How Using "Beesweet1.3b8.dll" with Delphi Prog ?

[DSPguru]

Originally posted by Nic
@DSPGuru:

Hiya :)

Forgot to mention, exception faults in the BeSweet.DLL can occur if very small buffer sizes (<1000) are sent to AC3Burst...This isn't a problem because ive made it so DVD2AVI will always buffer & send 8000 bytes. But I thought id mention it just in case you hear similar reports :)i'll check it out in the future. very busy lately..

Thanks again for all your help :) (I see movmasty guessed your avatar :) (btw i guessed wrong myself :( :) )

Cheers,
-Nic time to pick an avatar for yourself.. ;)

Dg.

[Dark-Cracker]

All the problemes are they solved ? thank u all for your help :-)
does it exist a final relase of besweet.dll , and the delphi exemple :-) ?

[CyberDemonII]

I have a problem using besweet.dll while creating an ogg file.

When i use the example2.c as example it is possible to create 2 ogg files in the same function.

But when i use the code in a mfc application the 2nd ogg file is invalid!

I use the following code :

hDLL=LoadLibrary("BeSweet.dll");
BeSweet = (BESWEET) GetProcAddress(hDLL, "BeSweet");
AC3_Bursts = (BURSTS) GetProcAddress(hDLL, "AC3_Bursts");

char buffer[AC3_BUFFER] = {0}; // AC3 bursts
int count = 1;
BS_Record info; // BeSweet info


BeSweet(ARGC,ARGV);

// Start Transcoding
while(count)
{
count = fread(buffer,1,AC3_BUFFER,fp);
info = AC3_Bursts(count,buffer);
if (!info->err)
{
}
}

strcpy(ARGV[0],"deinit");
BeSweet(ARGC,ARGV);
FreeLibrary(hDLL);

And the besweet commandline is like :
-core( -input ac3input -output file.ogg ) -boost( /b=2 /l=0.98 ) -ota( -G max ) -azid( -s surround2 ) -ogg( -b 320 )

Can somebody please help me ?

[DSPguru]

can you give more details about how you use BeSweet.dll to create 2 ogg streams ?

[CyberDemonII]

Hi DSPguru,

I am implemteting a sort of batch mode in my application. So its possible to do more conversions in a batch.
So for example when i have 2 video's , the tool must use besweet to convert twice an ac3 to ogg file.

I use the code as you saw in the previous message.

But ....

I did some testing and it seems that i have located the problem :

When i use -ota( -G max) , it crashes (Or the muxing from avi+ogg goes wrong because the ogg file in invalid)

But when i use -ota( -g max), it works perfect !

Could you please tell and investige how this can be ?
( I have no problems with mp3 , only ogg is wrong)

And as you saw in the previous message, i am using "example2.c" as base. But DO i have to do this, or can I just do it the simple way ?)

thanx for helping !!!

grtz,
CyberDemonII

[DSPguru]

i'm sorry. i don't have the time to help you today.
i'll reply in a couple of days.

stay tuned.

[mmgrover]

Using the 1.41 dll with the following:
char* ARGV[23]={"deinit","-core(","-input","ac3input","-output","test.mp2","-logfile ","BeSweet.txt",")","-ota(","-d","0","-G","max",")","-mp2enc(", "-b", "192", ")","-ssrc(","--rate","44100",")"};

trying to encode ac3 to mp2 I get a error "Invalid Floating Point operation"

Any Ideas?
Are my Arguments right?


mike

[DSPguru]

imho. "deinit" doesn't belong in here.

[DSPguru]

@CyberDemonII
as far as i remember, -g max has no effect on BeSweet.dll. you need to run BeSweet.dll twice by yourself.
first pass is used to find the maximal gain using the returned struct from BeSweet and the second pass is used for gain assertion. (since gain value is known in this point).
anyway, PostGain is much more advised. same goes for Azid's boosting comparing to -boost().
i suggest that you take a look at my advised commandline that can be found in BeSweetGUI's profiles.

[mmgrover]

@DspGuru

I had changed the line in example2.c that came from the dll
from:

char* ARGV[25]={"deinit","-core(","-input","ac3input","-output","test.mp3","-logfile ","BeSweet.txt",")","-ota(","-d","0","-G","max",")","-lame(","--alt-preset","128","--scale","1",")","-ssrc(","--rate","44100",")"};

to

char* ARGV[23]={"deinit","-core(","-input","ac3input","-output","test.mp2","-logfile ","BeSweet.txt",")","-ota(","-d","0","-G","max",")","-mp2enc(","-b","192",")","-ssrc(","--rate","44100",")"};

and

int ARGC=21; // 25

to

int ARGC=19; // 23


was that wrong?


mike

[DSPguru]

"deinit" should only come at the end of the process.

[TFM_TheMask]

Hi you all,

I want to use BeSweet dll to convert avi compressed audio streams to mpa audio. Thanks to this thread I got the ac3burst working, but what if the avi audio is a mp3 stream or another stream. How do I get this to work in delphi? Must I use AC3Burst or MPABurst or another burst?

Now I use the console application with this commandline:


"BeSweet.exe" -core( -input "Extracted_Audio.wav" -output "Encoded_Audio.mp2" -logfile "Encoded_Audio.log" ) -ota( -r 960 1001 -g max ) -shibatch( --rate 44100 ) -2lame( -e -b 224 -m s )

Can anyone help with this.

[DSPguru]

MPABurst

[TFM_TheMask]

That doesn't work. I use this commandline:

0: BeSweet.dll
1: -core(
2: -input
3: ac3input
4: -output
5: Test.mp2
6: -logfile
7: Test.txt
8: )
9: -ota(
10: -r
11: 960
12: 1001
13: -g
14: max
15: )
16: -shibatch(
17: --rate
18: 44100
19: )
20: -2lame(
21: -e
22: -b
23: 224
24: -m
25: s
26: )
besweet_argc: 27

I feed mpaburst a stream like BeStrip does with Ac3Burst.

By the way must I use AC3input in the commandline or does this trigger something else.

The logfile is:
BeSweet v1.5b6 by DSPguru.
--------------------------
Using azid.dll v1.8 (b825) by Midas (midas@egon.gyaloglo.hu).
Using Shibatch.dll v0.2 by Naoki Shibata & DSPguru (shibatch.sourceforge.net).
Using MP2enc.dll v1.129 (23/5/2002), Engine 1.129 <http://www.cdex.n3.net>.

Logging start : 05/08/03 , 14:24:03.

BeSweet.dll -core( -input ac3input -output Test.mp2 -logfile Test.txt ) -ota( -r 960 1001 -g max ) -shibatch( --rate 44100 ) -2lame( -e -b 224 -m s )

[00:00:00:000] +------- BeSweet -----
[00:00:00:000] | Input : ac3input
[00:00:00:000] | Output: Test.mp2
[00:00:00:000] | Floating-Point Process: Yes
[00:00:00:000] +-------- AZID -------
[00:00:00:000] | Output Stereo mode: Dolby surround compatible
[00:00:00:000] | Total Gain: 0.000dB, Compression: None
[00:00:00:000] | LFE levels: To LR -INF, To LFE 0.0dB
[00:00:00:000] | Center mix level: BSI
[00:00:00:000] | Surround mix level: BSI
[00:00:00:000] | Dialog normalization: No
[00:00:00:000] | Rear channels filtering: No
[00:00:00:000] +------ Shibatch -----
[00:00:00:000] | Source Sample-Rate: 48.0KHz
[00:00:00:000] | Dest. Sample-Rate: 44.1KHz
[00:00:00:000] | Attenuation : 0.0db
[00:00:00:000] +-------- FRC --------
[00:00:00:000] | Source Frame-Rate: 960
[00:00:00:000] | Dest. Frame-Rate: 1001
[00:00:00:000] +------- MP2ENC ------
[00:00:00:000] | Bitrate method : CBR
[00:00:00:000] | MP2 bitrate : 224
[00:00:00:000] | Channels Mode : Stereo
[00:00:00:000] | Error Protection: Yes
[00:00:00:000] +---------------------

and then I got an access violation when I call mpabursts.

