Post All Winamp 5.32 Official Plugin Bug Reports In This Thread

[DJ Egg]

Report ALL Winamp 5.32 Plugin bugs in this thread ONLY.

ALL OTHER REPORTS WILL BE DELETED (YOU HAVE BEEN WARNED).

INCOMPLETE\LAME REPORTS (AKA: IMPOSSIBLE TO REPRODUCE) WILL BE IGNORED.


Here is a LIST of the Official Plugins and their Versions that shipped with 5.32

Input:
Nullsoft CD plug-in (CDDB2) v3.3 [in_cdda.dll]
Nullsoft DirectShow Decoder v1.0 [in_dshow.dll]
Nullsoft FLAC Decoder v1.1.2c [in_flac.dll]
Nullsoft LineIn plugin v3.11 [in_linein.dll]
Nullsoft MIDI player 3.13 [in_midi.dll]
Nullsoft Module Decoder v2.2.3 [in_mod.dll]
Nullsoft MPEG Audio Decoder 4.03 [in_mp3.dll]
Nullsoft MPEG-4 Audio Decoder v1.1 [in_mp4.dll]
Nullsoft NSV Decoder v1.07 [in_nsv.dll]
Nullsoft Vorbis Decoder v1.47 [in_vorbis.dll]
Nullsoft Waveform Decoder v3.1 [in_wave.dll]
Nullsoft Windows Media Decoder 3.23 [in_wm.dll]

Output:
DirectSound Output v2.46 (d) [out_ds.dll]
Waveout output v2.1 (d) [out_wave.dll]
Nullsoft Disk Writer Plug-in v2.11 [out_disk.dll]

Visualization:
Advanced Visualization Studio v2.81c [vis_avs.dll] (v2.81d update available)
MilkDrop 1.04d [vis_milk.dll]
Nullsoft Tiny Fullscreen v2.08 [vis_nsfs.dll]

DSP/FX:
Nullsoft Signal Processing Studio DSP v0.37 [dsp_sps.dll]

CD Ripping Encoders:
AAC/aacPlus Encoder v1.23 [enc_aacplus.dll]
FLAC Encoder (Flake SVN r108) [enc_flac.dll]
MP4 AAC/aacPlus Encoder v1.23 [enc_aacplus.dll]
MP3 Encoder v1.32 [enc_lame.dll] (with lame_enc.dll 3.97)
WAV Output (built-in)
WMA Encoder v1.1 [enc_wma.dll]

General Purpose:
Nullsoft Tray Control v2.0 [gen_tray.dll]
Nullsoft Global Hotkeys v1.31 [gen_hotkeys.dll]
Nullsoft Winamp Library v3.04 [gen_ml.dll]
Nullsoft Modern Skins Support v1.24 [gen_ff.dll]
Jump To File Extra v0.97.7h (Build 57) [gen_jumpex.dll]

Media Library:
Nullsoft Bookmarks v1.02 [ml_bookmarks.dll]
Nullsoft Dashboard v1.0 [ml_dash.dll]
Nullsoft Format Converter v2.01 [ml_transcode.dll]
Nullsoft History v1.01 [ml_history.dll]
Nullsoft Local Media v1.1 [ml_local.dll]
Nullsoft Now Playing v2.01 [ml_nowplaying.dll]
Nullsoft Online Services v1.2 [ml_online.dll]
Nullsoft Playlists v1.05 [ml_playlists.dll] (v1.06 update available)
Nullsoft Portable Music Player Support v1.12 [ml_pmp.dll]
Nullsoft Replay Gain Analyzer v1.01 [ml_rg.dll]
Nullsoft Rip & Burn v1.11 [ml_disc.dll]
Nullsoft SHOUTcast Wire v1.06 [ml_wire.dll]
ml_xpdxs (Predixis MusicMagic) v1.0.4 [ml_xpdxs.dll]

Portables:
Nullsoft Creative NJB Plug-in v0.51 [pmp_njb.dll]
Nullsoft iPod Plug-in v0.54 [pmp_ipod.dll]
Nullsoft Microsoft PlaysForSure Plug-in v0.53 [pmp_p4s.dll]
Nullsoft USB Device Plug-in v0.52 [pmp_usb.dll]
Nullsoft ActiveSync Plug-in v0.14 [pmp_activesync.dll]

____________________________________________________________


Provide the following Information with every Report,
and with every CONFIRMATION of someone else's report:

Winamp 5.32 - clean install (with NO 3rd-party plugins)
Windows OS, including Service Pack
System specs (mobo, cpu, ram)
Video Card, Video Card Drivers version
Sound Card, Sound Card Drivers version
DirectX version
Plugin and Plugin version affected by/causing the bug

A full, detailed, step-by-step method of reproduction

If applicable, also include Error Logs and/or FULL Error Messages.

