Questions from a small experiment with various JackTrip clients
121 views
Skip to first unread message
Jason Tiller
Jan 5, 2021, 6:29:02 PM1/5/21
to jacktrip-users
Hi, all,
My name is Jason Tiller and I'm a member of a small community chorus in Palo Alto, CA. We have been completely silent for the last 9+ months other than a small "sing-your-own-track" video collaboration. We would very much like to get back to singing together in real time, and we've been looking to JackTrip to help us.
There were four members who tested JackTrip last Saturday. We set up an extra-small test server with bog standard settings.
1. Me, running a VirtualStudio box via ethernet
2. My director, running a VirtualStudio box via ethernet
3. Another chorister, who installed JamTrip on various Windows 10 laptops
4. A third chorister, who installed JamTrip on a single Windows 10 laptop
There were some promising outcomes and some that were less so:
1. The two VirtualStudio box users were able to register their devices and connect. We could both hear each other.
2. One chorister installed JT via JamTrip on 3 systems, 2 of which were completely unresponsive (couldn't hear or be heard); the one system that connected and could hear/be heard was almost unintelligibly garbled. We agreed it sounded as though samples were missing and not simply distorted.
3. One chorister installed JT via JamTrip on his laptop and said he could hear us but was never able to be heard.
4. The other VS user had beautiful sound but frequently could not be heard as the microphone jack seemed highly unreliable and he had to frequently jiggle his mic connector.
5. The sound from my VS had static, perhaps more pronounced when I spoke more loudly. Decreasing the input gain didn't seem to have an effect. I also tried two different dynamic mics.
6. I was unable to change my samples-per-message setting - the control on the devices page was disabled and only displayed 128 s/m.
7. The server status interface displayed the "connected" headset icon in random colors and showed either all zeros or NaN for various connection/throughput stats. For example, my headset icon was green and had reasonable stats, but it was red for the other VS user with all 0s, despite him being able to hear others and be heard by others.
8. The self-volume slider seemed to do nothing, as far as we could tell.
I realize that the JamTrip UI hides many details, including some that could possibly be used to debug these situations, but the intent of the test was to see if we could roll out either a JamTrip client or VS boxes to other choristers without having to become full-time tech support.
We're going to keep at it, and perhaps in our next session I'll try using JamTrip myself to get a flavor for the API, but we're not there yet. :(
---Jason
My name is Jason Tiller and I'm a member of a small community chorus in Palo Alto, CA. We have been completely silent for the last 9+ months other than a small "sing-your-own-track" video collaboration. We would very much like to get back to singing together in real time, and we've been looking to JackTrip to help us.
There were four members who tested JackTrip last Saturday. We set up an extra-small test server with bog standard settings.
1. Me, running a VirtualStudio box via ethernet
2. My director, running a VirtualStudio box via ethernet
3. Another chorister, who installed JamTrip on various Windows 10 laptops
4. A third chorister, who installed JamTrip on a single Windows 10 laptop
There were some promising outcomes and some that were less so:
1. The two VirtualStudio box users were able to register their devices and connect. We could both hear each other.
2. One chorister installed JT via JamTrip on 3 systems, 2 of which were completely unresponsive (couldn't hear or be heard); the one system that connected and could hear/be heard was almost unintelligibly garbled. We agreed it sounded as though samples were missing and not simply distorted.
3. One chorister installed JT via JamTrip on his laptop and said he could hear us but was never able to be heard.
4. The other VS user had beautiful sound but frequently could not be heard as the microphone jack seemed highly unreliable and he had to frequently jiggle his mic connector.
5. The sound from my VS had static, perhaps more pronounced when I spoke more loudly. Decreasing the input gain didn't seem to have an effect. I also tried two different dynamic mics.
6. I was unable to change my samples-per-message setting - the control on the devices page was disabled and only displayed 128 s/m.
7. The server status interface displayed the "connected" headset icon in random colors and showed either all zeros or NaN for various connection/throughput stats. For example, my headset icon was green and had reasonable stats, but it was red for the other VS user with all 0s, despite him being able to hear others and be heard by others.
8. The self-volume slider seemed to do nothing, as far as we could tell.
I realize that the JamTrip UI hides many details, including some that could possibly be used to debug these situations, but the intent of the test was to see if we could roll out either a JamTrip client or VS boxes to other choristers without having to become full-time tech support.