What am I doing wrong?

[DSPguru]

-input mpainput :)

[TFM_TheMask]

Got a little bit further but now I got a "invalid floating point operation".

What I did is extracted a vbrmp3 from a avi-file. Because BeSweet has troubles with dealing with a vbrmp3 file I decoded the wav with madplay. Then I tried the conversion to mpa with BeSweet.exe with the same commands as described in my first post. It worked fine.

The logfile output is:

BeSweet v1.5b1 by DSPguru.
--------------------------
Using Shibatch.dll v0.2 by Naoki Shibata & DSPguru (shibatch.sourceforge.net).
Using MP2enc.dll v1.129 (23/5/2002), Engine 1.129 <http://www.cdex.n3.net>.

Logging start : 05/08/03 , 16:29:03.

BeSweet.exe -core( -input test.wav -output test.mpa -logfile besweetexe.log ) -ota( -r 960 1001 -g max ) -shibatch( --rate 44100 ) -2lame( -e -b 224 -m s )

[00:00:00:000] +------- BeSweet -----
[00:00:00:000] | Input : test.wav
[00:00:00:000] | Output: test.mpa
[00:00:00:000] | Floating-Point Process: Yes
[00:00:00:000] | Overall Track Gain: 1.499dB
[00:00:00:000] +------ Shibatch -----
[00:00:00:000] | Source Sample-Rate: 48.0KHz
[00:00:00:000] | Dest. Sample-Rate: 44.1KHz
[00:00:00:000] | Attenuation : 0.0db
[00:00:00:000] +-------- FRC --------
[00:00:00:000] | Source Frame-Rate: 960
[00:00:00:000] | Dest. Frame-Rate: 1001
[00:00:00:000] +------- MP2ENC ------
[00:00:00:000] | Bitrate method : CBR
[00:00:00:000] | MP2 bitrate : 224
[00:00:00:000] | Channels Mode : Stereo
[00:00:00:000] | Error Protection: Yes
[00:00:00:000] +---------------------
[01:45:58:293] Conversion Completed !
[01:45:58:293] Actual Avg. Bitrate : 214kbps
[00:17:31:000] <-- Transcoding Duration
Logging ends : 05/08/03 , 16:46:34.

Then I tried it using your dll and with mpainput and the log file output is:

BeSweet v1.5b6 by DSPguru.
--------------------------
Using hip.dll v1.13 by Myers Carpenter <myers@users.sf.net>
Using Shibatch.dll v0.2 by Naoki Shibata & DSPguru (shibatch.sourceforge.net).
Using MP2enc.dll v1.129 (23/5/2002), Engine 1.129 <http://www.cdex.n3.net>.

Logging start : 05/08/03 , 18:13:00.

BeSweet.dll -core( -input mpainput -output Test.mp2 -logfile Test.txt ) -ota( -r 960 1001 -g max ) -shibatch( --rate 44100 ) -2lame( -e -b 224 -m s )

[00:00:00:000] +------- BeSweet -----
[00:00:00:000] | Input : mpainput
[00:00:00:000] | Output: Test.mp2
[00:00:00:000] | Floating-Point Process: Yes
[00:00:00:000] +------ Shibatch -----
[00:00:00:000] | Source Sample-Rate: 48.0KHz
[00:00:00:000] | Dest. Sample-Rate: 44.1KHz
[00:00:00:000] | Attenuation : 0.0db
[00:00:00:000] +-------- FRC --------
[00:00:00:000] | Source Frame-Rate: 960
[00:00:00:000] | Dest. Frame-Rate: 1001
[00:00:00:000] +------- MP2ENC ------
[00:00:00:000] | Bitrate method : CBR
[00:00:00:000] | MP2 bitrate : 224
[00:00:00:000] | Channels Mode : Stereo
[00:00:00:000] | Error Protection: Yes
[00:00:00:000] +---------------------
[00:00:00:036] Stream error : Sync found after 428 bytes
[00:00:00:072] Stream error : Sync found after 632 bytes
[00:00:00:108] Stream error : Sync found after 119 bytes
[00:00:00:134] Stream error : Sync found after 451 bytes
[00:00:00:170] Stream error : Sync found after 1153 bytes
[00:00:00:222] Stream error : Sync found after 176 bytes
[00:00:00:258] Stream error : Sync found after 242 bytes
[00:00:00:282] Stream error : Sync found after 322 bytes
[00:00:00:308] Stream error : Sync found after 261 bytes
[00:00:00:344] Stream error : Sync found after 417 bytes
[00:00:00:370] Stream error : Sync found after 28 bytes
[00:00:00:394] Stream error : Sync found after 406 bytes
[00:00:00:420] Stream error : Sync found after 20 bytes
[00:00:00:444] Stream error : Sync found after 210 bytes
[00:00:00:468] Stream error : Sync found after 64 bytes
[00:00:00:494] Stream error : Sync found after 316 bytes
[00:00:00:518] Stream error : Sync found after 6 bytes
[00:00:00:544] Stream error : Sync found after 41 bytes
[00:00:00:580] Stream error : Sync found after 3 bytes
[00:00:00:607] Stream error : Sync found after 624 bytes
[00:00:00:633] Stream error : Sync found after 173 bytes
[00:00:00:669] Stream error : Sync found after 119 bytes
[00:00:00:705] Stream error : Sync found after 271 bytes
[00:00:00:729] Stream error : Sync found after 51 bytes
[00:00:00:755] Stream error : Sync found after 70 bytes
[00:00:00:781] Stream error : Sync found after 251 bytes
[00:00:00:807] Stream error : Sync found after 196 bytes
[00:00:00:833] Stream error : Sync found after 536 bytes
[00:00:00:869] Stream error : Sync found after 1122 bytes
[00:00:00:893] Stream error : Sync found after 523 bytes
[00:00:00:919] Stream error : Sync found after 544 bytes
[00:00:00:945] Stream error : Sync found after 544 bytes
[00:00:00:969] Stream error : Sync found after 10 bytes
[00:00:00:993] Stream error : Sync found after 219 bytes
[00:00:01:029] Stream error : Sync found after 22 bytes
[00:00:01:053] Stream error : Sync found after 564 bytes
[00:00:01:089] Stream error : Sync found after 379 bytes
[00:00:01:116] Stream error : Sync found after 147 bytes
[00:00:01:142] Stream error : Sync found after 888 bytes
[00:00:01:168] Stream error : Sync found after 557 bytes
[00:00:01:194] Stream error : Sync found after 10 bytes
[00:00:01:230] Stream error : Sync found after 692 bytes
[00:00:01:266] Stream error : Sync found after 94 bytes
[00:00:01:328] Stream error : Sync found after 23 bytes
[00:00:01:352] Stream error : Sync found after 1079 bytes
[00:00:01:378] Stream error : Sync found after 76 bytes
[00:00:01:402] Stream error : Sync found after 154 bytes
[00:00:01:428] Stream error : Sync found after 4 bytes
[00:00:01:452] Stream error : Sync found after 343 bytes
[00:00:01:478] Stream error : Sync found after 231 bytes
[00:00:01:505] Stream error : Sync found after 161 bytes
[00:00:01:541] Stream error : Sync found after 644 bytes
[00:00:01:565] Stream error : Sync found after 724 bytes
[00:00:01:589] Stream error : Sync found after 376 bytes
[00:00:01:613] Stream error : Sync found after 1419 bytes
[00:00:01:685] Stream error : Sync found after 687 bytes
[00:00:01:721] Stream error : Sync found after 1402 bytes
[00:00:01:747] Stream error : Sync found after 657 bytes
[00:00:01:783] Stream error : Sync found after 1255 bytes
[00:00:01:807] Stream error : Sync found after 48 bytes
[00:00:01:843] Stream error : Sync found after 599 bytes
[00:00:01:903] Stream error : Sync found after 317 bytes
[00:00:01:939] Stream error : Sync found after 1004 bytes
[00:00:01:975] Stream error : Sync found after 238 bytes
[00:00:02:011] Stream error : Sync found after 983 bytes
[00:00:02:035] Stream error : Sync found after 34 bytes
[00:00:02:071] Stream error : Sync found after 20 bytes
[00:00:02:095] Stream error : Sync found after 36 bytes
[00:00:02:131] Stream error : Sync found after 910 bytes
[00:00:02:167] Stream error : Sync found after 431 bytes
[00:00:02:203] Stream error : Sync found after 1029 bytes
[00:00:02:265] Stream error : Sync found after 580 bytes
[00:00 at this point I got a invalid floating point operation.

