http://foobar2000.hydrogenaudio.org/0.6_beta26.zip (http://foobar2000.hydrogenaudio.org/0.6_beta26.zip)
http://foobar2000.hydrogenaudio.org/0.6_beta24_sdk.zip (http://foobar2000.hydrogenaudio.org/0.6_beta24_sdk.zip)
Please report only actual bugs in this thread.
Any useless posts [span style='font-size:7pt;line-height:100%'](eg. "This Version Is Slow", "This Version Has No Bass", "My Seekbar Plugin Doesn't Work", "Shuffle Doesn't Work", "Zip Compression Sucks", "Kernel Streaming BSODs My WinXP", etc)[/span] will be deleted.
[edit] beta26 posted, pre-b11 SDKs do not work
Seems fine so far.
Is the SDK final, or will we be seeing another 0.6 SDK when foobar 0.6 final is released?
Now I've got to go recompile foo_shizzle for 0.6.
yes, SDK is final (unless bugs show up) and it is intended to work with at least all 0.6x releases.
Good to know.
On a completely unrelated note, foo_shizzle doesn't work at all under 0.6. I'll have to figure out what's going on.
you most likely need to configure database, it's disabled by default now (only caches files being in use).
Yep. I noticed that right after I posted. It works now.
By the way, should shortcut_actions::get_name() just return the name of the component? Also, is it acceptable to return NULL if the component doesn't need its own group? (i.e. Meta Search \ Meta Search is sort of overkill, but then, I don't know how you're planning to implement this, so maybe not.) I know it's not being used yet, but I figure I should plan ahead.
return null if you don't want your own group or whatever-group-name-you-want if you want one.
Thanks.
I seem to have found a bug. Meta Search registers a database callback because it keeps database-related information in memory. However, the on_new() method is not being called when I add new files. (this is after nuking and restarting, of course). The on_modify() method is being called, however.
P.S. I don't know if 0.5 had this behavior or not. I never thought to check it. (My function was tested by modifying files. My on_modify calls on_new.)
Also, I don't know if this is relevant information or not, but my callback is not being registered on_init(). It's registered the first time the search window is displayed.
On Win95, there are no menu buttons. Also, the shuffle and repeat menu items do not do anything at all. This behavior actually appeared in an earlier version... 0.53, maybe? Will you be addressing this?
Thanks.
I seem to have found a bug. Meta Search registers a database callback because it keeps database-related information in memory. However, the on_new() method is not being called when I add new files. (this is after nuking and restarting, of course). The on_modify() method is being called, however.
P.S. I don't know if 0.5 had this behavior or not. I never thought to check it. (My function was tested by modifying files. My on_modify calls on_new.)
Also, I don't know if this is relevant information or not, but my callback is not being registered on_init(). It's registered the first time the search window is displayed.
obsolete callbacks got dropped, read metadb.h
On Win95, there are no menu buttons. Also, the shuffle and repeat menu items do not do anything at all. This behavior actually appeared in an earlier version... 0.53, maybe? Will you be addressing this?
1. win95 is buggy, if you still use it, you get what you're asking for
2. can't reproduce, yet another problem on your side
obsolete callbacks got dropped, read metadb.h
Oh. So is there no longer any way to get notified when a new entry is added to the database?
Hmm. I'll look at playlist_loader_callback. Maybe That'll do what I want.
1st, 10x for the dropdowns (combo-boxes) everywhere
one bug though, i cant edit in them more then their length. meaning, unlike the previous textboxes which allowed to overflow the size, the new dropdowns don't.
1. win95 is buggy, if you still use it, you get what you're asking for
Some people don't have the option of upgrading. (Think "work computer", "test bed", etc.) The odd thing is that there's a lot of apps that *do* work on Win95.
2. can't reproduce, yet another problem on your side
... because you're not testing on Win95? That's a "duh" response. I know that others
can reproduce this; I've seen mention of it before. It's a Win95 thing (first clue) and popped up as a regression circa v0.53 (second clue... I can tell you the exact version number if you have an archive available). You know what code you've changed; it shouldn't be too hard to pinpoint.
And if you still want to brush people off, it could be done with a little more tact, ya know. "Win95 is not a supported platform. This behavior cannot be reproduced on any platform that
is supported and so has been classified as a fault in Win95." Ahhhhh..... customer service......
obsolete callbacks got dropped, read metadb.h
Oh. So is there no longer any way to get notified when a new entry is added to the database?
Hmm. I'll look at playlist_loader_callback. Maybe That'll do what I want.
you should use void metadb::enum_entries(metadb_enum_callback * cb) to track contents of database, like albumlist does
1. win95 is buggy, if you still use it, you get what you're asking for
Some people don't have the option of upgrading. (Think "work computer", "test bed", etc.) The odd thing is that there's a lot of apps that *do* work on Win95.
2. can't reproduce, yet another problem on your side
... because you're not testing on Win95? That's a "duh" response. I know that others can reproduce this; I've seen mention of it before. It's a Win95 thing (first clue) and popped up as a regression circa v0.53 (second clue... I can tell you the exact version number if you have an archive available). You know what code you've changed; it shouldn't be too hard to pinpoint.
And if you still want to brush people off, it could be done with a little more tact, ya know. "Win95 is not a supported platform. This behavior cannot be reproduced on any platform that is supported and so has been classified as a fault in Win95." Ahhhhh..... customer service......
then use programs that were written in win95 age and tested on win95, instead of whining about software you don't even pay for.
retard.
one bug though, i cant edit in them more then their length. meaning, unlike the previous textboxes which allowed to overflow the size, the new dropdowns don't.
fixed
you should use void metadb::enum_entries(metadb_enum_callback * cb) to track contents of database, like albumlist does
Well, the point of doing it the other way was to avoid the overhead of interfacing with the database for searches.
The "Stardard Search" option of the plugin maintains a list of all meta-info in memory. That way, it can be sorted and binary-searched. (though the binary search actually turned on in the released version, due to a bug)
In addition, keeping a list of entries allows me to keep associated pre-formatted strings, which avoids the overhead of creating strings for display. (Just extended playlist_entry and added a new field)
I guess it's your call, but why is the metadb_callback included in the API at all? It doesn't seem very useful if it only reports when entries are modified or movied, and never when they are added.
P.S. The plugin does use a metadb_enum_callback to get the original information.
as you have probably noticed, metadb_callback is for getting notified about files being edited or moved, which is very important eg. in playlist.
keeping a copy of database produces really big memory overhead on your side, i don't see any serious reason to do that.
I suppose it could be important for things like the playlist. It still seems like a gaping hole in the functionality, but as I said, that's your call.
And yes, keeping that copy of the database in memory does add a pretty big memory overhead. But it also adds a significant speed increase. That's why it's just one of the three search options.
I guess I'll be changing some things up for the 0.6 release of Meta Search. (i.e. No more call back. Build list on dialog display. That means that the dialog will have to be closed and re-openned to reflect database changes, but I guess that's the way it's going to have to be.)
whats wrong about adding "refresh" button like the one in albumlist ?
That works, too. I just liked the fact that it was automatic.
It's not like it actually affects me in the least. I add new songs maybe once every month. The only reason that I even noticed that the callback wasn't working properly (or rather, the way I expected) is because I had to rebuild the database after reinstalling. I just like the thought of it being automatic. Plus now I'm going to have to squeeze a 6th button in there.
I do like the bigger-by-default albumlist.
$blend(a,b,c,d) - generates a color in c/d between colors a and b
There goes my fading script !
Seriously though, it seems to only have three parameters and implements d as the total time itself.
e.g. im using
$blend($rgb(152,96,150),$rgb(212,146,10),%_time_elapsed_seconds%)
which works
from the description i thought it would be
$blend($rgb(152,96,150),$rgb(212,146,10),%_time_elapsed_seconds%,%_time_total_seconds%)
but that doesnt work, returns nothing.
edit: also tray max/restore controls are still broken..
Using Win2k & rar achives.
Does not get stored in database.
Folders in rar file show up as tracks.
beta2 posted, see links in first post
installer (http://foobar2000.hydrogenaudio.org/foobar_0.6b26.exe) for beta26.
foo_mp4 (http://www.saunalahti.fi/cse/foobar2000/foo_mp4_0.6.zip), foo_speex (http://www.saunalahti.fi/cse/foobar2000/foo_speex_0.6.exe), foo_mpcenc (http://www.saunalahti.fi/cse/foobar2000/foo_mpcenc.zip), foo_vorbisenc (http://www.saunalahti.fi/cse/foobar2000/foo_vorbisenc.zip), foo_clienc (http://www.saunalahti.fi/cse/foobar2000/foo_clienc.zip).
All components compiled for beta11+.
Is it intentional that foobar does not check anymore if added (dropped) file can be played? I can add directory full of jpegs to playlist...
Is it intentional that foobar does not check anymore if added (dropped) file can be played? I can add directory full of jpegs to playlist...
preferences / core / restrict incoming files
preferences / core / restrict incoming files
Yes I know the option, but I find it quite inconvinient - to enable _all_ formats foobar can play it requires some folk knowlegde about all extensions all plugins accept (not mentioning it requires reconfiguration each time new plugin shows up). May I suggest exclusion list instead (Do not enqueue files matching: *.jpeg,...)?
edit: ok, sorry it's more new function request than bug report..
the link to beta2 is still listed as beta1, i.e. the link is http://foobar2000.hydrogenaudio.org/0.6_beta.zip (http://foobar2000.hydrogenaudio.org/0.6_beta.zip) instead of http://foobar2000.hydrogenaudio.org/0.6_beta2.zip (http://foobar2000.hydrogenaudio.org/0.6_beta2.zip)
@case: how to enable mppenc ?
Requires mppenc.exe in path or in foobar's dir, it's somewhat buggy as I don't seem to get older than 1.15r to work.
Nice work zZzZzZz! Useful new features and stable/fast as always.
Already replaced the stable version for me.
dev0
[delete me]
@case: how to enable mppenc ?
Requires mppenc.exe in path or in foobar's dir, it's somewhat buggy as I don't seem to get older than 1.15r to work.
tnx, i missed that pulldown in diskwriter actualy, working fine with 1.15r on wint, i tested some mp3->mpc just for fun and the desired file got strange rg value (+64 db )
edit: cant reproduce now, everything else got transcoded fine (without any rg tags), i guess that was lonely problem.
[/delete me]
I'm stuck with WinNT4.0, I'm afraid (work PC, hoping for Win2k soon!) but:
Testing beta 2: [edit: this was actually beta1 - sorry!]
1. Display / "Show buttons" still not working (like Win95, I'm not expecting a fix, and happily use keyboard shortcuts or menus)
[edit: this works perfectly on 0.6 beta 5 - thanks!]
2. Unlike v0.586, the Playlist/Add Directory... adds non-audio files (which then error, of course) such as .html .jpg .png .doc .exe .txt Seems like I'd have to manually restrict the file formats added (in Core) to get this version working like the old one.
[edit] ah, while I was playing & occasionally, typing I see this was reported. Feel free to delete parts of this post or delete the whole thing [/edit]
[edit: This is fixed in 0.6b5 - thankyou!]
3. foo_dsp_crossfeed didn't load because it's built with an outdated version of the SDK and will need recompiling (not a bug in fb2k). Suddenly noticed why I so like crossfeed when I shuffled into a 1965 track with total stereo separation while wearing headphones.