We're going to keep at it, and perhaps in our next session I'll try using JamTrip myself to get a flavor for the API, but we're not there yet. :(
---Jason
Vic Wong
Jan 5, 2021, 7:13:10 PM1/5/21
to jacktri...@googlegroups.com
Happy to help with the Jamtrip side of things. Admittedly, I have been testing it primarily with direct peer-to-peer setups.
The issues you're describing sound like Jamtrip might be missing some audio auto-patching settings. If you're familiar with qjackctl, you can still run it while Jamtrip is running, and see which patches are missing, and manually connect them. By default Jamtrip routes input 1 and 2 to the send_1 of a jacktrip server, and receive_1 to both channels of the local output, but there could be something missed. I'd be curious to know what.
Also, are you using Jamtrip to connect to hosted VS servers or are you using Jamtrip to host the servers locally?
Feel free to file tickets on the jamtrip github project with any logging output you can provide. I do have a ticket pending around windows not connecting to VS servers: https://github.com/vicwomg/jamtrip/issues/9
--
You received this message because you are subscribed to the Google Groups "jacktrip-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to jacktrip-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/jacktrip-users/9bd68de3-54f6-49cf-b1a8-e5bfd36ba943%40www.fastmail.com.
neal rhodes
Jan 5, 2021, 7:26:40 PM1/5/21
to jacktri...@googlegroups.com
In October we were where you are now. Don't give up.
https://www.dropbox.com/s/u7yibmhg2771puz/River%20of%20Jordan%20Thursday%20Recording%202%20-%20Reverb.mp3?dl=0 is a recording we made last week with $15 PC microphones and home-built VS boxes.
There are many gotchas, and many adjustments. from my list, the ones that might apply:
2) If you are testing two devices on the same home router, you must change the Port on the 2nd device. (otherwise one device will steal the audio from the other)
5) If you place your device not at your border router, but behind a Sonicwall TZ-190 firewall router, it WILL connect, but you will get no sound.
6) If Using DAC+ ADC Pro with PC type mics, you must close jumpers J1 and J3 to provide bias voltage. And..... amixer set 'ADC Mic Bias',0 'Mic Bias on'
Fiddle around between CD quality (jacktrip) and lower (jamulus) and see what happens. Be PATIENT and wait 15 seconds to see the effect.
Some PI power packs are crap, and produce AC ripple in addition to the +5V DC. Seems like more the Pi 3; The stock Pi 4 supply doesn't seem to have same problems. Measure AC voltage from the shield on the RCA jack to house ground.
The Web UI changes to volume take a LONG time to be recognized, eg, maybe 10-15 seconds.
You also do not mention any separate testing of each user's internet connection for ping time, jitter, and bandwidth. And if someone was watching football in the same house at the same time. Unfortunately, internet connection is the LAST thing people check, and the hardest thing to fix.
Keep real careful notes on each VS box, test results from each location, and try the same box using different internet at different locations. Play "odd man out" to see how the problems travel.
Unless someone lives on Google Fiber and has truly fantastic low internet latency, I'd give up on the PC approaches. If you lose 15 seconds in the PC, and have 30ms network latency, you are sunk. FYI, we are 600 miles from the Virginia server and typical latency is 35+ms. It's bearable.
Regards,
Neal Rhodes
Dynamic Range
Trinity Lutheran Church
Lilburn, Georgia.
Jason Tiller
Jan 6, 2021, 2:25:29 PM1/6/21
to 'Jon Raskin' via jacktrip-users
Hi, Vic,
On Tue, Jan 5, 2021, at 4:12 PM, Vic Wong wrote:
Happy to help with the Jamtrip side of things. Admittedly, I have been testing it primarily with direct peer-to-peer setups.
We will eventually have 40+ choristers to connect, so we never considered P2P solutions.
The issues you're describing sound like Jamtrip might be missing some audio auto-patching settings. If you're familiar with qjackctl, you can still run it while Jamtrip is running, and see which patches are missing, and manually connect them. By default Jamtrip routes input 1 and 2 to the send_1 of a jacktrip server, and receive_1 to both channels of the local output, but there could be something missed. I'd be curious to know what.
I will pass this along to our JamTrip users. I'm glad that QJackCtl can look "under the hood" without resorting to reading logfiles.
Also, are you using Jamtrip to connect to hosted VS servers or are you using Jamtrip to host the servers locally?
We used a default extra-small test server that we set up on the San Francisco host.