Maybe the size of the buffer is wrong. I give mpabursts a buffer of 2048 bytes. The Bs_info record the function mpabursts returns says [00:00:2048:000] peak: 0,00. It looks that the bs_info.seconds is giving back the size of the buffer.

[DSPguru]

a. which problem does BeSweet have with vbrmp3 ?
b. what were you inputting BeSweet.dll ? the original mpa stream (mpainput) or the decoded stream (pcminput) ?

[TFM_TheMask]

a)BeSweet.exe very rarely hangs when converting extracted wav(mp3) to mpa. This is also why DVD2SVCD uses MadPlay. I experienced this ones in my program The FilmMachine and that's why I use MadPlay before executing BeSweet. It's more stable.
b)You're right I feed the pcm output stream from madplay to youre dll. But when I feed the extracted wav (mp3) directly to the dll with mpainput it gives also an "invalid floating point operation".

The log output is:

BeSweet v1.5b6 by DSPguru.
--------------------------
Using hip.dll v1.13 by Myers Carpenter <myers@users.sf.net>
Using Shibatch.dll v0.2 by Naoki Shibata & DSPguru (shibatch.sourceforge.net).
Using MP2enc.dll v1.129 (23/5/2002), Engine 1.129 <http://www.cdex.n3.net>.

Logging start : 05/09/03 , 14:27:03.

BeSweet.dll -core( -input mpainput -output Test.mpa -logfile Test.txt ) -ota( -r 960 1001 -g max ) -shibatch( --rate 44100 ) -2lame( -e -b 224 -m s )

[00:00:00:000] +------- BeSweet -----
[00:00:00:000] | Input : mpainput
[00:00:00:000] | Output: Test.mpa
[00:00:00:000] | Floating-Point Process: Yes
[00:00:00:000] +------ Shibatch -----
[00:00:00:000] | Source Sample-Rate: 48.0KHz
[00:00:00:000] | Dest. Sample-Rate: 44.1KHz
[00:00:00:000] | Attenuation : 0.0db
[00:00:00:000] +-------- FRC --------
[00:00:00:000] | Source Frame-Rate: 960
[00:00:00:000] | Dest. Frame-Rate: 1001
[00:00:00:000] +------- MP2ENC ------
[00:00:00:000] | Bitrate method : CBR
[00:00:00:000] | MP2 bitrate : 224
[00:00:00:000] | Channels Mode : Stereo
[00:00:00:000] | Error Protection: Yes
[00:00:00:000] +--------------------- here I get an invalid floating point operation

[DSPguru]

let's try to clear things out..
- the pcminput works correctly ?
- now, lets only talk about the mpainput :
-- please post a logfile where you feed BeSweet.exe with your source mpafile. report problems if exists.
-- please post a logfile where you feed BeSweet.dll with mpa chunks using the minimalist commandline :
-core( -input mpainput -output Test.mpa -logfile Test.txt )

[TFM_TheMask]

1. What do you mean with pcminput. Must I use another function call for this, and use pcminput instead of mpainput?

2. Here's a logfile from your dll with minimum commands with the original extracted wav (mp3). It works, only found 1 stream error (sync found after 58 bytes). By the way your Bs_info record still doesn't give the time, it only gives 00:00:2048:000 every 2048 bytes (if I use an ac3-file and ac3bursts the bs_info record gives the correct time).

BeSweet v1.5b6 by DSPguru.
--------------------------
Using hip.dll v1.13 by Myers Carpenter <myers@users.sf.net>
Using lame_enc.dll v1.28 (18/4/2002), Engine 3.92 <http://www.mp3dev.org/>.

Logging start : 05/09/03 , 15:06:03.

BeSweet.dll -core( -input mpainput -output Test.mpa -logfile Test.txt )

[00:00:00:000] +------- BeSweet -----
[00:00:00:000] | Input : mpainput
[00:00:00:000] | Output: Test.mpa
[00:00:00:000] | Floating-Point Process: Yes
[00:00:00:000] | Source Sample-Rate: 48.0KHz
[00:00:00:000] +-------- LAME -------
[00:00:00:000] | Bitrate method : CBR
[00:00:00:000] | MP3 bitrate : 128
[00:00:00:000] | Channels Mode : Joint Stereo
[00:00:00:000] | Error Protection: No
[00:00:00:000] +---------------------
[00:00:00:024] Stream error : Sync found after 58 bytes
[01:45:51:936] Conversion Completed !
[01:45:51:936] Actual Avg. Bitrate : 127kbps
[00:22:13:000] <-- Transcoding Duration
Logging ends : 05/09/03 , 15:28:16.

3. Here is a logfile from your exe with the commands I always use with the original extracted wav (mp3). No errors, worked fine.

BeSweet v1.5b1 by DSPguru.
--------------------------
Using hip.dll v1.13 by Myers Carpenter <myers@users.sf.net>
Using Shibatch.dll v0.2 by Naoki Shibata & DSPguru (shibatch.sourceforge.net).
Using MP2enc.dll v1.129 (23/5/2002), Engine 1.129 <http://www.cdex.n3.net>.

Logging start : 05/09/03 , 16:09:33.