On the subject of Crossfeed, it's probably a minority interest, but is it feasible for fb2k's architecture to support a custom APEv2 tag to indicate "binaural recording" (which requires headphones and no crossfeed for the proper effect) or to indicate "crossfeed defeat", and then turn crossfeed off just during those tracks where it should be defeated (or leave it off it's already off, such as when playing over loudspeakers). I guess it's the message passing from the tag reader to the DSP plugin and the synchronisation with the audio stream that might be problematic, but perhaps this is easier than I think?
I love fb2k! Many thanks to all who have been putting all the tools for my listening pleasure in one great package. And being able to replaygain, crossfeed and equalize audio CDs on a PC player is awesome too.
Best regards,
Dick Darlington
Slight display bug in the Display Config. The "max" label on the transparency slider is placed a bit to high, and it sometimes doesn't repaint very prettily.
http://gelaed.com/images/display_config.png (http://gelaed.com/images/display_config.png)
I get an illegal operation whenever I try to load large amounts of files at once or open my large playlist from fb .58c. The standard input and output are the only components I have loaded and I'm using entirely default options.
It appears to crash at approximately the same place each time but not always. (i.e. mostly the same folder, sometimes the same file)
.6 beta 2
win98se
stripe
Music from 'Psycho' kills fb2k 0.6 beta 2.
It's not my rip, just an MP3 I downloaded. Plays fine in WinAMP and WMP. The only odd thing I can see about it is that Tag reports:
Details: 44100 Hz Stereo, 4290576851 kbps
Crashes fb2k when I attempt to play it or add it to the playlist. Probably some weird old encoder. When I decoded to WAV and then re-encoded with LAME, fb2k plays it fine.
Just FYI...
It appears to crash at approximately the same place each time but not always. (i.e. mostly the same folder, sometimes the same file)
I had the same thing happening with my large playlist, that's how I discovered that it was the music from 'Psycho' causing the crash. Try loading directory by directory until you find the one causing it, then narrow it down to a single file. I'll bet you'll find some weirdness about that file.
In my case, I reencoded the thing, and then all was well.
I have a minor problem with 0.6 beta2 on Win98SE: When I remove entries from the playlist, the playlist is not properly repainted. It still shows the removed entries, until fb2k is forced to repaint them.
Did anyone notice this on WinNT/2K?
EDIT: fixed in beta3, thanks Peter!
[delete me]
@dickd: #1 check if it is the correct version, i have buttons under winnt4 (should show '..v0.6 beta 2')
[/delete me]
when i add folder to the playlist, it adds everything in the dir, playlists, images, etc.
2ndly when i click on follow cursor, i have to click it twice before it selects.
The statusbar flickers when %_time_elapsed% or %_time_remaining% is used in it (Win2k).
The statusbar flickers when %_time_elapsed% or %_time_remaining% is used in it (Win2k).
Always has on my system XP.
matrix, read some of the other posts (look for the ones by messer).
beta3 posted
i'm looking for some samples of those problematic mp3 files
i'm looking for some samples of those problematic mp3 files
I saved copies of the ones that were crashing on me. The smallest is 3MB. I can send it where ever you would like.
EDIT: still crashes with beta 3
Minor interface quirk:
if both "selection total time display" and "volume display" are enabled there's enough space to hold "total time" less then 24h (which should be enough for most playlists), but with "volume display" hidden, "total time" cannot even hold 1:00:00
Few little things:
Foobar overwrote my Diskwriter Output file name formatting string.
Foobar overwrote my old database! Lots of custom metadata lost.
Balloon misspelled in System tray configuration. (At least as far as I know, according to my Canadian dictionary.)
As a little side-bit: What was the equivalent "Desired volume level" for 0.5x's Replaygain?
@Matrix: Can't reproduce the follow cursor bug. OS?
@DickD: Metadata-specified DSP settings would be interesting to play with. Good idea, although I don't know the feasibility of it.
The access key (http://www.hydrogenaudio.org/forums/index.php?act=ST&f=24&t=7258) conflict still remains, but now it's conflicting with Maximize instead. Maybe use 'e' instead of 'x'?
Other than that, seems pretty good.
As a little side-bit: What was the equivalent "Desired volume level" for 0.5x's Replaygain?
There wasn't
Canar, I believe that all of the meta-data should be written into the files, unless you're talking about CD replaygain data.
Canar, I believe that all of the meta-data should be written into the files, unless you're talking about CD replaygain data.
Not when it comes to CUE files.
Testing with fb2k 0.6b3, but probably applicable to previous versions...
When shuffle is enabled and the current song is stopped, selecting play causes the current song to change, rather than playing the current song.
IMHO, the shuffle shouldn't occur from stop when play is selected. Rather, the current song should be started and any subsequent plays would then trigger the shuffle.
There wasn't
I know the option didn't exist. I'm asking: what was the default used by 0.5x?
There wasn't
I know the option didn't exist. I'm asking: what was the default used by 0.5x?
I would guess 0.0db. And 0.0db = 89db? Sort of confused
Canar, I believe that all of the meta-data should be written into the files, unless you're talking about CD replaygain data.
Not when it comes to CUE files.
That's entirely possible.
$blend does not work (as noted before, this time in beta 3) as advertised. For example,
$blend(000000,FFFFFF,50,100)
does not return a gray color, but instead returns white. When 50 is changed to 0 (??), the function gives gray. Or am I misinterpreting the docs?
Just DL'd Beta 3, started it, then opened up the config dialog. And, well, don't you think it's kinda, well, huge? No offence, but having a config dialog the size of your screen is a little overkill? Eh, who knows, it might just be Win98SE playing games on me again. It's done that before. Just thought you should know though.
Update:
Ahh, just found another bug, the File Info dialog seems to be re-sizable, but doesn't resize correctly. What I mean is that the Tags list gets wider, but the File info list stays the same width, so it's a little cut off. Also, the close button is cut off as well. Height resizing works though.
Can anyone tell me what the purpose of 'Desired volume level' inside Replay-gain is?
I've set to some value like -0.1dB and then scan my mp3. But the replay gain value has no different from that with setting of 0dB.
Think of it as a "ReplayGain preamp" where 0.0dB = "±0.0dB (89dB)".
This is like preamp, but it only applies when ReplayGain data is present.
Think of it as a "ReplayGain preamp" where 0.0dB = "±0.0dB (89dB)".
This is like preamp, but it only applies when ReplayGain data is present.
Ah. So, if I'm understanding properly it does not affect the RG data as written to the file?
Thanks MTRH!
if the default RG value adjusts to 89.0 db, why does the slider allow up to 12db "preamp" onto that? shouldn't it be max 11 db to level it at 100? what am i missing?
Kernel Streaming didn't seem to work for me on 0.6b3. It gave me an error:
INFO (CORE) : opening file for playback : file://D:\mp3s\09-linkin_park_-_breaking_the_habit.mp3
ERROR (foo_out_ks) : KS output error: unsupported data format.
However, it worked fine for me in 0.586. Bug?
Kernel Streaming didn't seem to work for me on 0.6b3. It gave me an error:
INFO (CORE) : opening file for playback : file://D:\mp3s\09-linkin_park_-_breaking_the_habit.mp3
ERROR (foo_out_ks) : KS output error: unsupported data format.
However, it worked fine for me in 0.586. Bug?
you really should give us some more info than that :B
Kernel Streaming didn't seem to work for me on 0.6b3. It gave me an error:
INFO (CORE) : opening file for playback : file://D:\mp3s\09-linkin_park_-_breaking_the_habit.mp3
ERROR (foo_out_ks) : KS output error: unsupported data format.
However, it worked fine for me in 0.586. Bug?
you really should give us some more info than that :B
Okay, I just tried reinstalling it and it worked. must be something wrong I did previously. heh, anyway thanks loads.
[edited] I found out it was due to my Output Data Format settings that had the problem to arise.
albumlist doesn't work, click refresh don't see anything!
database enabled, nothing in albumlist box
albumlist doesn't work, click refresh don't see anything!
database enabled, nothing in albumlist box
uhm. what settings are you using? did you change anything from default? etc. etc.
albumlist doesn't work, click refresh don't see anything!
database enabled, nothing in albumlist box
uhm. what settings are you using? did you change anything from default? etc. etc.
she's a virgin install, just have "d:\my music" and * in the two database boxes.
there she goes!
just have d:\my music (no quotes) and the second box blank.
oh snap!
how do i get this thing working with multi-cd albums! (with dirs like 'artist - album\cd1' etc)
i want albumlist to show multi-cd albums as 'artist - album - cd1'!
i tried putting %_directoryname% in both boxes but that makes a 'cd1' album in albumlist that, when clicked, playlists ALL the mp3s in my mp3 dir which are under 'cd1' (or disc1/cd2/disc2 etc) subdirs!
anyone figure out the right uhm syntax/code/whathaveyou to do this?
oh and it would be kool if albumlist loaded with foobar!
%artist%[ - %album%][ - %disc% ]|%title%
That would work if you have all the tags correct. It shows it as "artist - album - cd1", and when expanded it shows songs.
beta4 posted, crash when opening mp3 file with xing header not containing frame count should be fixed now
if the default RG value adjusts to 89.0 db, why does the slider allow up to 12db "preamp" onto that? shouldn't it be max 11 db to level it at 100? what am i missing?
'fraid you are! dB = decibels , they are a measure of power ratio. They are not a percentage. They involve logarithms.
dBs = 10 log (Power 1 / Power 2)
here (http://www.sospubs.co.uk/sos/1994_articles/feb94/decibels.html) is an ok (not great) explanation of decibels:
Hi.
Can you change "reset all" button to "reset current page", and use shif+click this button to "reset all".
Thank you.
About the 'show balloon tooltip on song change'. Very nice option, but;
if you change the song again within the time the balloon is visable, the new balloon isn't visable until the old one dissapears. it would be nice if the new balloon comes on top.
(i'm using win2k sp2)
edit:
I second Pack27's suggestion about the reset button!
Hi.
Can you change "reset all" button to "reset current page", and use shif+click this button to "reset all".
Thank you.
no (reasons not worth explaining here).
About the 'show balloon tooltip on song change'. Very nice option, but;
if you change the song again within the time the balloon is visable, the new balloon isn't visable until the old one dissapears. it would be nice if the new balloon comes on top.
(i'm using win2k sp2)
I have the same problem with win2k sp2. However, it works fine with winxp sp1. The balloon is updated immediately when the song is changed with winxp.
[edit] Whoops, I'm using win2k sp3, not sp2.
I want to see 'File Size = xxxxx bytes' in File Info / Technical Info. I think, implementing this is one-minute work.
BTW, 0.6b4 eats 51% of my CPU (Celeron 500) when 0.586 - just 35%... I'm using Win2kSP3, Output is KS on AC97, and just EQ on DSP.
[EDIT]
Now i'm feeling a noticeable, annoying slowdowns even on DS or WaveOut... Maybe because a new 64bit core? Can be data processing bitdepth optional, e.g. 16/32/64 bits somewhere in options?
Seems like better for me is to stay on the 0.586 because of CPU usage. So sad that folks stops supporting plugins for 0.5x
[/EDIT]
I think this can be reported as a bug:
screenshot (http://koti.mbnet.fi/antsau/pic/db.png)
Is there something fucked up with the database? It does this all the time, and both (foo_shizzle and foo_dbsearch) show up like this, so I believe it's not a bug in either one of the plugins.
I haven't seen that happen before. Have you added those files and checked the paths? Perhaps you've just got some duplicate files. (Or perhaps you need to run the "remove dead entries" option.)
P.S. If those aren't duplicate files, then it probably is something with the database if foo_dbsearch is doing it. I'd be willing to accept that I might have a bug in my code, but foo_dbsearch is really clean and simple.
beta4 posted, crash when opening mp3 file with xing header not containing frame count should be fixed now
Confirmed, it now plays all of my problem files wthout crashing.
