Sorry, the compiler hangs on the following file:
newmdct.c
----- Original Message -----
From: "John Zitterkopf" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Thursday, April 03, 2003 12:25 AM
Subject: Write VBR header bug w/ 3.93.1 DLL code - Fix here
> I've been noticing that when using precompiled binaries for lame_enc.dll;
> when I set
> format.LHV1.bWriteVBRHeader
> to true under Delphi 4.0; the VBRHeader is not writen as shown by the
debug
> log:
>
> Lame_enc configuration options:
> ==========================================================
> version =1
> Layer =3
> mode =Stereo
> Input sample rate =44.1 kHz
> Output sample rate =44.1 kHz
> bitrate =160 kbps
> Quality Setting =5
> Low pass frequency =21421
> Low pass width =-1
> High pass frequency =0
> High pass width =-1
> No short blocks =0
> Force short blocks =0
> de-emphasis =0
> private flag =0
> copyright flag =1
> original flag =0
> CRC =off
> Fast mode =enabled
> Force mid/side stereo =disabled
> Padding Type =0
> Disable Resorvoir =0
> Allow diff-short =1
> Temporal masking effect=0
> Strict ISO Encoding =No
> Scale = 0.00
> VBR =enabled, VBR_q =4, VBR method =vbr_abr (average
> bitrate 192 kbps)
> Vbr Min bitrate =160 kbps
> Vbr Max bitrate =192 kbps
> Write VBR Header =No
> VBR Hard min =0
> ATH Only =0
> ATH short =0
> ATH short =0
> ATH no =0
> ATH type =2
> ATH lower =0
> ATH aa =-1
> ATH aa loudapprox =2
> ATH aa sensitivity =0
> ATHtype =2
> Experimental nspsytune =0
> Experimental X =0
> Experimental Y =0
> Experimental Z =0
>
> I noticed that the following line in the
> beInitStream()
> function. The line which reads:
>
> // check if the VBR tag is required
> if ( TRUE == lameConfig.format.LHV1.bWriteVBRHeader)
> {
> lame_set_bWriteVbrTag( gfp, 1 );
> }
>
> is different than other boolean checks like lameConfig.format.LHV1.bNoRes
> // check if the VBR tag is required
> if ( /*TRUE ==*/ lameConfig.format.LHV1.bWriteVBRHeader)
> {
> lame_set_bWriteVbrTag( gfp, 1 );
> }
>
> If I comment out the TRUE == line and recompile; the VBR header is written
> as I'd expect:
>
>
> Lame_enc configuration options:
> ==========================================================
> version =1
> Layer =3
> mode =Stereo
> Input sample rate =44.1 kHz
> Output sample rate =44.1 kHz
> bitrate =96 kbps
> Quality Setting =2
> Low pass frequency =14315
> Low pass width =-1
> High pass frequency =0
> High pass width =-1
> No short blocks =0
> Force short blocks =0
> de-emphasis =0
> private flag =0
> copyright flag =1
> original flag =0
> CRC =off
> Fast mode =enabled
> Force mid/side stereo =disabled
> Padding Type =0
> Disable Resorvoir =1
> Allow diff-short =1
> Temporal masking effect=0
> Strict ISO Encoding =No
> Scale = 0.00
> VBR =enabled, VBR_q =4, VBR method =vbr_abr (average
> bitrate 133 kbps)
> Vbr Min bitrate =96 kbps
> Vbr Max bitrate =192 kbps
> Write VBR Header =Yes
> VBR Hard min =0
> ATH Only =0
> ATH short =0
> ATH short =0
> ATH no =0
> ATH type =2
> ATH lower =0
> ATH aa =-1
> ATH aa loudapprox =2
> ATH aa sensitivity =0
> ATHtype =2
> Experimental nspsytune =0
> Experimental X =0
> Experimental Y =0
> Experimental Z =0
>
> The other thing I've noticed with 3.93.1 is that when compiling with MSVC
> for the Pentium using MSVC++5.0 the compiler hangs:
> nmake -f Makefile.MSVC COMP=MS GTK=NO DLL CPU=P1
> but if I compile for P2 or P3; it compiles fine. This happens with both
the
> original source and the modified sourcecode.
>
> I don't know the cause for this. Anyone?
>
> John
>
_______________________________________________
mp3encoder mailing list
[EMAIL PROTECTED]
http://minnie.tuhs.org/mailman/listinfo/mp3encoder