To reproduce some bugs, we may also need a link to a zipped sample file
(eg. an .mp3, .aac, .wav, .mid file, etc).

The more information you provide US, the Faster we can or will help you.


Do NOT report ANY 3rd-party plugin bugs/issues here.
Report them to the author(s) instead (and point them to the 5.32 sdk)



Maintained by DJ Egg

[YtseJam]

Clean install
Windows XP-SP2
AMD Athlon XP 1.13GHz, 512MB DDRAM
ATI Radeon 9600 256MB DDRAM
Creative Audigy
DirectX 9c
Nullsoft Format Converter v2.01 [ml_transcode.dll]

Options >> Preferences >> Media Library >> Nullsoft Format Converter >> Configure selected plug-in >> CRASH

I did not have my iPod plugged-in at the time.
Didn't try with the iPod plugged-in, irrelevent anyhow.

[Edit --> DJ Egg]
Hmmm, I can't reproduce this. Anyone else?

[pkasting]

This is a followup to a bug I reported in the 5.31 core thread. I had reported that a track which played in iTunes 7 did not play in Winamp 5.31. DJ Egg correctly noted that the problem was that the album replaygain for the track was set to -51.62 dB, and commented that calculating the replaygain again for that track should fix the problem.

I have found that the problem happens when I calculate the replaygain for the entire album this file is part of, and still occurs in Winamp 5.32. To reproduce:
(1) Download http://www.sendspace.com/file/cdejqw (warning, 60+ MB)
(2) Add whole album to playlist, select, and send to->calculate replaygain
(3) Once calculation is complete, note that the album gain is supposedly around -9 dB, but the "album peak" is suspiciously high, at over 380.
(4) Save as album data
(5) Check any one of the files' metadata (alt-3), and note that the gain values appear sane
(6) Change Winamp's title to "%gain%", turn on replaygain playback and select album gain, and try to play one of the tracks. The title is "-51.62 dB" and the track is silent.

Thus there are multiple bugs here: first, that the album peak and album gain appear to be wrong, and second, that the gain used during playback and the gain displayed on the ID3 tag page do not match.

Winamp 5.32 - clean install (with NO 3rd-party plugins)
Windows XP SP2
Athlon XP "Barton" 2500+ (1.84 GHz), 512 MB RAM, Shuttle MN31N (from memory, could be wrong)
Radeon 9800 PRO (ati2dvag.dll v6.14.0010.6631)
Onboard NVIDIA nForce audio (nvapu.sys v6.14.0442.0030)
DirectX 9.0c

[pkasting]

This is a repost of a bug I posted on the 5.31 plugins thread, which has not been fixed in 5.32.

Files whose pathnames contain extended characters do not appear in the media library.

Steps to reproduce:
(1) Create a subdirectory whose name contains U+03A9 (Greek capital letter Omega). (I have tried this with various other extended characters as well, such as U+2665 black heart suit, so it's not specific to this character.)
(2) Place some audio files in this subdirectory.
(3) Enqueue or play the files. Note that Winamp plays them successfully.
(3) Attempt to add the audio files to Winamp's media library (for example, by dragging to the ML). Note files do not appear.

As a side note, getting the file properties (alt-3) on an .m4a in such a directory results in a "short" pathname at the top of the window, e.g.:
C:\Music\Own\DISSEC~1\(09)DI~1.M4A
...whereas, when placed in a "normal" subdirectory, the full "long" pathname is shown. For .mp3 files, the full path is shown in both cases.

Winamp 5.32 - clean install (with NO 3rd-party plugins)
Windows XP SP2
Athlon XP "Barton" 2500+ (1.84 GHz), 512 MB RAM, Shuttle MN31N (from memory, could be wrong)
Radeon 9800 PRO (ati2dvag.dll v6.14.0010.6631)
Onboard NVIDIA nForce audio (nvapu.sys v6.14.0442.0030)
DirectX 9.0c
Problematic plugin: gen_ml.dll

[DJ Egg]

Re: Error calculating album replaygain

Reasons why never to use Xing Encoder - #1
http://stashbox.org/5595/rg_results.png

Yeah, I can reproduce the problem, but there's definitely something wrong with those files. What software did you use to rip them with? I suggest that you switch to something else which uses Lame Encoder instead of Xing (eg. Winamp, CDex, EAC).