I want to see 'File Size = xxxxx bytes' in File Info / Technical Info. I think, implementing this is one-minute work.
BTW, 0.6b4 eats 51% of my CPU (Celeron 500) when 0.586 - just 35%... I'm using Win2kSP3, Output is KS on AC97, and just EQ on DSP.
[EDIT]
Now i'm feeling a noticeable, annoying slowdowns even on DS or WaveOut... Maybe because a new 64bit core? Can be data processing bitdepth optional, e.g. 16/32/64 bits somewhere in options?
Seems like better for me is to stay on the 0.586 because of CPU usage. So sad that folks stops supporting plugins for 0.5x
[/EDIT]
No slowdowns here, ~14% CPU here with EQ.
Using either Win2k or WinXP.
CPU is Celeron 466MHz
I haven't seen that happen before. Have you added those files and checked the paths? Perhaps you've just got some duplicate files. (Or perhaps you need to run the "remove dead entries" option.)
P.S. If those aren't duplicate files, then it probably is something with the database if foo_dbsearch is doing it. I'd be willing to accept that I might have a bug in my code, but foo_dbsearch is really clean and simple.
This happens to me all the time. I can't even remember anymore how many times I've removed dead entries and nuked the database. I've just added the files by adding my whole mp3-directory, and I'm 110% sure they're not duplicate files. It's weird thing, as when I've nuked the database and added the files, first the plugins tell everything works great, no duplicate entries. And then some time when I open up either one of the plugins, the duplicates are there. Weird.
Wierd. Like I said, if they are showing up with foo_dbsearch (and they aren't duplicates), it's probably a database bug.
Maybe Peter should look at a copy of your database to see what's going on.
Well here's a new shot (http://koti.mbnet.fi/antsau/pic/db2.png) (and as the playlist shows, I've only got 3061 songs). Actually right now both plugins show that there's 3656 entries.. It seems to grow everytime I run foobar
My database: grab it here (http://koti.mbnet.fi/antsau/database.rar)
Sounds like a bug.
edit: Err, "looks" like a bug.
Just noted that Album list works fine. Peter said that it is quite surely a bug in the plugins.
<DEATH> anza`: do you get this with album list or not ?
<anza`> let's see
<anza`> nope
<DEATH> then blame gay plugins
<DEATH> ah, and tell them that albumlist source is included with SDK
.
.
.
<DEATH> but im pretty sure those plugins do something wrong
<DEATH> such as rely on database items being sorted by filename
I'll look into it later. (Though I don't have this problem, and don't know what i should be looking for. :\)
And I already knew that the albumlist source is included.
With your database, I get multiple entries. I guess I at least have a starting point now.
Albumlist does not work fine. This is what happens when I add "broken" to the playlist from the albumlist:
http://gelaed.com/images/albumlist.png (http://gelaed.com/images/albumlist.png)
The albumlist only shows one instance, but when I add it, I get 7 instances in the playlist.
P.S. That's with anza's database. I have no trouble with my database.
because foobar create duplicates in database only when "restrict directory to :" is used.
because foobar create duplicates in database only when "restrict directory to :" is used.
Not true. Happens to me and I'm not restricting any directories.
Edit: Duplicate problems should be fixed in beta 5.
beta5 posted, duplicate entries in database fixed (@#! typo in sorting func)
beta5 posted, duplicate entries in database fixed (@#! typo in sorting func)
might it be a debug build? the files all seem to be significantly larger than beta4 (almost 2x). no mileage in it being debug for us since we can't trace it?
I'm getting a bug when loading mods (using kode's latest in_mod) that is similar to the one I used to get when loading mp3s (mp3s loading mp3s seems to be fixed). I don't know if the problem is with the plugin or fb itself, but it occurs in the exact same 'way' that it crashed when loading mp3s.
I get an illegal operation whenever I try to load large amounts of files at once or open my large playlist from fb .58c. The standard input and output are the only components I have loaded and I'm using entirely default options.
It appears to crash at approximately the same place each time but not always. (i.e. mostly the same folder, sometimes the same file)
i'm using win98se
this is with beta5
also, this isn't technically a bug, but it would be really nice if you add a /random command line switch.
stripe
$blend does not work (as noted before, this time in beta 3)
yeh. now (from b2) it takes (requires) the fourth parameter, but ignores it. i.e. you could enter 0 or 999 for it but the result is the same.
also it there a reason that the tray max/restore controls arent fixed? Its been there as long as I can remember. Click restore,maximise,restore etc. only by right clicking the tray icon . either straight away or after a few max/restores it will be greyed out. (im on xp, i hope thats not the reason)
edit: teenee: installer is UPX'd = half size files.
While testing foo_dbsearch, I noticed that case conversion functions in fb2k did not work on non-ASCII characters (like german umlauts or accented characters). I included the following in my on_init() function:
setlocale(LC_ALL, "");
The plugin now works, but the fb2k core (and therefore the titleformatting) still seems to use "C" locale. The MS runtime reference does not say that towlower depends on locale (or I've missed it), but apparently it does.
I'm running Win98SE.
RTFM: FB2K is all UTF-8 internally, use supplied string conversion routines from pfc/string.h
I'm finding that 0.6 takes a ridiculously long time to load large playlists - something which didn't happen in 0.5. A playlist of 7088 files takes about 30 seconds to load or open on 0.6 beta5 but only 2 seconds with 0.586.
OS: Windows XP
Activate the db
I just noticed a problem with the 'resume playback after restart' in 0.6b5. If the same song (rather, same file) occurs more than once in a playlist, fb2k will resume the song in the same spot, as desired, however it will resume in the first instance of the song in the playlist...
Example..
020 Blah - Some artist
342 Blah - Some artist
Song #343 will be playing. fb2k is closed at 0:56. When reopened, song 020 begins to play at 0:56.
Ok, this build seems to work fine, great!
Activate the db
If you mean check the box - 'Database enabled' - I have already done that. Please, somebody must be having the same problem with large playlists in 0.6???
Using WinNT4.0, fb2k v0.6 beta 5 (accidentally had beta 1 before, not beta 2 as I thought, D'oh! Previous post edited)
Thanks, zZzZz... This is great.
• Now have buttons working in NT
• no need to manually specify restriction to audio file types to avoid a playlist full of PNG, JPG and TXT files!
Also, I like the Album List's Double-click Sends to Playlist option, but this means you can't expand that part of the album tree without starting the album playing and deleting what's currently on the playlist (at least, that's how I interpret the GUI - no plus sign to expand it, so you have to double-click). I'd really like it if you could expand the tree in some other way, but I'll just turn off that option. It would also be good to undo changes to the playlist (such as overwriting it when you meant to add an album instead of sending it)
I'll keep testing new betas, but for normal use with headphones I'll stick with 0.586 until I can get a foo_dsp_crossfeed.dll that supports the new SDK.
Thanks for the few comments on the idea of using metadata to allow crossfeed to be defeated (or other DSPs, I guess)
I'm finding that 0.6 takes a ridiculously long time to load large playlists - something which didn't happen in 0.5. A playlist of 7088 files takes about 30 seconds to load or open on 0.6 beta5 but only 2 seconds with 0.586.
OS: Windows XP
I've got this one too. Not as bad as you, for me it takes "only" a few seconds to load up a playlist with ~3000 songs. And yes, the db is enabled, read the topic if you don't believe me
Also, I like the Album List's Double-click Sends to Playlist option, but this means you can't expand that part of the album tree without starting the album playing and deleting what's currently on the playlist (at least, that's how I interpret the GUI - no plus sign to expand it, so you have to double-click). I'd really like it if you could expand the tree in some other way, but I'll just turn off that option. It would also be good to undo changes to the playlist (such as overwriting it when you meant to add an album instead of sending it)
Use arrow/cursor keys: left expands selected node, right collapses it.
RTFM: FB2K is all UTF-8 internally, use supplied string conversion routines from pfc/string.h
Sorry to disappoint you, but I do use pfc/string.h, and case conversion does not work without the call to setlocale.
I also tried "$lower(Überraschung)" in my title formatting string in 0.6b5 and 0.586. This produced "Überraschung" instead of "überraschung" (btw. Überraschung means surprise in English). Again, this all applies to the ANSI version running on Win98SE.
and wath about KS on this v0.6 beta 5 ?
the librarie foo_out_ks.dll failed to load
now i can use only directsound ....i would like to use KS or ASIO......
and wath about KS on this v0.6 beta 5 ?
the librarie foo_out_ks.dll failed to load
now i can use only directsound ....i would like to use KS or ASIO......
Extract the files from "extra" folder to the same dir as foobar2000.exe
beta 5 is pretty good only a few problems
1. loading playlists takes forever
2 "add directory" takes ages to load
3. the options page is too big (IMHO)
4. how do i remove entries from "Restrict Directories to" in the datbase settings page???
I also tried "$lower(Überraschung)" in my title formatting string in 0.6b5 and 0.586. This produced "Überraschung" instead of "überraschung" (btw. Überraschung means surprise in English). Again, this all applies to the ANSI version running on Win98SE.
Same goes for Unicode version (on WinXP) & tolower_utf8() & add_string_utf8_tolower() etc..
and wath about KS on this v0.6 beta 5 ?
the librarie foo_out_ks.dll failed to load
now i can use only directsound ....i would like to use KS or ASIO......
Extract the files from "extra" folder to the same dir as foobar2000.exe
i have yet done it ...in the "extra" folder thereinst any KS dll ....and however i had extract every file from this folder ... :confused:
Sorry to disappoint you, but I do use pfc/string.h, and case conversion does not work without the call to setlocale.
I also tried "$lower(Überraschung)" in my title formatting string in 0.6b5 and 0.586. This produced "Überraschung" instead of "überraschung" (btw. Überraschung means surprise in English). Again, this all applies to the ANSI version running on Win98SE.
if you try to use international characters on win9x, you basically get what you are asking for.
also, apparently crt towlower() function doesn't handle unicode, i'll look for some workaround.
beta 5 is pretty good only a few problems
1. loading playlists takes forever
2 "add directory" takes ages to load
3. the options page is too big (IMHO)
4. how do i remove entries from "Restrict Directories to" in the datbase settings page???
1. Yes, yes, yes my problem exactly!
2. Not having that problem
3. Alright by me
4. ??? just hit the delete key or backspace...
beta 5 is pretty good only a few problems
1. loading playlists takes forever
2 "add directory" takes ages to load
3. the options page is too big (IMHO)
4. how do i remove entries from "Restrict Directories to" in the datbase settings page???
1. Yes, yes, yes my problem exactly!
2. Not having that problem
3. Alright by me
4. ??? just hit the delete key or backspace...
1,2 : R.T.F.changelog, you need to configure database
4. you don't "remove" entries, dropdown list just keeps history of things you typed before.
0.6 beta6 posted, with updated SDK, case-conversion functions should be fixed now (uses ugly hack to remove #define UNICODE dependency but seems to work fine so far)
@case:
either something wrong with flaccer or is there a hidden config somewhere?
default cli flac vs foobar flaccer:
26.883.404 We_Use_The_Pain.wav
18.467.407 We_Use_The_Pain_cmd.flac
24.913.775 We_Use_The_Pain_foobar.flac
I'll keep testing new betas, but for normal use with headphones I'll stick with 0.586 until I can get a foo_dsp_crossfeed.dll that supports the new SDK.
http://www.naivesoftware.com/files/exp/foo...p_crossfeed.dll (http://www.naivesoftware.com/files/exp/foo_dsp_crossfeed.dll)
I use english version win2000 and I want to see the unicode names of chinese songs in the status bar.
