From xvid-devel-bounces@xvid.org Thu Dec 8 08:08:38 2005 Return-Path: X-Original-To: itdp@fh-biergarten.de Delivered-To: itdp@localhost Received: from edu.bnhof.de (edu.bnhof.de [213.167.167.52]) by mail.kliche.org (Postfix) with ESMTP id D5600126D82 for ; Thu, 8 Dec 2005 08:08:37 +0100 (CET) Received: from edu.bnhof.de (localhost [127.0.0.1]) by edu.bnhof.de (Postfix) with ESMTP id 52AC11EEC5; Thu, 8 Dec 2005 08:08:19 +0100 (CET) X-Original-To: xvid-devel@edu.bnhof.de Delivered-To: xvid-devel@edu.bnhof.de Received: from s6.lansco.de (s6.lansco.de [213.239.218.110]) by edu.bnhof.de (Postfix) with ESMTP id D8C1A1EEC1 for ; Thu, 8 Dec 2005 08:08:05 +0100 (CET) Received: from mail-ihug.icp-qv1-irony5.iinet.net.au (ihug-mail.icp-qv1-irony5.iinet.net.au [203.59.1.199]) by s6.lansco.de (Postfix) with ESMTP id 38CB227CF0 for ; Thu, 8 Dec 2005 08:08:02 +0100 (CET) Received: from 203-173-11-165.dyn.iinet.net.au (HELO [10.1.1.3]) ([203.173.11.165]) by mail-ihug.icp-qv1-irony5.iinet.net.au with ESMTP; 08 Dec 2005 15:07:59 +0800 X-BrightmailFiltered: true X-Brightmail-Tracker: AAAAAA== Message-ID: <4397DC12.9030109@syskin.cjb.net> Date: Thu, 08 Dec 2005 17:39:06 +1030 From: Radek Czyz User-Agent: Thunderbird 1.5 (Windows/20051201) MIME-Version: 1.0 To: xvid-devel@xvid.org References: <1132929315.4387212367133@www.lansco.de> <438731E1.9050905@syskin.cjb.net> <1132934488.4387355800550@www.lansco.de> <1132934559.4387359f82e20@www.lansco.de> <43880598.7010605@syskin.cjb.net> <1133001253.43883a25b075c@www.lansco.de> <43884191.3030502@syskin.cjb.net> <1133004637.4388475da551b@www.lansco.de> <43885821.801@syskin.cjb.net> <1133013826.43886b42c6de9@www.lansco.de> In-Reply-To: <1133013826.43886b42c6de9@www.lansco.de> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Subject: [XviD-devel] lambda patch X-BeenThere: xvid-devel@xvid.org X-Mailman-Version: 2.1.4 Precedence: list Reply-To: xvid-devel@xvid.org List-Id: xvid-devel.xvid.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: xvid-devel-bounces@xvid.org Errors-To: xvid-devel-bounces@xvid.org Hi everyone, I finally found some time to tweak lambdas and mode decisions back. Effect: - slowdown introduced by Michael's patch remains, but is now translated to PSNR gain at all quants, so it's worth it. - I changed lambdas to be linear for low quants too. I understand Michael's reasoning for lower-than-psnr-best values (and probably agree with the theory). However, I believe that such HVS experiments belong to 1.2.-127 tree, and need some testing. With linear quants everywhere, at least we can explain slowdown as PSNR-boosting. - I tweaked constants for much better VHQ 0 mode decision. Patch is short, I put it online at http://syskin.is.dreaming.org/lambda.txt I strongly think this is the best thing to do for 1.1.0, and the best starting point for 1.2.-127. Comments? If no one writes anything (why that crossed my mind? lol) I'll commit to both cvs head and 1.1.0. If you disagree, say so now. Regards, Radek _______________________________________________ XviD-devel mailing list XviD-devel@xvid.org http://list.xvid.org/mailman/listinfo/xvid-devel From xvid-devel-bounces@xvid.org Thu Dec 8 10:31:48 2005 Return-Path: X-Original-To: itdp@fh-biergarten.de Delivered-To: itdp@localhost Received: from edu.bnhof.de (edu.bnhof.de [213.167.167.52]) by mail.kliche.org (Postfix) with ESMTP id D9E4D126D82 for ; Thu, 8 Dec 2005 10:31:47 +0100 (CET) Received: from edu.bnhof.de (localhost [127.0.0.1]) by edu.bnhof.de (Postfix) with ESMTP id BB88D1EEF3; Thu, 8 Dec 2005 10:31:01 +0100 (CET) X-Original-To: xvid-devel@edu.bnhof.de Delivered-To: xvid-devel@edu.bnhof.de Received: from s6.lansco.de (s6.lansco.de [213.239.218.110]) by edu.bnhof.de (Postfix) with ESMTP id B53121EEEF for ; Thu, 8 Dec 2005 10:30:21 +0100 (CET) Received: from s5.lansco.de (s5.lansco.de [212.63.130.170]) by s6.lansco.de (Postfix) with ESMTP id 2440F27B35 for ; Thu, 8 Dec 2005 10:30:20 +0100 (CET) Received: from localhost (s5.lansco.de [212.63.130.170]) by s5.lansco.de (Postfix) with ESMTP id 05AEF258CA for ; Thu, 8 Dec 2005 10:30:19 +0100 (CET) Received: from p54B5FB29.dip.t-dialin.net (p54B5FB29.dip.t-dialin.net [84.181.251.41]) by www.lansco.de (IMP) with HTTP for ; Thu, 8 Dec 2005 10:30:19 +0100 Message-ID: <1134034219.4397fd2bcc3ea@www.lansco.de> Date: Thu, 8 Dec 2005 10:30:19 +0100 From: Michael Militzer To: xvid-devel@xvid.org Subject: Re: [XviD-devel] lambda patch References: <1132929315.4387212367133@www.lansco.de> <438731E1.9050905@syskin.cjb.net> <1132934488.4387355800550@www.lansco.de> <1132934559.4387359f82e20@www.lansco.de> <43880598.7010605@syskin.cjb.net> <1133001253.43883a25b075c@www.lansco.de> <43884191.3030502@syskin.cjb.net> <1133004637.4388475da551b@www.lansco.de> <43885821.801@syskin.cjb.net> <1133013826.43886b42c6de9@www.lansco.de> <4397DC12.9030109@syskin.cjb.net> In-Reply-To: <4397DC12.9030109@syskin.cjb.net> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit User-Agent: Internet Messaging Program (IMP) 3.2.3 X-Originating-IP: 84.181.251.41 X-BeenThere: xvid-devel@xvid.org X-Mailman-Version: 2.1.4 Precedence: list Reply-To: xvid-devel@xvid.org List-Id: xvid-devel.xvid.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: xvid-devel-bounces@xvid.org Errors-To: xvid-devel-bounces@xvid.org Hi Radek, indeed, it's a short patch ;) I have no objections against it: it's reasonable to give a higher penalty to 8x8 modes, especially at VHQ=0. Also, I have no problem with switching back to a pure linear lambda scaling also for the low quants. As said, it won't make much of a difference either way. As a side note: I've also ran some tests recently because of the lambdas. However, aside from 8x8/16x16 mode decision it also seemed to me that inter/intra decision isn't working perfectly with VHQ=0. On some sequences I had large gains from lowering the value of MV16_INTER_BIAS, e.g. to 400. You may want to give it a try if it works for your test sequences too... Regards, Michael Quoting Radek Czyz : > Hi everyone, > > I finally found some time to tweak lambdas and mode decisions back. > Effect: > > - slowdown introduced by Michael's patch remains, but is now translated > to PSNR gain at all quants, so it's worth it. > > - I changed lambdas to be linear for low quants too. I understand > Michael's reasoning for lower-than-psnr-best values (and probably agree > with the theory). However, I believe that such HVS experiments belong to > 1.2.-127 tree, and need some testing. With linear quants everywhere, at > least we can explain slowdown as PSNR-boosting. > > - I tweaked constants for much better VHQ 0 mode decision. > > Patch is short, I put it online at http://syskin.is.dreaming.org/lambda.txt > > I strongly think this is the best thing to do for 1.1.0, and the best > starting point for 1.2.-127. > > Comments? If no one writes anything (why that crossed my mind? lol) I'll > commit to both cvs head and 1.1.0. If you disagree, say so now. > > Regards, > Radek > _______________________________________________ > XviD-devel mailing list > XviD-devel@xvid.org > http://list.xvid.org/mailman/listinfo/xvid-devel > _______________________________________________ XviD-devel mailing list XviD-devel@xvid.org http://list.xvid.org/mailman/listinfo/xvid-devel From xvid-devel-bounces@xvid.org Fri Dec 9 06:12:51 2005 Return-Path: X-Original-To: itdp@fh-biergarten.de Delivered-To: itdp@localhost Received: from edu.bnhof.de (edu.bnhof.de [213.167.167.52]) by mail.kliche.org (Postfix) with ESMTP id DCE92126D82 for ; Fri, 9 Dec 2005 06:12:50 +0100 (CET) Received: from edu.bnhof.de (localhost [127.0.0.1]) by edu.bnhof.de (Postfix) with ESMTP id 11E3015815; Fri, 9 Dec 2005 06:12:25 +0100 (CET) X-Original-To: xvid-devel@edu.bnhof.de Delivered-To: xvid-devel@edu.bnhof.de Received: from s6.lansco.de (s6.lansco.de [213.239.218.110]) by edu.bnhof.de (Postfix) with ESMTP id E88FCA83A for ; Fri, 9 Dec 2005 06:12:05 +0100 (CET) Received: from mail-ihug.icp-qv1-irony1.iinet.net.au (ihug-mail.icp-qv1-irony1.iinet.net.au [203.59.1.195]) by s6.lansco.de (Postfix) with ESMTP id D352121B85 for ; Fri, 9 Dec 2005 06:12:02 +0100 (CET) Received: from 203-173-11-165.dyn.iinet.net.au (HELO [10.1.1.3]) ([203.173.11.165]) by mail-ihug.icp-qv1-irony1.iinet.net.au with ESMTP; 09 Dec 2005 13:12:00 +0800 X-BrightmailFiltered: true X-Brightmail-Tracker: AAAAAA== Message-ID: <43991242.3030608@syskin.cjb.net> Date: Fri, 09 Dec 2005 15:42:34 +1030 From: Radek Czyz User-Agent: Thunderbird 1.5 (Windows/20051201) MIME-Version: 1.0 To: xvid-devel@xvid.org Subject: Re: [XviD-devel] lambda patch References: <1132929315.4387212367133@www.lansco.de> <438731E1.9050905@syskin.cjb.net> <1132934488.4387355800550@www.lansco.de> <1132934559.4387359f82e20@www.lansco.de> <43880598.7010605@syskin.cjb.net> <1133001253.43883a25b075c@www.lansco.de> <43884191.3030502@syskin.cjb.net> <1133004637.4388475da551b@www.lansco.de> <43885821.801@syskin.cjb.net> <1133013826.43886b42c6de9@www.lansco.de> <4397DC12.9030109@syskin.cjb.net> <1134034219.4397fd2bcc3ea@www.lansco.de> In-Reply-To: <1134034219.4397fd2bcc3ea@www.lansco.de> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-BeenThere: xvid-devel@xvid.org X-Mailman-Version: 2.1.4 Precedence: list Reply-To: xvid-devel@xvid.org List-Id: xvid-devel.xvid.