Log in
Register
Menu
Log in
Register
Home
What's new
Latest activity
Authors
Forums
New posts
Search forums
What's new
New posts
Latest activity
Members
Current visitors
New posts
Search forums
Menu
Log in
Register
Install the app
Install
Forums
Satellite TV receivers & systems support forums
Satellite PC Card Receivers, Internet by Satellite
neumoDVB
JavaScript is disabled. For a better experience, please enable JavaScript in your browser before proceeding.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an
alternative browser
.
Reply to thread
Message
<blockquote data-quote="deeptho" data-source="post: 1117597" data-attributes="member: 215446"><p>Just in case it is not clear: the drivers are here:</p><p><a href="https://github.com/deeptho/linux_media" target="_blank">GitHub - deeptho/linux_media: TBS linux open source drivers</a></p><p>They can be built with the well known media_build, but I placed a version that works here:</p><p><a href="https://github.com/deeptho/media_build" target="_blank">GitHub - deeptho/media_build: media_build for compiling media tree on this site</a></p><p></p><p>In principle it should work with any version, but once in a while changes in the generic linux kernel </p><p>causes incompatibilities with linux_media, which then fails to compile. Usually the problems are in other unrelated </p><p>drivers (which is what you experience).</p><p></p><p>I update the drivers from time to time by integrating changes from the official linux_media. It would of course be </p><p>better if my changes were integrated there or if they could be compiled "out of tree", but even then keeping up with </p><p>kernel changes is a continuous effort.</p><p></p><p>I am personally running kernel 5.10.19 on fedora 34. That kernel or any older one should work in theory.</p><p>Newer ones are sometimes a problem when some incompatible changes are made in the standard linux kernel.</p><p> </p><p>What kernel version is your ubuntu using? </p><p></p><p>Perhaps you should just try compiling it and let me know if it works or not (things have changed since the initial version,</p><p>so problems may have gone or new ones may have grown).</p><p></p><p>In the longer term it would probably be best to support a few well chosen kernels, e.g., for the most used ubuntu and fedora versions</p><p>and then update them from time to time.</p><p></p><p></p><p></p><p></p><p>Well. my own experience on windows with these cards is quite poor. I have multiple hardware versions of them and</p><ul> <li data-xf-list-type="ul">ebspro, which is a pretty nice program does not work with them: no blindscan, no spectrum</li> <li data-xf-list-type="ul">crazyscan, another nice program, works with some but not all of the hardware versions of the tbs6903x and tbs6909x but only on the first tuner. The reason is probably that the driver is made for the tbs6903x and does not take into account the extra tuners on tbs6909x. The spectrum is obviously wrong (spectrum shows artificial jumps), blindscan is sometimes working, but more often not.</li> <li data-xf-list-type="ul">Some hardware versions ar enot working at all (probably because of an unoffficial hardware version number)</li> <li data-xf-list-type="ul">No success with any other programs: crashes, or all kinds of problems</li> </ul><p>I have tried ebsPRO and crazyscan on other cards, with good results. I like ebsPRO best. In any case, I avoid windows when possible.</p><p></p><p>Of course the above trouble makes it impossible to compare speeds.</p><p>In general, about the speed of blindscan: you have to be careful when comparing:</p><ul> <li data-xf-list-type="ul">Some types of blindscan give less information, or miss more transponders. </li> <li data-xf-list-type="ul">Some types of blindscan are faster than others on some satellites and slower on others</li> <li data-xf-list-type="ul">Some satellites have weird or incompliant muxes which can cause slow downs or even some software to get stuck on them</li> </ul><p>So you will have to try for yourself. In any case, there is always a balance between quality of the scan and speed.</p><p>By the way: during development I scanned most satellites between 30.0W and 53E. On several of them weird problems occurred,</p><p>which required workarounds in the code. I even encountered it mux which claimed to be on 99.9 Ghz, for example. And then </p><p>there is the problem of spill-over: one one sat position receiving data from another sat. So this full arc scan was a good exercise</p><p>to track down (and fix) problems.</p><p></p><p>The fastest version of my blindscan code is in `neumo-blindscan`, which is already on github. </p><p>It does the bare minimum: finding muxes that can be locked ad discovering which multi-streams (if any) exist. </p><p>It tries to not miss any mux (and finds most, but with some exceptions).