BeSweet.exe -core( -input test.wav -output test.mpa -logfile besweetexe.log ) -ota( -r 960 1001 -g max ) -shibatch( --rate 44100 ) -2lame( -e -b 224 -m s )

[00:00:00:000] +------- BeSweet -----
[00:00:00:000] | Input : test.wav
[00:00:00:000] | Output: test.mpa
[00:00:00:000] | Floating-Point Process: Yes
[00:00:00:000] | Overall Track Gain: 1.499dB
[00:00:00:000] +------ Shibatch -----
[00:00:00:000] | Source Sample-Rate: 48.0KHz
[00:00:00:000] | Dest. Sample-Rate: 44.1KHz
[00:00:00:000] | Attenuation : 0.0db
[00:00:00:000] +-------- FRC --------
[00:00:00:000] | Source Frame-Rate: 960
[00:00:00:000] | Dest. Frame-Rate: 1001
[00:00:00:000] +------- MP2ENC ------
[00:00:00:000] | Bitrate method : CBR
[00:00:00:000] | MP2 bitrate : 224
[00:00:00:000] | Channels Mode : Stereo
[00:00:00:000] | Error Protection: Yes
[00:00:00:000] +---------------------
[01:45:51:904] Conversion Completed !
[01:45:51:904] Actual Avg. Bitrate : 214kbps
[00:27:28:000] <-- Transcoding Duration
Logging ends : 05/09/03 , 16:37:01.

[DSPguru]

1. for wav (pcm inputs), you naturally need to set "-input pcminput", and use the PCM16S_Bursts function.

2. regarding the bugs you reported :
2.1. as you can see, there's no problem for BeSweet to process vbrmp3 file. dvd2svcd choosed to use madplay prior to the fixes i put into BeSweet.
2.2. the bs_info bug and the combination of sample-rate-conversion with frame-rate-conversion issue had already been fixed in the latter BeSweet versions.

conclusion : i just need to release a newer BeSweet.dll (v1.5b17)


now, i understand from you that you use BeSweet.dll, implementing a transcode process from AVI into mp2.
may i ask why don't you just use BeSweet.exe to do that directly ?

[TFM_TheMask]

1. What other functions can you call in the dll. In your example file you only speak of ac3bursts and mpabursts. Is the function declaration of PCM16S_Bursts the same as ac3bursts and mpabursts.

2.1. Ok I will skip MadPlay for the time being to test the new release of BeSweet.exe directly with vbrmp3's.

2.2. When will you release the new dll, and will it work with the commands I had troubles with. The commands also won't work on AC3 with ac3input. Only the commands used in the earlier posts in this thread are working (that is the one that is used in BeStrip, -ota( -d 0 -G max ) -lame( --alt-preset 128 --scale 1 )).

By the way I use the same commands on all extracted avi audiostreams with youre BeSweet.exe (so it doesn't matter if it is AC3, MP2, MP3, PCM).

At this time I use besweet.exe because it's the best at audio conversions. But since my Program is designed for windows I like to to show your bs_info record graphical. Consoles and CLI apps are so unfancy.

[DSPguru]

1. same decleration. open BeSweet.dll with "dependency walker" for more info on its functions.
2. i'll publish dll v1.5b17 in near future.
3. fancy gui is cool :cool: !

[TFM_TheMask]

Tip: I was bored and tested a little bit. I found out that -2lame is the problem. If I use the commands without -2lame it's working great.

By the way is toolame the replacement of 2lame in your latter versions?

Also tested the pcmbursts function, it worked fine without -2lame. The Bs_info record also gives false values back with this function.

Hope you will inform me when you publish youre new dll.

[DSPguru]

zmAn forced me to create a new BeSweet.dll build for his new tool (azidts), so you can get it from its package.

[TFM_TheMask]

Here are my test results of your new dll (v1.5b17).

1. AC3Bursts with ac3 file
a. BS_info record : peak is always 0, time is displayed correctly.
b. commands that work: shibatch, ota and toolame.
c. commands that won't work: mp2enc and 2lame. (floating point error).

2. MPABursts with vbrmp3 file
a. BS_info record : time is always 0, peak is displayed correctly.
b. commands that work: shibatch, ota and toolame.
c. commands that won't work: mp2enc and 2lame. (floating point error).

3. PCMBursts with decoded madplay file
a. BS_info record : is displayed correctly.
b. commands that work: shibatch, ota and toolame.
c. commands that won't work: mp2enc and 2lame. (floating point error).

Hope this is of some use. Conclusion: your dll is not working with mp2enc.dll (I only tested the commands I send in my first post).

By the way the logfile that is created is always appended. Also with the command -logfile.


4 Question
With BeSweet.exe there is always a two pass process. First it will find the maximum gain and after that the real transcoding begins.
With the dll the transcoding begins immediately. Is it also possible to do a two-pass like in the exe?

[DSPguru]

1. i'll check the peak issue
2. i'll check the time indicator for mp2 inputs
3. i'll check the 2lame/mp2enc problem

4. logfile is indeed appended, that's because i assume that the main tool will always write to the logfile before BeSweet.dll
5. BeSweet cannot run twice on the inputfile, because it has no control on the input file - the input is being burst, remeber ;) ?

so either you use a PostGain/HybridGain methods (if you don't know what that is, search in the audio encoding forum :) ), or either you implement two-passes in the main program :
first pass would come after calling "no_output", in the second pass you'll be asserting gain according to the PEAK value that was recieved in BS_Record. the two commandlines (1st & 2nd pass) should only differ in the "-g" switch. all others (but to "-logfile", i guess) should stay the same.

[TFM_TheMask]

new finding

1. AC3burst: if I use -G max the peak is displayed but there is no output. if I use -g max than the peak is not displayed, but there is output.

2. Mpaburst: if I use -G max I get a floating point error. if I use -g max there is output.

3. Pcmburst: if I use -G max there is no output. if I use -g max than there is output.

Is this normal?

[DSPguru]

:D :logfile: :D

[TFM_TheMask]

Here it is.

BeSweet v1.5b17 by DSPguru.
--------------------------
Using azid.dll v1.8 (b825) by Midas (midas@egon.gyaloglo.hu).
Using Shibatch.dll v0.24 by Naoki Shibata & DSPguru (shibatch.sourceforge.net).
Using tooLame.dll v0.2l (Apr 24 2003) by Mike Cheng <http://tooLame.sf.net>

Logging start : 05/12/03 , 23:06:45.

BeSweet.dll -core( -input ac3input -output Test.mp3 -logfile Test.txt ) -ota( -r 960 1001 -G max ) -shibatch( --rate 44100 ) -toolame( -m s -b 224 -e )

[00:00:00:000] +------- BeSweet -----
[00:00:00:000] | Input : ac3input
[00:00:00:000] | Output: Test.mp3
[00:00:00:000] | Floating-Point Process: No
[00:00:00:000] | PostGain normalize to : 6662.00
[00:01:48:448] Conversion Completed !
[00:00:10:000] <-- Transcoding Duration

Logging ends : 05/12/03 , 23:06:55.


BeSweet v1.5b17 by DSPguru.
--------------------------
Using azid.dll v1.8 (b825) by Midas (midas@egon.gyaloglo.hu).
Using Shibatch.dll v0.24 by Naoki Shibata & DSPguru (shibatch.sourceforge.net).
Using tooLame.dll v0.2l (Apr 24 2003) by Mike Cheng <http://tooLame.sf.net>

Logging start : 05/12/03 , 23:07:02.