I also tried to calculate RG with Foobar2000, but it comes up with some "8 out of 12 items could not be processed. If they were a part of an album, album scan results are inaccurate" error, and the 8 bad files say "unsupported format or corrupted file":
http://stashbox.org/5598/rg_results_fb.png


I'm honestly not sure if anything can be done about this, heh.
All I can suggest for now is that you remove the RG info with an app like Foobar2000.

[Benski]

Originally posted by pkasting
Thus there are multiple bugs here: first, that the album peak and album gain appear to be wrong,

This is happening because the files appear to be corrupt.

Originally posted by pkasting
and second, that the gain used during playback and the gain displayed on the ID3 tag page do not match.

This is happening because of the crazy replaygain values. Clipping prevention is turned on by default, and since the album peak is 380, it must lower the gain by -51dB [20 * log10(1/380)]

[Jimbolino]

Not really a bug, just a small strangeness.

Winamp 5.32 - clean install (with NO 3rd-party plugins)
Windows XP SP2
System specs (Laptop, P4-2ghz, 512mb)
Ati 7000 mobility
Realtek AC97, Giving a driver error due to wrong drivers from Windows Update
DirectX 9.0c
out_wave.dll

This is how i got the bug:

1. Open winamp without a valid/working soundcard
2. Select waveout plugin
3. Play a video and behold the following strange url in the error message:

---------------------------
waveOut output v2.1 (d) error
---------------------------
No sound devices found. Please buy a soundcard first (or install proper drivers if you think that you have one).
More info can be found at http://www.geocities.com/mmsystem002/

Error code: 2
Windows error message:
"A device ID has been used that is out of range for your system."
---------------------------
OK
---------------------------

Ofcourse there is not a working geocities page there...

[Edit --> DJ Egg]
Yup. That error message has been in out_wave for more than 5 years.
Back then, the geocities url was valid, but it died a year or so ago.
Maybe we should change it to this, or just remove the line totally?

[PrinceDhampir]

there are a bug in the Percentage of the progress of the song when it is being transfered to the player.

I have a Portable music player conventional is a Markvision player

I found the bug in this way:

1. Clean Install
2. Open Winamp
3. I plug my Portable
4. I select an Album And i Click on Sen to>Portable
5. When i see the Window of the Conversion Progress the
Percentage give crazy numbers like 233%, 3330% and so...

In other Way when i Begin to Transfer the Music to the Player
it Crash
And Sometimes Winamp Show The percentage well...

[Logy]

bugs...

still present in media library on latest winamp 5.32 with only essential pack install and win xp sp 2, when you removed the usb drive and connect agains another portable exist (unusable aka useless)), but this is only happens when you click "hide this icons when no portables attached"!

adios!

[angelofmoon]

Creative ZVM (30GB) streaming will skip over files and filenames with Unicode in them calling them [Http:/1.1 404 NOT FOUND] etc etc. Will play the rest of everything fine.

[a1studmuffin]

I've always had an m3u file like this:

D:\MP3s\.

And in previous versions of Winamp this would recursively add my entire MP3s folder to the playlist. In 5.32 it doesn't work.

[DrO]

is already known about and is apparently fixed for the next winamp version (was a thread about this in the tech support forum yesterday but i don't have the time to find it out)

-daz

[Edit --> DJ Egg]
It was actually posted in the 5.32 Released thread
and yes, it is already fixed for 5.33

[blade06]

Winamp 5.32 Clean Install
WindowsXP Pro Service Pack 2 (formatted a few days ago)
Pentium 4 1.3GHz, 512MB Ram
The rest is irrelevant considering the issue

How to recreate:
Options > Preferences > Media Library > History
Uncheck all 4 boxes
Play a file
Refresh Library View (I use Local Media > Audio in the screenshot)
Play Count increments and played last time shows

Winamp 5 was clean installed, the entire Winamp directory was deleted first, no previous media library files remained. History was turned off before playing the very first file (other than the demo) after installing Winamp.

