IPTV streams > 10 Mbps = video/audio stuttering

[Oualid2013]

Try Dolphin Browser with the custom Flash plug in.

Who i can find custom flash plugin?

ty

[no_spam_for_me]

Who i can find custom flash plugin?

ty

At another thread I gave you the link to xda and you told me you don't need it, AND you also post the same at different threads all over the forum, and at past you do the same and I asked you to stop these spamming...

[Oualid2013]

At another thread I gave you the link to xda and you told me you don't need it, AND you also post the same at different threads all over the forum, and at past you do the same and I asked you to stop these spamming...

Yes same time web dont work and the link you give me the flash play is not there

[lawin1]

IPTV streams > 10 Mbps = video/audio stuttering [partially solved]

Good news! Finally after a few weeks I discovered what causes to play live IPTV streams with pauses, buffering, stuttering, etc. In fact this defect can be observed not only in XBMC, but in all media players which support live IPTV streams (I haven‘t tested „Video on Demand“ services).

One good fellow helped me to track it down. He was on Finless custom ROM and I made some test on official 104K4 ROM. He had no problems with internet speeds in Finless ROM, so I made a presumption that this is somehow connected with network settings in official firmware. It turns out that firmware developers at Beelink still thinks that it is OK to leave network settings untouched as if it is a mobile device . So Tronsmart S89 was not prepared to deal with high internet speeds from the beginning. It appears that we have to enlarge TCP buffer sizes.

Theory:

http://www.onlamp.com/pub/a/onlamp/2005/11/17/tcp_tuning.html

https://community.freescale.com/docs/DOC-98322

http://www.cyberciti.biz/faq/linux-tcp-tuning/

http://fasterdata.es.net/network-tuning/tcp-issues-explained/

There is a temporal and imperfect solutions of this problem if you use an official firmware. Root is required in both cases:

I method. Install Bionic TCP Buffer Tweak - https://play.google.com/store/apps/d...ilis.bionictcp
Launch this app, press „Large TCP buffers“. You will need to repeat this procedure every time after reboot (sleep and wake up doesn‘t count).

II method. Install Android Tuner Free - https://play.google.com/store/apps/d...=ccc71.at.free
Launch this app and browse to the left to the very first page until you see „Tune your device‘s performance and battery life“. Now scroll down and press „Kernel configuration (sysctl)“ and change default values to these:
net.core.rmem_max 2.00MB
net.core.wmem_max 2.00MB
net.ipv4.tcp_rmem 4096 1048576 2097152
net.ipv4.tcp_wmem 4096 524288 2097152

Press “Save” at the bottom and then in the upper part of this page near “Apply on boot” select “boot”.

There is more elegant ways to change TCP values to normal ones, i. e. to run custom script after the boot using Script Manager or init.d. Does somebody need custom script? I can write it, but don’t expect that I’ll explain to you how to use it with Script Manager or init.d. I have wasted too much of my free time already …

In summary, Beelink needs to acknowledge this problem and to provide better TCP settings in the future ROMs.

[no_spam_for_me]

But these changes are only for stock rom?
Finless do some other but maybe same changes (r/w in one statement):

# ReadMin, ReadInitial, ReadMax, WriteMin, WriteInitial, WriteMax
setprop net.tcp.buffersize.default 524288,1048576,2097152,262144,524288,1048576

I don't know 'net.core' (also no output at getprop)...

[lawin1]

Yes, only for stock ROM.

Stock ROM:
# ReadMin, ReadInitial, ReadMax, WriteMin, WriteInitial, WriteMax,
setprop net.tcp.buffersize.default 4096,87380,110208,4096,16384,110208

Bionic TCP will tweak stock ROM:
setprop net.tcp.buffersize.default 4096,1048576,2097152,4096,524288,2097152
and then also executes:
echo 4096 1048576 2097152 > /proc/sys/net/ipv4/tcp_rmem
echo 4096 524288 2097152 > /proc/sys/net/ipv4/tcp_wmem
echo 1 > /proc/sys/net/ipv4/route/flush
echo 2097152 > /proc/sys/net/core/rmem_max
echo 2097152 > /proc/sys/net/core/wmem_max
echo 262144 > /proc/sys/net/core/rmem_default
echo 262144 > /proc/sys/net/core/wmem_default