BeSweet.dll -core( -input ac3input -output Test.mp3 -logfile Test.txt ) -ota( -r 960 1001 -g max ) -shibatch( --rate 44100 ) -toolame( -m s -b 224 -e )

[00:00:00:000] +------- BeSweet -----
[00:00:00:000] | Input : ac3input
[00:00:00:000] | Output: Test.mp3
[00:00:00:000] | Floating-Point Process: No
[00:00:00:000] +-------- AZID -------
[00:00:00:000] | Output Stereo mode: Dolby surround compatible
[00:00:00:000] | Total Gain: 0.000dB, Compression: None
[00:00:00:000] | LFE levels: To LR -INF, To LFE 0.0dB
[00:00:00:000] | Center mix level: BSI
[00:00:00:000] | Surround mix level: BSI
[00:00:00:000] | Dialog normalization: No
[00:00:00:000] | Rear channels filtering: No
[00:00:00:000] +------ Shibatch -----
[00:00:00:000] | Source Sample-Rate: 48.0KHz
[00:00:00:000] | Dest. Sample-Rate: 44.1KHz
[00:00:00:000] | Attenuation : 0.0db
[00:00:00:000] +-------- FRC --------
[00:00:00:000] | Source Frame-Rate: 960
[00:00:00:000] | Dest. Frame-Rate: 1001
[00:00:00:000] +------ tooLame ------
[00:00:00:000] | Bitrate method : CBR
[00:00:00:000] | MP2 bitrate : 224
[00:00:00:000] | Channels Mode : Stereo
[00:00:00:000] | Error Protection: Yes
[00:00:00:000] +---------------------
[00:01:48:448] Conversion Completed !
[00:01:48:448] Actual Avg. Bitrate : 214kbps
[00:00:21:000] <-- Transcoding Duration

Logging ends : 05/12/03 , 23:07:23.


BeSweet v1.5b17 by DSPguru.
--------------------------
Using hip.dll v1.19 by Myers Carpenter <myers@users.sf.net>
Using Shibatch.dll v0.24 by Naoki Shibata & DSPguru (shibatch.sourceforge.net).
Using tooLame.dll v0.2l (Apr 24 2003) by Mike Cheng <http://tooLame.sf.net>

Logging start : 05/12/03 , 23:13:21.

BeSweet.dll -core( -input mpainput -output Test.mpa -logfile Test.txt ) -ota( -r 960 1001 -G max ) -shibatch( --rate 44100 ) -toolame( -m s -b 224 -e )

[00:00:00:000] +------- BeSweet -----
[00:00:00:000] | Input : mpainput
[00:00:00:000] | Output: Test.mpa
[00:00:00:000] | Floating-Point Process: No
[00:00:00:000] | PostGain normalize to : 6662.00
[00:00:00:106] Conversion Completed !
[00:00:28:000] <-- Transcoding Duration

Logging ends : 05/12/03 , 23:13:49.


BeSweet v1.5b17 by DSPguru.
--------------------------
Using hip.dll v1.19 by Myers Carpenter <myers@users.sf.net>
Using Shibatch.dll v0.24 by Naoki Shibata & DSPguru (shibatch.sourceforge.net).
Using tooLame.dll v0.2l (Apr 24 2003) by Mike Cheng <http://tooLame.sf.net>

Logging start : 05/12/03 , 23:14:26.

BeSweet.dll -core( -input mpainput -output Test.mpa -logfile Test.txt ) -ota( -r 960 1001 -g max ) -shibatch( --rate 44100 ) -toolame( -m s -b 224 -e )

[00:00:00:000] +------- BeSweet -----
[00:00:00:000] | Input : mpainput
[00:00:00:000] | Output: Test.mpa
[00:00:00:000] | Floating-Point Process: No
[00:00:00:000] +------ Shibatch -----
[00:00:00:000] | Source Sample-Rate: 48.0KHz
[00:00:00:000] | Dest. Sample-Rate: 44.1KHz
[00:00:00:000] | Attenuation : 0.0db
[00:00:00:000] +-------- FRC --------
[00:00:00:000] | Source Frame-Rate: 960
[00:00:00:000] | Dest. Frame-Rate: 1001
[00:00:00:000] +------ tooLame ------
[00:00:00:000] | Bitrate method : CBR
[00:00:00:000] | MP2 bitrate : 224
[00:00:00:000] | Channels Mode : Stereo
[00:00:00:000] | Error Protection: Yes
[00:00:00:000] +---------------------
[00:00:00:000] Stream error : Sync found after 126 bytes
[00:00:00:106] Conversion Completed !
[00:00:00:106] Actual Avg. Bitrate : 608879kbps
[00:00:41:000] <-- Transcoding Duration

Logging ends : 05/12/03 , 23:15:07.


BeSweet v1.5b17 by DSPguru.
--------------------------
Using Shibatch.dll v0.24 by Naoki Shibata & DSPguru (shibatch.sourceforge.net).
Using tooLame.dll v0.2l (Apr 24 2003) by Mike Cheng <http://tooLame.sf.net>

Logging start : 05/12/03 , 23:18:12.

BeSweet.dll -core( -input pcminput -output Test.mpa -logfile Test.txt ) -ota( -r 960 1001 -G max ) -shibatch( --rate 44100 ) -toolame( -m s -b 224 -e )

[00:00:00:000] +------- BeSweet -----
[00:00:00:000] | Input : pcminput
[00:00:00:000] | Output: Test.mpa
[00:00:00:000] | Floating-Point Process: No
[00:00:00:000] | PostGain normalize to : 6662.00
[01:45:50:575] Conversion Completed !
[00:01:34:000] <-- Transcoding Duration

Logging ends : 05/12/03 , 23:19:46.

[DSPguru]

1. oopsi, i forgot you use BeSweet to encode mp2.
PostGain is only applicable for MP3 or Ogg Vorbis.
i guess you'll have to implement the two passes.

2. please post the logfiles of mp2enc.dll

3. what happens if you only encode with mp2enc and don't do FRC,SSRC ?

[TFM_TheMask]

I am trying to do the 2-pass. What I do is: first I use this command on the function NoOutput: BeSweet.dll -core( -input pcminput -output Test.mpa -logfile Test.txt ) -ota( -r 960 1001 -g max ) -shibatch( --rate 44100 ) -toolame( -m s -b 224 -e )

By the way I assume that the function NoOutput also gives back the BS_Info record and has the same parameters as the other functions. (I also saw that you added some functions to the dll, should I use some for my purpose?)

The BS_Info record looks like this:
hours: 1479
minutes: 29600
seconds: 253
milliseconds: -15328
peak: 0

So there is no peak information to use in the 2 pass.

If there were a peak value I would use it in the next commandline:
BeSweet.dll -core( -input pcminput -output Test.mpa -logfile Test.txt ) -ota( -r 960 1001 -g peakvalue ) -shibatch( --rate 44100 ) -toolame( -m s -b 224 -e )


Am I write?

You asked for the logfiles for mp2enc.dll. Here it is:

BeSweet v1.5b17 by DSPguru.
--------------------------
Using azid.dll v1.8 (b825) by Midas (midas@egon.gyaloglo.hu).
Using Shibatch.dll v0.24 by Naoki Shibata & DSPguru (shibatch.sourceforge.net).
Using MP2enc.dll v1.129 (23/5/2002), Engine 1.129 <http://www.cdex.n3.net>.

Logging start : 05/13/03 , 22:12:32.

