by someone else.

https://app.box.com/s/vmhm6dtzt1cnp4v7k4oc

So any help – even if it to say – "that waveform proves you are not actually feeding 192kHZ samples to RDSSpy.

Thanks

David

Quote from Old Forum on 7. 11. 2018, 10:24

admin wrote:

Looking at your screenshot: Although the data is coming on the 192 kHz sampling rate, there's no RDS information included in the signal. The difference between peak signal level (complete demodulated MPX signal) and relative RDS level is usually about 25 dB. This ratio is almost constant because it is given by broadcast standards and appropriate FM deviation caused by MPX and RDS.

On your screenshot this ratio is over 60 dB. The RDS has been filtered out somewhere, it seems there's only an ADC noise. There are several reasons possible:
1) 192 kHz not supported by the audio codec hardware (I was unable to find a difference between ALC272 and ALC272X)
2) There's a low-pass filter in analogue path on the board so sampling is 192 kHz but no RDS goes into the chip
3) There's some "audio effect" enabled in the driver
4) A resampling occurs somewhere inside the driver/mixer.

Try to run some audio editing software, for example Gold Wave and monitor the recording. You'll see how the original MPX signal is filtered. In correct case the FFT window should look like on the picture attached. Based on the filter slope, the reason may be found.

 

Quote from Old Forum on 7. 11. 2018, 10:25

DaveB wrote:

Thanks for reply – nothing showing for the 57kHz sub carrier. So I googled "SDR# and RDSSpy" which yielded the inoframtion that basically the audio is filtered using a 16KHz lpf and the chances of finding an RDS signal were zero. That was in July 2012 – there is an option to have filter/unfiltered audio but ticking/unticking this option makes no difference to the audio spectrum showing in Goldwave. So I suppose the m=next move is to try a different piece of SDR software – one which is known to work with RDSSpy.

I'll post my results when I get the chance to do it.

David

Quote from Old Forum on 7. 11. 2018, 10:28

DaveB wrote:

Goldwave has allowed me to see what is going on. The microphone input and Stereo Mixer are definitely sampling at 192kHz – when I set sampling rate to 96kHz the audio spectrum display cuts off at 48kHz, set it to 48kHz and the audio spectrum cuts off at 24 kHz.
But at 192khz although the audio spectrum goes to 96kHz there is no sign of the 57kHz RDS subcarrier – it looks like there is a sharp roll-off above 48 kHz audio frequency with the max attenuation around 60 KHz. All audio enhancements were turned off on the microphone and Stereo Mixer – so it looks like there is an LPF in the analogue chain.

How do you find out if a soundcard has internal audio filtering unless you try it out?

More annoying is the HDSDR sounds much better than SDR# and has a much lower CPU usage.

Thanks for help

David

Quote from Old Forum on 7. 11. 2018, 10:28

admin wrote:

The X axis is going up to 96 kHz so the GoldWave is sampling at the 192 kHz rate. But if the real roll-off is around 48 kHz and it is sharp, it is almost sure that this is not due to analogue filtering but due to the fact that original sampling rate is only 96 kHz, half of the required. Usually the driver allows sampling rate of up to 192 kHz also if the original source does not support it. It's probably for compatibility purposes, if you record audio only, it is not a real problem. How it works? The original signal is resampled by software, additional samples are inserted between existing samples so it results in the 192 kHz rate. Of course, this method cannot create any new information in the signal so the upper side of the spectrum contains only aliasing products which are finally filtered by digital low-pass filter.
In your case, it may be a driver issue, hard to say…

Quote from Old Forum on 7. 11. 2018, 10:28

DaveB wrote:

Thanks for reply – a post in Skywaves made me think I needed to use VAC. I found some instructions on the web of how to connect HDSDR to RDSSpy via VAC.

I ended up with setting the output from HDSDR to 192kHz, setting up a virtual audio cable, and sending the audio output from the HDSDR to the VAC (192kHz in and out. Using Audio Repeater (KS) that comes with VAC I set the input to the virtual audio cable and the output to Realtek HD Audio Device with 192kHz sampling rate. That gave me back the sound – next I set RDSSpy to the Stereo Mixer input and it worked! I checked the waveform using Goldwave and a nice strong 57KHz double peak with the spectrum plummetting to zero above 76kHX (roughly). According to the status information page the Relative RDS signal is about 25dB less than the input signal level.

So there we have it – any laptop with Realtek HD audio should be able to be interface HDSDR and RDSSpy.

Thanks for all your help.

David

Quote from Old Forum on 7. 11. 2018, 10:28

LXS wrote:

Hi all,

I'm just trying to get my radio connected to the soundcard, the setting is a Grundig G5 with MPX-output and direct MPX-input to a 192 kHz-soundcard (ASUS Xonar DX). Yet I don't get any RDS signals to RDS Spy.

What I would like to know: Which contacts of the 3,5 mm-plug do I have to connect to the MPX-signal? Both left and right? Actually only the left one is connected to MPX-signal, the right one is still connected to line out/right cannel of the radio.

Hope that someone can help, thanks to all!

Quote from Old Forum on 7. 11. 2018, 10:28

admin wrote:

The Grundig G5 is not equipped with MPX output. It's the Line output, providing signal after stereo decoder. In that case it is absolutely necessary to connect both left and right.