0.6beta can't display unicode characters in the status bar. But 0.586 can.
I'll keep testing new betas, but for normal use with headphones I'll stick with 0.586 until I can get a foo_dsp_crossfeed.dll that supports the new SDK.
http://www.naivesoftware.com/files/exp/foo...p_crossfeed.dll (http://www.naivesoftware.com/files/exp/foo_dsp_crossfeed.dll)
Thanks!
0.6 beta6 posted, with updated SDK, case-conversion functions should be fixed now (uses ugly hack to remove #define UNICODE dependency but seems to work fine so far)
$caps still doesn't work with unicode strings in playlist here.
BTW is it possible for $caps to 'ignore' non-letters (ehm.. say, brackets, slashes and punctuation)? If they are directly before word with no space between, it's not capitalized.
Thanks for the new crossfeed, Trelane!
That 1965 track, "Lies", sounds wonderful now it's outside my head!
The Albumlist is borked somehow. It's reporting that I have no entries in the database, but I have the database turned on, and it has 3526 entries, as Meta Search shows. My "Add All" plugin also finds those 3526 entries with no problem.
http://gelaed.com/images/albumlist2.png (http://gelaed.com/images/albumlist2.png)
move your database out of the foobar directory..
start foobar and let it create default empty db
close foobar and put back your full db
should work
The thing is, the database isn't empty. It's not a database problem, because my plugins (as well as fooshion's) can see the database info just fine. It's apparently an albumlist problem.
Nevermind. Apparently the albumlist really doesn't like it when the "tree hierarchy" string contains invalid data. I was missing a percent sign after one of the meta-names.
My bad.
beta 5 is pretty good only a few problems
1. loading playlists takes forever
2 "add directory" takes ages to load
3. the options page is too big (IMHO)
4. how do i remove entries from "Restrict Directories to" in the datbase settings page???
1. Yes, yes, yes my problem exactly!
2. Not having that problem
3. Alright by me
4. ??? just hit the delete key or backspace...
1,2 : R.T.F.changelog, you need to configure database
4. you don't "remove" entries, dropdown list just keeps history of things you typed before.
OK I have R.T.F.changelog. I have 'Database enabled'. 'Album list' works fine. 'Add directory' works fine. 'foo-dbsearch' & 'Meta Search' work fine (now surely none of these would function if I DIDN'T have 'Datadase enabled') BUT loading large playlists is slow. A playlist of 2800 files = 10 sec, 7000 files = >20 seconds and if I open 0.6 with a large playlist on board (i.e. 'playlist.fpl' = 2,302 KB), 0.6 takes 20-30 seconds to open. All these take < 2 secs with 0.586. If I doing something wrong I'd really love to know what it is??
(http://www.users.bigpond.com/c.dimartino/database.png)
Loading large playlists seems to be slower here, too, but I'm not certain, because I haven't reinstalled 0.586 to confirm.
When adding tons of mp3s into playlist i get this error. I also recently nuked the database, could be a problem with adding tons of files to db?
Microsoft Visual C++ Error
Program: <path to foobar2000.exe>
R6025
- pure virtual function call
Then you click ok, it closes Foobar.
I dunno what the hell this could mean.
WinXP Pro
Foobar 0.6 b6
First post on Hydrogen...
With v.6beta6, I'm having trouble using the resample/kernelstreaming combination with some very low-quality source files. I have no problems with "normal" MP3 or Vorbis files, but when I attempt to play a set of MP3s in 22050Hz Mono at 24kbit, the output when using KS+resampler is at extremely high speeds, as if the resampler was disabled. I've had to start Winamp for the first time in weeks.
The files in question: http://jameswong.com/ykproject/disc/standa...lone/index.html (http://jameswong.com/ykproject/disc/standalone/index.html)
I'm using a Hercules GameTheater XP (v4.10 drivers) under Windows 2k SP3 on an Athlon 1.1Ghz. I have FB2k set to Kernel Streaming, and the Resampler DSP to 48kHz (fast mode); this produces correct output with normal-quality MP3 and Vorbis files. With the sample files from that site above, the output is very very fast, about what you'd expect if you played a 22050Hz file at 48kHz; turning the Resampler off doesn't make a difference, so I'm guessing that the Resampler DSP doesn't work with these files for some reason.
I can play the files in FB2k normally if I use DirectSound or WaveOut.
i found a wierd one...
i just added a few albums to my collection, and i highlighted the whole playlist and selected to do track replaygain as i always do... and it acted like nothing needed it... but when i select the files that need it, it works...
in the past, i could just select the entire playlist and it would do any that needed it...
i'm using beta 6...
beta 5 is pretty good only a few problems
1. loading playlists takes forever
2 "add directory" takes ages to load
3. the options page is too big (IMHO)
4. how do i remove entries from "Restrict Directories to" in the datbase settings page???
1. Yes, yes, yes my problem exactly!
2. Not having that problem
3. Alright by me
4. ??? just hit the delete key or backspace...
1,2 : R.T.F.changelog, you need to configure database
4. you don't "remove" entries, dropdown list just keeps history of things you typed before.
OK I have R.T.F.changelog. I have 'Database enabled'. 'Album list' works fine. 'Add directory' works fine. 'foo-dbsearch' & 'Meta Search' work fine (now surely none of these would function if I DIDN'T have 'Datadase enabled') BUT loading large playlists is slow. A playlist of 2800 files = 10 sec, 7000 files = >20 seconds and if I open 0.6 with a large playlist on board (i.e. 'playlist.fpl' = 2,302 KB), 0.6 takes 20-30 seconds to open. All these take < 2 secs with 0.586. If I doing something wrong I'd really love to know what it is??
(http://www.users.bigpond.com/c.dimartino/database.png)
yeah same here..sorry peter...
I'm just wondering...
Isn't it because of the new playlist format ??
It now seems to contain tags of the files... Doesn't foobar2000 compare thoses tags to the database when you load this playlist ??
What happens when you try to load a big .pls playlist (old playlist format) ??
I'm just wondering...
Isn't it because of the new playlist format ??
It now seems to contain tags of the files... Doesn't foobar2000 compare thoses tags to the database when you load this playlist ??
What happens when you try to load a big .pls playlist (old playlist format) ??
ok..well m3u playlists load the same way they did in previous versions. but m3u8 playlists load the same way new playlist format
SLOWLY
how about RW/FF into adjacent tracks?
beta7 posted, playlist loading speed problems should be fixed now
Ah, much better.
just installed 0.6b7 (was using 0.586) .. when i restore fb from the system tray, the screen starts drawing very slowly ... don't know if it's been allready reported ...
Still getting an illegal operation crash when loading mod files in large amounts, much like the crash I was originally getting loading mp3s. Should I keep reporting this here since it's not technically included in the beta package?
loading my large m3u8 playlist (which includes lots mod files) take a long time and then crashes
(win98 se, beta7, in_mod compile from april 3)
stripe
Ah, much better.
Yep, playlist loading fixed here too.
just installed 0.6b7 (was using 0.586) .. when i restore fb from the system tray, the screen starts drawing very slowly ... don't know if it's been allready reported ...
Can't reproduce.
I know the play button behaviour has changed with shuffle, but because of this when you open a playlist from command line it will play the first track even when shuffle is on.. in this instance you cant press the random button instead.
Yeah I hear you, I have the same problem, its all good...
$caps and $caps2 in 0.6b7 (and earlier betas) skip the first word. That is, instead of "We Suck Young Blood", I get "we Suck Young Blood".
edit: To clarify, the original string is "we suck young blood"
I think the database still has some bugs. A friend just installed and added files, but neither Meta Search nor the albumlist was finding the files. Nuking the database fixed it, though. To the best of my memory:
A friend installed Meta Search. Then I realized that he had .586 so he installed the newest beta. He added a directory. I told him to turn on the database. He did so. Set the directory exclusion to the only folder he actually has mp3s in. Then he added that same folder to the playlist. Tried Meta Search; zero results. Zero results in the album list. Checked and database search was indeed enabled. Closed and restarted foobar. Still not working. However, adding the files again was quite fast, indicating that the database was operating, at least partially. Removed the directory "limiter". Closed again. Started. Re-added files. Still no dice. Nuked the database. Restarted. Added all the files again. Finally worked.
That's all the information I can recall. He's running XP pro, and this was with foobar6b7.
I think the database still has some bugs. A friend just installed and added files, but neither Meta Search nor the albumlist was finding the files. Nuking the database fixed it, though. To the best of my memory:
A friend installed Meta Search. Then I realized that he had .586 so he installed the newest beta. He added a directory. I told him to turn on the database. He did so. Set the directory exclusion to the only folder he actually has mp3s in. Then he added that same folder to the playlist. Tried Meta Search; zero results. Zero results in the album list. Checked and database search was indeed enabled. Closed and restarted foobar. Still not working. However, adding the files again was quite fast, indicating that the database was operating, at least partially. Removed the directory "limiter". Closed again. Started. Re-added files. Still no dice. Nuked the database. Restarted. Added all the files again. Finally worked.
That's all the information I can recall. He's running XP pro, and this was with foobar6b7.
uninteresting, cant reproduce.
uninteresting, cant reproduce.
well, i can.. same happening here... after deleting the database file completely then it worked....
I think the database still has some bugs. A friend just installed and added files, but neither Meta Search nor the albumlist was finding the files. Nuking the database fixed it, though. To the best of my memory:
A friend installed Meta Search. Then I realized that he had .586 so he installed the newest beta. He added a directory. I told him to turn on the database. He did so. Set the directory exclusion to the only folder he actually has mp3s in. Then he added that same folder to the playlist. Tried Meta Search; zero results. Zero results in the album list. Checked and database search was indeed enabled. Closed and restarted foobar. Still not working. However, adding the files again was quite fast, indicating that the database was operating, at least partially. Removed the directory "limiter". Closed again. Started. Re-added files. Still no dice. Nuked the database. Restarted. Added all the files again. Finally worked.
That's all the information I can recall. He's running XP pro, and this was with foobar6b7.
adding new entries working "faster" second time is usually caused by windows file caching.
this still sounds like some sort of user error to me, EVERY change of database settings needs restart to take effect.
I've got a problem a bit like George's friend. The add directory command just won't work well. I nuked database, restarted foobar and then I added my whole mp3 -directory with the add directory function. Kazaam! It only adds a few of the files. It always either reads infos from only a few files (but still adds every file to the pl) or then it freezes to some file. I have to always do a windows search for *.mp3 in the mp3-directory and then drag-and-drop the files to foobar, to get them work well. This has been happening since.. about beta4 or 5.
Maybe small bug in .6, still happening in b7.
With sort incoming files checked I get consistently out of sort tracks when sending to fb2k. Noticed this only starting with .6 series, worked OK with .5's.
Here's a minor glitch that I've noticed.
Start a song, then pause it.
Open album list window, select something and click "Send to playlist".
Instead of going to the first song in the new playlist, the song that was playing before will be unpaused.
Slow seeking in MP3 files playing via WLAN.
When I seek 1 minute forward in an MP3 file it takes about 10 secound before it continues to play. If I to the same in an flac or Vorbis file it plays instantly...
I just checked with version 0.58 and no problem there!
A Bug maybe?
Need more info?
0.6 Changelog
- new mp3 seeking code, slower but sample-exact
perhaps it's because of that.
0.6 Changelog
- new mp3 seeking code, slower but sample-exact
perhaps it's because of that.
Maybe it`s the new MP3 seeking code..
I just found out it plays instantly if I play MP3 files from my HD, but takes 10 secounds if I play from LAN.