BeSweet.dll -core( -input ac3input -output Test.mp3 -logfile Test.txt ) -ota( -r 960 1001 -g max ) -shibatch( --rate 44100 ) -2lame( -m s -b 224 -e )

[00:00:00:000] +------- BeSweet -----
[00:00:00:000] | Input : ac3input
[00:00:00:000] | Output: Test.mp3
[00:00:00:000] | Floating-Point Process: Yes
[00:00:00:000] +-------- AZID -------
[00:00:00:000] | Output Stereo mode: Dolby surround compatible
[00:00:00:000] | Total Gain: 0.000dB, Compression: None
[00:00:00:000] | LFE levels: To LR -INF, To LFE 0.0dB
[00:00:00:000] | Center mix level: BSI
[00:00:00:000] | Surround mix level: BSI
[00:00:00:000] | Dialog normalization: No
[00:00:00:000] | Rear channels filtering: No
[00:00:00:000] +------ Shibatch -----
[00:00:00:000] | Source Sample-Rate: 48.0KHz
[00:00:00:000] | Dest. Sample-Rate: 44.1KHz
[00:00:00:000] | Attenuation : 0.0db
[00:00:00:000] +-------- FRC --------
[00:00:00:000] | Source Frame-Rate: 960
[00:00:00:000] | Dest. Frame-Rate: 1001
[00:00:00:000] +------- MP2ENC ------
[00:00:00:000] | Bitrate method : CBR
[00:00:00:000] | MP2 bitrate : 224
[00:00:00:000] | Channels Mode : Stereo
[00:00:00:000] | Error Protection: Yes
[00:00:00:000] +--------------------- Here invalid floating point operation.

The Bs_info.milliseconds is 160.

This is also the case with pcmbursts and mpabursts

[DSPguru]

1. here's the best way to do it :
first pass
1. call BeSweet with arguments :
BeSweet.dll -core( -input pcminput -output Test.mpa -logfile Test.txt )
2. call NoOutput (has no input or output arguments)
3. call XXXburst as long as you wish
4. call BeSweet with deinit

second pass is trivial, i believe...


2. please post a logfile when encoding with mp2enc, omitting FRC&SSRC

3. the new functions are for pcm inputs (8bit / mono / 6 channels / etc'..)

[TFM_TheMask]

Here is the logfile of mp2enc, same result only bs_info.millisec is 064.

BeSweet v1.5b17 by DSPguru.
--------------------------
Using azid.dll v1.8 (b825) by Midas (midas@egon.gyaloglo.hu).
Using MP2enc.dll v1.129 (23/5/2002), Engine 1.129 <http://www.cdex.n3.net>.

Logging start : 05/13/03 , 22:33:54.

BeSweet.dll -core( -input ac3input -output Test.mp3 -logfile Test.txt ) -2lame( -m s -b 224 -e )

[00:00:00:000] +------- BeSweet -----
[00:00:00:000] | Input : ac3input
[00:00:00:000] | Output: Test.mp3
[00:00:00:000] | Floating-Point Process: Yes
[00:00:00:000] +-------- AZID -------
[00:00:00:000] | Output Stereo mode: Dolby surround compatible
[00:00:00:000] | Total Gain: 0.000dB, Compression: None
[00:00:00:000] | LFE levels: To LR -INF, To LFE 0.0dB
[00:00:00:000] | Center mix level: BSI
[00:00:00:000] | Surround mix level: BSI
[00:00:00:000] | Dialog normalization: No
[00:00:00:000] | Rear channels filtering: No
[00:00:00:000] | Source Sample-Rate: 48.0KHz
[00:00:00:000] +------- MP2ENC ------
[00:00:00:000] | Bitrate method : CBR
[00:00:00:000] | MP2 bitrate : 224
[00:00:00:000] | Channels Mode : Stereo
[00:00:00:000] | Error Protection: Yes
[00:00:00:000] +--------------------- Here invalid floating point operation.

Will try the 2-pass.

[DSPguru]

invalid operation @ BeSweet.dll or @ mp2enc.dll ?

[TFM_TheMask]

How can I check this?

[DSPguru]

it should appear at the same window that tells you about the wrong opeartion.


ps.. this isn't ready yet.. but it you feel like experimenting will the latest build :
http://BeSweet.notrace.dk/latest.zip

[TFM_TheMask]

The debugger window gives me no name of where it's going wrong?

Will check it out tomorrow. Thanks for the latest build.

[TFM_TheMask]

1. Tested your latest dll (b18).

The peak was fixed with ac3. So the bs_info record for ac3burst is complete.

With the mpaburst I still don't get the h, m, s and ms. It shows only the peak.

Also still a floating point in all functions with mp2enc.dll.

2.I also tried to find the floating point error, but no luck. I think it is in the mp2enc.dll because all other commands are working.

Does mp2enc.dll in combination with the besweet.dll work with you?

3. Can you tell me what the difference is between -toolame, -2lame and -mp2enc. The all have the same arguments.

4. 2pass

You said:
1. call BeSweet with arguments :
BeSweet.dll -core( -input pcminput -output Test.mpa -logfile Test.txt )

Should I also put -ota( -r 960 1001 -g max ) -shibatch( --rate 44100 ) -toolame( -m s -b 224 -e ) in this call?

[DSPguru]

1. no timestamps for mpa ? weird...
2. azidts correctly encodes mp2 (BeSweet.dll+mp2enc.dll).
3. toolame is toolame.dll, mp2enc/2lame is mp2enc.dll . toolame has much more arguments (4 psy models, vbr, etc'.. )
4. for the first pass, you should only care about the input plugins, and that's only two sections : -core & -azid.

[TFM_TheMask]

1. Do you perhaps know how azidts has implemented your dll with mp2enc? I think he is not using your burst functions. Isn't he using your dll like example1(complete file input) and not like example2 with bursts.

2. Am I correct when I use the peak value of the first pass (for example 0.87) in this statement, -ota( -r 960 1001 -g 0.87 ). if that is so the 2 pass works.

3. Which command can I use best for my purpose: -toolame, -2lame or -mp2enc.

[DSPguru]

1. he is using 166Bursts (for six channels), and 16SBursts (for stereo downmixed)
2. almost correct. the gain value should be 1/peak rather than the peak itself.
3. hmm.. don't know. anyway, this belongs (and probably had been discussed) in the audio encoding forum and svcd forums :)


so.. currently we only know of two flaws left. right :) ?
mp2enc & mpaburst timestamps.

Cheers !
Dg.

[TFM_TheMask]

Youre right, two flaws.

I will check azidts out, and try the 2-pass.

Do you know if there is any sourcecode of a program that uses your dll and mp2enc? Because I could check that out then.

cheers!

[DSPguru]

int main(int argc,char **argv)
{

#define BURST_SIZE 1000


char buffer[BURST_SIZE]; // bursts of mpa
int count =BURST_SIZE; // burst size

char* mpainput="mpainput";
char* deinit[1]={"deinit"};

HINSTANCE hDLL = LoadLibrary("BeSweet.dll");
BESWEET BeSweet = (BESWEET) GetProcAddress(hDLL, "BeSweet");
BURSTS Bursts = (BURSTS) GetProcAddress(hDLL, "MPA_Bursts");
VERSION BS_VERSION = (VERSION) GetProcAddress(hDLL, "BS_VERSION");

FILE* inputfile = fopen(argv[3],"rb");

BS_Record info;


// here we init BeSweet.dll with the commandline arguments for mpa
argv[3]=mpainput;
BeSweet(argc,argv);

// transcode
while(1)
{
count=fread(buffer,1,BURST_SIZE,inputfile);
if(!count) break;
info=Bursts(count,buffer);
fprintf(stdout,"[%02d:%02d:%02d:%03d] transcoding! Max gain : %2.1fdB \r",info->hours,info->minutes,info->seconds,info->millis,-20*log10(info->peak));
if (info->err) break;
}

// deinit BeSweet.dll
BeSweet(1,deinit);
FreeLibrary(hDLL);

fclose(inputfile);
}btw, it seems you've got the wrong mp2enc.dll release, as for me, it prints.. :Using MP2enc.dll v1.129 (7/6/2002), Engine 1.129 <http://www.cdex.n3.net>

[TFM_TheMask]

You were right. I think I got the wrong version. I found an older version v1.11 (big size 147.456 bytes) and it worked, but when I tried the conversed file in mplayer it didn't play.

