BSID Bug?

[Trip]

Nick,

I'm not sure where you're getting the BSID from in plpinfo or streaminfo, but it doesn't show for any of my local stations even though it's in the SLT opening tags on three of my four 3.0 stations.

Code: Select all

0: sfi=0 mod=qam16 cod=9/15 layer=core ti=hti lls=1 lock=1
1: sfi=0 mod=qam256 cod=11/15 layer=core ti=hti lls=1
16: sfi=0 mod=qpsk cod=2/15 layer=core ti=hti lls=0

5002: 145.1 WBFF (no data)
5003: 111.1 WBAL (no data)
5004: 102.1 WMAR (no data)
5006: 122.1 WMPT (no data)
5007: 154.1 WNUV (no data)

<SLT bsid="1408">

Code: Select all

0: sfi=0 mod=qpsk cod=6/15 layer=core ti=cti lls=1 lock=1
1: sfi=0 mod=qam64 cod=6/15 layer=enhanced ti=cti lls=0

5001: 158.1 WIAV (no data)
5006: 158.5 24/7MMT (no data)

<SLT bsid="9100" {http://www.w3.org/2001/XMLSchema-instance}schemaLocation="tag:atsc.org,2016:XMLSchemas/ATSC3/Delivery/SLT/1.0/ SLT-1.0-20211209.xsd">

Code: Select all

0: sfi=0 mod=qam64 cod=11/15 layer=core ti=hti lls=1 lock=1
16: sfi=0 mod=qpsk cod=2/15 layer=core ti=hti lls=0
1: sfi=1 mod=qam256 cod=11/15 layer=core ti=hti lls=0

1: 132.1 WHUT (no data)
2: 107.1 WJLA
3: 105.1 WTTG (no data)
4: 104.1 WRC (no data)
5: 109.1 WUSA (no data)

<SLT bsid="540">

Meanwhile, I can't get a PCAP to look directly at the SLT for my tuner in Virginia Beach, but:

Code: Select all

0: sfi=0 mod=qpsk cod=9/15 layer=core ti=hti lls=1 lock=1
1: sfi=1 mod=qam256 cod=8/15 layer=core ti=hti lls=0
bsid=0x0bf0

1: 110.1 WAVY (no data)
2: 143.1 WVBT (no data)
3: 145.1 WNLO-CD (no data)
4: 133.1 WTVZ (no data)
8: 143.2 WVBT-CW (no data)
tsid=0x0BF0

Not sure if this is something that can be updated in a future firmware release.

- Trip

[nickk]

Just got back to AZ - storm canceled 40+ flights out of DC including mine

The issue is that the HDHomeRun is not reporting the bsid for some stations correct?

Nick

[nickk]

Found the code - the firmware gets the bsid from l1detail.

[drmpeg]

Found the code - the firmware gets the bsid from l1detail.

The bsid only appears in the L1_Detail if L1D_version is >= than 1. I'm gonna guess that there are plenty of transmitters sending a L1_Detail with L1D_version = 0 and no bsid.

[Trip]

Just got back to AZ - storm canceled 40+ flights out of DC including mine

The issue is that the HDHomeRun is not reporting the bsid for some stations correct?

Nick

Yikes. Glad you made it back safely. It was a good meeting.

Yes. Would it be possible to go to the L1D first and if that's empty, fall back on the SLT?

- Trip

[nickk]

I have the following working...

1) Added a new var /tuner<n>/l1detail to TECH and DEV models that returns the l1basic+l1detail binary data (base64 encoded).

2) No change to plpinfo - it returns the bsid from l1detail (if present). This preserves layering - all plpinfo information comes from l1detail.

3) streaminfo now returns bsid=0xnnnn from the SLT. This preserves layering - all streaminfo information comes from the SLT.

4) streaminfo no longer returns tsid=0xnnnn on ATSC3 channels (this was previously the bsid from l1detail... badly named and from the wrong layer).

Will this work for your usage?

[Trip]

I have the following working...

1) Added a new var /tuner<n>/l1detail to TECH and DEV models that returns the l1basic+l1detail binary data (base64 encoded).

2) No change to plpinfo - it returns the bsid from l1detail (if present). This preserves layering - all plpinfo information comes from l1detail.