Feel free to file tickets on the jamtrip github project with any logging output you can provide. I do have a ticket pending around windows not connecting to VS servers: https://github.com/vicwomg/jamtrip/issues/9
Thanks again, and I'll get back to you with any feedback from our next session! I appreciate your insights very much!
---Jason
To view this discussion on the web visit https://groups.google.com/d/msgid/jacktrip-users/CAC306cW20WftWhU5qU2uv7i-GaOpRCtK5T4qbA01b8GRW1bBRg%40mail.gmail.com.
Jason Tiller
Jan 6, 2021, 4:52:16 PM1/6/21
to jacktrip-users
Hi, Neal,
On Tue, Jan 5, 2021, at 4:26 PM, neal rhodes wrote:
In October we were where you are now. Don't give up.
Yeah, we're gonna keep beating on it. :)
https://www.dropbox.com/s/u7yibmhg2771puz/River%20of%20Jordan%20Thursday%20Recording%202%20-%20Reverb.mp3?dl=0 is a recording we made last week with $15 PC microphones and home-built VS boxes.
I love it! That's really encouraging.
There are many gotchas, and many adjustments. from my list, the ones that might apply:2) If you are testing two devices on the same home router, you must change the Port on the 2nd device. (otherwise one device will steal the audio from the other)
This was not our case - every member had their own, single device (PC or VS) behind a single router.
5) If you place your device not at your border router, but behind a Sonicwall TZ-190 firewall router, it WILL connect, but you will get no sound.
Given that our test users were in their own homes and are not network engineers, I have a feeling they weren't using their own network security appliances. Still, I'll check!
6) If Using DAC+ ADC Pro with PC type mics, you must close jumpers J1 and J3 to provide bias voltage. And..... amixer set 'ADC Mic Bias',0 'Mic Bias on'
Our VS users had no problems hearing/being heard (except for the unreliability of the 3.5mm input connector).
Fiddle around between CD quality (jacktrip) and lower (jamulus) and see what happens. Be PATIENT and wait 15 seconds to see the effect.
We tried both a jacktrip+jamulus server as well as a jacktrip-only server with no observable difference.
A few more questions that I haven't seen addressed:
1. Any ideas on why my microphones (a Shure PGA48 https://d24z4d3zypmncx.cloudfront.net/Pubs/PGA48/pga48-specification-sheet-english.pdf and a Behringer XMB500 https://mediadl.musictribe.com/media/sys_master/h7e/h90/8849608572958.pdf) would exhibit static or crackly sound? I need to find out what our other VS user had for a mic because his rig didn't exhibit static. Note that the Behringer is SPECIFICALLY recommended by the JackTrip documentation and no mention was made of this device requiring an "unbalancing" accessories.
2. Neither of the datasheets for these devices mention "balance" anywhere, nor is that feature listed in their documentation. How am I supposed to know if a microphone's output is balanced or not?
3. Any idea on why the device controls for changing my samples-per-message were disabled? The other VS user was able to change his setting from 128 to 256, which is what the server was set at.
Some PI power packs are crap, and produce AC ripple in addition to the +5V DC. Seems like more the Pi 3; The stock Pi 4 supply doesn't seem to have same problems. Measure AC voltage from the shield on the RCA jack to house ground.
We were using the recommended CanaKit power supplies with the Virtual Studio boxen - all direct from JackTrip.
The Web UI changes to volume take a LONG time to be recognized, eg, maybe 10-15 seconds.
Why would that be the case?
You also do not mention any separate testing of each user's internet connection for ping time, jitter, and bandwidth. And if someone was watching football in the same house at the same time. Unfortunately, internet connection is the LAST thing people check, and the hardest thing to fix.Keep real careful notes on each VS box, test results from each location, and try the same box using different internet at different locations. Play "odd man out" to see how the problems travel.Unless someone lives on Google Fiber and has truly fantastic low internet latency, I'd give up on the PC approaches. If you lose 15 seconds in the PC, and have 30ms network latency, you are sunk. FYI, we are 600 miles from the Virginia server and typical latency is 35+ms. It's bearable.
We were all within 30 miles of the test server - my pings were in the 15-25ms range. Others had NaNs or 0s listed for their latencies. For me, the only other active internet usage was Zoom. Next time I'll try to nail people down on that as well.
Excellent ideas, Neal! Thanks! And keep 'em coming! I can't wait to sing with friends again.
BTW, has anyone done any JT experimentation with a PERCUSSION ensemble? My group is contracted to play a gig in July, but being that we're here in California, who knows when we'll get to practice in person...
Thanks again you wonderful, helpful people!
