From job at kimmeringer.de  Wed Feb  5 13:57:27 2020
From: job at kimmeringer.de (Lothar Kimmeringer)
Date: Wed, 5 Feb 2020 14:57:27 +0100
Subject: <Sound Dev>
 com.sun.media.sound.SoftMainMixer.processAudioBuffers
 stoppedMixers-handling-bug?
In-Reply-To: <ccf18acc-7590-6e0e-654e-8e21e2312010@oracle.com>
References: <1d58ba9e-5ea0-9f20-28d8-803177d68058@kimmeringer.de>
 <ccf18acc-7590-6e0e-654e-8e21e2312010@oracle.com>
Message-ID: <f40ec36f-bc6c-ee0e-63db-ddc4953d0083@kimmeringer.de>



Am 30.01.2020 um 04:40 schrieb Sergey Bylokhov:

> Thank you for this bug report, can you please file it:
> https://bugs.java.com/bugdatabase

I seem to have missed the link to that site from the OpenJDK-site for some
reason but I remember the link from the Good Old Days[TM] ;-)

Filed a bug report, thanks for pointing me to the form. Should be an
easy fix in my eyes.


Thanks and cheers, Lothar


From izmaelverhage at gmail.com  Sun Feb  9 08:32:22 2020
From: izmaelverhage at gmail.com (Izmar Verhage)
Date: Sun, 9 Feb 2020 09:32:22 +0100
Subject: <Sound Dev> WASAPI
Message-ID: <CAMyM1U4Ve_hLbCRSANS02rvH6cAwMGLd5hv6vKWenS0Dy-y6nQ@mail.gmail.com>

Hi all,

I recently joined OpenJDK after concluding that the Java Sound API seems to
have no way to speak WASAPI. In the OpenJDK sources that I checked I was
only able to find PLATFORM_API_WinOS_DirectSound.cpp and no WASAPI variant
of a similar file.

To my knowledge WASAPI should give us much lower latency than DirectSound,
regardless of whether shared or exclusive WASAPI mode is used.

Additionally DirectSound has been deprecated for quite some time.

If the only impediment to have WASAPI is that someone should build it, I
would like to work on this.

Cheers,
Izmar
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.java.net/pipermail/sound-dev/attachments/20200209/1b7c2a43/attachment.htm>

From Sergey.Bylokhov at oracle.com  Mon Feb 10 07:03:04 2020
From: Sergey.Bylokhov at oracle.com (Sergey Bylokhov)
Date: Sun, 9 Feb 2020 23:03:04 -0800
Subject: <Sound Dev> RFR: 8238738 AudioSystem.getMixerInfo() takes about 30
 sec to report a gone audio device
Message-ID: <730d5ba4-b467-4168-58a0-a14b539e9825@oracle.com>

Hello.
Please review the fix for jdk/client.

Bug: https://bugs.openjdk.java.net/browse/JDK-8238738
Fix: http://cr.openjdk.java.net/~serb/8238738/webrev.00

I propose to decrease the constant "WAIT_BETWEEN_CACHE_REFRESH_MILLIS"
and update the list of devices more often. The 30 seconds delay was
used from the very beginning when we cache the service providers as well,
but that cache was removed back in 2013.

Note that if the data is cached then the method works x10 times
faster, this is why I decreased the value but did not delete it
completely, 3 seconds should be fine.

-- 
Best regards, Sergey.

From philip.race at oracle.com  Mon Feb 10 20:12:14 2020
From: philip.race at oracle.com (Phil Race)
Date: Mon, 10 Feb 2020 12:12:14 -0800
Subject: <Sound Dev> RFR: 8238738 AudioSystem.getMixerInfo() takes about
 30 sec to report a gone audio device
In-Reply-To: <730d5ba4-b467-4168-58a0-a14b539e9825@oracle.com>
References: <730d5ba4-b467-4168-58a0-a14b539e9825@oracle.com>
Message-ID: <89adb4cc-f48e-dc78-3311-b1e7e3b9060b@oracle.com>

+1

-phil

On 2/9/20 11:03 PM, Sergey Bylokhov wrote:
> Hello.
> Please review the fix for jdk/client.
>
> Bug: https://bugs.openjdk.java.net/browse/JDK-8238738
> Fix: http://cr.openjdk.java.net/~serb/8238738/webrev.00
>
> I propose to decrease the constant "WAIT_BETWEEN_CACHE_REFRESH_MILLIS"
> and update the list of devices more often. The 30 seconds delay was
> used from the very beginning when we cache the service providers as well,
> but that cache was removed back in 2013.
>
> Note that if the data is cached then the method works x10 times
> faster, this is why I decreased the value but did not delete it
> completely, 3 seconds should be fine.
>


From Sergey.Bylokhov at oracle.com  Thu Feb 13 08:06:42 2020
From: Sergey.Bylokhov at oracle.com (Sergey Bylokhov)
Date: Thu, 13 Feb 2020 00:06:42 -0800
Subject: <Sound Dev> WASAPI
In-Reply-To: <CAMyM1U4Ve_hLbCRSANS02rvH6cAwMGLd5hv6vKWenS0Dy-y6nQ@mail.gmail.com>
References: <CAMyM1U4Ve_hLbCRSANS02rvH6cAwMGLd5hv6vKWenS0Dy-y6nQ@mail.gmail.com>
Message-ID: <45b82167-33b8-c15b-ec26-e9c10d773ff5@oracle.com>

Hi Izmar,

You are very welcome to the OpenJDK team! And thank you for sharing your ideas on how to improve JavaSound. Before you start, could you tell us more about WASAPI(shared)? What are the advantages over DirectSound? I am not sure we can use WASAPI(exclusive) and block everything else on the system.

On 2/9/20 12:32 am, Izmar Verhage wrote:
> I recently joined OpenJDK after concluding that the Java Sound API seems to have no way to speak WASAPI. In the OpenJDK sources that I checked I was only able to find PLATFORM_API_WinOS_DirectSound.cpp and no WASAPI variant of a similar file.
> 
> To my knowledge WASAPI should give us much lower latency than DirectSound, regardless of whether shared or exclusive WASAPI mode is used.
> 
> Additionally DirectSound has been deprecated for quite some time.
> 
> If the only impediment to have WASAPI is that someone should build it, I would like to work on this.


-- 
Best regards, Sergey.