Does not bother me much. I just wonder what it was, because everything went smoothly with version 0.58..
Thanks!
Maybe it`s the new MP3 seeking code..
I just found out it plays instantly if I play MP3 files from my HD, but takes 10 secounds if I play from LAN.
Does not bother me much. I just wonder what it was, because everything went smoothly with version 0.58..
Thanks!
Hmm... maybe the new seeking code performs more file seeks. HD data is cached by the OS, so latency is negligible. Network drives may not be cached.
Well, a guess anyway. Perhaps Peter could comment on this.
Baloon tooltips doesn't work on my WinXP prof polish version. I've tested almost all system settings and still nothing. It didn't work on betas from 5 to 7 on other not tested. I don;t know now if its fault of my sys (probably it is) and is there something I can do about it
It works here on Win XP, but then I am running the US version of it...
Edit: Yup, I'm an idiot. It was by design, and frankly, much more useful than how Windows handles it by default.
- Dane
ctrl+click to toggle selected or not
does foobar use WM_APPCOMMAND for multimedia keyboards?? im asuming it does, because foobar only uses them when foobar is the active app... if u change the WM_APPCOMMAND to HSHELL_APPCOMMAND, it will pick up the keyboard commands even if foobar isnt the active app... just a sugestion..
Bizzy
ctrl+click to toggle selected or not
I know, but that's not typical behavior. I didn't know that that was by design. My mistake.
Edit: Actually, you're right. That's a lot more useful than the default behavior. I guess I should edit my original post before I look like an idiot (too late, but still...)
does foobar use WM_APPCOMMAND for multimedia keyboards?? im asuming it does, because foobar only uses them when foobar is the active app... if u change the WM_APPCOMMAND to HSHELL_APPCOMMAND, it will pick up the keyboard commands even if foobar isnt the active app... just a sugestion..
Bizzy
With logitech keyboards, you can just set the multimedia keys as global hotkeys in foobar.
but not everyone uses logitech.
Well, I thought it was possible that you might, and that the information might therefore be useful. Also, it's likely that since it works with Logitech, it might work with some other keyboards as well. You might want to give it a try. Unistall any special keyboard software (because it'll catch those keys) and then open up the keyboard shortcuts window, and instead of pressing standard keys, just press the multimedia keys. It's worth a shot if you haven't tried it already.
Is it normal for the advanced limiter to be chewing up 16% of the CPU on a 550Mhz/512 megabyte XP Pro system? I thought only the resampler was that hungry.
Peter, you might want to give the metadb::file_removed() call a once over. I was trying to use it earlier, and I think I found a bug. If I call metadb::file_removed(), then database seems to remove the file properly. However, when I re-open foobar, the entries re-appear in the database, as though they were never removed.
edit: metadb::user_requested_remove() seems to be having the same problem.
Ok, I figured out what's going on. The entries are being deleted, but then they are being re-added to the database when foobar is re-openned, because they were still in the playlist. I guess it's your call if you want to leave that behavior, but it seems to me that playlist.fpl entries shouldn't be added back to the database. It's kind of bizarre to have the clear the playlist before removing dead entries.
Besides, I thought that the metadb_callback was supposed to be used by things like the playlist. Why isn't it removing those dead entries when they are deleted from the database?
First an MPCenc, now a FLACenc called "flaccer"
If only somebody could make an Oggenc diskwriter too!
First an MPCenc, now a FLACenc called "flaccer"
It'd be much cooler if it was called "FLACcid". That's just a cool word.
how is flaccid a cool word? What are you a lesbian or something?
If only somebody could make an Oggenc diskwriter too!
There is foo_vorbisenc source code in SDK, it's just not compiled
In foo_delete I'm using metadb::user_requested_remove() to clear dead entries associated with files user just erased.
When I recycle file and call user_requested_remove(), entry disappears from albumlist and other "database browser" plugins. But if I restore the file from recycle bin and add it back to playlist, none of those plugins show it (refresh, reload foobar - nothing helps).
When instead of restoring old file I'll make another one with the same path and name (but different metadata) and add it to playlist, foobar shows _old_ ("deleted") metadata instead of reloading with new one... Of course still it does not put it back into albumlist and similiar plugins.
I'm getting a bug when loading mods (using kode's latest in_mod) that is similar to the one I used to get when loading mp3s (mp3s loading mp3s seems to be fixed). I don't know if the problem is with the plugin or fb itself, but it occurs in the exact same 'way' that it crashed when loading mp3s.
I get an illegal operation whenever I try to load large amounts of files at once or open my large playlist from fb .58c. The standard input and output are the only components I have loaded and I'm using entirely default options.
It appears to crash at approximately the same place each time but not always. (i.e. mostly the same folder, sometimes the same file)
i'm using win98se
this is with beta5
also, this isn't technically a bug, but it would be really nice if you add a /random command line switch.
stripe
Hi, I uploaded a debug build here (http://www.cqasys.com/projects/kode54/foo_mod_debug.zip). You'll need a debug viewer like this one (http://www.sysinternals.com/ntw2k/freeware/debugview.shtml), though.
The purpose is to find which file(s) are causing bass.dll to crash. You will see a bunch of output like this:
[1824] file://D:\E\MUSIC\MODULES\reduz\FORSWA.IT
[1824] Loading
[1824] loaded
[1824] buffers away
[1824] setting info
[1824] name
[1824] length
for each file you add. The number will be the process ID of Foobar2000, and the paths will vary. When it hits whatever file is causing the crash, the debug info should cut off at "Loading" with the crash module path in the line preceding it. Verify that this file crashes when loaded alone, then rename it (blah.mod -> blah_mod.crash, or something that changes the extension) and proceed with adding your collection.
Once you have found all of the crashing files, please try to verify them with XMPlay, and if that works, try to feed the module(s) to writewav.exe from the BASS C example programs, or the masm example player.exe.
Please report all files which are confirmed to crash Bass.dll or XMPlay to Ian Luck at un4seen.com (http://www.un4seen.com), either by e-mail or through his forum. (PM to user
admin)
I wouldn't recommend that debug build for more than locating files which cause crashes, though. For one, all that extra debug info being sent to debug console. (Or nowhere, if you don't have a debug viewer or console connected.) Two, it keeps BASS.dll initialized for the entire session, so may be using more memory at all times, whereas normal build only initializes BASS when it needs to play a file or collect info.
Ok, it turns out that most of the files were corrupt, and apparently in .586 the reason they worked was because I hadn't loaded my playlist from scratch since switching to the bass-based in_mod.dll. The modplug-based dll (and modplug player as well) doesn't crash when loading them; instead they just play all screwed up.
According to the debug program, I got a crash when loading this (http://people.whitman.edu/~millerbc/files/pneumatic.mod) mod, but it seems to load ok when i try it again.
If you want the corrupt files (to try and make the .dll more durable or something like that) I'll be happy to put them online for you. None of them play using XMPlay (though some of them play fine in ModPlugPlayer), however, so I guess I'd probably be better off sending them to Ian Luck.
stripe/ben
how is flaccid a cool word? What are you a lesbian or something?
Better than Flaccer, IMHO. At least it's a real word. And I don't have any issues with "flaccid"... A little insecure, are we? Alternately, when that ugly/fat/undesirable chick walks by, it could be a good thing.
Alternately, you could go for FLACjacket or something.
Hmmm, FLACjacket. Might work. Still would love that oggenc to be compiled though. Case? John33? I know you're out there...
In foo_delete I'm using metadb::user_requested_remove() to clear dead entries associated with files user just erased.
FYI, user_requested_remove() is for telling database that user wants some files to be removed from it, NOT for notifying about files that got deleted, use entry_dead() or file_removed() instead (though their implementation in beta7 is probably the same).
FYI, user_requested_remove() is for telling database that user wants some files to be removed from it, NOT for notifying about files that got deleted, use entry_dead() or file_removed() instead
Ok, thanks. I'm changing it then.
beta8 posted, some types of services compiled with previous SDK no longer work because of API changes, new SDK coming later today or tomorrow
Peter, the link in the first post points to:
http://foobar2000.hydrogenaudio.org/0.6_beta7.zip (http://foobar2000.hydrogenaudio.org/0.6_beta7.zip)
Shouldn't is point to 0.6_beta8.zip?
Thanks
btw $fix_eol() is documented in title formatting help but it isnt a function (results in "[UNKNOWN FUNCTION]" ).
Balloon tip doesn't work anymore with beta8. Worked in beta7. Already tried with a clean foobar2000.cfg, doesn't work...
(WinXP Pro SP-1 NL)
I have a problem, when i play .cue files. Foobar2000 Beta8 (and all the other beta's) allmost frezes. I don't know why.
Can anyone look in to my problem.
// Lars
beta9 posted
beta9 posted
Balloon tip fixed, thnx.
beta9 > baloon tooltips still not workin on my winXP pl + sp1 but probably it's my system fault becouse it didnt work on any of betas
beta9 > baloon tooltips still not workin on my winXP pl + sp1 but probably it's my system fault becouse it didnt work on any of betas
check reg key:
HKCU\Software\Microsoft\Windows\CurrentVersion\Explorer\Advanced\EnableBalloonTips
edit: also preview in title formatting seems to be broken in beta9.
and i get ak's replaygain edit crash too.
There seems to be a minor glitch with per-track rg.
After reloading info from file (or restarting foobar), album_peak value appears (equal to track_peak).
Happens with mpc's, replaygain.exe shows that files have only track rg values.
BTW trying to 'edit rg info' crashes beta8 and 9 here.
check reg key:
HKCU\Software\Microsoft\Windows\CurrentVersion\Explorer\Advanced\EnableBalloonTips
edit: also preview in title formatting seems to be broken in beta9.
and i get ak's replaygain edit crash too.
thanks helped
There seems to be a minor glitch with per-track rg.
After reloading info from file (or restarting foobar), album_peak value appears (equal to track_peak).
Happens with mpc's, replaygain.exe shows that files have only track rg values.
BTW trying to 'edit rg info' crashes beta8 and 9 here.
1) cant reproduce, is it mpc-specific perhaps ? what exactly did you do ?
2) fixed, thanks, expect beta10 soon
edit: also preview in title formatting seems to be broken in beta9.
can't reproduce
are the rg values still stored ok (nd just being showed incorrectly), or are we going to have to re-rg all the values since the beta introduction? or am i not understanding the problem/
can't reproduce
yeh i worked it out. i started foobar with an empty playlist (not something i usually do) and so the preview showed nothing..
cant reproduce, is it mpc-specific perhaps ? what exactly did you do ?
I scanned mpc's per-track, then open file info, hit 'reload info from file', album peak value appears.
Scrnshot (http://www.hot.ee/nafigator/pics/rg.png)
It just seems to repeat track peak value, while rg.exe shows that there are no album_peak values stored in those mpc's.
cant reproduce, is it mpc-specific perhaps ? what exactly did you do ?
I scanned mpc's per-track, then open file info, hit 'reload info from file', album peak value appears.
Scrnshot (http://www.hot.ee/nafigator/pics/rg.png)
It just seems to repeat track peak value, while rg.exe shows that there are no album_peak values stored in those mpc's.
ok, i've found it (it was mpc-specific after all), thanks.
actually it was a minor problem with reading RG info from mpc files (using track gain for album gain when album gain is not present), your files most likely don't contain any invalid info.
beta10 really soon.
There seems to be something new in regards to the way 0.6 beta handles the clearing playlist when adding files from command line. When the option is disabled, I exit fb with a loaded playlist. If I start it using the play key on my logitech keyboard, it starts with no playlist, and plays nothing. In 0.586, it loaded the playlist that it quit with, and started playing the first entry.