---Jason
To view this discussion on the web visit https://groups.google.com/d/msgid/jacktrip-users/CAPf%2BAdYDq9HU1tRBBn%3DC0V0n9oWHJNFfOxJz-32cteynJ8vWWg%40mail.gmail.com.
Bill van Melle
Jan 6, 2021, 6:50:24 PM1/6/21
to jacktrip-users
> If you are testing two devices on the same home router, you must change the Port on the 2nd device.
Do you know how to do that with the command-line jacktrip client? I thought it might be the -o (port offset) or --bindport options, but if I try either of them, jacktrip quickly crashes.
Bill van Melle
Jan 6, 2021, 11:07:16 PM1/6/21
to jacktrip-users
Hi, Vic. We spoke last week over Facebook.
On Tuesday, January 5, 2021 at 4:13:10 PM UTC-8 vic...@gmail.com wrote:
The issues you're describing sound like Jamtrip might be missing some audio auto-patching settings. If you're familiar with qjackctl, you can still run it while Jamtrip is running, and see which patches are missing, and manually connect them. By default Jamtrip routes input 1 and 2 to the send_1 of a jacktrip server, and receive_1 to both channels of the local output, but there could be something missed. I'd be curious to know what.
This was somewhat instructive, in that it explains why I can hear myself but nobody else can. It's because I'm not actually connected! It looks like JamTrip doesn't detect when the connection fails, but goes ahead and connects the local capture ports to the local playback ports. I include a typical log below.
So I think it would reduce confusion a whole lot if JamTrip detected that the jacktrip process has already exited. Beyond that, I'd love to know why jacktrip isn't succeeding. I find when I run it at the command line, it fails as below with messages about mismatched buffer size/sampling rate in over 90% of my attempts.
This is all talking to servers at jacktrip.org.
Setting JACK Process Callback...
SUCCESS
---------------------------------------------------------
The Sampling Rate is: 48000
---------------------------------------------------------
The Audio Buffer Size is: 256 samples
or: 1024 bytes
---------------------------------------------------------
The Number of Channels is: 1
---------------------------------------------------------
The RTAudio device ID is: 0
---------------------------------------------------------
Using UDP Protocol
---------------------------------------------------------
Server Address set to: 54.151.95.242 Port: 61002
---------------------------------------------------------
UDP Socket Receiving in Port: 4464
---------------------------------------------------------
Waiting for Peer...
Received Connection from Peer!
Stopping JackTrip...
ERROR: Peer Buffer Size is : 13016
Local Buffer Size is : 256
Make sure both machines use same buffer size
---------------------------------------------------------
ERROR: Peer Sampling Rate is : 0
Local Sampling Rate is : 48000
Make sure both machines use the same Sampling Rate
---------------------------------------------------------
ERROR: Peer Audio Bit Resolution is : 0
Local Audio Bit Resolution is : 16
Make sure both machines use the same Bit Resolution
---------------------------------------------------------
JackTrip Processes STOPPED!
---------------------------------------------------------
Exiting JackTrip...
** Connection established, patching channels...
** Connecting '54.151.95.242:receive_1' to 'system:playback_1'.
stderr: ERROR 54.151.95.242:receive_1 not a valid port
** Connecting '54.151.95.242:receive_1' to 'system:playback_2'.
stderr: ERROR 54.151.95.242:receive_1 not a valid port
** Connecting 'system:capture_1' to '54.151.95.242:send_1'.
stderr: ERROR 54.151.95.242:send_1 not a valid port
** Connecting 'system:capture_2' to '54.151.95.242:send_1'.
stderr: ERROR 54.151.95.242:send_1 not a valid port
Bonnie Kwong
Jan 7, 2021, 12:57:16 AM1/7/21
to jacktri...@googlegroups.com
Hi Bill,
This issue is documented but not entirely understood.
Follow these instructions to turn off printer and file sharing.
Bonnie
--
You received this message because you are subscribed to the Google Groups "jacktrip-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to jacktrip-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/jacktrip-users/92087bdb-4c68-43da-9f70-773f55d4436an%40googlegroups.com.
Bonnie Kwong
Jan 7, 2021, 12:57:20 AM1/7/21
to jacktri...@googlegroups.com
Bill van Melle
Jan 7, 2021, 1:25:45 AM1/7/21
to jacktri...@googlegroups.com
Thanks for the info! Unfortunately, turning off file sharing is really not an option for me. Is JackTrip 1.2 still available somewhere? It doesn't have the problem, at least according to the filer of the bug report. Or is there a notable downside to the older version?