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: xvid-devel-bounces@xvid.org Errors-To: xvid-devel-bounces@xvid.org Heyhey, I committed the small patch to HEAD and 1_0_branch, and tagged the two files as release_1_1_0. Hopefully I did this correctly, I was mostly guessing what I need to do ;)) I also committed the "expose VHQ and trellis lambda to plugins" patch (to HEAD), with one modification over what I shown you before (there was an xvidcore bug that was harmless until now). The patch is a NOP if no lambda plugins are connected. Radek Michael Militzer wrote: > Hi Radek, > > indeed, it's a short patch ;) I have no objections against it: it's > reasonable to give a higher penalty to 8x8 modes, especially at VHQ=0. > Also, I have no problem with switching back to a pure linear lambda > scaling also for the low quants. As said, it won't make much of a > difference either way. > > As a side note: I've also ran some tests recently because of the lambdas. > However, aside from 8x8/16x16 mode decision it also seemed to me that > inter/intra decision isn't working perfectly with VHQ=0. On some sequences > I had large gains from lowering the value of MV16_INTER_BIAS, e.g. to 400. > You may want to give it a try if it works for your test sequences too... > > Regards, > Michael > > > Quoting Radek Czyz : > >> Hi everyone, >> >> I finally found some time to tweak lambdas and mode decisions back. >> Effect: >> >> - slowdown introduced by Michael's patch remains, but is now translated >> to PSNR gain at all quants, so it's worth it. >> >> - I changed lambdas to be linear for low quants too. I understand >> Michael's reasoning for lower-than-psnr-best values (and probably agree >> with the theory). However, I believe that such HVS experiments belong to >> 1.2.-127 tree, and need some testing. With linear quants everywhere, at >> least we can explain slowdown as PSNR-boosting. >> >> - I tweaked constants for much better VHQ 0 mode decision. >> >> Patch is short, I put it online at http://syskin.is.dreaming.org/lambda.txt >> >> I strongly think this is the best thing to do for 1.1.0, and the best >> starting point for 1.2.-127. >> >> Comments? If no one writes anything (why that crossed my mind? lol) I'll >> commit to both cvs head and 1.1.0. If you disagree, say so now. >> >> Regards, >> Radek >> _______________________________________________ >> XviD-devel mailing list >> XviD-devel@xvid.org >> http://list.xvid.org/mailman/listinfo/xvid-devel >> > > > > > _______________________________________________ > XviD-devel mailing list > XviD-devel@xvid.org > http://list.xvid.org/mailman/listinfo/xvid-devel > > _______________________________________________ XviD-devel mailing list XviD-devel@xvid.org http://list.xvid.org/mailman/listinfo/xvid-devel From xvid-devel-bounces@xvid.org Fri Dec 9 16:27:21 2005 Return-Path: X-Original-To: itdp@fh-biergarten.de Delivered-To: itdp@localhost Received: from edu.bnhof.de (edu.bnhof.de [213.167.167.52]) by mail.kliche.org (Postfix) with ESMTP id C01B5126D82 for ; Fri, 9 Dec 2005 16:27:20 +0100 (CET) Received: from edu.bnhof.de (localhost [127.0.0.1]) by edu.bnhof.de (Postfix) with ESMTP id E4E3A1991D; Fri, 9 Dec 2005 16:26:44 +0100 (CET) X-Original-To: xvid-devel@edu.bnhof.de Delivered-To: xvid-devel@edu.bnhof.de Received: from s6.lansco.de (s6.lansco.de [213.239.218.110]) by edu.bnhof.de (Postfix) with ESMTP id 5B9301916A for ; Fri, 9 Dec 2005 16:26:31 +0100 (CET) Received: from cmsout02.mbox.net (cmsout02.mbox.net [165.212.64.32]) by s6.lansco.de (Postfix) with ESMTP id DAA432194B for ; Fri, 9 Dec 2005 16:26:28 +0100 (CET) Received: from cmsout02.mbox.net (cmsout02.mbox.net [165.212.64.32]) by cmsout02.mbox.net (Postfix) with ESMTP id 0839B4C8EA for ; Fri, 9 Dec 2005 15:26:27 +0000 (GMT) Received: from cmsapps01.cms.usa.net [165.212.11.136] by cmsout02.mbox.net via smtad (C8.MAIN.3.27I); Fri, 09 Dec 2005 15:26:26 GMT X-USANET-Source: 165.212.11.136 IN avcp-xvidmail@usa.net cmsapps01.cms.usa.net X-USANET-MsgId: XID413JLiPAb2021X02 Received: from mumba [205.247.13.101] by cmsapps01.cms.usa.net (ASMTP/codepro@usa.net) via mtad (C8.MAIN.3.27E) with ESMTP id 772JLiPAy0132M36; Fri, 09 Dec 2005 15:26:24 GMT X-USANET-Auth: 205.247.13.101 AUTH codepro@usa.net mumba From: "Alex Volkov" To: Date: Fri, 9 Dec 2005 10:25:21 -0500 MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_NextPart_000_0035_01C5FCAA.DCA33310" X-Mailer: Microsoft Office Outlook, Build 11.0.6353 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1506 Thread-Index: AcX81MTqkPsnHLtYTGSYKhL2N6NrSA== Message-ID: <772JLiPAy0132M36@cmsapps01.cms.usa.net> Z-USANET-MsgId: XID772JLiPAZ0132X36 Subject: [XviD-devel] Bug+patch: mem corruption w/ unaligned bitstream writes, take 2 X-BeenThere: xvid-devel@xvid.org X-Mailman-Version: 2.1.4 Precedence: list Reply-To: xvid-devel@xvid.org List-Id: xvid-devel.xvid.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: xvid-devel-bounces@xvid.org Errors-To: xvid-devel-bounces@xvid.org This is a multi-part message in MIME format. ------=_NextPart_000_0035_01C5FCAA.DCA33310 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Hello, I am reposting this as I never got any response the first time. There is a bug in 1.1-HEAD Bitstream init/writing functions that causes memory corruption when a bitstream passed to enc_encode() is not aligned on 32bit boundary. This can wreak havoc on encoder output buffer. If you do not write out the data immediately after enc_encode() returns it, and instead, simply move the bitstream ptr, on the next enc_encode() operation, the previous couple bytes (the unaligned ones) get clobbered. For example, you may choose to forego flushing the dx50-compat empty marker, which is 7 bytes long, and simply advance the output ptr by 7 bytes. On the next enc_encode() operation, the last 3 bytes (which contain the time_increment) get clobbered. This of course causes severe block artifacts during decoding. The patch to fix unaligned bitstream inits is attached. -Alex. ------=_NextPart_000_0035_01C5FCAA.DCA33310 Content-Type: text/plain; name="bitstream.patch.txt" Content-Transfer-Encoding: quoted-printable Content-Disposition: attachment; filename="bitstream.patch.txt" Index: bitstream.h=0A= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=0A= RCS file: /xvid/xvidcore/src/bitstream/bitstream.h,v=0A= retrieving revision 1.22=0A= diff -u -u -p -r1.22 bitstream.h=0A= --- bitstream.h 23 May 2005 09:29:43 -0000 1.22=0A= +++ bitstream.h 9 Dec 2005 15:17:22 -0000=0A= @@ -193,8 +193,12 @@ BitstreamInit(Bitstream * const bs,=0A= #endif=0A= bs->bufb =3D tmp;=0A= =0A= - bs->buf =3D 0;=0A= bs->pos =3D bs->initpos =3D bitpos*8;=0A= + /* preserve the intervening bytes */=0A= + if (bs->initpos > 0)=0A= + bs->buf =3D bs->bufa & ~((1 << (32 - bs->initpos)) - 1);=0A= + else=0A= + bs->buf =3D 0;=0A= bs->length =3D length;=0A= }=0A= =0A= @@ -220,7 +224,11 @@ BitstreamReset(Bitstream * const bs)=0A= #endif=0A= bs->bufb =3D tmp;=0A= =0A= - bs->buf =3D 0;=0A= + /* preserve the intervening bytes */=0A= + if (bs->initpos > 0)=0A= + bs->buf =3D bs->bufa & ~((1 << (32 - bs->initpos)) - 1);=0A= + else=0A= + bs->buf =3D 0;=0A= bs->pos =3D bs->initpos;=0A= }=0A= =0A= ------=_NextPart_000_0035_01C5FCAA.DCA33310 Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ XviD-devel mailing list XviD-devel@xvid.org http://list.xvid.org/mailman/listinfo/xvid-devel ------=_NextPart_000_0035_01C5FCAA.DCA33310-- From xvid-devel-bounces@xvid.org Sat Dec 17 10:18:58 2005 Return-Path: X-Original-To: itdp@fh-biergarten.de Delivered-To: itdp@localhost Received: from edu.bnhof.de (edu.bnhof.de [213.167.167.52]) by mail.kliche.org (Postfix) with ESMTP id C08D3126D85 for ; Sat, 17 Dec 2005 10:18:58 +0100 (CET) Received: from edu.bnhof.de (localhost [127.0.0.1]) by edu.bnhof.de (Postfix) with ESMTP id 7DF991EE92; Sat, 17 Dec 2005 10:18:43 +0100 (CET) X-Original-To: xvid-devel@edu.bnhof.de Delivered-To: xvid-devel@edu.bnhof.de Received: from s6.lansco.de (s6.lansco.de [213.239.218.110]) by edu.bnhof.de (Postfix) with ESMTP id 04BF61994C for ; Sat, 17 Dec 2005 10:18:34 +0100 (CET) Received: from mail-ihug.icp-qv1-irony1.iinet.net.au (ihug-mail.icp-qv1-irony1.iinet.net.au [203.59.1.195]) by s6.lansco.de (Postfix) with ESMTP id 4EE35277E1 for ; Sat, 17 Dec 2005 10:18:31 +0100 (CET) Received: from 203-173-11-165.dyn.iinet.net.au (HELO [10.1.1.3]) ([203.173.11.165]) by mail-ihug.icp-qv1-irony1.iinet.net.au with ESMTP; 17 Dec 2005 17:18:28 +0800 X-BrightmailFiltered: true X-Brightmail-Tracker: AAAAAA== Message-ID: <43A3D7DE.5040308@syskin.cjb.net> Date: Sat, 17 Dec 2005 19:48:22 +1030 From: Radek Czyz User-Agent: Thunderbird 1.5 (Windows/20051201) MIME-Version: 1.0 To: xvid-devel@xvid.org Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Subject: [XviD-devel] footprint patch X-BeenThere: xvid-devel@xvid.org X-Mailman-Version: 2.1.4 Precedence: list Reply-To: xvid-devel@xvid.org List-Id: xvid-devel.xvid.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: xvid-devel-bounces@xvid.org Errors-To: xvid-devel-bounces@xvid.org Hi everyone, I just made a big ugly patch for one small thing - I reduced memory taken but each MACROBLOCK instance from 460 to 374 bytes. That's 100kb less data for PAL encoding (times two+number of bframes), which hopefully reduces some cache usage. The idea is to use union and keep b-vop specific stuff (8 vectors) and p/i-vop specific stuff (ac/dc prediction, mostly) in one memory space. patch at http://syskin.is.dreaming.org/sizeof.patch If you don't like this patch, yell. Otherwise, I'll commit it tomorrow. Radek _______________________________________________ XviD-devel mailing list XviD-devel@xvid.org http://list.xvid.org/mailman/listinfo/xvid-devel From xvid-devel-bounces@xvid.org Sun Dec 18 14:33:55 2005 Return-Path: X-Original-To: itdp@fh-biergarten.de Delivered-To: itdp@localhost Received: from edu.bnhof.de (edu.bnhof.de [213.167.167.52]) by mail.kliche.org (Postfix) with ESMTP id 417B5126D9B for ; Sun, 18 Dec 2005 14:33:55 +0100 (CET) Received: from edu.bnhof.de (localhost [127.0.0.1]) by edu.bnhof.de (Postfix) with ESMTP id C907F1EEFF; Sun, 18 Dec 2005 14:33:28 +0100 (CET) X-Original-To: xvid-devel@edu.bnhof.de Delivered-To: xvid-devel@edu.bnhof.de Received: from s6.lansco.de (s6.lansco.de [213.239.218.110]) by edu.bnhof.de (Postfix) with ESMTP id 27B6C1EEF8 for ; Sun, 18 Dec 2005 14:33:05 +0100 (CET) Received: from smtp1.libero.it (smtp1.libero.it [193.70.192.51]) by s6.lansco.de (Postfix) with ESMTP id 2961D28D0A for ; Sun, 18 Dec 2005 14:32:58 +0100 (CET) Received: from localhost (172.16.1.84) by smtp1.libero.it (7.0.027-DD01) id 439D8FC30092EDC8 for xvid-devel@xvid.org; Sun, 18 Dec 2005 14:32:57 +0100 Received: from smtp2.libero.it ([172.16.1.97]) by localhost (asav5.libero.it [193.70.192.154]) (amavisd-new, port 10024) with ESMTP id 17022-08-2 for ; Sun, 18 Dec 2005 14:32:57 +0100 (CET) Received: from libero.it (172.16.1.86) by smtp2.libero.it (7.0.027-DD01) id 431C3B200118B509 for xvid-devel@xvid.org; Sun, 18 Dec 2005 14:32:57 +0100 Date: Sun, 18 Dec 2005 14:32:57 +0100 Message-Id: MIME-Version: 1.0 X-Sensitivity: 3 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable From: "carlo.bramix" To: "xvid-devel" X-XaM3-API-Version: 4.3 (R1) (B3pl11) X-SenderIP: 151.25.139.105 X-Scanned: with antispam and antivirus automated system at libero.it Subject: [XviD-devel] Problem with BS_VERSION_BUGGY_CHROMA_ROUNDING? X-BeenThere: xvid-devel@xvid.org X-Mailman-Version: 2.1.4 Precedence: list Reply-To: xvid-devel@xvid.org List-Id: xvid-devel.xvid.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: xvid-devel-bounces@xvid.org Errors-To: xvid-devel-bounces@xvid.org Hello, I'm having some bugs when playing some movies with XVID. The colors of the image are diffused like a strange fluid effect. The problem disappears if I used the DECODER.C from the 1.1.0 BETA 2 sour= ces. By the way, there is a thing that made the debugging much more difficult.= Into the newest DECODER.C all tab chars have been changed with two spaces= and this makes impossible a decent file comparison! I spent one day for making the two sources comparable. I also downloaded the history of that file. The bug on DECODER.C happens when I switch from version 1.71 to 1.72. Perhaps into a piece of code like this one: if (dec->quarterpel) { if (dec->bs_version <=3D BS_VERSION_BUGGY_CHROMA_ROUNDING) { uv_dx =3D (uv_dx>>1) | (uv_dx&1); uv_dy =3D (uv_dy>>1) | (uv_dy&1); } else { uv_dx /=3D 2; uv_dy /=3D 2; } } the "<=3D" symbol into the second IF statement should be converted to ">"= , otherwise older versions will fall wrongly into the first code and vice= versa Into those (old) movie clips, the bs_version member is ZERO. By applying the described change, the graphic output is correct. What do you think about it? Sincerely, Carlo Bramini _______________________________________________ XviD-devel mailing list XviD-devel@xvid.org http://list.xvid.org/mailman/listinfo/xvid-devel From xvid-devel-bounces@xvid.org Sun Dec 18 16:21:31 2005 Return-Path: X-Original-To: itdp@fh-biergarten.de Delivered-To: itdp@localhost Received: from edu.bnhof.de (edu.bnhof.de [213.167.167.52]) by mail.kliche.org (Postfix) with ESMTP id 99D0E126D9B for ; Sun, 18 Dec 2005 16:21:31 +0100 (CET) Received: from edu.bnhof.de (localhost [127.0.0.1]) by edu.bnhof.de (Postfix) with ESMTP id 621211EF04; Sun, 18 Dec 2005 16:20:55 +0100 (CET) X-Original-To: xvid-devel@edu.bnhof.de Delivered-To: xvid-devel@edu.bnhof.de Received: from s6.lansco.de (s6.lansco.de [213.239.218.110]) by edu.bnhof.de (Postfix) with ESMTP id 24C991EF02 for ; Sun, 18 Dec 2005 16:20:06 +0100 (CET) Received: from mail.gmx.net (mail.gmx.net [213.165.64.21]) by s6.lansco.de (Postfix) with SMTP id 2D5DE279F9 for ; Sun, 18 Dec 2005 16:20:04 +0100 (CET) Received: (qmail invoked by alias); 18 Dec 2005 15:20:04 -0000 Received: from cm239-124.liwest.at (EHLO njuuus.desdemona.ssw.uni-linz.ac.at) [81.10.239.124] by mail.gmx.net (mp010) with SMTP; 18 Dec 2005 16:20:04 +0100 X-Authenticated: #1641027 Received: from localhost (HELO gmx.at) [127.0.0.1] by njuuus.desdemona.ssw.uni-linz.ac.at (81.10.239.124) with ESMTP (Classic Hamster Version 2.0 Build 2.0.0.1) ; Sun, 18 Dec 2005 16:20:02 +0100 Message-ID: <43A57E22.3F05BEE7@njuuus.desdemona.ssw.uni-linz.ac.at> Date: Sun, 18 Dec 2005 16:20:02 +0100 From: Kurt Bernhard Pruenner Organization: Sysadmins Anonymous X-Mailer: Mozilla 4.79 [en] (Windows NT 5.0; U) MIME-Version: 1.0 To: xvid-devel@xvid.org Subject: Re: [XviD-devel] Problem with BS_VERSION_BUGGY_CHROMA_ROUNDING? References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Posting-Agent: Hamster/2.0.0.1 (with KorrNews) X-Message-Flag: Outlook not so good. (Magic 8-Ball) Lines: 17 X-Y-GMX-Trusted: 0 X-BeenThere: xvid-devel@xvid.org X-Mailman-Version: 2.1.4 Precedence: list Reply-To: xvid-devel@xvid.org List-Id: xvid-devel.xvid.