[Benski]

Originally posted by blade06
Winamp 5.32 Clean Install
WindowsXP Pro Service Pack 2 (formatted a few days ago)
Pentium 4 1.3GHz, 512MB Ram
The rest is irrelevant considering the issue

How to recreate:
Options > Preferences > Media Library > History
Uncheck all 4 boxes
Play a file
Refresh Library View (I use Local Media > Audio in the screenshot)
Play Count increments and played last time shows

Winamp 5 was clean installed, the entire Winamp directory was deleted first, no previous media library files remained. History was turned off before playing the very first file (other than the demo) after installing Winamp.

uncheck Prefs->Local Media->Enable 'Recently Played' view in Media Library

[DevelX]

Tryied on 2 computers:
1. computer:
Winamp 5.32 - clean install Bundle
Windows XP SP2 Pro
3Ghz HyperThreading CPU - Intel Pentium4, 1GB Ram
MSO Nvidia Graphic 6600GT, newest drivers aviable
soundcanrd onboard, unknown, HD audio controlled by Realtek High Definition audio, 16bit I guess.
DirectX version: 9.0c
in_vawe 3.1

2. computer (this is my main reason):
Winamp 5.32 Clean install Bundle
WinXP SP2 Pro
1.7 Ghz Intel Mobile
Intel Graphic chipset
Sound card onboard, Intel High Definition Audio, 196kHz, 32bit.
DirectX: 9.0c
in_wave: 3.1

So - the problem is: I can't configure in_wave to make 32 bit output, so even 32bit waves are played 16bit (out_wave showes me). So - for what purpose I have 32bit soundcard, if I can play only 16bit sounds? There is great difference in sound quality and it was configurable in older winamps, so why not now?

[Edit --> DJ Egg]
Wrong forum/thread. This is not a bug.
in_wave now uses the global 'allow 24bit' setting in Prefs > Playback.
When enabled, 32bit WAV's will play at 32bit.

[Sliver85]

Problem with playing mpg's. (Plugin problem?)
when playing a .mpg file, if you seek to any time at all, playback restarts at the beginning of the file. this happens when using the arrow keys for seeking +/- 5seconds, when using the seek bar in the main player window and when using the 'jump to time' function. the time display shows what the progress of the file would be if the playback had not restarted, and the file does play to completion.

[pkasting]

Originally posted by Benski
This is happening because the files appear to be corrupt.

Fair enough. If the RG algorithm can be made more robust to handle the corruption, that would be awesome (after all, the MP3 decoding itself is robust enough to play these files back), but if not, then something like what Foobar does (detecting bogus results and displaying an error instead of using them) would be a good second choice.

This is happening because of the crazy replaygain values. Clipping prevention is turned on by default, and since the album peak is 380, it must lower the gain by -51dB [20 * log10(1/380)]

Does that mean that clip prevention lowers the album by the max of (album loudness, album peak)? That seems strange; why not lower by the album replaygain, and if potential clipping is detected during decoding, just do some limiting at that point? In any case, it seems like the ID3 tag should tell the user what the program is actually going to do; I would have figured out the true nature of this problem more quickly had I not checked it and seen that the reported gain was within a reasonable range.

[pkasting]

Apologies in advance for the vague steps to reproduce here.

I opened my media library, which is on the "Audio" page. I then right-clicked the columns in the main view, selected "customize", and added Album Gain and Track Gain entries. I then clicked the Album Gain column to sort by that. After a few seconds, Winamp crashed. After shutting down the process, I restarted it and found my Media Library no longer had the gain columns and was showing an old search I had done. When I deleted the search (so that all my audio tracks would appear instead of just those matching the search), Winamp crashed again while the main listing read "Scanning... /". This now happens every time I repeat these steps.

I pulled up the crash in VS .NET 2003 and found it was a NULL pointer dereference in ml_local.dll. The callstack looks like:

> ml_local.dll!031775c2()
ml_local.dll!03177968()
nscrt.dll!7c363113()
nscrt.dll!7c342f71()
nscrt.dll!7c342f9c()
nscrt.dll!7c34aefe()
ml_local.dll!0317d591()
ntdll.dll!7c90d529()
user32.dll!77d4a15c()
ntdll.dll!7c90eae3()
user32.dll!77d48b8c()
user32.dll!77d4cbd4()
ml_local.dll!03174f76()
nscrt.dll!7c34e9a3()
nde.dll!031a6862()
nde.dll!031a686b()
ml_local.dll!03163059()
ml_local.dll!03177b3c()
ml_local.dll!031780b0()
ml_local.dll!03178142()
kernel32.dll!7c80b683()