BTW, why is issue 139 marked "Closed"? The existence of a workaround doesn't mean it's not still a bug that should be fixed.
To view this discussion on the web visit https://groups.google.com/d/msgid/jacktrip-users/CAEiUSmuKoCKG6bpj0HudkU-L5u4v1NBo%2B38LSUYW2jkv7jSxDA%40mail.gmail.com.
Synthia Cynthia Payne
Jan 7, 2021, 2:22:25 AM1/7/21
to jacktri...@googlegroups.com
some folks turn off file sharing just for the session and then turn it
back on afterwards, but if it's a bug it should probably be opened.
synth
On 1/6/2021 10:25 PM, Bill van Melle wrote:
> Thanks for the info! Unfortunately, turning off file sharing is really
> not an option for me. Is JackTrip 1.2 still available somewhere? It
> doesn't have the problem, at least according to the filer of the bug
> report. Or is there a notable downside to the older version?
>
> BTW, why is issue 139 marked "Closed"? The existence of a workaround
> doesn't mean it's not still a bug that should be fixed.
>
>
> On Wed, Jan 6, 2021 at 9:57 PM Bonnie Kwong <sibilant...@gmail.com
> <mailto:sibilant...@gmail.com>> wrote:
>
> https://github.com/jacktrip/jacktrip/issues/139
> <mailto:jacktrip-user...@googlegroups.com>.
> <https://groups.google.com/d/msgid/jacktrip-users/CAEiUSmuKoCKG6bpj0HudkU-L5u4v1NBo%2B38LSUYW2jkv7jSxDA%40mail.gmail.com?utm_medium=email&utm_source=footer>.
> <https://groups.google.com/d/msgid/jacktrip-users/CAAaiUA5YEY1Uj0TgwMZaYR%3DJnhZM7Z%2BFCOB_P3zHH-RAXRUxtg%40mail.gmail.com?utm_medium=email&utm_source=footer>.
back on afterwards, but if it's a bug it should probably be opened.
synth
On 1/6/2021 10:25 PM, Bill van Melle wrote:
> Thanks for the info! Unfortunately, turning off file sharing is really
> not an option for me. Is JackTrip 1.2 still available somewhere? It
> doesn't have the problem, at least according to the filer of the bug
> report. Or is there a notable downside to the older version?
>
> BTW, why is issue 139 marked "Closed"? The existence of a workaround
> doesn't mean it's not still a bug that should be fixed.
>
>
> On Wed, Jan 6, 2021 at 9:57 PM Bonnie Kwong <sibilant...@gmail.com
>
> https://github.com/jacktrip/jacktrip/issues/139
> <https://github.com/jacktrip/jacktrip/issues/139>
>
> On Wed, Jan 6, 2021 at 9:57 PM Bonnie Kwong
> <sibilant...@gmail.com <mailto:sibilant...@gmail.com>> wrote:
>
> Hi Bill,
>
> This issue is documented but not entirely understood.
>
> Follow these instructions to turn off printer and file sharing.
>
> Bonnie
>
> On Wed, Jan 6, 2021 at 8:07 PM Bill van Melle
> <bill.va...@gmail.com <mailto:bill.va...@gmail.com>> wrote:
>
> Hi, Vic. We spoke last week over Facebook.
>
> On Tuesday, January 5, 2021 at 4:13:10 PM UTC-8
>
> On Wed, Jan 6, 2021 at 9:57 PM Bonnie Kwong
> <sibilant...@gmail.com <mailto:sibilant...@gmail.com>> wrote:
>
> Hi Bill,
>
> This issue is documented but not entirely understood.
>
> Follow these instructions to turn off printer and file sharing.
>
> Bonnie
>
> On Wed, Jan 6, 2021 at 8:07 PM Bill van Melle
> <bill.va...@gmail.com <mailto:bill.va...@gmail.com>> wrote:
>
> Hi, Vic. We spoke last week over Facebook.
>
> On Tuesday, January 5, 2021 at 4:13:10 PM UTC-8
> vic...@gmail.com <mailto:vic...@gmail.com> wrote:
>
> The issues you're describing sound like Jamtrip might be
> missing some audio auto-patching settings. If you're
> familiar with qjackctl, you can still run it while
> Jamtrip is running, and see which patches are missing,
> and manually connect them. By default Jamtrip routes
> input 1 and 2 to the send_1 of a jacktrip server, and
> receive_1 to both channels of the local output, but
> there could be something missed. I'd be curious to know
> what.