However, if I disable the option, it works as before, but now trying to play something from explorer adds it to the playlist, instead of replacing it (obviously, but this was not the case in 0.568). Can anyone reproduce?
nevermind, I'm dumb.
Woohoo! Vorbisenc now available! Thanks Case!
(Updated MPCenc to work with older encoders too)
http://www.saunalahti.fi/cse/foobar2000/?M=A (http://www.saunalahti.fi/cse/foobar2000/?M=A)
When "show buttons" is turned on, you get an extra "button" thing on the title bar. Here is a screenshot
(http://www.worldwide-fx.com/scott/foobartitle.jpg)
When "show buttons" is turned on, you get an extra "button" thing on the title bar. Here is a screenshot
(http://www.worldwide-fx.com/scott/foobartitle.jpg)
This is just a spacer, UTFS
What's "UTFS"? Also, maybe it was there, but I never noticed this in earlier versions. I just upgraded from version 6 beta 7.
Hi all,
maybe there is a small problem with using $transition(a,b,c) as it seems it doesn't like another color in a, resulting in a weird output of, ahem, color codes
sample code for the status bar:
$transition($progress(%_time_elapsed_seconds%,%_time_total_seconds%,15,A64B42#,'='),80FFFF,679D4A)
giving something like
7FFBF77EF7EF7DF2E77CEEDF7BE9D67AE5CE79E0C6#==============
thx
Peter, I found a limitation in the commandline specification. Please have a look at it. More info on it in this topic (http://www.hydrogenaudio.org/forums/index.php?act=ST&f=29&t=8195&st=0&#entry81416) and here on Total Commander forum (http://ghisler.ch/board/viewtopic.php?p=4978#4978)...
editLimitation was/is within the tool I used to parse the files into fb2k (tool called lst2str). Sorry for bothering you with this one...
What's "UTFS"? Also, maybe it was there, but I never noticed this in earlier versions. I just upgraded from version 6 beta 7.
UTFS == Use The F*cking Search
Another thing:
I want to play a folder in fb2k and use the following commandline:
foobar2000.exe E:\MUSIC\SYSTEM~1\STEALT~1
The problem that occurs is that the incoming files are not sorted properly ('sort incoming files' enabled).
after foo_mpcenc and foo_vorbisenc when foo_lamenc ???
(foo_mpcenc and foo_vorbisenc work fine here)
Volume control indicator in status bar does not change when the volume is changed from my keyboard (Dell). It did before. I can change the volume indicator value by double clilcking the "volume" and then using the slider. But, say I lower the volume to -24.60 db I can then raise it from the keyboard volume control and the indicated value does not change.
Anyone else have this happening?
foobar 0.6 up to beta 10
Bug with GUI speeddowns (I wrote about it in support forum) still exist.
Only minor font glitch (using b10):
Playlist\Add location - textbox font is MS Sans Serif instead of Tahoma.
Kernel streaming does not seem to work with my onboard sound (mb = asus a7n8x, nforce2 chipset)
maybe it's me, maybe it is indeed a bug.
i won't rule out neither
Kernel streaming does not seem to work with my onboard sound (mb = asus a7n8x, nforce2 chipset)
maybe it's me, maybe it is indeed a bug.
i won't rule out neither
Can you be more specific. What exactly happens? You should note that KS is still experimental and some soundcard drivers are buggy or otherwise don't support it.
oh yeah, more info coming up!
INFO (CORE) : startup time: 94 ms
INFO (CORE) : opening file for playback : file://F:\-------_M_P_3_--------\bla\bla.mp3
ERROR (foo_out_ks) : KS output error: unsupported data format.
and no sound of course
i'm guessing it's the drivers .... ?
reports on KS not working is highly uninteresting just as stated zillion times.
$caps() doesn't work properly with certain polish characters, for example: (http://tomkooo.ds.pg.gda.pl/foobar1.png)
i'm using win2000pro and foobar0.6 beta10
it appears that these characters are not considered as a letters.
Foobar 0.6 Beta 10
FPL Playlist Bug:
1) Disable database (and nuke it), restart foobar2000
2) Open a RAR archive for playback
3) Save an fpl playlist
4) Clear playlist and restart foobar2000
5) Try to load the new playlist and receive following error
INFO (CORE) : startup time: 60 ms
INFO (CORE) : opening file for playback : unpack://rar|78|file://C:\Music\My Releases\Long Beach Dub Allstars - Wonders Of The World.rar|Long Beach Dub Allstars - Wonders Of The World\Long_Beach_Dub_Allstars_-_01_-_Wonders_Dub_I.mp3
INFO (foo_output_std) : attempting to create DirectSound buffer : 44100hz 16bps 2ch fixed-point
INFO (foo_output_std) : DirectSound initialized successfully, using hardware mixing
INFO (CORE) : opening file for playback : unpack://zip|78|file://C:\Music\My Releases\Long Beach Dub Allstars - Wonders Of The World.rar|Long Beach Dub Allstars - Wonders Of The World\Long_Beach_Dub_Allstars_-_01_-_Wonders_Dub_I.mp3
ERROR (CORE) : error opening file for playback : unpack://zip|78|file://C:\Music\My Releases\Long Beach Dub Allstars - Wonders Of The World.rar|Long Beach Dub Allstars - Wonders Of The World\Long_Beach_Dub_Allstars_-_01_-_Wonders_Dub_I.mp3
I theorize that the problem lies in this line: "unpack://zip" which should read "unpack://rar"
Am I correct?
Note 1: PROBLEM FOR ARCHIVED FILES ONLY!
Note 2: This exact procedure works fine with either enabling the database or using m3u8, m3u, pls playlists.
Note 3: Any of these playlists created (m3u, m3u8, pls included) with 0.6 are incompatible with 0.586.
~Alfred Ska
PS. On a side note, it's been a long time since I've laughed as hard as I did when I saw MTRH's avatar, a beautiful mockery of Peter's cat. My thanks for the good times!
Foobar 2000 0.6 beta 10 bugs:
1. When the mass tag or mass rename windows are open, playback stops after the current track finishes (not sure if this is by design or not).
2. The list box in the file info dialog uses MS Sans Serif instead of Tahoma.
Foobar 2000 0.6 beta 10 bugs:
1. When the mass tag or mass rename windows are open, playback stops after the current track finishes (not sure if this is by design or not).
2. The list box in the file info dialog uses MS Sans Serif instead of Tahoma.
1. uninteresting, 2. uninteresting too, open the config and change it to whatever you want
Foobar 0.6 Beta 10
FPL Playlist Bug:
1) Disable database (and nuke it), restart foobar2000
2) Open a RAR archive for playback
3) Save an fpl playlist
4) Clear playlist and restart foobar2000
5) Try to load the new playlist and receive following error
INFO (CORE) : startup time: 60 ms
INFO (CORE) : opening file for playback : unpack://rar|78|file://C:\Music\My Releases\Long Beach Dub Allstars - Wonders Of The World.rar|Long Beach Dub Allstars - Wonders Of The World\Long_Beach_Dub_Allstars_-_01_-_Wonders_Dub_I.mp3
INFO (foo_output_std) : attempting to create DirectSound buffer : 44100hz 16bps 2ch fixed-point
INFO (foo_output_std) : DirectSound initialized successfully, using hardware mixing
INFO (CORE) : opening file for playback : unpack://zip|78|file://C:\Music\My Releases\Long Beach Dub Allstars - Wonders Of The World.rar|Long Beach Dub Allstars - Wonders Of The World\Long_Beach_Dub_Allstars_-_01_-_Wonders_Dub_I.mp3
ERROR (CORE) : error opening file for playback : unpack://zip|78|file://C:\Music\My Releases\Long Beach Dub Allstars - Wonders Of The World.rar|Long Beach Dub Allstars - Wonders Of The World\Long_Beach_Dub_Allstars_-_01_-_Wonders_Dub_I.mp3
I theorize that the problem lies in this line: "unpack://zip" which should read "unpack://rar"
Am I correct?
Note 1: PROBLEM FOR ARCHIVED FILES ONLY!
Note 2: This exact procedure works fine with either enabling the database or using m3u8, m3u, pls playlists.
Note 3: Any of these playlists created (m3u, m3u8, pls included) with 0.6 are incompatible with 0.586.
~Alfred Ska
PS. On a side note, it's been a long time since I've laughed as hard as I did when I saw MTRH's avatar, a beautiful mockery of Peter's cat. My thanks for the good times!
fixed, thanks
$caps() doesn't work properly with certain polish characters, for example:
(...)
i'm using win2000pro and foobar0.6 beta10
it appears that these characters are not considered as a letters.
fixed for beta11, thanks
When opening a folder from commandline, the files aren't sorted in the playlist ('sort incoming files' enabled).
beta 11 (and previous)
beta11 posted.
In beta 11, whenever I go try to go to "Title Formatting" in the pereferences screen, a "foobar2000.exe has encountered a problem and needs to close. We are sorry for the inconvenience." message and it closes.
I'm using XP+SP1
In beta 11, whenever I go try to go to "Title Formatting" in the pereferences screen, a "foobar2000.exe has encountered a problem and needs to close. We are sorry for the inconvenience." message and it closes.
I'm using XP+SP1
Jep, crashes here too...
Can´t reproduce (P3@667Mhz, Win2kSP3, all plugins from the extra folder)
It crashes on my Win2kSP3 too (beta10 works OK)
psy_jc
Posted on Apr 11 2003 - 05:33 AM
In beta 11, whenever I go try to go to "Title Formatting" in the pereferences screen, a "foobar2000.exe has encountered a problem and needs to close. We are sorry for the inconvenience." message and it closes.
I'm using XP+SP1
working fine on my XPpl + sp1
In beta 11, whenever I go try to go to "Title Formatting" in the pereferences screen, a "foobar2000.exe has encountered a problem and needs to close. We are sorry for the inconvenience." message and it closes.
I'm using XP+SP1
Tried using fresh config (deleting your config and reconfiguring foobar)?
crashes here too... (XP without SP1)
New-found (for me) minor bug in both beta10 and also reproduced in v0.586, running on WinNT4.0.
I've recorded an album from cassette tape, stored as Monkey's Audio .APE
I've only noise-reduced and made .MPC files from 2 of the tracks that aren't on the CD version, but wanted to add Album Gain info, so I scanned the .APE files as an album to find the correct values. The replaygain_album_gain was -0.8900 dB and I knew the album peak was in Track13 (by quite a wide margin) - one of the two tracks I made an .MPC from (I'll omit 'replaygain_' from the front of metadata field names from now on)
I scanned each .MPC for Track Gain:
Track13.mpc (Heavy psychedelic rock)
track_gain = -2.5000 dB
track_peak = 0.8025
Track14.mpc (Quiet duet)
track_gain = +3.3400 dB
track_peak = 0.4551
So, to add the album gain data and preserve the intended emotional effect I used fb2k to 'Edit replaygain info (advanced)':
Track13.mpc
track_gain = -2.5200 dB
album_gain = (blank) so I typed in -0.8900 dB as I'd measured on the .APE file
track_peak = 0.8025
album_peak = (blank) so I typed in 0.8025 (same as Track13 peak) and hit OK
Track14.mpc
track_gain = +3.3400 dB
album_gain = (blank) so I typed in -0.8900 dB as I'd measured on the .APE file
track_peak = 0.4551
album_peak = (blank) so I typed in 0.8025 (same as Track13 peak) and hit OK
Now, when I open Show File Info, all seems well, but if I then hit 'Reload info from file', the album_gain has increased (become less negative) to -0.8800 dB (was -0.89) and the track_gain for Track13.mpc has increased to -2.5100 dB (was -2.52) and decreased for track14 to +3.3300 dB (was +3.34 dB). I.e., in all cases the manitssa excluding the sign was decreasing by 0.01. The _peak values remained correct.