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: xvid-devel-bounces@xvid.org Errors-To: xvid-devel-bounces@xvid.org "carlo.bramix" wrote: > By the way, there is a thing that made the debugging much more > difficult. > Into the newest DECODER.C all tab chars have been changed with two > spaces and this makes impossible a decent file comparison! Well, I don't know how decent diff is for you, but those options of it help a lot: -E --ignore-tab-expansion Ignore changes due to tab expansion. -b --ignore-space-change Ignore changes in the amount of white space. -w --ignore-all-space Ignore all white space. -B --ignore-blank-lines Ignore changes whose lines are all blank. -- Kurt Bernhard Pruenner --- Haendelstrasse 17 --- 4020 Linz --- Austria .......It might be written "Mindfuck", but it's spelt "L-A-I-N"....... np: David Holmes - Let's Get Killed (Let's Get Killed) _______________________________________________ XviD-devel mailing list XviD-devel@xvid.org http://list.xvid.org/mailman/listinfo/xvid-devel From xvid-devel-bounces@xvid.org Sun Dec 18 16:40:28 2005 Return-Path: X-Original-To: itdp@fh-biergarten.de Delivered-To: itdp@localhost Received: from edu.bnhof.de (edu.bnhof.de [213.167.167.52]) by mail.kliche.org (Postfix) with ESMTP id 9485F126D9B for ; Sun, 18 Dec 2005 16:40:28 +0100 (CET) Received: from edu.bnhof.de (localhost [127.0.0.1]) by edu.bnhof.de (Postfix) with ESMTP id 51ACA1EF0E; Sun, 18 Dec 2005 16:39:48 +0100 (CET) X-Original-To: xvid-devel@edu.bnhof.de Delivered-To: xvid-devel@edu.bnhof.de Received: from s6.lansco.de (s6.lansco.de [213.239.218.110]) by edu.bnhof.de (Postfix) with ESMTP id 733271EF0A for ; Sun, 18 Dec 2005 16:39:24 +0100 (CET) Received: from mail-ihug.icp-qv1-irony1.iinet.net.au (ihug-mail.icp-qv1-irony1.iinet.net.au [203.59.1.195]) by s6.lansco.de (Postfix) with ESMTP id 7227628D04 for ; Sun, 18 Dec 2005 16:39:21 +0100 (CET) Received: from 203-173-11-165.dyn.iinet.net.au (HELO [10.1.1.3]) ([203.173.11.165]) by mail-ihug.icp-qv1-irony1.iinet.net.au with ESMTP; 18 Dec 2005 23:39:18 +0800 X-BrightmailFiltered: true X-Brightmail-Tracker: AAAAAA== Message-ID: <43A5829D.7070106@syskin.cjb.net> Date: Mon, 19 Dec 2005 02:09:09 +1030 From: Radek Czyz User-Agent: Thunderbird 1.5 (Windows/20051201) MIME-Version: 1.0 To: xvid-devel@xvid.org Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Subject: [XviD-devel] multithreaded pvop-ME X-BeenThere: xvid-devel@xvid.org X-Mailman-Version: 2.1.4 Precedence: list Reply-To: xvid-devel@xvid.org List-Id: xvid-devel.xvid.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: xvid-devel-bounces@xvid.org Errors-To: xvid-devel-bounces@xvid.org Hi everyone :) I just finished my multithreaded ME for p-frames project ^___^ How it works: each thread gets every second (third, fourth etc, depending on # of threads) row of macroblocks. It keeps record of how many MBs has already been processed in any line, and it also has access to the same record for the line above it. Every time a thread needs something to do, it reads the record of line above it and remembers how many macroblocks it can process now. Once it finishes, it reads the record again. Thing to notice: if the record is stale and too old, it just means smaller data chunk at a time. No need for mutexes. Patch at: http://syskin.is.dreaming.org/smp.patch Test results (defaults, no b-frames, vhq4): 1 thread : 1:48 2 threads : 1:05 3 threads : 1:06 4 threads : 1:08 16 threads: 1:11 That's on amd X2 4200+. Clip is 608x256 (-> 16 rows of macroblocks). Tomorrow, bframes. Yet another thread can be introduced to encode data to the bitstream. That needs some extra code changes, so I'll leave it for later. Radek _______________________________________________ XviD-devel mailing list XviD-devel@xvid.org http://list.xvid.org/mailman/listinfo/xvid-devel From xvid-devel-bounces@xvid.org Sun Dec 18 17:07:53 2005 Return-Path: X-Original-To: itdp@fh-biergarten.de Delivered-To: itdp@localhost Received: from edu.bnhof.de (edu.bnhof.de [213.167.167.52]) by mail.kliche.org (Postfix) with ESMTP id 812E7126D9B for ; Sun, 18 Dec 2005 17:07:53 +0100 (CET) Received: from edu.bnhof.de (localhost [127.0.0.1]) by edu.bnhof.de (Postfix) with ESMTP id 2AA251EF02; Sun, 18 Dec 2005 17:07:00 +0100 (CET) X-Original-To: xvid-devel@edu.bnhof.de Delivered-To: xvid-devel@edu.bnhof.de Received: from s6.lansco.de (s6.lansco.de [213.239.218.110]) by edu.bnhof.de (Postfix) with ESMTP id 756EB1EF02 for ; Sun, 18 Dec 2005 17:06:39 +0100 (CET) Received: from mail.ln9.net (rrcs-24-73-193-138.se.biz.rr.com [24.73.193.138]) by s6.lansco.de (Postfix) with ESMTP id 90ADB27B01 for ; Sun, 18 Dec 2005 17:06:37 +0100 (CET) Received: from [192.168.2.67] (62-117.35-65.tampabay.res.rr.com [65.35.117.62]) by mail.ln9.net (Postfix) with ESMTP id 1FE954B0626 for ; Sun, 18 Dec 2005 11:06:32 -0500 (EST) Message-ID: <43A58907.1060102@leoninedev.com> Date: Sun, 18 Dec 2005 11:06:31 -0500 From: Bryan Mayland User-Agent: Thunderbird 1.5 (Windows/20051025) MIME-Version: 1.0 To: xvid-devel@xvid.org Subject: Re: [XviD-devel] multithreaded pvop-ME References: <43A5829D.7070106@syskin.cjb.net> In-Reply-To: <43A5829D.7070106@syskin.cjb.net> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit X-BeenThere: xvid-devel@xvid.org X-Mailman-Version: 2.1.4 Precedence: list Reply-To: xvid-devel@xvid.org List-Id: xvid-devel.xvid.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: xvid-devel-bounces@xvid.org Errors-To: xvid-devel-bounces@xvid.org Radek Czyz wrote: > How it works: each thread gets every second (third, fourth etc, > depending on # of threads) row of macroblocks. It keeps record of how > many MBs has already been processed in any line, and it also has > access to the same record for the line above it. I wrote something similar, except I do it where any thread can do any block based on the requirement that the topright and left block are already done. Here's my numbers on an X2 4600+ (gotta love these processors). Clip is MPEG2 640x480, encoding to no b-frames, VHQ4 Original: 33:26 (50% utilization) BryyMT: 20:08 (83% utilization) So we both see the same 39% improvement it seems. Mine requires a mutex and a signal though. _______________________________________________ XviD-devel mailing list XviD-devel@xvid.org http://list.xvid.org/mailman/listinfo/xvid-devel From xvid-devel-bounces@xvid.org Mon Dec 19 10:40:10 2005 Return-Path: X-Original-To: itdp@fh-biergarten.de Delivered-To: itdp@localhost Received: from edu.bnhof.de (edu.bnhof.de [213.167.167.52]) by mail.kliche.org (Postfix) with ESMTP id AA09A126DA0 for ; Mon, 19 Dec 2005 10:40:10 +0100 (CET) Received: from edu.bnhof.de (localhost [127.0.0.1]) by edu.bnhof.de (Postfix) with ESMTP id 162B31EEC1; Mon, 19 Dec 2005 10:39:35 +0100 (CET) X-Original-To: xvid-devel@edu.bnhof.de Delivered-To: xvid-devel@edu.bnhof.de Received: from s6.lansco.de (s6.lansco.de [213.239.218.110]) by edu.bnhof.de (Postfix) with ESMTP id 318891EEBD for ; Mon, 19 Dec 2005 10:39:20 +0100 (CET) Received: from customer-domains.icp-qv1-irony7.iinet.net.au (customer-domains.icp-qv1-irony7.iinet.net.au [203.59.1.128]) by s6.lansco.de (Postfix) with ESMTP id 57FAD27B1A for ; Mon, 19 Dec 2005 10:39:17 +0100 (CET) Received: from 203-173-11-165.dyn.iinet.net.au (HELO [10.1.1.3]) ([203.173.11.165]) by customer-domains.icp-qv1-irony7.iinet.net.au with ESMTP; 19 Dec 2005 17:39:15 +0800 X-BrightmailFiltered: true X-Brightmail-Tracker: AAAAAA== Message-ID: <43A67FB8.6010101@syskin.cjb.net> Date: Mon, 19 Dec 2005 20:09:04 +1030 From: Radek Czyz User-Agent: Thunderbird 1.5 (Windows/20051201) MIME-Version: 1.0 To: xvid-devel@xvid.org Subject: Re: [XviD-devel] multithreaded pvop-ME References: <43A5829D.7070106@syskin.cjb.net> In-Reply-To: <43A5829D.7070106@syskin.cjb.net> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit X-BeenThere: xvid-devel@xvid.org X-Mailman-Version: 2.1.4 Precedence: list Reply-To: xvid-devel@xvid.org List-Id: xvid-devel.xvid.