From mboxrd@z Thu Jan 1 00:00:00 1970 From: Takashi Iwai Subject: Re: Noise in begnning of alsa application Date: Mon, 21 Jan 2019 22:07:14 +0100 Message-ID: References: <4a800a2a-50ce-8bda-3b5c-26d782c40810@ivitera.com> Mime-Version: 1.0 (generated by SEMI 1.14.6 - "Maruoka") Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Received: from mx1.suse.de (mx2.suse.de [195.135.220.15]) by alsa0.perex.cz (Postfix) with ESMTP id 507C72667E7 for ; Mon, 21 Jan 2019 22:07:15 +0100 (CET) In-Reply-To: <4a800a2a-50ce-8bda-3b5c-26d782c40810@ivitera.com> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: alsa-devel-bounces@alsa-project.org Sender: alsa-devel-bounces@alsa-project.org To: Pavel Hofman Cc: alsa-devel@alsa-project.org, Chakravarthi Pradeep List-Id: alsa-devel@alsa-project.org On Mon, 21 Jan 2019 19:07:09 +0100, Pavel Hofman wrote: > > > Dne 21. 01. 19 v 18:46 Chakravarthi Pradeep napsal(a): > > On Wed, Jan 16, 2019 at 5:07 PM Chakravarthi Pradeep wrote: > >> > >> I have ALSA driver for capture device which works fine with VLC. when > >> I try to capture and render using own application having some issues. > >> When I start play, it has some noise in beginning and audio is > >> discontinuous in between. Discontinuous means, some audio frame are > >> getting dropped in between. cut cut sound for sine wave. I have tried > >> snd_pcm_drop and snd_pcm_drain but no use. Kindly suggest how to > >> remove the starting noise in ALSA. > > I have experienced the same issue, with ice172 driver. IMO part of the > first period when opening the stream contains samples from the > previous run. In one of my my projects I have to drop initial samples > in order to get correct and stable results from FFT analysis. > > I do not know if the problem is in the driver, alsa-lib or if the > application (in my case the SDL library) uses alsa-lib API > incorrectly. To tell the truth I did not investigate details and > "fixed" it by dropping the initial samples. It's likely a driver-specific problem, but still wondering how it happens. Does the whole first period contain the previous samples, or just a few beginning samples? Also is it on the x86 PC, or on other architecture? thanks, Takashi