Index of /programs/tsdecrypt/

File Name  ↓ File Size  ↓ Date  ↓ 
5.5 KiB2013-Sep-12 10:53
6.3 KiB2013-Sep-12 11:12
14.7 KiB2013-Sep-16 09:09
137.7 KiB2013-Sep-12 10:54
68.7 KiB2011-Sep-20 09:34
73.3 KiB2011-Sep-30 10:44
73.7 KiB2011-Nov-11 23:59
73.8 KiB2011-Nov-14 10:32
77.5 KiB2011-Nov-18 16:31
83.0 KiB2011-Dec-23 10:39
90.8 KiB2012-Feb-24 11:26
85.8 KiB2012-Mar-07 08:26
128.8 KiB2012-Apr-11 22:19
129.2 KiB2012-Apr-19 10:51
135.0 KiB2012-Sep-10 19:38
20.4 KiB2013-Sep-12 10:56
tsdecrypt reads incoming mpeg transport stream over UDP/RTP and then
decrypts it using code words obtained from OSCAM or similar cam
server. tsdecrypt communicates with CAM server using cs378x (camd35
over tcp) protocol or newcamd protocol.

tsdecrypt is released under GNU GPL v2.

tsdecrypt depends on openssl. It uses rand, MD5, DES and AES functions.
You probably already have openssl installed. If you are seeing compilation
errors related to missing openssl headers you should install openssl-dev
or similar packet depending on your Linux distribution.

If you are not using shipped FFdecsa then the other tsdecrypt dependency
is libdvbcsa. libdvbcsa can be downloaded from

FFdecsa vs libdvbcsa
tsdecrypt supports two different CSA decryption libraries.

libdvbcsa is an external library from VideoLAN project released under
LGPL, FFdecsa is shipped with tsdecrypt.

libdvbcsa is better tested and have served tsdcrypt for a long time so
it is still the default decryption library. Also it is faster with file

To compile tsdecrypt with libdvbcsa run 'make' or if you have tested
FFdecsa compilation and want to switch to libdvbcsa run 'make dvbcsa'.
'make distclean' switches decryption library back to libdvbcsa.

FFdecsa support was added after tsdecrypt v7.0. FFdecsa can be up to 40%
faster than libdvbcsa especially on 32-bit hardware. If you want to use
FFDecsa you have to run 'make ffdecsa'.

The setup script will test which FFDecsa algorithm is the fastest for
your CPU and will configure tsdecrypt compilation to stop using libdvbcsa.
Most likely the script will choose PARALLEL_128_SSE or PARALLEL_128_SSE2.
Once the script have finished it stores its decision in FFdecsa.opts file
and as long as this file exists tsdecrypt will be compiled with FFdecsa.
You can force regeneration by deleting running 'make ffdecsa_force' or
'make distclean ffdecsa'.

FFDecsa is considerably slower than libdvbcsa when tsdecrypt's input
is a file or stdin (single packet decryption mode). If you are
decrypting already recorded files you are better of using libdvbcsa.

When tsdecrypt is run it reports which decryption library is used in its
version string. You can use --bench option to test decryption speed.

The development is done using git. tsdecrypt repository is hosted

To clone the repository issue the following commands:

   git clone git://
   cd tsdecrypt
   git submodule init
   git submodule update

The code is developed and tested under modern Linux. It is also
compiled from time to time under OS X but is not tested there.

To see all Makefile targets run 'make help'.

Updating the code
To update cloned tsdecrypt, go to the directory where the repository
is cloned and run the following commands:

   git fetch origin
   git merge origin/master
   git submodule update
   make clean all

tsdecrypt's master branch should always be useful so it is safe to
use it instead of official release. The master branch will always
be better than any released version.

Command line parameters
tsdecrypt is controlled using command line parameters. For more information
about the parameters see the man page. Here is a list of supported command
line parameters:

Main options:
 -i --ident <server>        | Format PROVIDER/CHANNEL. Default: empty
 -d --daemon <pidfile>      | Daemonize program and write pid file.
 -N --notify-program <prg>  | Execute <prg> to report events. Default: empty

Input options:
 -I --input <source>        | Where to read from. File or multicast address.
                            .    -I    (v4 multicast)
                            .    -I [ff01::1111]:5000 (v6 multicast)
                            .    -I file://in.ts      (read from file)
                            . By default the input is stdin.
 -1 --input-source <ipaddr> | Set multicast input source ip.
 -R --input-rtp             | Enable RTP input
 -z --input-ignore-disc     | Do not report discontinuty errors in input.
 -M --input-service <srvid> | Choose service id when input is MPTS.
 -T --input-buffer <ms>     | Set input buffer time in ms. Default: 0
 -W --input-dump <filename> | Save input stream in file.