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: xvid-devel-bounces@xvid.org Errors-To: xvid-devel-bounces@xvid.org Heyhey, New patch. This time, b-frames multithreaded too - just simply divided into slices. Unlike pframes, this changes bistream a bit because at the slice boundary there's no motion prediction from top, but the loss is insignificant. full patch (pframes and bframes) : http://syskin.is.dreaming.org/smp2.patch Both dispatcher functions should probably go to encoder.c so that they can add another thread, actually doing the encoding. I'll find best place later, for now, this kludge is good enough. Question, what's pthread equivalent of win32's Sleep(0)? (and BeOS too). Regards, Radek _______________________________________________ XviD-devel mailing list XviD-devel@xvid.org http://list.xvid.org/mailman/listinfo/xvid-devel From xvid-devel-bounces@xvid.org Mon Dec 19 17:12:34 2005 Return-Path: X-Original-To: itdp@fh-biergarten.de Delivered-To: itdp@localhost Received: from edu.bnhof.de (edu.bnhof.de [213.167.167.52]) by mail.kliche.org (Postfix) with ESMTP id CBFC5126DA2 for ; Mon, 19 Dec 2005 17:12:34 +0100 (CET) Received: from edu.bnhof.de (localhost [127.0.0.1]) by edu.bnhof.de (Postfix) with ESMTP id B523F1EE95; Mon, 19 Dec 2005 17:11:54 +0100 (CET) X-Original-To: xvid-devel@edu.bnhof.de Delivered-To: xvid-devel@edu.bnhof.de Received: from s6.lansco.de (s6.lansco.de [213.239.218.110]) by edu.bnhof.de (Postfix) with ESMTP id 98E151994C for ; Mon, 19 Dec 2005 17:11:40 +0100 (CET) Received: from mail.ln9.net (rrcs-24-73-193-138.se.biz.rr.com [24.73.193.138]) by s6.lansco.de (Postfix) with ESMTP id D82E42870F for ; Mon, 19 Dec 2005 17:11:37 +0100 (CET) Received: from [192.168.100.164] (yarr.ad.ln9.net [192.168.100.164]) by mail.ln9.net (Postfix) with ESMTP id 829F64B019F for ; Mon, 19 Dec 2005 11:11:36 -0500 (EST) Message-ID: <43A6DBB8.7030604@leoninedev.com> Date: Mon, 19 Dec 2005 11:11:36 -0500 From: Bryan Mayland Organization: Leonine Development, Inc. User-Agent: Thunderbird 1.5 (Windows/20051025) MIME-Version: 1.0 To: xvid-devel@xvid.org Subject: Re: [XviD-devel] multithreaded pvop-ME References: <43A5829D.7070106@syskin.cjb.net> <43A67FB8.6010101@syskin.cjb.net> In-Reply-To: <43A67FB8.6010101@syskin.cjb.net> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit X-BeenThere: xvid-devel@xvid.org X-Mailman-Version: 2.1.4 Precedence: list Reply-To: xvid-devel@xvid.org List-Id: xvid-devel.xvid.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: xvid-devel-bounces@xvid.org Errors-To: xvid-devel-bounces@xvid.org Radek Czyz wrote: > Question, what's pthread equivalent of win32's Sleep(0)? (and BeOS too). sched_yield(), although I hardly think continuously looping on a sleep(0)/sched_yield() is a good wait scheme. _______________________________________________ XviD-devel mailing list XviD-devel@xvid.org http://list.xvid.org/mailman/listinfo/xvid-devel From xvid-devel-bounces@xvid.org Tue Dec 20 05:45:16 2005 Return-Path: X-Original-To: itdp@fh-biergarten.de Delivered-To: itdp@localhost Received: from edu.bnhof.de (edu.bnhof.de [213.167.167.52]) by mail.kliche.org (Postfix) with ESMTP id 964F4126D82 for ; Tue, 20 Dec 2005 05:45:16 +0100 (CET) Received: from edu.bnhof.de (localhost [127.0.0.1]) by edu.bnhof.de (Postfix) with ESMTP id 2B5A318B6C; Tue, 20 Dec 2005 05:45:01 +0100 (CET) X-Original-To: xvid-devel@edu.bnhof.de Delivered-To: xvid-devel@edu.bnhof.de Received: from s6.lansco.de (s6.lansco.de [213.239.218.110]) by edu.bnhof.de (Postfix) with ESMTP id 812D0158C9 for ; Tue, 20 Dec 2005 05:44:52 +0100 (CET) Received: from mail.gmx.net (mail.gmx.net [213.165.64.21]) by s6.lansco.de (Postfix) with SMTP id CFE761791F for ; Tue, 20 Dec 2005 05:44:23 +0100 (CET) Received: (qmail invoked by alias); 20 Dec 2005 04:44:23 -0000 Received: from chello080109116125.4.15.vie.surfer.at (EHLO localhost) [80.109.116.125] by mail.gmx.net (mp027) with SMTP; 20 Dec 2005 05:44:23 +0100 X-Authenticated: #3831892 Date: Tue, 20 Dec 2005 05:44:16 +0100 From: Michael Niedermayer To: xvid-devel@xvid.org Message-ID: <20051220044416.GJ6501@MichaelsNB> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.5.10i X-Y-GMX-Trusted: 0 Subject: [XviD-devel] possible qpel direct chroma MV rounding bug X-BeenThere: xvid-devel@xvid.org X-Mailman-Version: 2.1.4 Precedence: list Reply-To: xvid-devel@xvid.org List-Id: xvid-devel.xvid.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: xvid-devel-bounces@xvid.org Errors-To: xvid-devel-bounces@xvid.org Hi after some heated flam .. ehh i mean disscussion about libavcodec ability to decode xvid encoded files binary identically to xvid i found some differences, when using qpel + b frames, one of these is caused by the following code in xvid, which doesnt seem to match the mpeg4 spec but its 5oclock in the morning so please forgive me if iam wrong uv_dx = pMB->mvs[0].x + pMB->mvs[1].x + pMB->mvs[2].x + pMB->mvs[3].x; uv_dy = pMB->mvs[0].y + pMB->mvs[1].y + pMB->mvs[2].y + pMB->mvs[3].y; b_uv_dx = pMB->b_mvs[0].x + pMB->b_mvs[1].x + pMB->b_mvs[2].x + pMB->b_mvs[3].x; b_uv_dy = pMB->b_mvs[0].y + pMB->b_mvs[1].y + pMB->b_mvs[2].y + pMB->b_mvs[3].y; if (dec->quarterpel) { if (dec->bs_version <= BS_VERSION_BUGGY_CHROMA_ROUNDING) { uv_dx = (uv_dx>>1) | (uv_dx&1); uv_dy = (uv_dy>>1) | (uv_dy&1); b_uv_dx = (b_uv_dx>>1) | (b_uv_dx&1); b_uv_dy = (b_uv_dy>>1) | (b_uv_dy&1); } else { uv_dx /= 2; uv_dy /= 2; b_uv_dx /= 2; b_uv_dy /= 2; } } what i think is wrong on this is that the /2 should be done before the summation of the 4 vectors mpeg4 spec: For the motion compensation of both chrominance blocks, the forward motion vector (MVFx_chro, MVFy_chro) is calculated by the sum of K forward luminance motion vectors dividing by 2K and then rounding toward the nearest half sample position as defined in Table 7-6 to Table 7-9. In quarter sample mode the vectors are divided by 2 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ before summation. The backward motion vector (MVBx_chro, MVBy_chro) is derived in the same way. The rest ^^^^^^^^^^^^^^^^ process is the same as the chrominance motion compensation of the bi-directional mode described in subclause -- Michael _______________________________________________ XviD-devel mailing list XviD-devel@xvid.org http://list.xvid.org/mailman/listinfo/xvid-devel From xvid-devel-bounces@xvid.org Wed Dec 21 07:19:00 2005 Return-Path: X-Original-To: itdp@fh-biergarten.de Delivered-To: itdp@localhost Received: from edu.bnhof.de (edu.bnhof.de [213.167.167.52]) by mail.kliche.org (Postfix) with ESMTP id EEC5F126D8B for ; Wed, 21 Dec 2005 07:18:59 +0100 (CET) Received: from edu.bnhof.de (localhost [127.0.0.1]) by edu.bnhof.de (Postfix) with ESMTP id CBE301EEF6; Wed, 21 Dec 2005 07:18:45 +0100 (CET) X-Original-To: xvid-devel@edu.bnhof.de Delivered-To: xvid-devel@edu.bnhof.de Received: from s6.lansco.de (s6.lansco.de [213.239.218.110]) by edu.bnhof.de (Postfix) with ESMTP id E3BE119928 for ; Wed, 21 Dec 2005 07:18:41 +0100 (CET) Received: from localhost.localdomain (ASte-Genev-Bois-151-1-19-143.w83-114.abo.wanadoo.fr [83.114.17.143]) by s6.lansco.de (Postfix) with ESMTP id 92F1C27996 for ; Wed, 21 Dec 2005 07:18:39 +0100 (CET) Received: from localhost.localdomain (localhost.localdomain [127.0.0.1]) by localhost.localdomain (8.12.11/8.12.11) with ESMTP id jBL6FRFj003352 for ; Wed, 21 Dec 2005 07:16:12 +0100 Subject: Re: [XviD-devel] possible qpel direct chroma MV rounding bug From: Pascal Massimino To: xvid-devel@xvid.org In-Reply-To: <20051220044416.GJ6501@MichaelsNB> References: <20051220044416.GJ6501@MichaelsNB> Content-Type: text/plain Message-Id: <1135145706.3200.3.camel@localhost.localdomain> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.4.5 (1.4.5-14) Date: Wed, 21 Dec 2005 07:15:07 +0100 Content-Transfer-Encoding: 7bit X-BeenThere: xvid-devel@xvid.org X-Mailman-Version: 2.1.4 Precedence: list Reply-To: skal@planet-d.net, xvid-devel@xvid.org List-Id: xvid-devel.xvid.