Search - User list
Full Version: Tutorial: Getting low-latency audio on Windows
Root » Technical support » Tutorial: Getting low-latency audio on Windows
1 2
stefanha
jammr supports several different audio systems on Windows, depending on the sound card your computer has. The older sound systems are compatible with most machines but have noticable lag - they are bad at real-time audio.

This is a quick tutorial on setting up low-latency audio:

1. If possible, choose the “WDM Kernel Streaming” audio system from File | Settings…. This usually produces good results.
2. Otherwise, choose the “WASAPI” audio system and select the recording (input) and playback (output) devices you wish to use.
3. If you are unable to join jams or see an “Failed to start audio” error message, right-click on the speaker icon in the bottom right corner of the screen in the task bar and select “Playback devices”.
4. Choose your device from the list and click “Properties”.
5. Then click “Advanced” and note the audio format (for example, “24 bit, 44100 Hz”).
6. Now repeat this for “Recording devices” instead of “Playback devices” and select formats on the recording and playback devices that match. This will ensure that jammr to use the playback and recording devices together.

We recommend selecting 24 bit and 48000 Hz. If the devices do not support this format, try 16 bit and 44100 Hz. The important thing is to use the same format for both the playback and recording devices.

When you have this set up you should notice that you feel no delay between when your fingers play notes on your instrument and when you hear the notes.
NakajimaYusuke
Thank you for your e-mail, Stefan.

I am sorry that I did not reply sooner to you.
Because,I am very poor at English. Translation is time-consuming work for me.

I write to a forum so that a problem and its solution can be shared.

This is the action reporting in an old version.
These problems may be solved in the new version.
I have not tried the new version in detail. I will try it, soon.

My first problem is this.

I chose “WASAPI” as “Audio system”.
Then, “Input device” and “Output device” were not displayed.

http://forum.jammr.net/topic/4/
I followed these directions.

About “Sample rate” and “Bit rate”, I arranged “Recording devices” and “Playback devices” equally.
However, the result was the same.

Moreover, I tried various combination for “Sample rate” and “Bit rate”.

Example,

Playback devices 44100 Hz 16 bit
Recording devices 44100 Hz 16 bit

Playback devices 48000 Hz 16 bit
Recording devices 48000 Hz 16 bit

Playback devices 44100 Hz 24 bit
Recording devices 44100 Hz 24 bit

Playback devices 48000 Hz 24 bit
Recording devices 48000 Hz 24 bit

However, the result was the same.
“Input device” and “Output device” do not still appear.

If I select blank“Input device” and “Output device” and enter the room,The result is this.


I checked that the WASAPI operated normally in “Reaper” and “foobar2000”.

And, I checked log.txt.(In new version.)
It saids this.
6 09 2013 14:00:26 DEBUG: Trying Pa_OpenStream() with sampleRate 44100 inputLatency 0.0029 outputLatency 0.0029 innch 1 outnch 1
6 09 2013 14:00:26 DEBUG: Input device:  (Windows WASAPI)
6 09 2013 14:00:26 DEBUG: Output device:  (Windows WASAPI)
6 09 2013 14:00:26 CRIT: Pa_OpenStream() failed: Invalid number of channels
6 09 2013 14:00:26 CRIT: create_audioStreamer_PortAudio() failed


My second problem is this.

I chose “WDM-KS” as “Audio system”.
Then, I felt the intolerable noise.

This noise resembled the noise when the buffer size of ASIO was lowered.
I think that the cause of this noise is because the buffer size of “WDM-KS” is too small.

Reaper can adjust the buffer size of “WDM-KS”.
This noise will be able to be erased if this adjustment is possible at jammr.


I am using these equipment.
win7 64bit
FocusriteSaffire 6 USB(Old Type)
VOX Tonelab ST
Line6 Toneport UX1

I expect that this project will become good.
please do your best.
stefanha
NakajimaYusuke
Because,I am very poor at English. Translation is time-consuming work for me.

Thank you for replying and sharing details! The WDM KS problem should be fixed now, see below for more information.

NakajimaYusuke
My first problem is this.

I chose “WASAPI” as “Audio system”.
Then, “Input device” and “Output device” were not displayed.