>
>
> This was somewhat instructive, in that it explains why I can
> hear myself but nobody else can. It's because I'm not
> actually connected! It looks like JamTrip doesn't detect
> when the connection fails, but goes ahead and connects the
> local capture ports to the local playback ports. I include
> a typical log below.
>
> So I think it would reduce confusion a whole lot if JamTrip
> detected that the jacktrip process has already exited.
> Beyond that, I'd love to know why jacktrip isn't succeeding.
> I find when I run it at the command line, it fails as below
> with messages about mismatched buffer size/sampling rate in
> over 90% of my attempts.
>
> This is all talking to servers at jacktrip.org
> <http://jacktrip.org>.
>
> The issues you're describing sound like Jamtrip might be
> missing some audio auto-patching settings. If you're
> familiar with qjackctl, you can still run it while
> Jamtrip is running, and see which patches are missing,
> and manually connect them. By default Jamtrip routes
> input 1 and 2 to the send_1 of a jacktrip server, and
> receive_1 to both channels of the local output, but
> there could be something missed. I'd be curious to know
> what.
>
>
> This was somewhat instructive, in that it explains why I can
> hear myself but nobody else can. It's because I'm not
> actually connected! It looks like JamTrip doesn't detect
> when the connection fails, but goes ahead and connects the
> local capture ports to the local playback ports. I include
> a typical log below.
>
> So I think it would reduce confusion a whole lot if JamTrip
> detected that the jacktrip process has already exited.
> Beyond that, I'd love to know why jacktrip isn't succeeding.
> I find when I run it at the command line, it fails as below
> with messages about mismatched buffer size/sampling rate in
> over 90% of my attempts.
>
> This is all talking to servers at jacktrip.org
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/jacktrip-users/92087bdb-4c68-43da-9f70-773f55d4436an%40googlegroups.com
> <https://groups.google.com/d/msgid/jacktrip-users/92087bdb-4c68-43da-9f70-773f55d4436an%40googlegroups.com?utm_medium=email&utm_source=footer>.
> https://groups.google.com/d/msgid/jacktrip-users/92087bdb-4c68-43da-9f70-773f55d4436an%40googlegroups.com
>
> --
> You received this message because you are subscribed to the Google
> Groups "jacktrip-users" group.
> To unsubscribe from this group and stop receiving emails from it,
> send an email to jacktrip-user...@googlegroups.com
> <mailto:jacktrip-user...@googlegroups.com>.
> --
> You received this message because you are subscribed to the Google
> Groups "jacktrip-users" group.
> To unsubscribe from this group and stop receiving emails from it,
> send an email to jacktrip-user...@googlegroups.com
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/jacktrip-users/CAEiUSmuKoCKG6bpj0HudkU-L5u4v1NBo%2B38LSUYW2jkv7jSxDA%40mail.gmail.com
> <https://groups.google.com/d/msgid/jacktrip-users/CAEiUSmuKoCKG6bpj0HudkU-L5u4v1NBo%2B38LSUYW2jkv7jSxDA%40mail.gmail.com?utm_medium=email&utm_source=footer>.
>
> --
> You received this message because you are subscribed to the Google
> Groups "jacktrip-users" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to jacktrip-user...@googlegroups.com
> <mailto:jacktrip-user...@googlegroups.com>.
> --
> You received this message because you are subscribed to the Google
> Groups "jacktrip-users" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to jacktrip-user...@googlegroups.com
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/jacktrip-users/CAAaiUA5YEY1Uj0TgwMZaYR%3DJnhZM7Z%2BFCOB_P3zHH-RAXRUxtg%40mail.gmail.com
> <https://groups.google.com/d/msgid/jacktrip-users/CAAaiUA5YEY1Uj0TgwMZaYR%3DJnhZM7Z%2BFCOB_P3zHH-RAXRUxtg%40mail.gmail.com?utm_medium=email&utm_source=footer>.
neal rhodes
Jan 7, 2021, 7:51:06 AM1/7/21
to jacktri...@googlegroups.com
My understanding/observation is that the VS boxes poll the mothership on a periodic basis to see if they are supposed to change something. When I switch from CD quality to high quality, it takes several seconds to see the result. When I change input volumes, it may take 10-15 seconds to see the result.