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: xvid-devel-bounces@xvid.org Errors-To: xvid-devel-bounces@xvid.org Hi Michael! On Tue, 2005-12-20 at 05:44, Michael Niedermayer wrote: > Hi > > after some heated flam .. ehh i mean disscussion about libavcodec ability > to decode xvid encoded files binary identically to xvid i found some > differences, when using qpel + b frames, one of these is caused by the > following code in xvid, which doesnt seem to match the mpeg4 spec but > its 5oclock in the morning so please forgive me if iam wrong Indeed, i think you're perfectly right (even at 5am;). The code doesn't follow the spec. Now, even if the fix is trivial, we'll have to think of something for the backward compatibility... later! Skal _______________________________________________ XviD-devel mailing list XviD-devel@xvid.org http://list.xvid.org/mailman/listinfo/xvid-devel From xvid-devel-bounces@xvid.org Wed Dec 21 07:34:52 2005 Return-Path: X-Original-To: itdp@fh-biergarten.de Delivered-To: itdp@localhost Received: from edu.bnhof.de (edu.bnhof.de [213.167.167.52]) by mail.kliche.org (Postfix) with ESMTP id 986C9126D8B for ; Wed, 21 Dec 2005 07:34:52 +0100 (CET) Received: from edu.bnhof.de (localhost [127.0.0.1]) by edu.bnhof.de (Postfix) with ESMTP id 6D5B31EEE7; Wed, 21 Dec 2005 07:34:37 +0100 (CET) X-Original-To: xvid-devel@edu.bnhof.de Delivered-To: xvid-devel@edu.bnhof.de Received: from s6.lansco.de (s6.lansco.de [213.239.218.110]) by edu.bnhof.de (Postfix) with ESMTP id 056BF1EEF9 for ; Wed, 21 Dec 2005 07:34:28 +0100 (CET) Received: from smtp10.wanadoo.fr (smtp10.wanadoo.fr [193.252.22.21]) by s6.lansco.de (Postfix) with ESMTP id 6BFC82796C for ; Wed, 21 Dec 2005 07:34:26 +0100 (CET) Received: from me-wanadoo.net (localhost [127.0.0.1]) by mwinf1004.wanadoo.fr (SMTP Server) with ESMTP id 1B1BE240016D for ; Wed, 21 Dec 2005 07:34:26 +0100 (CET) Received: from localhost.localdomain (ASte-Genev-Bois-151-1-19-143.w83-114.abo.wanadoo.fr [83.114.17.143]) by mwinf1004.wanadoo.fr (SMTP Server) with ESMTP id C5CFD2400167 for ; Wed, 21 Dec 2005 07:34:25 +0100 (CET) X-ME-UUID: 20051221063425810.C5CFD2400167@mwinf1004.wanadoo.fr Subject: Re: [XviD-devel] multithreaded pvop-ME From: Pascal Massimino To: xvid-devel@xvid.org In-Reply-To: <43A6DBB8.7030604@leoninedev.com> References: <43A5829D.7070106@syskin.cjb.net> <43A67FB8.6010101@syskin.cjb.net> <43A6DBB8.7030604@leoninedev.com> Content-Type: text/plain Message-Id: <1135146718.3200.18.camel@localhost.localdomain> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.4.5 (1.4.5-14) Date: Wed, 21 Dec 2005 07:31:58 +0100 Content-Transfer-Encoding: 7bit X-BeenThere: xvid-devel@xvid.org X-Mailman-Version: 2.1.4 Precedence: list Reply-To: skal@planet-d.net, xvid-devel@xvid.org List-Id: xvid-devel.xvid.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: xvid-devel-bounces@xvid.org Errors-To: xvid-devel-bounces@xvid.org Hi Bryan and Radek! On Mon, 2005-12-19 at 17:11, Bryan Mayland wrote: > Radek Czyz wrote: > > Question, what's pthread equivalent of win32's Sleep(0)? (and BeOS too). > sched_yield(), although I hardly think continuously looping on a > sleep(0)/sched_yield() is a good wait scheme. Depends, but i'd tend to agree in our situation. Actually, for our problem of racing MBs, i think i would be better to hard-code lightweight spin locks instead of going back through the kernel for suspending the contending thread. That is, we should use: while( mb_pos==top_mb_pos ) { /* do nothing and wait for the other thread to advance 'top_mb_pos' */ } But only when you're sure you're running on a real MT system! With two real threads running physically separated. Otherwise it would be an overkill, sure... The idea is that contention time is very low for our MB race (it's not like we were dealing with listeners or something), so it's better to simply stand still burning CPU cycles for a short while, instead of re-entering the kernel. Now, a middle solution could be: int count; for(count=200; count>0 && !condition; --count) { } if (count==0 && !condition) { sched_yield(); etc... the full-monty } Of course, '200' needs to be adjusted... In any case, that's great you could add MT !!! later! Skal _______________________________________________ XviD-devel mailing list XviD-devel@xvid.org http://list.xvid.org/mailman/listinfo/xvid-devel From xvid-devel-bounces@xvid.org Wed Dec 21 07:52:58 2005 Return-Path: X-Original-To: itdp@fh-biergarten.de Delivered-To: itdp@localhost Received: from edu.bnhof.de (edu.bnhof.de [213.167.167.52]) by mail.kliche.org (Postfix) with ESMTP id 9BCE3126D8B for ; Wed, 21 Dec 2005 07:52:58 +0100 (CET) Received: from edu.bnhof.de (localhost [127.0.0.1]) by edu.bnhof.de (Postfix) with ESMTP id 6CE4B1EEFC; Wed, 21 Dec 2005 07:52:43 +0100 (CET) X-Original-To: xvid-devel@edu.bnhof.de Delivered-To: xvid-devel@edu.bnhof.de Received: from s6.lansco.de (s6.lansco.de [213.239.218.110]) by edu.bnhof.de (Postfix) with ESMTP id C6B651EEF5 for ; Wed, 21 Dec 2005 07:52:39 +0100 (CET) Received: from mail-ihug.icp-qv1-irony2.iinet.net.au (ihug-mail.icp-qv1-irony2.iinet.net.au [203.59.1.196]) by s6.lansco.de (Postfix) with ESMTP id 191C42879E for ; Wed, 21 Dec 2005 07:52:36 +0100 (CET) Received: from 203-173-11-165.dyn.iinet.net.au (HELO [10.1.1.3]) ([203.173.11.165]) by mail-ihug.icp-qv1-irony2.iinet.net.au with ESMTP; 21 Dec 2005 14:52:33 +0800 X-BrightmailFiltered: true X-Brightmail-Tracker: AAAAAA== Message-ID: <43A8FBA2.8040508@syskin.cjb.net> Date: Wed, 21 Dec 2005 17:22:18 +1030 From: Radek Czyz User-Agent: Thunderbird 1.5 (Windows/20051201) MIME-Version: 1.0 To: xvid-devel@xvid.org Subject: Re: [XviD-devel] multithreaded pvop-ME References: <43A5829D.7070106@syskin.cjb.net> <43A67FB8.6010101@syskin.cjb.net> <43A6DBB8.7030604@leoninedev.com> <1135146718.3200.18.camel@localhost.localdomain> In-Reply-To: <1135146718.3200.18.camel@localhost.localdomain> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit X-BeenThere: xvid-devel@xvid.org X-Mailman-Version: 2.1.4 Precedence: list Reply-To: xvid-devel@xvid.org List-Id: xvid-devel.xvid.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: xvid-devel-bounces@xvid.org Errors-To: xvid-devel-bounces@xvid.org Pascal Massimino wrote: > That is, we should use: > > while( mb_pos==top_mb_pos ) { /* do nothing and wait for the other thread to advance 'top_mb_pos' */ } I tried that and it was very slow. My reasoning is that if current thread has nothing to do, then surely there exists another thread that does. It didn't exactly work for me yet (3 threads were slower than 2) but I plan to add third thread, which is doing the MC and bitstream encoding, and which can always jump in and fill any gaps on any ME threads. Also note that your solution encourages processing only one MB at a time and then resynchronize again. If one thread is so much ahead of the other, it's probably best for this thread to purposefully delay itself and grab larger MB chunk next time. I plan to experiment with Sleep() more and try this: if a thread grabbed very few macroblocks at one time (say below 3), process the MBs but then Sleep() before you even try to resynchronize again. More complex solution involves using one thread for more than one line of macroblocks at a time, to reduce number of stalls (same logic as using more threads than CPUs, but without involving real OS threads). Thanks for your comments everyone, my experience with multithreading is basically nil so I'm grateful :) Oh, guess what: there's a demand for multithreaded b-frame ME that produces completely identical bitstream. I would have to use the same logic as pvop-ME. I'll definitely try this because this method has its interesting advantages. I'll see if it's faster or not (guess what, it might be). Radek _______________________________________________ XviD-devel mailing list XviD-devel@xvid.org http://list.xvid.org/mailman/listinfo/xvid-devel From xvid-devel-bounces@xvid.org Wed Dec 21 16:54:22 2005 Return-Path: X-Original-To: itdp@fh-biergarten.de Delivered-To: itdp@localhost Received: from edu.bnhof.de (edu.bnhof.de [213.167.167.52]) by mail.kliche.org (Postfix) with ESMTP id 6A1B9126D83 for ; Wed, 21 Dec 2005 16:54:22 +0100 (CET) Received: from edu.bnhof.de (localhost [127.0.0.1]) by edu.bnhof.de (Postfix) with ESMTP id 8201118B73; Wed, 21 Dec 2005 16:54:09 +0100 (CET) X-Original-To: xvid-devel@edu.bnhof.de Delivered-To: xvid-devel@edu.bnhof.de Received: from s6.lansco.de (s6.lansco.de [213.239.218.110]) by edu.bnhof.de (Postfix) with ESMTP id 264051564E for ; Wed, 21 Dec 2005 16:54:01 +0100 (CET) Received: from mail.gmx.net (mail.gmx.de [213.165.64.21]) by s6.lansco.de (Postfix) with SMTP id 2FCAA28DDC for ; Wed, 21 Dec 2005 16:53:59 +0100 (CET) Received: (qmail invoked by alias); 21 Dec 2005 15:53:54 -0000 Received: from chello080109116125.4.15.vie.surfer.at (EHLO localhost) [80.109.116.125] by mail.gmx.net (mp029) with SMTP; 21 Dec 2005 16:53:54 +0100 X-Authenticated: #3831892 Date: Wed, 21 Dec 2005 16:53:49 +0100 From: Michael Niedermayer To: mencoder-users@mplayerhq.hu Message-ID: <20051221155349.GD5154@MichaelsNB> References: <20051220212142.GF23367@MichaelsNB> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20051220212142.GF23367@MichaelsNB> User-Agent: Mutt/1.5.10i X-Y-GMX-Trusted: 0 Cc: xvid-devel@xvid.org Subject: [XviD-devel] Re: [MEncoder-users] libavcodec vs. xvid for decoding xvid X-BeenThere: xvid-devel@xvid.org X-Mailman-Version: 2.1.4 Precedence: list Reply-To: xvid-devel@xvid.org List-Id: xvid-devel.xvid.