Similar results can be achieved with Android Tuner as described above. In fact, now I am using custom script with Script Manager.

no_spam_for_me, how do you know that Finless ROM has that values below you have mentioned? I extracted from Finless ROM boot.img and init.rc file has this entry which is different from yours:

# ReadMin, ReadInitial, ReadMax, WriteMin, WriteInitial, WriteMax,
setprop net.tcp.buffersize.default 4096,87380,256960,4096,16384,256960

But these changes are only for stock rom?
Finless do some other but maybe same changes (r/w in one statement):

# ReadMin, ReadInitial, ReadMax, WriteMin, WriteInitial, WriteMax
setprop net.tcp.buffersize.default 524288,1048576,2097152,262144,524288,1048576

I don't know 'net.core' (also no output at getprop)...

[no_spam_for_me]

You are right! (I'm wrong )
I'm at minix and there I find it (at one of the cpu_fixes for s8? by finless I also find your values (no idea why))

At minix

    setprop net.tcp.buffersize.default 524288,1048576,2097152,262144,524288,1048576
    setprop net.tcp.buffersize.wifi    524288,1048576,2097152,262144,524288,1048576
    setprop net.tcp.buffersize.lte     524288,1048576,2097152,262144,524288,1048576
    setprop net.tcp.buffersize.umts    4094,87380,110208,4096,16384,110208
    setprop net.tcp.buffersize.hspa    4094,87380,262144,4096,16384,262144
    setprop net.tcp.buffersize.hsupa   4094,87380,262144,4096,16384,262144
    setprop net.tcp.buffersize.hsdpa   4094,87380,262144,4096,16384,262144
    setprop net.tcp.buffersize.hspap   4094,87380,1220608,4096,16384,1220608
    setprop net.tcp.buffersize.edge    4093,26280,35040,4096,16384,35040
    setprop net.tcp.buffersize.gprs    4092,8760,11680,4096,8760,11680
    setprop net.tcp.buffersize.evdo    4094,87380,262144,4096,16384,262144

[MINIX]

DLNA test: http://www.youtube.com/watch?v=dXmfcXHBndo

(BETA 4 firmware)

[Ichijoe]

If this really has been a "Problem" since 103k4.... Are they ever going to fix this? It doesn't seem to be that big of a deal to me... The fact that haven't send to me the wrong message...

[lawin1]

If this really has been a "Problem" since 103k4.... Are they ever going to fix this? It doesn't seem to be that big of a deal to me... The fact that haven't send to me the wrong message...

They will never fix this. I suspect that someone responsible for Tronsmart firmware intentionally ignores this problem. Maybe someone needs Tronsmart to be much worse than Minix. Don't laugh but I suspect that this is a case of corruption.

[Ichijoe]

They will never fix this. I suspect that someone responsible for Tronsmart firmware intentionally ignores this problem. Maybe someone needs Tronsmart to be much worse than Minix. Don't laugh but I suspect that this is a case of corruption.

you mean like how they all seemingly have omitted changing media.amplayer.widevineenable= from FALSE to TRUE?! (e.g. re-enabling the Widevine DRM), Opting instead to push "other" Boxes beanded as GSF certified, that eventually cost more?! Yeah.... Every time some Marketing Shrill pops up 'round here.. I make it a point to, point this 'lil gem out to them... Of course I'm still waiting for any number of 'em to respond back.

Widevine, unfortunately IS NOT the final word in all things DRM... But, a number of Apps sans Google Play Movies also need this to make their Apps work. The joke here being that the New Guy Tronfy actually packs those other "apps" with their current Firmware.... Along with the same old boring build.prop with Widevine set to false!

So corruption... What corruption?! /sarcasem.
More like a case of not giving a sh-- The MXIII hasn't even hit the One Year mark yet... And, its looking increasingly likely that, its already gone "legacy". Mark-speak for dead.
Here we are still waiting for Lollipop... And, Google's already moved on to M now... Thats telling in its own way.