I could not find the version you are using. Can you tell me where to get it. I've been to cdex site but I got a newer version v1.13 but also gave me an floating point error.

By the way only the date is different, not the version.

If It's the solution, only one flaw to go (mpa timestamps).

Thanks

[DSPguru]

i fixed the timestamp issue but didn't upload it yet.
mp2enc.dll can be found on the regular BeSweet package, and it's probably also in the azidts package.

[TFM_TheMask]

1. Tried the mp2enc.dll from azidts and your stable release and the both give me a floatingpoint error.

The version of mp2enc in azidts and your stable release is:

Using MP2enc.dll v1.129 (23/5/2002), Engine 1.129 <http://www.cdex.n3.net>.

Please could you send me the one dated 7/6/2002 that you have.

2. 2-pass
I tried as you said, first pass only -core commands and second pass the full commandline. I got a floating point error. I think this is because in the first pass different dll's are loaded by besweet as in the second pass. Seems that in the first pass your dll has to know which encoder it must use for the second pass. See the log:

BeSweet v1.5b18 by DSPguru.
--------------------------
Using lame_enc.dll v1.28 (4/6/2002), Engine 3.92 <http://www.mp3dev.org/>.

Logging start : 05/15/03 , 14:47:03.

BeSweet.dll -core( -input pcminput -output testpcm.mp2 -logfile testpcm.txt )

[00:00:00:000] +------- BeSweet -----
[00:00:00:000] | Input : pcminput
[00:00:00:000] | Output: testpcm.mp2
[00:00:00:000] | Floating-Point Process: Yes
[00:00:00:000] | Source Sample-Rate: 48.0KHz
[00:00:00:000] +-------- LAME -------
[00:00:00:000] | Bitrate method : CBR
[00:00:00:000] | MP3 bitrate : 128 pass 1
[00:00:00:000] | Channels Mode : Joint Stereo
[00:00:00:000] | Error Protection: No
[00:00:00:000] +---------------------
[00:05:00:504] Conversion Completed !
[00:05:00:504] Actual Avg. Bitrate : 0kbps
[00:00:02:000] <-- Transcoding Duration

Logging ends : 05/15/03 , 14:47:05.


BeSweet v1.5b18 by DSPguru.
--------------------------
Using Shibatch.dll v0.24 by Naoki Shibata & DSPguru (shibatch.sourceforge.net).
Using MP2enc.dll v1.11 (0/0/1999), Engine 1.11 <http://www.cdex.n3.net>.

Logging start : 05/15/03 , 14:47:10.

BeSweet.dll -core( -input pcminput -output testpcm.mp2 -logfile testpcm.txt ) -ota( -r 960 1001 -g 1.15 ) -shibatch( --rate 44100 ) -2lame( -m s -b 224 -e )

[00:00:02:000] +------- BeSweet -----
[00:00:02:000] | Input : pcminput
[00:00:02:000] | Output: testpcm.mp2
[00:00:02:000] | Floating-Point Process: No
[00:00:00:000] | Overall Track Gain: 1.214dB
[00:00:00:000] +------ Shibatch -----
[00:00:00:000] | Source Sample-Rate: 48.0KHz
[00:00:00:000] | Dest. Sample-Rate: 44.1KHz pass 2
[00:00:00:000] | Attenuation : 0.0db
[00:00:00:000] +-------- FRC --------
[00:00:00:000] | Source Frame-Rate: 960
[00:00:00:000] | Dest. Frame-Rate: 1001
[00:00:00:000] +------- MP2ENC ------
[00:00:00:000] | Bitrate method : CBR
[00:00:00:000] | MP2 bitrate : 224
[00:00:00:000] | Channels Mode : Stereo
[00:00:00:000] | Error Protection: Yes
[00:00:00:000] +---------------------

Then I used the full command in the first and second pass and used the peak from the first pass. It worked perfect.

At this time I only used pcmburst with -toolame, -ota and -shibatch.

Tomorrow I will test ac3burst and mpaburst.

3. Still don't know what is wrong with mp2enc!!

4. 2-pass with ac3burst and mpaburst.

mpaburst: got an access violation in the second pass because your dll uses hip.dll in the first-pass and not in the second pass. See log:

BeSweet v1.5b18 by DSPguru.
--------------------------
Using hip.dll v1.19 by Myers Carpenter <myers@users.sf.net>
Using Shibatch.dll v0.24 by Naoki Shibata & DSPguru (shibatch.sourceforge.net).
Using tooLame.dll v0.2l (Apr 24 2003) by Mike Cheng <http://tooLame.sf.net>

Logging start : 05/18/03 , 13:44:35.

BeSweet.dll -core( -input mpainput -output testmp3.mp2 -logfile testmp3.txt ) -ota( -r 960 1001 -g max ) -shibatch( --rate 44100 ) -toolame( -m s -b 224 -e )

[00:00:00:000] +------- BeSweet -----
[00:00:00:000] | Input : mpainput
[00:00:00:000] | Output: testmp3.mp2
[00:00:00:000] | Floating-Point Process: No
[00:00:00:000] +------ Shibatch -----
[00:00:00:000] | Source Sample-Rate: 48.0KHz
[00:00:00:000] | Dest. Sample-Rate: 44.1KHz
[00:00:00:000] | Attenuation : 0.0db
[00:00:00:000] +-------- FRC --------
[00:00:00:000] | Source Frame-Rate: 960
[00:00:00:000] | Dest. Frame-Rate: 1001
[00:00:00:000] +------ tooLame ------
[00:00:00:000] | Bitrate method : CBR
[00:00:00:000] | MP2 bitrate : 224
[00:00:00:000] | Channels Mode : Stereo
[00:00:00:000] | Error Protection: Yes
[00:00:00:000] +---------------------
[00:00:00:000] Stream error : Sync found after 126 bytes
[00:00:00:106] Conversion Completed !
[00:00:04:000] <-- Transcoding Duration

Logging ends : 05/18/03 , 13:44:39.


BeSweet v1.5b18 by DSPguru.
--------------------------
Using Shibatch.dll v0.24 by Naoki Shibata & DSPguru (shibatch.sourceforge.net).
Using tooLame.dll v0.2l (Apr 24 2003) by Mike Cheng <http://tooLame.sf.net>

Logging start : 05/18/03 , 13:44:51.

BeSweet.dll -core( -input mpainput -output testmp3.mp2 -logfile testmp3.txt ) -ota( -r 960 1001 -g 1.15 ) -shibatch( --rate 44100 ) -toolame( -m s -b 224 -e )

[00:00:04:000] +------- BeSweet -----
[00:00:04:000] | Input : mpainput
[00:00:04:000] | Output: testmp3.mp2
[00:00:04:000] | Floating-Point Process: No
[00:00:00:000] | Overall Track Gain: 1.214dB
[00:00:00:000] +------ Shibatch -----
[00:00:00:000] | Source Sample-Rate: 48.0KHz
[00:00:00:000] | Dest. Sample-Rate: 44.1KHz
[00:00:00:000] | Attenuation : 0.0db
[00:00:00:000] +-------- FRC --------
[00:00:00:000] | Source Frame-Rate: 960
[00:00:00:000] | Dest. Frame-Rate: 1001
[00:00:00:000] +------ tooLame ------
[00:00:00:000] | Bitrate method : CBR
[00:00:00:000] | MP2 bitrate : 224
[00:00:00:000] | Channels Mode : Stereo
[00:00:00:000] | Error Protection: Yes
[00:00:00:000] +---------------------

ac3burst: got an access violation in the second pass because your dll uses azid.dll in the first-pass and not in the second pass. See log:

BeSweet v1.5b18 by DSPguru.
--------------------------
Using azid.dll v1.8 (b825) by Midas (midas@egon.gyaloglo.hu).
Using Shibatch.dll v0.24 by Naoki Shibata & DSPguru (shibatch.sourceforge.net).
Using tooLame.dll v0.2l (Apr 24 2003) by Mike Cheng <http://tooLame.sf.net>

Logging start : 05/18/03 , 13:49:02.