On Track14.mpc I tried repeatedly hitting 'Update' then 'Reload info from file' and got the album_gain to -0.8700 dB before it stopped changing by 0.01 dB per step, and the track_gain to +3.2500 dB before it stopped changing. For Track13.mpc, the track_gain reached -2.5000 dB and stopped changing and album_gain stopped at the same -0.8700 dB.
I know this is a negligible and inaudible difference from the intended gain, but I thought I'd point out the apparent bug.
P.S. Loving 0.6 (buttons work on NT ). I particularly like Case's diskwriter encoders and clienc (as well as wapet for EAC) - if I ever get a hardware player I'll be able to enable Crossfeed DSP for earbud listening and ReplayGain (Album) and set it encoding MP3s (or maybe Vorbis) that even on a dumb player sound just how I'd have them if I had embedded fb2k + crossfeed! I'm also very glad to be able to use the player's UI as normal even while it's RG-scanning or converting files, which is a big step up from v0.586 which locked me out and fell silent after the current track. Thanks zZzZzZz, Case and everyone else!
beta12.
New-found (for me) minor bug in both beta10 and also reproduced in v0.586, running on WinNT4.0.
I've recorded an album from cassette tape, stored as Monkey's Audio .APE
I've only noise-reduced and made .MPC files from 2 of the tracks that aren't on the CD version, but wanted to add Album Gain info, so I scanned the .APE files as an album to find the correct values. The replaygain_album_gain was -0.8900 dB and I knew the album peak was in Track13 (by quite a wide margin) - one of the two tracks I made an .MPC from (I'll omit 'replaygain_' from the front of metadata field names from now on)
I scanned each .MPC for Track Gain:
Track13.mpc (Heavy psychedelic rock)
track_gain = -2.5000 dB
track_peak = 0.8025
Track14.mpc (Quiet duet)
track_gain = +3.3400 dB
track_peak = 0.4551
So, to add the album gain data and preserve the intended emotional effect I used fb2k to 'Edit replaygain info (advanced)':
Track13.mpc
track_gain = -2.5200 dB
album_gain = (blank) so I typed in -0.8900 dB as I'd measured on the .APE file
track_peak = 0.8025
album_peak = (blank) so I typed in 0.8025 (same as Track13 peak) and hit OK
Track14.mpc
track_gain = +3.3400 dB
album_gain = (blank) so I typed in -0.8900 dB as I'd measured on the .APE file
track_peak = 0.4551
album_peak = (blank) so I typed in 0.8025 (same as Track13 peak) and hit OK
Now, when I open Show File Info, all seems well, but if I then hit 'Reload info from file', the album_gain has increased (become less negative) to -0.8800 dB (was -0.89) and the track_gain for Track13.mpc has increased to -2.5100 dB (was -2.52) and decreased for track14 to +3.3300 dB (was +3.34 dB). I.e., in all cases the manitssa excluding the sign was decreasing by 0.01. The _peak values remained correct.
On Track14.mpc I tried repeatedly hitting 'Update' then 'Reload info from file' and got the album_gain to -0.8700 dB before it stopped changing by 0.01 dB per step, and the track_gain to +3.2500 dB before it stopped changing. For Track13.mpc, the track_gain reached -2.5000 dB and stopped changing and album_gain stopped at the same -0.8700 dB.
I know this is a negligible and inaudible difference from the intended gain, but I thought I'd point out the apparent bug.
P.S. Loving 0.6 (buttons work on NT ). I particularly like Case's diskwriter encoders and clienc (as well as wapet for EAC) - if I ever get a hardware player I'll be able to enable Crossfeed DSP for earbud listening and ReplayGain (Album) and set it encoding MP3s (or maybe Vorbis) that even on a dumb player sound just how I'd have them if I had embedded fb2k + crossfeed! I'm also very glad to be able to use the player's UI as normal even while it's RG-scanning or converting files, which is a big step up from v0.586 which locked me out and fell silent after the current track. Thanks zZzZzZz, Case and everyone else!
looks like rounding errors caused by the way MPC stores RG info.
It it indended behavior that metadb_handle::handle_query_locked can return a NULL pointer?
It it indended behavior that metadb_handle::handle_query_locked can return a NULL pointer?
yes, i'm going to add more comments to headers about that.
creating a metadb_handle never fails (actually a lot of things rely on metadb_handles just for storing playlist_entries), but errors may occur when reading info from file.
Cool. I was just wondering. I hunted a bug for a while before I realized that I was getting a NULL file_info pointer for one file.
When I try to add an item to the playlist, I get the debug message "Error: this function can be called only from main thread."
metadb_handle_list list;
list.add_item(metadb::get()->handle_create(make_playlist_entry(file,0)));
playlist_oper::get()->add_items(list);
list.delete_all();
I'm using Foobar2000 0.6b12 + SDK 0.6b11. Any ideas or suggestions?
I assume that you already know this, but the functions in play_control.h cannot be called from outside the main thread. If you're trying to do things in a background thread, you can't (not if they call any of the play_control.h functions, at least).
I assume that you already know this, but the functions in play_control.h cannot be called from outside the main thread. If you're trying to do things in a background thread, you can't (not if they call any of the play_control.h functions, at least).
Which main thread - the component's thread? I assume this new behavior in recent 0.6 beta builds.
Edit: Ok, I see the warning at the top of play_control.h
important: you should call api commands declared in this header ONLY from main thread !!
A better question might be: why? I assume it is the new way of dealing with multithreaded components, but wouldn't it be possible to have a lock/unlock method in play_control and playlist_oper?
I have just noticed either a but, or bizzare operation depending on how you view things.
If I drag and drop a .PLS file into the main foobar window, it will open it ok.
Now, if the .PLS has just filenames in it, then the added file entries to the playlist are put in with the _last path that I broswed to_, not the directory that contains the PLS file.
If there are paths in the PLS then its fine, but this is not really doable when accessing them on different drive letters on different machines.
<uninteresting posts removed, use the F. search before posting>
Aero: actually, i do have "playlist_lock()"-like function inside the exe, but i am afraid of adding more stuff like that to API since i've already seen nice deadlock when eg. trying to stop playback from inside database_lock() section (actually hackfixed it on my side but it may still deadlock in rare cases).
I'll rather prevent any external threads from modifying playlist, it seems more safe to me.
Here is something I found using foobar .6 b13... I managed to freeze it and had to force shutdown the program.
I was playing music (APE files) while going thru a playlist retagging items. Note, I was playing different files than the ones I was retagging.
A song got to the end and right as it was about to switch to the next song (showed 1 second left to play in the window) I hit "RUN" on the masstagger plugin to add a single tag to a few song of which I was not playing or going to play.
After this, the whole program froze. It looks as if I had hit run right as foobar was trying to load the next file which killed it....
Nice work on this player Peter. I am converting all my co-workers from winamp to foobar B)
Grant
Edit: Not a bug. Turns out that loading a playlist with 7 weeks of music takes a while. (I'm going thru all my music working on tags)
One more bug I found in beta 13: (Although it could be said this is not proper usage)
I had a playlist of music loaded up in foobar then exited foobar. I went to my foobar directory and deleted my foobar database.foo file since I wanted to start a new one from scratch.
I loaded foobar up again and it never loaded but instead the process took 100% CPU until I force quit the program using task manager.
I deleted the playlist.fpl file after force quiting the process and it loaded right back up. It looks as if you have a playlist but no database, it will kill it.
--Grant
cant reproduce either of them :\
[edit] do you have any weird components (eg. id3v2 readers) installed ? if so, remove them and see if that still happens.
Peter,
I will see if I can reproduce the first problem I found. The only 'extra' plugins I have are the mppenc and clientc plugin installed from Case.
The second one was my fault and not a bug...
EDIT: I reproduced the freeze using masstagger.
Basically, I have an APE file playing, and I have masstagger plug open and ready to edit about 10 files. When the song playing hits 1 second before the end of the track, I hit RUN. It tags about 3 or 4 files then totally freezes up.... I can reproduce this on any number of different albums I have (all of them are APE tho, not sure if that makes a difference)
--Grant
One more bug I found in beta 13: (Although it could be said this is not proper usage)
I had a playlist of music loaded up in foobar then exited foobar. I went to my foobar directory and deleted my foobar database.foo file since I wanted to start a new one from scratch.
I loaded foobar up again and it never loaded but instead the process took 100% CPU until I force quit the program using task manager.
I deleted the playlist.fpl file after force quiting the process and it loaded right back up. It looks as if you have a playlist but no database, it will kill it.
--Grant
I've had this for a quite long time. I haven't considered it as a bug, because I've thought foobar only loads up the tags and that's why it does this. It has always loaded the program, but usually it takes about a minute or some times even more.
EDIT: So you found it out yourself then
Also I have had this kind of a minor bug: I sometimes exit foobar and open it up instantly again (updated a plugin etc, so foobar has to be restarted), but it never loads up. After clicking the icon again, it loads up well. This is while having about 3000 entries in the database, and I've noticed that if the database doesn't contain any entries, foobar loads up just fine.
Also I have had this kind of a minor bug: I sometimes exit foobar and open it up instantly again (updated a plugin etc, so foobar has to be restarted), but it never loads up. After clicking the icon again, it loads up well. This is while having about 3000 entries in the database, and I've noticed that if the database doesn't contain any entries, foobar loads up just fine.
not a bug, new instance cant start up when old one is still active, look at task manager.
Basically, I have an APE file playing, and I have masstagger plug open and ready to edit about 10 files. When the song playing hits 1 second before the end of the track, I hit RUN. It tags about 3 or 4 files then totally freezes up.... I can reproduce this on any number of different albums I have (all of them are APE tho, not sure if that makes a difference)
still cant reproduce but i believe i know what it is, fixing
beta14 posted, see if that still happens (i actually merged playlist and metadb critical sections to prevent two critical section deadlock snafu from repeating since i see more conditions where such problem could occur)
beta14 posted, see if that still happens (i actually merged playlist and metadb critical sections to prevent two critical section deadlock snafu from repeating since i see more conditions where such problem could occur)
Its all fixed now. I can't reproduce the bug anymore. NICE WORK!
--Grant
Aero: actually, i do have "playlist_lock()"-like function inside the exe, but i am afraid of adding more stuff like that to API since i've already seen nice deadlock when eg. trying to stop playback from inside database_lock() section (actually hackfixed it on my side but it may still deadlock in rare cases).
I'll rather prevent any external threads from modifying playlist, it seems more safe to me.
Hmmm...this makes writing multi-threaded Foobar components very hard, or in my case, next to impossible. Do you have any suggestions on how a multi-threaded component could accomodate your restrictions? I have one or two, but they all boil down to basically duplicating the SDK in the component, which is nearly as bad as writing Winamp 2.x plugins using messages to control everything.
So is there any chance at exposing your playlist_lock() functionality to the SDK? Especially in light of your progress in beta14 in removing the deadlock condition?
What did you have in mind?
What did you have in mind?
kode54 - Was this question for me?
None of my ideas about handling the multi-threaded playlist modification thing are really very good. Basically, you would need some way for the child threads of the component to signal the parent thread and pass the information to it. If the child thread just processes some data and exits, you could have it return a pointer to some memory containing your data structure (duplicating the SDK, in other words), and have the parent thread act on that data. Or you could have the parent thread WaitForMultipleObjects, and use an Event or whatever to signal when there is some command that needs to be acted upon.