Regarding mics, if It's an XLR connector, then I'd expect it to be a balanced connection. Some conversion necessary. Note that the round 3.5 plug is more of an engineering abomination than the 1/4" plug; they all are prone to poor contact.
We have been using several difference desktop microphones, and have settled on these: https://www.amazon.com/gp/product/B08GZSQNT7/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1 as they are well suited to multiple people 12" away, have their own little desktop stand, and provide us with a minimal footprint to go into people's homes. Not everyone wants a big clonking mic, which needs a stand, and ... That said, desktop/PC mics require 2.5V Bias voltage, which the HiFiBerry will do, but you have to set jumper pins and do some command line setup.
Regarding routers, yes, maybe not a factor for you, but we had our singers plugging into WiFi extenders upstairs, and had to be REALLY CLEAR about finding out where their actual border router was.
I'm not an expert, but if you have reasonable latency, not sure why you'd need 256; We use 128 and have 24-39ms latency. Higher numbers would cause more total delay.
Stepping back, I'm thinking you are seeing about 3-4 different problems. You have to tackle one at a time. As in:
1st) Verify each box and microphone and headset actually work by themselves. There is a way of using SSH to login into a local box. (on windows get the Putty SSH client) The Jacktrip Foundation staff have written examples of commands to record and playback audio. This would also let you evaluate crackle, line noise, etc. Independent of the networking and server setup. Sorry, I don't have the details on that.
2nd) get two people connected to the server. I'll admit we did this with two boxes in our home, one upstairs and one downstairs. That's where that client port number comes in. I find it less confusing to just deal with two boxes myself.
3rd) Then take that 2nd box to someone else's house.
This may be a lot more diddling than you had expected.
On Wed, Jan 6, 2021 at 6:50 PM Bill van Melle <bill.va...@gmail.com> wrote:
> If you are testing two devices on the same home router, you must change the Port on the 2nd device.Do you know how to do that with the command-line jacktrip client? I thought it might be the -o (port offset) or --bindport options, but if I try either of them, jacktrip quickly crashes.
--
You received this message because you are subscribed to the Google Groups "jacktrip-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to jacktrip-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/jacktrip-users/68a4a99e-6f18-4d89-84de-1b7a29e9170dn%40googlegroups.com.
Bonnie Kwong
Jan 7, 2021, 1:26:39 PM1/7/21
to jacktri...@googlegroups.com
Hi Bill,
Here's a Win 10 installer Chris Chafe made last June for v 1.2:
If you could let us know whether you still see the problem after switching to this version, it would help us troubleshoot.
Bonnie
To view this discussion on the web visit https://groups.google.com/d/msgid/jacktrip-users/CAPf%2BAdaAHzDWiSTB3RvaAZa7zTo-AJ%3D3fc9FWfVr4A_W_%3DXZuw%40mail.gmail.com.
Mike Dickey
Jan 7, 2021, 3:09:02 PM1/7/21
to jacktri...@googlegroups.com
Hi Jason,
Static in your audio is often caused by your Internet connection,
in particular having gaps when packets don't reach the server fast
enough (jitter) or at all (loss). This can usually be resolved by
changing the quality setting of their VSD device from "High" to
"Medium." Note that this requires use of JackTrip+Jamulus servers.
Not to let the cat out of the bag, but Chris is working on some
JackTrip updates that will likely solve it for "High" as well (in
the future / next release).
Inability to change samples on the device page may be caused by the device having an older firmware that doesn't support this feature. What is the version number displayed in the lower left corner (it should be "2020102003")? Normally, the devices will update themselves when they first start up, but there is a known issue with specific releases having trouble with it. This page has a few steps that usually helps: Device Not Updating Firmware – JackTrip (zendesk.com)
The HiFiBerry cards require unbalanced input. The most common problem to look out for with balanced vs unbalanced is people can't hear one another in certain conditions (balanced mono can result in the signal being canceled out during mixing). Note that this will not cause static but rather no audio at all. Typically, this is caused by the cable connecting your microphone to the 3.5" input device on the HiFiBerry card, or in particular how it has the tip & ring wired. If you are using the recommended cables, it should not be a problem. See more info at Sometimes unable to hear a musician – JackTrip (zendesk.com)
Latency statistics require having ICMP open on your router. If you see NaN etc it's because your firewall is blocking this protocol.
Take care,
-Mike
To view this discussion on the web visit https://groups.google.com/d/msgid/jacktrip-users/f04d0d20-4e84-411c-9239-bcb18390c64f%40www.fastmail.com.