Winamp 5.32 - clean install (with NO 3rd-party plugins)
Windows XP SP2
Athlon XP "Barton" 2500+ (1.84 GHz), 512 MB RAM, Shuttle MN31N (from memory, could be wrong)
Radeon 9800 PRO (ati2dvag.dll v6.14.0010.6631)
Onboard NVIDIA nForce audio (nvapu.sys v6.14.0442.0030)
DirectX 9.0c
Problematic plugin: ml_local.dll

[CraigF]

The crash by sorting with album gain is a known bug corrected for the next version. Thanks.

[flocksoft]

(tested on Winamp 5.32 clean install, no third party plugins on a fresh Windows XP SP2 Pro; the mp3 id3 tag frame for the publisher is TPUB)

maybe this is the expected behaviour, I don't know...

Winamp is able to read TPUB frame, but selecting Edit selected items... in Media Library to change the publisher, such value is changed only in Media Library, not in the file

test:

• add the attached file in Media Library (the publisher in the file is xxxpublisher)
• change with the Media Library the publisher to a new value and save
• the Media Library is correctly updated with the new value
• opening the file with another program like Mp3tag or foobar2000 the publisher tag is xxxpublisher, not the new value

this behaviour occurs also when the file has no TPUB frame

tested only with mp3...

can someone confirms this?

bye

[Edit --> DJ Egg]
Reproduced. Thanks

[abu]

clean install of 5.32.
Whenever you do a ML_IPC_DB_UPDATEITEM from within a plugin for a ML item that has a TUID2 extended info (which it gets by Winamp during ripping or showing the "NowPlaying" screen), Winamp crashes.

This can only be reproduced from a ML plugin, you can write a simple one that uses the ML_IPC_DB_UPDATEITEM message, see example below.

code:


void testIt()

{

  char query[]="type=0";

  mlQueryStruct mqs={query,0,};

  SendMessage(plugin.hwndLibraryParent,WM_ML_IPC,(WPARAM)&mqs,ML_IPC_DB_RUNQUERY); //run the query

  int lr=mqs.results.Size;

  for(i=0; i<lr; i++)

  {

    itemRecord *ice=&mqs.results.Items[i];

    // here we crash if ice has a TUID2 value:

    SendMessage(plugin.hwndLibraryParent,WM_ML_IPC,(WPARAM)ice,ML_IPC_DB_UPDATEITEM);

  }

  SendMessage(plugin.hwndLibraryParent,WM_ML_IPC,(WPARAM)&mqs,ML_IPC_DB_FREEQUERYRESULTS); //free memory

}

So you see, all I do is writing the old info back, no changes are made. That crashes with the TUID2 info. In the debugger, I can see that the crash happens somewhere in NSCRT, called from NDE, called from ml_local.

Obviously Winamp should not crash if we give it back the same information that we got from the Winamp API.

This was posted for the 5.23 version already, but has not been fixed.

[abu]

One more oddity: Updating a ML item which has an extended info "albumartist", where the value starts with a digit (0..9), Winamp crashes at ML_IPC_DB_UPDATEITEM or ML_IPC_DB_ADDORUPDATEITEM!

e.g. getRecordExtendedItem(ice,"albumartist") yields "2 unlimited" in my case.

More on this in this thread, mainly the last couple of posts.

[abu]

Winamp 5.32 - clean install (with NO 3rd-party plugins)
Windows XP SP2System specs (mobo, cpu, ram)
Nullsoft iPod Plug-in v0.54 pmp_ipod.dll

The iPod Nano 2G is capable of gapless playback of MP3 files, it works fine with iTunes.
With pmp_ipod, simply transfer some files from a live album (or something like "The wall" from Pink Floyd) to the iPod and play them: You hear little clicks and gaps between the tracks. Gapless playback does not work at all.

Technical hint: This is caused because in the MHIT at offset 248 there must be a "magical" number, that is quite close to the number of bytes that the track has (it's always a bit less). I don't know what this number really is, it is only written by iTunes 7. But if it is set to zero (as does pmp_ipod), the gapless does not work. This seems to affect only the iPod Nano, iPod 5.5G seem to work (I don't have one myself).

This has been reported to Will via PM already, just wanted to make sure it's in the records. As the Nano is a millionseller, I hope you can support it soon for Gapless. Thanks.