BeSweet.dll -core( -input ac3input -output tfm.mp2 -logfile tfm.txt ) -ota( -r 960 1001 -g max ) -shibatch( --rate 44100 ) -toolame( -m s -b 224 -e )

[00:00:00:000] +------- BeSweet -----
[00:00:00:000] | Input : ac3input
[00:00:00:000] | Output: tfm.mp2
[00:00:00:000] | Floating-Point Process: No
[00:00:00:000] +-------- AZID -------
[00:00:00:000] | Output Stereo mode: Dolby surround compatible
[00:00:00:000] | Total Gain: 0.000dB, Compression: None
[00:00:00:000] | LFE levels: To LR -INF, To LFE 0.0dB
[00:00:00:000] | Center mix level: BSI
[00:00:00:000] | Surround mix level: BSI
[00:00:00:000] | Dialog normalization: No
[00:00:00:000] | Rear channels filtering: No
[00:00:00:000] +------ Shibatch -----
[00:00:00:000] | Source Sample-Rate: 48.0KHz
[00:00:00:000] | Dest. Sample-Rate: 44.1KHz
[00:00:00:000] | Attenuation : 0.0db
[00:00:00:000] +-------- FRC --------
[00:00:00:000] | Source Frame-Rate: 960
[00:00:00:000] | Dest. Frame-Rate: 1001
[00:00:00:000] +------ tooLame ------
[00:00:00:000] | Bitrate method : CBR
[00:00:00:000] | MP2 bitrate : 224
[00:00:00:000] | Channels Mode : Stereo
[00:00:00:000] | Error Protection: Yes
[00:00:00:000] +---------------------
[00:00:00:032] Stream error : Sync found after 334 bytes
[00:01:19:456] Conversion Completed !
[00:00:03:000] <-- Transcoding Duration

Logging ends : 05/18/03 , 13:49:05.


BeSweet v1.5b18 by DSPguru.
--------------------------
Using Shibatch.dll v0.24 by Naoki Shibata & DSPguru (shibatch.sourceforge.net).
Using tooLame.dll v0.2l (Apr 24 2003) by Mike Cheng <http://tooLame.sf.net>

Logging start : 05/18/03 , 13:49:22.

BeSweet.dll -core( -input ac3input -output tfm.mp2 -logfile tfm.txt ) -ota( -r 960 1001 -g 5.88 ) -shibatch( --rate 44100 ) -toolame( -m s -b 224 -e )

[00:00:03:000] +------- BeSweet -----
[00:00:03:000] | Input : ac3input
[00:00:03:000] | Output: tfm.mp2
[00:00:03:000] | Floating-Point Process: No

Hope this is of some kind of use to you.

[TFM_TheMask]

Hi DSPguru

I don't know if you read my last post but I found a little bit more on the mp2enc error.

I did a 2-pass with the pcm function with a full commandline with -2lame in it. The first pass went correctly, the logfile even said that it uses mp2enc. Only the second pass gave me a floating point error. See log:

BeSweet v1.5b18 by DSPguru.
--------------------------
Using Shibatch.dll v0.24 by Naoki Shibata & DSPguru (shibatch.sourceforge.net).
Using MP2enc.dll v1.129 (23/5/2002), Engine 1.129 <http://www.cdex.n3.net>.

Logging start : 05/18/03 , 20:00:07.

BeSweet.dll -core( -input pcminput -output testpcm.mp2 -logfile testpcm.txt ) -ota( -r 960 1001 -g max ) -shibatch( --rate 44100 ) -2lame( -m s -b 224 -e )

[00:00:00:000] +------- BeSweet -----
[00:00:00:000] | Input : pcminput
[00:00:00:000] | Output: testpcm.mp2
[00:00:00:000] | Floating-Point Process: Yes
[00:00:00:000] +------ Shibatch -----
[00:00:00:000] | Source Sample-Rate: 48.0KHz
[00:00:00:000] | Dest. Sample-Rate: 44.1KHz
[00:00:00:000] | Attenuation : 0.0db
[00:00:00:000] +-------- FRC --------
[00:00:00:000] | Source Frame-Rate: 960
[00:00:00:000] | Dest. Frame-Rate: 1001
[00:00:00:000] +------- MP2ENC ------
[00:00:00:000] | Bitrate method : CBR
[00:00:00:000] | MP2 bitrate : 224
[00:00:00:000] | Channels Mode : Stereo
[00:00:00:000] | Error Protection: Yes
[00:00:00:000] +---------------------
[01:45:50:575] Conversion Completed !
[00:02:33:000] <-- Transcoding Duration

Logging ends : 05/18/03 , 20:02:41.


BeSweet v1.5b18 by DSPguru.
--------------------------
Using Shibatch.dll v0.24 by Naoki Shibata & DSPguru (shibatch.sourceforge.net).
Using MP2enc.dll v1.129 (23/5/2002), Engine 1.129 <http://www.cdex.n3.net>.

Logging start : 05/18/03 , 20:03:32.

BeSweet.dll -core( -input pcminput -output testpcm.mp2 -logfile testpcm.txt ) -ota( -r 960 1001 -g 1.18 ) -shibatch( --rate 44100 ) -2lame( -m s -b 224 -e )

[00:02:33:000] +------- BeSweet -----
[00:02:33:000] | Input : pcminput
[00:02:33:000] | Output: testpcm.mp2
[00:02:33:000] | Floating-Point Process: Yes
[00:00:00:000] | Overall Track Gain: 1.438dB
[00:00:00:000] +------ Shibatch -----
[00:00:00:000] | Source Sample-Rate: 48.0KHz
[00:00:00:000] | Dest. Sample-Rate: 44.1KHz
[00:00:00:000] | Attenuation : 0.0db
[00:00:00:000] +-------- FRC --------
[00:00:00:000] | Source Frame-Rate: 960
[00:00:00:000] | Dest. Frame-Rate: 1001
[00:00:00:000] +------- MP2ENC ------
[00:00:00:000] | Bitrate method : CBR
[00:00:00:000] | MP2 bitrate : 224
[00:00:00:000] | Channels Mode : Stereo
[00:00:00:000] | Error Protection: Yes
[00:00:00:000] +---------------------

[DSPguru]

10q, TFM. sorry, i'm currently short in time, but didn't forget this thread.

* maybe there's a problem with mp2enc.dll dependencies with the dlls on your computer.

ps,
get the latest dll binary from my webpage.

[TFM_TheMask]

I used your latest dll I think in the latest test (v1.5b18).

Can you tell me what your dll is doing different in the first pass, because then everything with mp2enc works fine.

I also tried mp2enc on a different computer at work. Same floating point error.

By the way the timestamp of mpaburst works fine now.

Thanks