Output options:
 -O --output <dest>         | Where to send output. File or multicast address.
                            .    -O    (v4 multicast)
                            .    -O [ff01::2222]:5000 (v6 multicast)
                            .    -O file://out.ts     (write to file)
                            . By default the output is stdout.
 -o --output-intf <value>   | Set multicast output interface.
                            . Default for IPv4: (intf addr)
                            . Default for IPv6: -1      (intf number)
 -t --output-ttl <ttl>      | Set multicast ttl. Default: 1
 -r --output-rtp            | Enable RTP output.
 -k --output-rtp-ssrc <id>  | Set RTP SSRC. Default: 0
 -g --output-tos <tos>      | Set TOS value of output packets. Default: none
 -u --no-output-on-error    | Do not output data when the code word is missing.
 -p --no-output-filter      | Disable output filtering. Default: enabled
 -y --output-nit-pass       | Pass through NIT.
 -w --output-eit-pass       | Pass through EIT (EPG).
 -x --output-tdt-pass       | Pass through TDT/TOT.

CA options:
 -c --ca-system <ca_sys>    | Process input EMM/ECM from <ca_sys>.
                            | Valid systems are: CONAX (default), CRYPTOWORKS,
                            .   IRDETO, SECA (MEDIAGUARD), VIACCESS,
                            .   VIDEOGUARD (NDS), NAGRA, DRECRYPT, BULCRYPT,
                            .   GRIFFIN and DGCRYPT.
 -C --caid <caid>           | Set CAID. Default: Taken from --ca-system.
 -Y --const-cw <codeword>   | Set constant code word for decryption.
                            . Example cw: a1a2a3a4a5a6a7a8b1b2b3b4b5b6b7b8
 -Q --biss-key <biss-key>   | Set BISS key for decryption.
                            . Example key: 112233445566

CAMD server options:
 -A --camd-proto <proto>    | Set CAMD network protocol.
                            . Valid protocols are: CS378X (default) and NEWCAMD
 -s --camd-server <host>    | Set CAMD server address. Default port: 2233
                            . Example IPv4 addr and port:
                            . Example IPv6 addr and port: [2a00::1014]:2233
                            . Example hostname          :
                            . Example hostname and port :
                            . Example IPv4 hostname     :
                            . Example IPv6 hostname     :
 -U --camd-user <user>      | Set CAMD server user. Default: user
 -P --camd-pass <pass>      | Set CAMD server password. Default: pass
 -B --camd-des-key <key>    | Set DES key for newcamd protocol.
                            . Default: 0102030405060708091011121314
 -4 --ipv4                  | Use only IPv4 addresses of the camd server.
 -6 --ipv6                  | Use only IPv6 addresses of the camd server.

EMM options:
 -e --emm                   | Enable sending EMM's to CAMD. Default: disabled
 -E --emm-only              | Send only EMMs to CAMD, skipping ECMs and without
                            .   decoding the input stream.
 -Z --emm-pid <pid>         | Force EMM pid. Default: none
 -f --emm-report-time <sec> | Report each <sec> seconds how much EMMs have been
                            .   received/processed. Set <sec> to 0 to disable
                            .   the reports. Default: 60 sec
 -a --emm-filter <filter>   | Add EMM filter defined by <filter>.
                            . This option can be used multiple times (max:16).
                            . See FILTERING file for more info.

ECM options:
 -X --ecm-pid <pid>         | Force ECM pid. Default: none
 -v --ecm-only              | Send only ECMs to CAMD, skipping EMMs and without
                            .   decoding the input stream.
 -H --ecm-report-time <sec> | Report each <sec> how much ECMs and CWs have been
                            .   processed/skipped. Set <sec> to 0 to disable
                            .   the reports. Default: 60 sec
 -G --ecm-irdeto-type <int> | Process IRDETO ECMs with index X /0-255/
                            .   It is better to use --ecm-irdeto-chid option!
 -2 --ecm-irdeto-chid <int> | Set CHID to filter Irdeto ECMs (Default: 0x0000).
 -K --ecm-no-log            | Disable ECM and code words logging.
 -J --cw-warn-time <sec>    | Warn if no valid code word has been received.
                            .   Set <sec> to 0 to disable. Default: 60 sec

 -q --ecm-and-emm-only      | Send ECMs and EMMs to CAMD but do not decode
                            .   the input stream.

Logging options:
 -S --syslog                | Log messages using syslog.
 -l --syslog-host <host>    | Syslog server address. Default: disabled
 -L --syslog-port <port>    | Syslog server port. Default: 514
 -F --log-file <filename>   | Log to file <filename>.
 -D --debug <level>         | Message debug level.
                            .    0 = default messages
                            .    1 = show PSI tables
                            .    2 = show EMMs
                            .    3 = show duplicate ECMs
                            .    4 = packet debug
                            .    5 = packet debug + packet dump

Debugging options:
 -n --ecm-file <file.txt>   | Read ECM from text file.
 -m --emm-file <file.txt>   | Read EMM from text file.