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: xvid-devel-bounces@xvid.org Errors-To: xvid-devel-bounces@xvid.org Hi On Tue, Dec 20, 2005 at 10:21:42PM +0100, Michael Niedermayer wrote: [...] > the difference is caused by several bugs, the only one i found sofar, but > there must be more, is the chroma MV rounding for direct MBs in xvid, ive > reported that one to xvid-dev ... the remaining bug is also in xvid decoder_mbinter() is used to handle forward and backward MBs in b frames, it is also used to handle 1MV and 4MV for P frames, sadly it uses pMB->mode != MODE_INTER4V to test for 4MV, the problem is MODE_INTER4V == MODE_BACKWARD so backward vectors will use 8x8 qpel which doesnt match 16x16 qpel due to the block boundary mirroring in mpeg4 when these are fixed, xvid-clip.avi is decoded binary identical with xvid and lavc as long as -lavdopts bitexact -noslices is used, the reason why -noslices is needed is unknown, maybe a bug in mplayers md5sum vo maybe in lavc -- Michael _______________________________________________ XviD-devel mailing list XviD-devel@xvid.org http://list.xvid.org/mailman/listinfo/xvid-devel From xvid-devel-bounces@xvid.org Sat Dec 24 07:36:28 2005 Return-Path: X-Original-To: itdp@fh-biergarten.de Delivered-To: itdp@localhost Received: from edu.bnhof.de (edu.bnhof.de [213.167.167.52]) by mail.kliche.org (Postfix) with ESMTP id E71CB126D9A for ; Sat, 24 Dec 2005 07:36:27 +0100 (CET) Received: from edu.bnhof.de (localhost [127.0.0.1]) by edu.bnhof.de (Postfix) with ESMTP id 885531EEF8; Sat, 24 Dec 2005 07:36:16 +0100 (CET) X-Original-To: xvid-devel@edu.bnhof.de Delivered-To: xvid-devel@edu.bnhof.de Received: from s6.lansco.de (s6.lansco.de [213.239.218.110]) by edu.bnhof.de (Postfix) with ESMTP id 769291EEF2 for ; Sat, 24 Dec 2005 07:36:08 +0100 (CET) Received: from localhost (ASte-Genev-Bois-151-1-68-38.w86-203.abo.wanadoo.fr [86.203.173.38]) by s6.lansco.de (Postfix) with ESMTP id 3F3C7286BA for ; Sat, 24 Dec 2005 07:36:06 +0100 (CET) Received: from massimin by localhost with local (Exim 4.60) (envelope-from ) id 1EqPTn-0001ai-9b for xvid-devel@xvid.org; Sun, 25 Dec 2005 07:35:39 +0100 Subject: Re: [XviD-devel] Re: [MEncoder-users] libavcodec vs. xvid for decoding xvid From: skal To: xvid-devel@xvid.org In-Reply-To: <20051221155349.GD5154@MichaelsNB> References: <20051220212142.GF23367@MichaelsNB> <20051221155349.GD5154@MichaelsNB> Content-Type: text/plain Content-Transfer-Encoding: 7bit Date: Sun, 25 Dec 2005 07:32:07 +0100 Message-Id: <1135492327.6070.3.camel@localhost> Mime-Version: 1.0 X-Mailer: Evolution 2.2.3 X-BeenThere: xvid-devel@xvid.org X-Mailman-Version: 2.1.4 Precedence: list Reply-To: xvid-devel@xvid.org List-Id: xvid-devel.xvid.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: xvid-devel-bounces@xvid.org Errors-To: xvid-devel-bounces@xvid.org Hi Michael! On Wed, 2005-12-21 at 16:53 +0100, Michael Niedermayer wrote: > Hi > > On Tue, Dec 20, 2005 at 10:21:42PM +0100, Michael Niedermayer wrote: > [...] > > the difference is caused by several bugs, the only one i found sofar, but > > there must be more, is the chroma MV rounding for direct MBs in xvid, ive > > reported that one to xvid-dev ... > > the remaining bug is also in xvid > > decoder_mbinter() is used to handle forward and backward MBs in b frames, it is > also used to handle 1MV and 4MV for P frames, sadly it uses > pMB->mode != MODE_INTER4V to test for 4MV, the problem is > MODE_INTER4V == MODE_BACKWARD so backward vectors will use 8x8 qpel which > doesnt match 16x16 qpel due to the block boundary mirroring in mpeg4 Damned, this is quite a typo! Ok, fortunately, it seems to only affect the decoder side, and only for b-frames (=> hence the mismatch doesn't propagate and it would explain why nobody but you spotted it before ;) Isibaar has committed some fix for your reports in current cvs. Everything should be hopefully ok, now. > > when these are fixed, xvid-clip.avi is decoded binary identical with xvid > and lavc as long as -lavdopts bitexact -noslices is used Thanks again for the reports! bye -Skal _______________________________________________ XviD-devel mailing list XviD-devel@xvid.org http://list.xvid.org/mailman/listinfo/xvid-devel From xvid-devel-bounces@xvid.org Sat Dec 24 16:20:41 2005 Return-Path: X-Original-To: itdp@fh-biergarten.de Delivered-To: itdp@localhost Received: from edu.bnhof.de (edu.bnhof.de [213.167.167.52]) by mail.kliche.org (Postfix) with ESMTP id 63A20126D9A for ; Sat, 24 Dec 2005 16:20:41 +0100 (CET) Received: from edu.bnhof.de (localhost [127.0.0.1]) by edu.bnhof.de (Postfix) with ESMTP id CBF6F1994E; Sat, 24 Dec 2005 16:20:20 +0100 (CET) X-Original-To: xvid-devel@edu.bnhof.de Delivered-To: xvid-devel@edu.bnhof.de Received: from s6.lansco.de (s6.lansco.de [213.239.218.110]) by edu.bnhof.de (Postfix) with ESMTP id F13D11993D for ; Sat, 24 Dec 2005 16:20:17 +0100 (CET) Received: from smtp6.libero.it (smtp6.libero.it [193.70.192.59]) by s6.lansco.de (Postfix) with ESMTP id 1E21F2903B for ; Sat, 24 Dec 2005 16:20:15 +0100 (CET) Received: from localhost (172.16.1.15) by smtp6.libero.it (7.0.027-DD01) id 439D9193011B9F37 for xvid-devel@xvid.org; Sat, 24 Dec 2005 16:20:15 +0100 Received: from smtp1.libero.it ([172.16.1.96]) by localhost (asav11.libero.it [193.70.192.93]) (amavisd-new, port 10024) with ESMTP id 10786-03-3 for ; Sat, 24 Dec 2005 16:20:14 +0100 (CET) Received: from libero.it (172.16.1.79) by smtp1.libero.it (7.0.027-DD01) id 431C3A42012A3A3F for xvid-devel@xvid.org; Sat, 24 Dec 2005 16:20:14 +0100 Date: Sat, 24 Dec 2005 16:20:14 +0100 Message-Id: Subject: Re: [XviD-devel] Problem with BS_VERSION_BUGGY_CHROMA_ROUNDING? MIME-Version: 1.0 X-Sensitivity: 3 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable From: "carlo.bramix" To: "xvid-devel" X-XaM3-API-Version: 4.3 (R1) (B3pl11) X-SenderIP: 151.25.158.116 X-Scanned: with antispam and antivirus automated system at libero.it X-BeenThere: xvid-devel@xvid.org X-Mailman-Version: 2.1.4 Precedence: list Reply-To: xvid-devel@xvid.org List-Id: xvid-devel.xvid.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: xvid-devel-bounces@xvid.org Errors-To: xvid-devel-bounces@xvid.org I usually use "Tadzio's File compare". It has also an option for compressing white spaces, so it allows to easil= y compare files like these ones. However, I manually converted the file in the previous format with tab ch= aracters and I left some lines with the spaces if they are really differe= nt. How about the true problem that forced me to start this message? Sincerely, Carlo Bramini ---------- Initial Header ----------- >From : xvid-devel-bounces@xvid.org To : xvid-devel@xvid.org Cc : Date : Sun, 18 Dec 2005 16:20:02 +0100 Subject : Re: [XviD-devel] Problem with BS_VERSION_BUGGY_CHROMA_ROUNDING?