</p><p></p><p>It is pretty fast on satellites with strong signals (e.g., a few minutes in total), and slower on other satellites, because it wastes time on investigating </p><p>noise peaks. I could make it even 4 times faster by using tuners in parallel, but that is not implemented yet and would require</p><p>some more driver changes. I think in speed it probably compares to ebsPRO if ebsPRO would support this card (it does not).</p><p></p><p>The version in neumo-DVB uses the same approach as nuemo-blindscan, so will lock the same muxes. </p><p>However, it also checks SI data, which takes at least 30 seconds extra on each mux. So </p><p>I would consider it "slow and thorough". It could be made as fast as neumo-blindscan with 2 or 3 lines of code changes.</p><p>Personally I prefer to just use the spectrum and click on some trasnponders of interest. Within a few seconds I can see if</p><p>it locks, and if I wait a little more, I know what it contains.</p><p></p><p></p><p></p><p></p><p>All code is completely different. I considered using his code, and studied it, but it did not really work on the cards I had at</p><p>the time, and I needed some incompatible changes. Also his kernel-tree was not compatible with some other drivers I needed (same </p><p>type of problem as you are experiencing). So I gave up, not because that code would have problems, but because I could not reuse it for the cards</p><p>I was really interested in anyway.</p><p></p><p>PS. Even now, I am not sure if that software supports scanning a full band, or only "blind tuning"</p><p>(i.e., clicking on the graph to tune it). If I remember correctly, this was one reason why I needed incompatible changes (support </p><p>full band blindscan)</p><p></p><p>So you will probably use updatelee or my code depending on the cards you want to use.</p><p>It is a pity that the two kernels are not compatible, but that is the way it is. It would not be too difficult </p><p>to have both approaches co-exist in the same kernel, but that requires some effort.</p><p></p><p>At some point, I also checked what changes in updateDVB would be needed to support my drivers, but I got stuck at some point for lack </p><p>of time, and never continued the search. So instead, I included the graphs ... in my own program.</p><p></p><p></p><p></p><p></p><p>The documentation has a whole section on installation and compilation. I would suggest starting from that. If something is</p><p>not clear, ask, and I will then update the documentation. As the software has never been run on Ubuntu, the first time Ubuntu-user</p><p>will probably need to make some extra effort, e.g., to find out exactly the names of the packages for prerequirsite software that neumo depends on. So if that first user takes some notes, I will add them to the documentation.</p><p></p><p>Making video is yet another effort. I might do it at some point, but isn't written version easier?</p></blockquote><p></p>
[QUOTE="deeptho, post: 1117597, member: 215446"] Just in case it is not clear: the drivers are here: [URL="https://github.com/deeptho/linux_media"]GitHub - deeptho/linux_media: TBS linux open source drivers[/URL] They can be built with the well known media_build, but I placed a version that works here: [URL="https://github.com/deeptho/media_build"]GitHub - deeptho/media_build: media_build for compiling media tree on this site[/URL] In principle it should work with any version, but once in a while changes in the generic linux kernel causes incompatibilities with linux_media, which then fails to compile. Usually the problems are in other unrelated drivers (which is what you experience). I update the drivers from time to time by integrating changes from the official linux_media. It would of course be better if my changes were integrated there or if they could be compiled "out of tree", but even then keeping up with kernel changes is a continuous effort. I am personally running kernel 5.10.19 on fedora 34. That kernel or any older one should work in theory. Newer ones are sometimes a problem when some incompatible changes are made in the standard linux kernel. What kernel version is your ubuntu using? Perhaps you should just try compiling it and let me know if it works or not (things have changed since the initial version, so problems may have gone or new ones may have grown). In the longer term it would probably be best to support a few well chosen kernels, e.g., for the most used ubuntu and fedora versions and then update them from time to time. Well. my own experience on windows with these cards is quite poor. I have multiple hardware versions of them and [LIST] [*]ebspro, which is a pretty nice program does not work with them: no blindscan, no spectrum [*]crazyscan, another nice program, works with some but not