[xxeonn]

Windows XP SP2 Home/Professional

Winamp 5.3x audio analyzer does not displays the entire audio spectrum. The bar version of the analyzer does not displays frequencies below 310Hz and the higher frequency bars remain idle. I could remember some older vesions of winamp having this problem but the were corrected, but it's back again. I don't think this has to do with my sound card or drivers because the same problem occurs on multiple computer i tested.

Any opinions?

[Mnietek]

Winamp 5.32,
XP SP2 (but I think it doesn't matter here).
gen_ff.dll v1.24

When using modern skin, winamp doesn't respond right to SendMessage(hwnd_winamp,WM_WA_IPC,IPC_GETWND_PE,IPC_IS_WNDSHADE).
It always returns 0 regardles of window state.
With classic-skinned winamp it works as it should.

[Ted [MSFT]]

Winamp 5.32 - clean install on XP SP2 - WMP11
K8VSEDX AMD64 3200+ 1G RAM
GeForce4 MX 440
ADI AC97
DX 9.0c
Nullsoft Microsoft PlaysForSure Plug-in v0.53 [pmp_p4s.dll]

It apears that the p4s plugin is not using the most recent list of WMDM format codes.

http://msdn.microsoft.com/library/de...formatcode.asp

The behavior is that WinAmp is not able to sync newer aditions to this list (FLAC, OGG, AAC) to devices which support those format codes. If you try it just give you the "unsupported" message.


Aditionally I hit an AV when trying to sync on first launch after install. I was not able to see the crash a second time. I created a full crash dump of this if you would like it for debugging (note: it is 132MB), but here's the basics:
(1454.13b8): Access violation - code c0000005 (!!! second chance !!!)
eax=05205070 ebx=00000000 ecx=00000017 edx=050d54c0 esi=7c34af62 edi=076efee8
eip=050c44fb esp=076ee514 ebp=076eecb4 iopl=0 nv up ei ng nz ac po cy
cs=001b ss=0023 ds=0023 es=0023 fs=003b gs=0000 efl=00000297
*** WARNING: Unable to verify checksum for C:\Program Files\Winamp\Plugins\pmp_p4s.dll
0:024> k
ChildEBP RetAddr
WARNING: Stack unwind information not available. Following frames may be wrong.
076eecb4 050c21a4 pmp_p4s!winampUninstallPlugin+0x2fdb
076eff24 04e4e957 pmp_p4s!winampUninstallPlugin+0xc84
7c36e3da f8458d51 ml_pmp!winampGetMediaLibraryPlugin+0x404d
51ec8b55 00000000 0xf8458d51

050c44b5 74e3 jz pmp_p4s!winampUninstallPlugin+0x2f7a (050c449a)
050c44b7 68bc540d05 push 0x50d54bc
050c44bc ff75f0 push dword ptr [ebp-0x10]
050c44bf ffd6 call esi
050c44c1 85c0 test eax,eax
050c44c3 59 pop ecx
050c44c4 59 pop ecx
050c44c5 750d jnz pmp_p4s!winampUninstallPlugin+0x2fb4 (050c44d4)
050c44c7 c6450b01 mov byte ptr [ebp+0xb],0x1
050c44cb c745e881b90000 mov dword ptr [ebp-0x18],0xb981
050c44d2 eb07 jmp pmp_p4s!winampUninstallPlugin+0x2fbb (050c44db)
050c44d4 c745e809300000 mov dword ptr [ebp-0x18],0x3009
050c44db 8a450b mov al,[ebp+0xb]
050c44de 8b4de4 mov ecx,[ebp-0x1c]
050c44e1 53 push ebx
050c44e2 83c120 add ecx,0x20
050c44e5 88472c mov [edi+0x2c],al
050c44e8 e87dcbffff call pmp_p4s+0x106a (050c106a)
050c44ed 8b8000010000 mov eax,[eax+0x100]
050c44f3 8b08 mov ecx,[eax]
050c44f5 50 push eax
050c44f6 a398bc0d05 mov [pmp_p4s!winampUninstallPlugin+0x1a778 (050dbc98)],eax
050c44fb ff5104 call dword ptr [ecx+0x4] ds:0023:0000001b=????????