And, I checked log.txt.(In new version.)
It saids this.
6 09 2013 14:00:26 DEBUG: Trying Pa_OpenStream() with sampleRate 44100 inputLatency 0.0029 outputLatency 0.0029 innch 1 outnch 1
6 09 2013 14:00:26 DEBUG: Input device:  (Windows WASAPI)
6 09 2013 14:00:26 DEBUG: Output device:  (Windows WASAPI)
6 09 2013 14:00:26 CRIT: Pa_OpenStream() failed: Invalid number of channels
6 09 2013 14:00:26 CRIT: create_audioStreamer_PortAudio() failed

Unfortunately this problem is probably still present in jammr 1.0.2. In the future jammr will support stereo and multi-channel sound cards. Until then, you may find WASAPI cannot be used with your hardware.

NakajimaYusuke
My second problem is this.

I chose “WDM-KS” as “Audio system”.
Then, I felt the intolerable noise.

This noise resembled the noise when the buffer size of ASIO was lowered.
I think that the cause of this noise is because the buffer size of “WDM-KS” is too small.

Reaper can adjust the buffer size of “WDM-KS”.
This noise will be able to be erased if this adjustment is possible at jammr.

I think that the WDM KS problem is now fixed in jammr 1.0.2 and later:
http://jammr.net/download.html

In jammr 1.0.2 you can set the sample rate and latency (buffer size). If you experience audio glitches (pops, clicks, noise) you can try increasing the latency setting to solve the problem.
him3
Hi, some of the people in our session are hearing a one bar echo of everything that is played. Have you come across this problem/do you know how to fix it? Thanks!
stefanha
him3
Hi, some of the people in our session are hearing a one bar echo of everything that is played. Have you come across this problem/do you know how to fix it? Thanks!

Hi him3,
Is one of the users using a microphone that is picking up the sound from the speakers?

Stefan
WS
Hi Stefaha,

This is a pretty old tutorial. ( 7 years now )

Does the info here still apply to newer computer systems or have there been some updates we should know about?

Thanks
stefanha
WS
Hi Stefaha,This is a pretty old tutorial. ( 7 years now ) Does the info here still apply to newer computer systems or have there been some updates we should know about?Thanks

Hi WS,
Yes, it is still useful info. I have updated it slightly. jammr has gained autodetection of the recommended settings since this post was published. For many users it will not be necessary to manually configure the details except to select the desired input and output devices.
joecarb
Hello, just got my USB audio box set up last night and “almost” had a successful jam with my keyboard bandmate. However, he claims that my vocals were nearly a measure behind what he was playing. I was singing when I heard what was coming from my speakers, however, as I said, to him, he was not hearing it in sync. Was one of us not properly configured?

Thanks,
Joe
stefanha
joecarb
Hello, just got my USB audio box set up last night and “almost” had a successful jam with my keyboard bandmate. However, he claims that my vocals were nearly a measure behind what he was playing. I was singing when I heard what was coming from my speakers, however, as I said, to him, he was not hearing it in sync. Was one of us not properly configured?Thanks,Joe

Hi Joe,
jammr is live but not real-time. It works by locking to the chord progression and keeping you in sync. You can read about it here:
https://jammr.net/howitworks.html

There is a guide to jamming successfully here: https://forum.jammr.net/topic/1724/
For a more in-depth explanation: https://forum.jammr.net/topic/1724/

Due to this approach it is not affected by latency and you can jam with people around the world, but the trade-off is that what you're playing needs to have a repeating structure. jammr is designed for improvising to a chord progression and works for rehearsing sections of songs. It is not suitable for performing or rehearsing through full songs that change chord progression, key, etc.

Hope this helps!
captaincancel
stefanha
It works by locking to the chord progression and keeping you in sync.

Can we elaborate on this a bit? When you say “lock”, does the latency compensation algorithm use the room's BPM/BPI to send adjustments?

I'm curious to know whether playing off the built-in metronome/BPI is better than listening to another click track/drum machine that a room member is feeding in as audio, but is not synced to the room (or even the same BPM/BPI).
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.
Powered by DjangoBB