all of the hardware versions of the tbs6903x and tbs6909x but only on the first tuner. The reason is probably that the driver is made for the tbs6903x and does not take into account the extra tuners on tbs6909x. The spectrum is obviously wrong (spectrum shows artificial jumps), blindscan is sometimes working, but more often not. [*]Some hardware versions ar enot working at all (probably because of an unoffficial hardware version number) [*]No success with any other programs: crashes, or all kinds of problems [/LIST] I have tried ebsPRO and crazyscan on other cards, with good results. I like ebsPRO best. In any case, I avoid windows when possible. Of course the above trouble makes it impossible to compare speeds. In general, about the speed of blindscan: you have to be careful when comparing: [LIST] [*]Some types of blindscan give less information, or miss more transponders. [*]Some types of blindscan are faster than others on some satellites and slower on others [*]Some satellites have weird or incompliant muxes which can cause slow downs or even some software to get stuck on them [/LIST] So you will have to try for yourself. In any case, there is always a balance between quality of the scan and speed. By the way: during development I scanned most satellites between 30.0W and 53E. On several of them weird problems occurred, which required workarounds in the code. I even encountered it mux which claimed to be on 99.9 Ghz, for example. And then there is the problem of spill-over: one one sat position receiving data from another sat. So this full arc scan was a good exercise to track down (and fix) problems. The fastest version of my blindscan code is in `neumo-blindscan`, which is already on github. It does the bare minimum: finding muxes that can be locked ad discovering which multi-streams (if any) exist. It tries to not miss any mux (and finds most, but with some exceptions). It is pretty fast on satellites with strong signals (e.g., a few minutes in total), and slower on other satellites, because it wastes time on investigating noise peaks. I could make it even 4 times faster by using tuners in parallel, but that is not implemented yet and would require some more driver changes. I think in speed it probably compares to ebsPRO if ebsPRO would support this card (it does not). The version in neumo-DVB uses the same approach as nuemo-blindscan, so will lock the same muxes. However, it also checks SI data, which takes at least 30 seconds extra on each mux. So I would consider it "slow and thorough". It could be made as fast as neumo-blindscan with 2 or 3 lines of code changes. Personally I prefer to just use the spectrum and click on some trasnponders of interest. Within a few seconds I can see if it locks, and if I wait a little more, I know what it contains. All code is completely different. I considered using his code, and studied it, but it did not really work on the cards I had at the time, and I needed some incompatible changes. Also his kernel-tree was not compatible with some other drivers I needed (same type of problem as you are experiencing). So I gave up, not because that code would have problems, but because I could not reuse it for the cards I was really interested in anyway. PS. Even now, I am not sure if that software supports scanning a full band, or only "blind tuning" (i.e., clicking on the graph to tune it). If I remember correctly, this was one reason why I needed incompatible changes (support full band blindscan) So you will probably use updatelee or my code depending on the cards you want to use. It is a pity that the two kernels are not compatible, but that is the way it is. It would not be too difficult to have both approaches co-exist in the same kernel, but that requires some effort. At some point, I also checked what changes in updateDVB would be needed to support my drivers, but I got stuck at some point for lack of time, and never continued the search. So instead, I included the graphs ... in my own program. The documentation has a whole section on installation and compilation. I would suggest starting from that. If something is not clear, ask, and I will then update the documentation. As the software has never been run on Ubuntu, the first time Ubuntu-user will probably need to make some extra effort, e.g., to find out exactly the names of the packages for prerequirsite software that neumo depends on. So if that first user takes some notes, I will add them to the documentation. Making video is yet another effort. I might do it at some point, but isn't written version easier? [/QUOTE]
Insert quotes…
Verification
Post reply
Forums
Satellite TV receivers & systems support forums
Satellite PC Card Receivers, Internet by Satellite
neumoDVB
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
Accept
Learn more…
Top