= > "carlo.bramix" wrote: > > By the way, there is a thing that made the debugging much more > > difficult. > > Into the newest DECODER.C all tab chars have been changed with two > > spaces and this makes impossible a decent file comparison! > > Well, I don't know how decent diff is for you, but those options of it > help a lot: > > -E --ignore-tab-expansion Ignore changes due to tab expansion. > -b --ignore-space-change Ignore changes in the amount of white space= . > -w --ignore-all-space Ignore all white space. > -B --ignore-blank-lines Ignore changes whose lines are all blank. > > -- > Kurt Bernhard Pruenner --- Haendelstrasse 17 --- 4020 Linz --- Austria > .......It might be written "Mindfuck", but it's spelt "L-A-I-N"....... > np: David Holmes - Let's Get Killed (Let's Get Killed) > _______________________________________________ > XviD-devel mailing list > XviD-devel@xvid.org > http://list.xvid.org/mailman/listinfo/xvid-devel > _______________________________________________ XviD-devel mailing list XviD-devel@xvid.org http://list.xvid.org/mailman/listinfo/xvid-devel From xvid-users-bounces@xvid.org Fri Dec 30 18:16:45 2005 Return-Path: X-Original-To: itdp@fh-biergarten.de Delivered-To: itdp@localhost Received: from edu.bnhof.de (edu.bnhof.de [213.167.167.52]) by mail.kliche.org (Postfix) with ESMTP id CAFE2126D90 for ; Fri, 30 Dec 2005 18:16:45 +0100 (CET) Received: from edu.bnhof.de (localhost [127.0.0.1]) by edu.bnhof.de (Postfix) with ESMTP id DFB2D1EEBE; Fri, 30 Dec 2005 18:16:35 +0100 (CET) X-Original-To: xvid-users@edu.bnhof.de Delivered-To: xvid-users@edu.bnhof.de Received: from s6.lansco.de (s6.lansco.de [213.239.218.110]) by edu.bnhof.de (Postfix) with ESMTP id BD5D11EEB9; Fri, 30 Dec 2005 18:16:32 +0100 (CET) Received: from s5.lansco.de (s5.lansco.de [212.63.130.170]) by s6.lansco.de (Postfix) with ESMTP id 62E3528DE4; Fri, 30 Dec 2005 18:16:26 +0100 (CET) Received: from localhost (s5.lansco.de [212.63.130.170]) by s5.lansco.de (Postfix) with ESMTP id 7FD0211716; Fri, 30 Dec 2005 18:16:26 +0100 (CET) Received: from p54B5F64B.dip.t-dialin.net (p54B5F64B.dip.t-dialin.net [84.181.246.75]) by www.lansco.de (IMP) with HTTP for ; Fri, 30 Dec 2005 18:16:26 +0100 Message-ID: <1135962986.43b56b6a181af@www.lansco.de> Date: Fri, 30 Dec 2005 18:16:26 +0100 From: Michael Militzer To: xvid-devel@xvid.org MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit User-Agent: Internet Messaging Program (IMP) 3.2.3 X-Originating-IP: 84.181.246.75 Cc: xvid-users@xvid.org Subject: [XviD-users] XviD 1.1.0 is available! X-BeenThere: xvid-users@xvid.org X-Mailman-Version: 2.1.4 Precedence: list Reply-To: xvid-users@xvid.org List-Id: xvid-users.xvid.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: xvid-users-bounces@xvid.org Errors-To: xvid-users-bounces@xvid.org Hello! This is XviD 1.1.0 release. This release is the long awaited 1.1.0. It is mostly API compatible with the previous stable release as we dropped support for reduced resolution coding. If your application didn't use that feature then the upgrade is totally compatible. Changes since 1.0.3: * xvidcore: - Improved Low bitrate quality. - Improved VBV support - Rate-Distortion mode decision for bvops - New postprocessing functions, brightness and deringing - New PowerPC port by Christoph Naegeli - Brand new amd64 Linux 64bit port by Andre Werthmann - Various decoder and encoder speedups - A few bugs squashed * VFW frontend - Mingw/CygWin support - Various small improvements - A few bugs squashed * DShow frontend - Mingw/CygWin support - Support for brightness control - Various small improvements - A few bugs squashed Changes since 1.1.0-beta2: * xvidcore - Field interlaced decoding - IEEE-1180 compliant SSE2 iDCT (disabled for safety) - Fixed misaligned reads on RISC platforms such as ARM - Fixed some minor decoder bugs - Completed GCC 4.0 support - Export only public API on GNU/Linux and Solaris - Work on the example apps. Support for AVS input in xvid_encraw * VFW frontend - Small updates * DShow frontend - Additional fourcc support The files are available in the download section of XviD.org: http://www.xvid.org/downloads.html -- The "XviD Team" _______________________________________________ XviD-users mailing list XviD-users@xvid.org http://list.xvid.org/mailman/listinfo/xvid-users From xvid-devel-bounces@xvid.org Sat Dec 31 17:57:23 2005 Return-Path: X-Original-To: itdp@fh-biergarten.de Delivered-To: itdp@localhost Received: from edu.bnhof.de (edu.bnhof.de [213.167.167.52]) by mail.kliche.org (Postfix) with ESMTP id 305B7126D90 for ; Sat, 31 Dec 2005 17:57:23 +0100 (CET) Received: from edu.bnhof.de (localhost [127.0.0.1]) by edu.bnhof.de (Postfix) with ESMTP id 708B619170; Sat, 31 Dec 2005 17:57:13 +0100 (CET) X-Original-To: xvid-devel@edu.bnhof.de Delivered-To: xvid-devel@edu.bnhof.de Received: from s6.lansco.de (s6.lansco.de [213.239.218.110]) by edu.bnhof.de (Postfix) with ESMTP id 8E7FC18BFD for ; Sat, 31 Dec 2005 15:28:32 +0100 (CET) Received: from smtp20.libero.it (smtp20.libero.it [193.70.192.147]) by s6.lansco.de (Postfix) with ESMTP id 45221277A7 for ; Sat, 31 Dec 2005 15:28:25 +0100 (CET) Received: from localhost (172.16.1.118) by smtp20.libero.it (7.0.027-DD01) id 439D916F0195E422 for xvid-devel@xvid.org; Sat, 31 Dec 2005 15:28:25 +0100 Received: from smtp20.libero.it ([172.16.1.77]) by localhost (asav14.libero.it [193.70.193.42]) (amavisd-new, port 10024) with ESMTP id 05064-06-2 for ; Sat, 31 Dec 2005 15:28:25 +0100 (CET) Received: from libero.it (172.16.1.13) by smtp20.libero.it (7.0.027-DD01) id 431C3BFF0130F6DC for xvid-devel@xvid.org; Sat, 31 Dec 2005 15:28:25 +0100 Date: Sat, 31 Dec 2005 15:28:25 +0100 Message-Id: Subject: Re: [XviD-devel] Problem with BS_VERSION_BUGGY_CHROMA_ROUNDING? MIME-Version: 1.0 X-Sensitivity: 3 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable From: "carlo.bramix" To: "xvid-devel" X-XaM3-API-Version: 4.3 (R1) (B3pl11) X-SenderIP: 151.25.221.185 X-Scanned: with antispam and antivirus automated system at libero.it X-BeenThere: xvid-devel@xvid.org X-Mailman-Version: 2.1.4 Precedence: list Reply-To: xvid-devel@xvid.org List-Id: xvid-devel.xvid.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: xvid-devel-bounces@xvid.org Errors-To: xvid-devel-bounces@xvid.org I just wanted to inform that this problem has been solved. I recompiled the new sources from "1.1.0 Final" and everything worked fin= e. Sincerely, Carlo Bramini _______________________________________________ XviD-devel mailing list XviD-devel@xvid.org http://list.xvid.org/mailman/listinfo/xvid-devel From xvid-devel-bounces@xvid.org Sat Dec 31 20:13:50 2005 Return-Path: X-Original-To: itdp@fh-biergarten.de Delivered-To: itdp@localhost Received: from edu.bnhof.de (edu.bnhof.de [213.167.167.52]) by mail.kliche.org (Postfix) with ESMTP id E23EF126D82 for ; Sat, 31 Dec 2005 20:13:49 +0100 (CET) Received: from edu.bnhof.de (localhost [127.0.0.1]) by edu.bnhof.de (Postfix) with ESMTP id 07C831992B; Sat, 31 Dec 2005 20:13:36 +0100 (CET) X-Original-To: xvid-devel@edu.bnhof.de Delivered-To: xvid-devel@edu.bnhof.de Received: from s6.lansco.de (s6.lansco.de [213.239.218.110]) by edu.bnhof.de (Postfix) with ESMTP id 21DDF1991D for ; Sat, 31 Dec 2005 19:09:13 +0100 (CET) Received: from s5.lansco.de (s5.lansco.de [212.63.130.170]) by s6.lansco.de (Postfix) with ESMTP id 98C4921F63 for ; Sat, 31 Dec 2005 19:09:07 +0100 (CET) Received: from localhost (s5.lansco.de [212.63.130.170]) by s5.lansco.de (Postfix) with ESMTP id 12BB225F3B for ; Sat, 31 Dec 2005 19:09:07 +0100 (CET) Received: from p54B5F3DD.dip.t-dialin.net (p54B5F3DD.dip.t-dialin.net [84.181.243.221]) by www.lansco.de (IMP) with HTTP for ; Sat, 31 Dec 2005 19:09:06 +0100 Message-ID: <1136052546.43b6c942dc86c@www.lansco.de> Date: Sat, 31 Dec 2005 19:09:06 +0100 From: Michael Militzer To: xvid-devel@xvid.org Subject: Re: [XviD-devel] Problem with BS_VERSION_BUGGY_CHROMA_ROUNDING? References: In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit User-Agent: Internet Messaging Program (IMP) 3.2.3 X-Originating-IP: 84.181.243.221 X-BeenThere: xvid-devel@xvid.org X-Mailman-Version: 2.1.4 Precedence: list Reply-To: xvid-devel@xvid.org List-Id: xvid-devel.xvid.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: xvid-devel-bounces@xvid.org Errors-To: xvid-devel-bounces@xvid.org Hi, yep, I've committed a fix for this recently. Good to hear it works... Regards, Michael Quoting "carlo.bramix" : > I just wanted to inform that this problem has been solved. > I recompiled the new sources from "1.1.0 Final" and everything worked fine. > > Sincerely, > > Carlo Bramini > > _______________________________________________ > XviD-devel mailing list > XviD-devel@xvid.org > http://list.xvid.org/mailman/listinfo/xvid-devel > _______________________________________________ XviD-devel mailing list XviD-devel@xvid.org http://list.xvid.org/mailman/listinfo/xvid-devel