[HiRMU]

Winamp 5.32 - Clean Install on WinXP SP2 or Win2000 SP4
Pentium 4 Prescott 3.0E GHz + 2 GB RAM + GeForce 6600 GT
Pentium III-S 1.4 GHz + 512 MB RAM + GeForce 4 Ti 4600
DirectX v9.0c "December 2006 Version" on both

Winamp 5.32's AAC Encoder apparently has some kind of compatibility issue. When I use the Transcoder feature (Right-click, send to Format Converter) to transcode tunes from MP3 to AAC for my Nokia E70 smartphone ( http://www.nokia.com/link?cid=EDITORIAL_5519 ), resulting files will hang phone's built-in music player. This did NOT use to happen with Winamp 5.24 + Transcoder v0.6 ( http://www.srcf.ucam.org/~wdhf2/tran...coder_v0.6.exe ) with exact same parameters (aacPlus Encoder @ 48kbps Parametric Stereo, MPEG-4 bitstream). There is sample of Winamp v5.32 built-in transcoder/format converter output (new_encoder.aac) and sample of Winamp v5.24 + Transcoder v0.6 output (old_encoder.aac) in a zipfile at http://www.sendspace.com/file/w7cdbx

Files are, curiously, exact same byte length but something about the 5.32 produced content isn't quite compatible. No other files have ever made the E70 player hung so thoroughly that reboot of phone is required. Am I doomed to use Winamp 5.24 + Transcoder 0.6 forever? The E70 is fully HE-AAC v2 (Nokia's designation for HE-AAC v2 is eAAC+) compliant.

[nkapur]

Winamp 5.32 (no ext. plugins)
XP SP2
Activesync plugin v0.14
Portable: HP IPAQ 6925 w/ Windows Mobile 5

The portable works fine except for Playlist sync. Winamp displays the playlist when transferred to portable but no playlist file is created on the device.

To recreate: Righclick a playlist and send to Portable. The portable will show the new playlist but when ejected, the device has no playlists on it.

The feature works fine with my IPOD (which of course uses a different plugin).

[scorpicon]

I currently use Easy CD-DA Extractor 10.0.3 (Build 1) to rip my CDs and embed a JPEG album art image saved from Photoshop CS2.

I used to rip my CD's and then add the art using MediaMonkey 2.5, once again, with JPEGs from Photoshop CS2.

Recently I have started to use Winamp 5.32's ReplayGain scanner on all of my music collection. Mp3's with art added by MediaMonkey get the ReplayGain data added perfectly. Those with art from Easy CD-DA Extractor get their JPEG's corrupted.

I'm not sure why the plugin Nullsoft Replay Gain Analyzer v1.01 [ml_rg.dll] is touching the images, but it's happening nonetheless.

I'm running:
Windows XP Professional SP2
IE7
DX9c
Pentium 4 2.53GHz
Creative Audigy 1
ATi Radeon X1600 Pro

[DJ Egg]

@ scorpicon

Please upload a sample file to sendspace.com and post the link here. The sample file should be an mp3 with an embedded jpeg added by ECD-DAE before it has been rg scanned by Winamp.

[scorpicon]

Originally posted by DJ Egg
@ scorpicon

Please upload a sample file to sendspace.com and post the link here. The sample file should be an mp3 with an embedded jpeg added by ECD-DAE before it has been rg scanned by Winamp.

Here's the link:
http://www.sendspace.com/file/rk4qh1
I made a copy of the folder, scanned the one in my library to ensure I would get the same results, tested it (yeah, it got corrupted), and uploaded the original copy I made.

In short, the link is a clean mp3 that should get the embedded art corrupted when you scan it using Winamp's included ReplayGain analyzer.

To the best of my knowledge, nothing else gets broken after the scan, just the art. But I'm not positive.

Also, I'm using the save as album button, rather than the save as track button for the analyzer.

[DJ Egg]

@ scorpicon

I can't see the album art for that file in any player, including wmp.
Therefore we can't even attempt to reproduce the problem.

mirror

[DrO]

aye, same here as well, nothing shows in the tag editor for any album art in wmp (file was never touched by winamp in anyway after downloading)

-daz

[scorpicon]

Originally posted by DJ Egg
@ scorpicon

I can't see the album art for that file in any player, including wmp.
Therefore we can't even attempt to reproduce the problem.