Misc options:
 -j --pid-report            | Report how much packets were received.
 -b --bench                 | Benchmark decrypton.
 -h --help                  | Show help screen.
 -V --version               | Show program version.

To get a quick start here are some example command lines. The default
CA system is set to CONAX, you can change it using --ca-system parameter,
see man page or program help for more options.


   # Decrypt multicast stream from using
   # as camd server and output decrypted result to
   tsdecrypt --camd-server \
       --input --output

   # Same as above but enable EMM processing
   tsdecrypt --emm --camd-server \
       --input --output

   # Same as above but do not filter output stream thus allowing
   # EIT/TOT/NIT, etc tables to passthrough
   tsdecrypt --no-output-filter --emm --camd-server \
       --input --output

   # Choose program/service_id to decrypt. Useful when the input is MPTS
   tsdecrypt --camd-server --input-service 1234 \
       --input --output

   # Read stream over RTP and process VIACCESS encoded channel
   tsdecrypt --ca-system VIACCESS --emm --camd-server \
       --input-rtp --input --output

   # Decrypt stream encypted with CAID 0x0963 (NDS, sky)
   tsdecrypt --camd-server --ca-system NDS --caid 0x0963 \
       --input --output

   # Send only EMMs to OSCAM for CAID 0x0963 (NDS, sky)
   tsdecrypt --camd-server --emm-only --caid 0x0963 \
       --input --output /dev/null

   # Decrypt stream encypted with CAID 0x5581 (Bulcrypt)
   tsdecrypt --camd-server --caid 0x5581 \
       --input --output

   # Decrypt BISS encrypted stream
   tsdecrypt --biss-key 0x112233445566 --input \

   # Decrypt file encypted with constant code word
   tsdecrypt --const-cw 0x00000000000000001111111111111111 \
       --input encrypted-file.ts --output file://decrypted-file.ts

   # Send ECM from file
   tsdecrypt --ecm-file ecm.txt --caid 0x5581 --input-service 12345 \

   # Decrypt IRDETO stream from Raduga (CHID == 0x0015)
   tsdecrypt --input --output \
             --camd-server \
             --ca-system IRDETO --caid 0x0652 --ecm-irdeto-chid 0x0015

OSCAM cs378x configuration
In order for tsdecrypt to communicate with OSCAM using cs378x (camd35
over tcp) protocol you can use configuration like the examples bellow.

# Example oscam.user file:
user     = user
pwd      = pass
group    = 1
au       = 1
uniq     = 0
monlevel = 4

# Example part of oscam.conf file:
port    = 2233

OSCAM newcamd configuration
In order for tsdecrypt to communicate with OSCAM using newcamd protocol
you can use configuration like the examples bellow.

# Example oscam.user file
user     = user
pwd      = pass
group    = 1
au       = 1
uniq     = 0
monlevel = 4
ident    = 0604:000000
caid     = 0604

# Example part of oscam.conf file:
port     = 1122@0604:000000
serverip =
key      = 0102030405060708091011121314

The above example allows tsdecrypt clients using newcamd to retrive keys
from OSCAM serving CAID 0x0604 (Irdeto) on port 1122.

Reporting bugs
If you think you have found bug in tsdecrypt, please report it to the
e-mail listed in Contact section (see below) of this README file.

When reporting bugs, please send the contents of tsdecrypt.log file
generated by using the following command (fill in **YOUR_OPTIONS**):

  tsdecrypt --debug 1 --log-file tsdecrypt.log **YOUR_OPTIONS**

Also write the full command line which you used and describe what
you think the tsdecrypt does wrong.

One commonly seen error is not an error in tsdecrypt at all. You
have to make sure that the software that streams channels from the
DVB card is sending ECMs and EMMs along with other streams. If these
streams are missing you'll probably get the following messages from

2012-04-02 22:02:12 | ECM | Received 0 (0 dup) and processed 0 in 3 seconds.
2012-04-02 22:03:09 | CW  | *ERR* No valid code word was received for 60 seconds!
2012-04-02 22:03:12 | ECM | Received 0 (0 dup) and processed 0 in 60 seconds.

Please configure your software to stream ECMs and EMMs. For dvblast
(version 2.0 and above) the correct options are --ecm-passthrough and
--emm-passthrough. Note that dvblast 2.1 have broken ecm passthrough
so use at least version 2.2.

Official releases can be downloaded from tsdecrypt home page which is

For patches, bug reports, complaints and so on send e-mail to

   Georgi Chorbadzhiyski <>