3) streaminfo now returns bsid=0xnnnn from the SLT. This preserves layering - all streaminfo information comes from the SLT.

4) streaminfo no longer returns tsid=0xnnnn on ATSC3 channels (this was previously the bsid from l1detail... badly named and from the wrong layer).

Will this work for your usage?

That sounds fantastic. Thanks so much. I really appreciate you taking the time on this!

EDIT: Does the change to streaminfo mean that the search API function will return bsid= instead of tsid=? I'm just wondering if it will break code in other people's projects that is looking specifically for the "tsid" field regardless of standard.

Also, would it be possible to add a var that dumps the entire SLT XML directly, preferably without needing to be TECH or DEV? Would be massively helpful for updating RabbitEars...

- Trip

[nickk]

lineup.json?tuning now reports the SLT BISD as TransportStreamID. Previously this was the L1Detail BSID (if present).

I have reverted the var rename in streaminfo back to tsid to ensure compatibility. The value comes from the SLT as discussed.

The HDHomeRun doesn't store the SLT today. I will see what I can do.

[Trip]

lineup.json?tuning now reports the SLT BISD as TransportStreamID. Previously this was the L1Detail BSID (if present).

I have reverted the var rename in streaminfo back to tsid to ensure compatibility. The value comes from the SLT as discussed.

The HDHomeRun doesn't store the SLT today. I will see what I can do.

Sounds great. I really appreciate this and will look forward to the next release!

- Trip

[nickk]

20250805beta1 firmware release is out...

[Trip]

20250805beta1 firmware release is out...

The links on the downloads page still seem to point to the current stable firmware.

- Trip

[nickk]

Direct firmware links updated and cache cleared... should be working now.

[drmpeg]

I'm in the process of writing a parser for the new L1_Detail function. Is the L1_Basic CRC supposed to be all zeroes?

Here's the BASE64 string from the HDHomeRun.

ACHgAAAAADIAFcYgYMGH////////AAAAAAEjCABAAAABRSzATlPXhAzif////w84uPc=

Which decodes to (in hex).

0021e000000000320015c62060c187ffffffffffff000000000123080040000001452cc04e53d7840ce27fffffff0f38b8f7

Here's what I have so far.

Code: Select all

./l1dump l1dump.txt
L1B_version = 0
L1B_mimo_scattered_pilot_encoding = Walsh-Hadamard pilots or no MIMO subframes
L1B_lls_flag = false
L1B_time_info_flag = Time information is not included in the current frame
L1B_return_channel_flag = 0
L1B_papr_reduction = No PAPR reduction used
L1B_frame_length_mode = symbol-aligned
L1B_time_offset = 3840
L1B_additional samples = 0
L1B_num_subframes = 1
L1B_preamble_num_symbols = 1
L1B_L1_Detail_content_tag = 0
L1B_L1_Detail_size_bytes = 25
L1B_L1_Detail_fec_type = Mode 1
L1B_L1_additional_parity_mode = K=0 (No additional parity used)
L1B_L1_Detail_total_cells = 2787
L1B_first_sub_mimo = No MIMO
L1B_first_sub_miso = No MISO
L1B_first_sub_fft_size = 32K
L1B_first_sub_reduced_carriers = 5.832 MHz
L1B_first_sub_guard_interval = GI3_512
L1B_first_sub_num_ofdm_symbols = 49
L1B_first_sub_scattered_pilot_pattern = SP24_2
L1B_first_sub_scattered_pilot_boost = 1.718
L1B_first_sub_sbs_first = present
L1B_first_sub_sbs_last = present
L1B_crc = 0

[nickk]

Ah, right...

The Sony demod driver returns the L1basic in a struct with everything split out. My tuner driver wrapper code puts it back together into the raw form before passing it to my ATSC 3 code. The CRC is the one thing that is missing so my conversion code sets it to zero.

[drmpeg]

Ah, right...

The Sony demod driver returns the L1basic in a struct with everything split out. My tuner driver wrapper code puts it back together into the raw form before passing it to my ATSC 3 code. The CRC is the one thing that is missing so my conversion code sets it to zero.

Okay, no problem. The CRC is not really that useful anyways.

I'll post a Github link when the parser is finished.