Really weird. It doesn't show up for me in WMP, but it shows up using:
Abander MP3 Image Extractor
Cover and TAG plugin for Winamp 5
MediaMonkey

I'm able to extract it using:
MP3 Image Tag Extractor

I've been continuing to do some further research. I think Easy CD-DA Extractor is using ID3V2.4, which I could imagine is storing things differently, and might cause the analyzer to incorrectly rewrite the data. I would be curious if this is correct, and if Winamp will eventually support this.

[Edit --> DJ Egg]
Aah, ok, I don't have any of those apps installed. We'll look into this further...

[Edit #2]
Hmm, it looks like an ID3v2.4 syncsafe bug in Easy CDDA Extractor, similar to the one in iTunes.
We might be able to implement some workaround... but don't hold your breath :/

[scorpicon]

[Edit #2]
Hmm, it looks like an ID3v2.4 syncsafe bug in Easy CDDA Extractor, similar to the one in iTunes.
We might be able to implement some workaround... but don't hold your breath :/

What are the details of this bug? Is it a matter of Easy CDDA Extractor making an error (I'd be happy to report this to it's author)? If there's some good further reading on the nature of this issue, could you point me towards it?

Thanks for the help. I won't hold my breath for a workaround, but it is nice to know you guys are doing your best to at least listen and give some feedback. And if you do end up providing a workaround, all the better!

[Benski]

Originally posted by scorpicon
What are the details of this bug? Is it a matter of Easy CDDA Extractor making an error (I'd be happy to report this to it's author)? If there's some good further reading on the nature of this issue, could you point me towards it?

Thanks for the help. I won't hold my breath for a workaround, but it is nice to know you guys are doing your best to at least listen and give some feedback. And if you do end up providing a workaround, all the better!

See Section 4 ("ID3v2 frame overview") of this document - http://id3.org/id3v2.4.0-structure

Of note is that it specifies that the frame sizes are "stored as a 32 bit synchsafe integer" (information on what that means can be found in section 6.2.

In id3v2.3, frame sizes were not syncsafe. This is a 2.4-specific change. For a full list of changes, see http://id3.org/id3v2.4.0-changes. Of particular note is the line "To avoid false synchronisations in the frame header the size description and flag field has been rewritten [S:4]"

iTunes behaves badly in the same regard, so it is not surprising that other software chooses to ignore the standard.

[Courtarro]

I'm running Windows XP SP2, 2.8Ghz Pentium 4 with an ATI Radeon 9800. I'm running Winamp 5.32.

When I play a song with 8-bit characters in the title or force the modern skin to show a TTF font (rather than the bitmap font), Winamp uses quite a lot of CPU (20% on my machine). When I revert to a 7-bit song with bitmap fonts enabled, Winamp CPU usage drops back down to practically nothing. It's quite easy to reproduce, and Process Explorer reveals that gen_ff.dll is often on the stack, which suggests that it's the Modern Skin engine causing the slowdown.

Edit: I should mention that I'm using the default Winamp modern skin.

[Benski]

Originally posted by Courtarro
I'm running Windows XP SP2, 2.8Ghz Pentium 4 with an ATI Radeon 9800. I'm running Winamp 5.32.

When I play a song with 8-bit characters in the title or force the modern skin to show a TTF font (rather than the bitmap font), Winamp uses quite a lot of CPU (20% on my machine). When I revert to a 7-bit song with bitmap fonts enabled, Winamp CPU usage drops back down to practically nothing. It's quite easy to reproduce, and Process Explorer reveals that gen_ff.dll is often on the stack, which suggests that it's the Modern Skin engine causing the slowdown.

Edit: I should mention that I'm using the default Winamp modern skin.

Yes. If you can deal with the slightly different looking fonts, go to Prefs->Modern Skins->Font Rendering and disable "Freetype internal TTF rendering". When disable, the skinning engine uses Windows' font renderer which is significantly faster.

[xzxzzx]

Just to add my two cents, I've encountered the same Unicode-causing-too-much-CPU-usage bug...

It was more than a little annoying, because it took me several minutes to kill Winamp (as I run it "High" CPU priority, (and taking 90%+ CPU time on a 3.6 GHz Celeron), as running at "normal" is too low, but there is no option for "above normal").

[ColdCold]

I'm not sure if this is a core product feature or part of some vis plugin... but I have to agree with xxeonn's post (1-1-2007), there is something wrong with the spectrum analyzer. the cut-off of 310Hz closely matches my observations. the SA display is messed-up.

please see my posts here:

http://forums.winamp.com/showthread....hreadid=247437

- the Coldest