Jason Tiller
Jan 7, 2021, 3:27:43 PM1/7/21
to jacktrip-users
Hi, Mike!
On Thu, Jan 7, 2021, at 12:08 PM, Mike Dickey wrote:
Static in your audio is often caused by your Internet connection, in particular having gaps when packets don't reach the server fast enough (jitter) or at all (loss). This can usually be resolved by changing the quality setting of their VSD device from "High" to "Medium." Note that this requires use of JackTrip+Jamulus servers. Not to let the cat out of the bag, but Chris is working on some JackTrip updates that will likely solve it for "High" as well (in the future / next release).
Hmm. That's extremely odd. I did not see any packet loss AT ALL and I have an ethernet connection and I'm quite close to the hosted server. The only other network activity at the time (unless I've been hijacked!) was the Zoom call. Then again, it is Comcast, so anything that can go wrong probably will.
Inability to change samples on the device page may be caused by the device having an older firmware that doesn't support this feature. What is the version number displayed in the lower left corner (it should be "2020102003")? Normally, the devices will update themselves when they first start up, but there is a known issue with specific releases having trouble with it. This page has a few steps that usually helps: Device Not Updating Firmware – JackTrip (zendesk.com)
This makes tremendous sense - my VS is of an older vintage than our other VS user, so it's entirely possible that's what's going on. I'll make sure I get the update. Thanks!
The HiFiBerry cards require unbalanced input. The most common problem to look out for with balanced vs unbalanced is people can't hear one another in certain conditions (balanced mono can result in the signal being canceled out during mixing). Note that this will not cause static but rather no audio at all. Typically, this is caused by the cable connecting your microphone to the 3.5" input device on the HiFiBerry card, or in particular how it has the tip & ring wired. If you are using the recommended cables, it should not be a problem. See more info at Sometimes unable to hear a musician – JackTrip (zendesk.com)
Yes, I am using the recomended XLR-female to 3.5mm-TRS-male cable from Hosa. I recommended that the other VS users also purchase this cable.
Latency statistics require having ICMP open on your router. If you see NaN etc it's because your firewall is blocking this protocol.
But... wouldn't that disallow any client on the network from using 'ping' and other ICMP-based utilities? I will mention this to the other users.
Take care,
You, too! Thank you very much for your help with this. We will get this going and sing again! :)
---Jason
To view this discussion on the web visit https://groups.google.com/d/msgid/jacktrip-users/3bb6b8c2-4cbf-8559-47e1-a71741fcb4d3%40jacktrip.org.
Bill van Melle
Jan 7, 2021, 5:50:09 PM1/7/21
to jacktri...@googlegroups.com
On Thu, Jan 7, 2021 at 10:26 AM Bonnie Kwong <sibilant...@gmail.com> wrote:
Here's a Win 10 installer Chris Chafe made last June for v 1.2:If you could let us know whether you still see the problem after switching to this version, it would help us troubleshoot.
Thank you! That worked fine, in at least the simple tests I did.
It's a bit of a weird installer--it creates a folder with a jacktrip stub, unlike v 1.2.1, which comes all bundled into a single executable. That means I couldn't simply drop this jacktrip.exe into the JamTrip resources folder. However, I used an obscure feature of Windows to do something resembling a Unix symlink, and that allows me to run JamTrip with this version as well. Yeah!
>cd C:\Users\Bill\AppData\Local\Programs\jamtrip\resources\resources\bin
>ren win32 win32-1.2.1
>mklink /d win32 C:\apps\jacktrip1.2IsHeresymbolic link created for win32 <<===>> C:\apps\jacktrip1.2IsHere>
--Bill
Vic Wong
Jan 7, 2021, 6:09:27 PM1/7/21
to jacktri...@googlegroups.com
This is all excellent information, I'll be sure to address it in Jamtrip soon. thanks Bill.
--
You received this message because you are subscribed to the Google Groups "jacktrip-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to jacktrip-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/jacktrip-users/CAAaiUA4HgdKMVxpwsQgyah44%2BWyYJMmcbLa7rc6ggX5tg83U7g%40mail.gmail.com.
Bill van Melle
Jan 11, 2021, 8:55:02 PM1/11/21
to jacktri...@googlegroups.com
One caveat, v 1.2 apparently doesn't do host lookup. So if I do
jacktrip -C jackloop256.stanford.edu
it fails with TCP Socket ERROR: Host not found. Using a raw IP address works fine.
Reply all
Reply to author
Forward
0 new messages