It all seems like far too much work just to doing simple modifications of the playlist. A simple playlist_lock/unlock would be much cleaner. Besides, I'm not sure what the difference between child and parent threads are, wrt to playlist modifications. Right now, two separate components could modify the playlist at the same time, unless Peter has some internal locking in Foobar. In that case, it really shouldn't matter which of the component's threads modifies the playlist.
why dont you just create a window (possibly invisible one) and PostMessage/SendMessage to it to execute code from main thread ?
beta16 posted, fixed rare annoying crash
why dont you just create a window (possibly invisible one) and PostMessage/SendMessage to it to execute code from main thread ?
Right, that was one of my ideas (which I hinted at with the Winamp 2.x slur). It is still suboptimal for a couple of reasons:
1. In my case, the main thread just fires off threads to do the actual Foobar related work. It doesn't know anything about Foobar, other than enough of initquit to start up. This keeps the main part portable between different audio players. Adding a message handler wouldn't be tough, but it would also mean...
2. The component has to implement another layer on top of Foobar's SDK to pass messages back and forth. If one has to do this, you might as well add a message handler interface to Foobar to save everyone the effort of creating their own. Make it compatable with Winamp's interface for bonus points!
3. It adds a dependancy for the child thread on the parent, where there might not have been one before.
I still don't understand the difference between a component's main thread and a child thread with respect to playlist modifications. If you could explain why this is important, it would be very helpful. Surely you must have something to lock the playlist so only 1 component (or the app itself) can change it at a time...what difference does it make which thread of a component makes the change?
"Add directory" hasn't been working well for a quite long time now. It adds the files to the playlist, but just won't add them to the database.
I've come across a couple of minor bugs:
Right aligned text in the playlist can't be coloured. It won't, for me, either take note of a selected or unselected colour, which is awkward when I want to have the current track shown a different colour to the others.
Also, formatting functions such as $len() count the colour codes as characters. This isn't normally a problem as I can just add the appropriate number of colour characters (or subtract, whatever), but this wouldn't be possible with transitions etc.
Thanks, it's all cool otherwise.
"Add directory" hasn't been working well for a quite long time now. It adds the files to the playlist, but just won't add them to the database.
you are not making any sense
Right aligned text in the playlist can't be coloured. It won't, for me, either take note of a selected or unselected colour, which is awkward when I want to have the current track shown a different colour to the others.
not a bug, you need to re-apply any color codes after each tab character
Right aligned text in the playlist can't be coloured. It won't, for me, either take note of a selected or unselected colour, which is awkward when I want to have the current track shown a different colour to the others.
not a bug, you need to re-apply any color codes after each tab character
Hell, sorry. I carefully set the colours again before the section, but on the wrong side of the tab.
With output set to '24 bit fixed point padded to 32 bit' I'm getting console info:
INFO (foo_output_std) : attempting to create DirectSound buffer : 44100hz 32bps 2ch floating-point
just like it would be 32 bit fp. Is it just a misplaced info or something else?
With output set to '24 bit fixed point padded to 32 bit' I'm getting console info:
INFO (foo_output_std) : attempting to create DirectSound buffer : 44100hz 32bps 2ch floating-point
just like it would be 32 bit fp. Is it just a misplaced info or something else?
reproduced (sort of) / fixed, thanks
zZzZzZz, well I didn't explain it really well, sorry. But what I was saying was that when I use add directory to add my mp3-directory with a bit over 3000 files, foobar reads infos from some of them but then just stops, like the other files would be in the database already. When I scroll down the list, it's updated really slowly and when I try to open album list, foo_dbsearch or foo_shizzle, foobar either crashes or just doesn't do a thing.
because someone smart disabled "immediately load info from new files" ?
Is there any reason $add(999999999,9999999999) gives -1884901890 ?
There seems to be some logic in $add etc. treatment of large numbers, but it aint useful (well at least i dont find it useful.)
The only reason Im asking is because $rand() produces numbers which $add etc. don't like. Well it all seems to revolve around 2^32 because $add(4294967296,4294967296) = 0 apparently.
If the sum of the two numbers is greater than 2^32 / 2, you'll get overflow errors (i.e. the wrong answer).
If the sum of the two numbers is greater than 2^32 / 2, you'll get overflow errors (i.e. the wrong answer).
ermm, ok then..
greater than 2^32 / 2
(the number formerly known as 2^31...? )
Sure.
Here comes the bitcounter. ;-)
The largest positive number a signed 32 bit integer can represent is 2^31 - 1. Add 1 and it wraps to -2^31.
I guess you run into trouble with something like $mod($add($rand(), foo), bar) which gives you negative values. This is a (stupid) property of $mod, which I believe translates to C's %. It would be a reasonable request to PP to make $mod(foo, bar) give values in the range [0, bar-1], not [-bar+1, bar-1].
Did I guess your situation correctly?
It's always in the range of [0, abs(bar)-1], but the sign of the result behaves just like multiplication and division.
Here comes the bitcounter. ;-)
The largest positive number a signed 32 bit integer can represent is 2^31 - 1. Add 1 and it wraps to -2^31.
Damn you. I always forget to subtract that one, because it's never actually affected a project I've worked on (though the 2^32 vs 2^31 has).
I just noticed jumping in time and seeking in large files (long playtime) is VERY slow near the end of the file. It gets slower as closer you get towards the end.
Don't know if it's a bug, or something else, just wanted you to know (if you don't know already).
Here comes the bitcounter. ;-)
The largest positive number a signed 32 bit integer can represent is 2^31 - 1. Add 1 and it wraps to -2^31.
I guess you run into trouble with something like $mod($add($rand(), foo), bar) which gives you negative values. This is a (stupid) property of $mod, which I believe translates to C's %. It would be a reasonable request to PP to make $mod(foo, bar) give values in the range [0, bar-1], not [-bar+1, bar-1].
Did I guess your situation correctly?
No, but you made me realise i could use $mod(), instead of what i was trying to do (times by 255 and divide by 2^32-1)..
I made a little script which will take the modulus of the number (in this case generating a random no between 0 and 255).
$puts(rand,$mod($rand(),256))
$if($strcmp($strchr($get(rand),-),1),$right($get(rand),$sub($len($get(rand)),1)),$get(rand))
Which works better than what i was trying to do.. So everything is ok
More about $mod(foo, bar)
@kode54: You just contradicted yourself. ;-) The sign is the same as the sign of the first argument. So if the first argument is negative, it's [-abs(bar) + 1, 0]. It does not behave like multiplication because the sign of bar is ignored.
@musicmusic: according to the programmer's feeling of elegance the last line should be
$ifgreater(rand, -1, rand, $add(rand, 256))
It will be completely unnecessary if $rand() does not output negative integers.
@zzz: Maybe $rand() should take a range parameter?
@musicmusic: according to the programmer's feeling of elegance the last line should be
$ifgreater(rand, -1, rand, $add(rand, 256))
I suppose that would work too for the purpose, mine would give the correct value for |x| (as |-90| is 90, mine returns 90 while yours would return 166)
Although,
$puts(rand,$mod($rand(),256))
$if($strcmp($left($get(rand),1),-),$mul($get(rand),-1),$get(rand))
makes more sense..
It will be completely unnecessary if $rand() does not output negative integers.
You mean $mod(), i pressume.
b17
pressing a hotkey assigned to "skip forward 4 minutes" (or skip back) actually skips by 4 minutes and 10 seconds. Not sure if this happened in previous betas, I only just noticed.
I just noticed jumping in time and seeking in large files (long playtime) is VERY slow near the end of the file. It gets slower as closer you get towards the end.
Perhaps the format does not support seeking (like mpc)?
I just noticed jumping in time and seeking in large files (long playtime) is VERY slow near the end of the file. It gets slower as closer you get towards the end.
Perhaps the format does not support seeking (like mpc)?
I was using a MP3 File "01. MooKee - The Omnipotent Hard DnB Mix (2002).mp3", 1h 15m 43s, bitrate = 192, channels = 2, samplerate = 44100.
I figured it may have something to do with my harddrive, but I think that is unlikely because its a fast fast, and not that old HD (Maxtor 80GB (D740X, 7200rpm)).
I'm using Windows 2K Pro Uk inc. SP3 and Foobar2000 v0.6 BETA 17.
Bug: diskwriter 'output file name formatting' seems to be using only metadata stored in source files (ignores metadb), so one cannot properly name output files when converting from WAV or other non-tagable formats...
beta18 posted
Have you read this thread (http://www.hydrogenaudio.org/show.php/act/ST/f/27/t/8352)? It's still happens with beta18. Surely this behaviour isn't intended.... is it?
Thanks for the %_ispaused% variable.
One slight problem, however... The title formatting help states that this variable can be used in the playlist and copy command only. In fact, it does not appear to be usable in the playlist. The variable can also be used in the title bar and status bar. It does not appear to be usable in the system tray variable (where I hoped it could be used). Is this intentional?
I'd like to throw a request out there, while I'm at it. It would be nice if the system tray balloon could popup when paused/unpaused, providing visual feedback. Any chance of throwing this request on the pile? (One way to implement this would be to pop up the balloon whenever the system tray title changes, assuming %_ispaused% was used in the string. This would handle track changes as well. You'd have to watch if people use the elapsed time in the string, though.)
Oh yeah, I'm using 0.6 beta 19 on Win2k sp3.
Don't you already know when you pause/unpause a track? What good is the balloon?
$Char() doesn't seem to want to display any of the high ascii characters in title formatting (specifically noticed this in the status bar). However, if I cut and paste the visible character from a character mapping program, it displays correctly.
Example (hoping it shows up)
Trying to show the character • (centered, filled dot)
$Char(149) shows an unknown character box. But if I paste • into the string it displays correctly.
This appears to be true for anything above ascii 127.
Using b20, XP w/ SP1, Verdana font
Usage: $puts(leftchar,•) works
$puts(leftchar,$char(149)) doesn't work
$get(leftchar)
When launching foobar using a command in a shortcut, it always begins at the first song in the playlist instead of using the command on the last song played (where the cursor currently resides).
If your playlist contains A.B,C,D and the cursor was on C when you last quit foobar, launching it using:
"C:\Program Files\foobar2000\foobar2000.exe" /Pause
Will start playing at the beginning of A instead of Pausing at the beginning of C.
When using the shortcuts with commands to launch foobar, it is ignoring both the cursor and the command.
Shortcuts with commands do work properly if foobar has already been launched.
Don't you already know when you pause/unpause a track? What good is the balloon?
I listen to music at work, quietly with headphones. Someone walks into my cube and wants to talk, so I take off the headphones and hit the pause hotkey. Since I don't have the phones on, I can't hear that it's paused. Maybe I hit the wrong key or something. I just would like a visual indication. I don't expect this to be a high priority or anything like that, I just thought I'd ask.
I guess there might be a bit of value in that case.
It would be nice if the system tray balloon could popup when paused/unpaused, providing visual feedback.
Look through keyboard shortcuts, just assign them the same key...
Look through keyboard shortcuts, just assign them the same key...
That would suit my needs just fine, if you could assign the same hotkey to different commands. Actually, I didn't know a keyboard shortcut was created for the balloon tooltip. I looked for it, but I didn't see it. Thanks for pointing it out. Even a separate hotkey would be fine, if you could use %_ispaused% in the system tray formatting. Oh well, I'll just put an indicator at the start of the window title so it shows up in the taskbar.
That would suit my needs just fine,
There's a Play/Pause hotkey, which frobs the playing state.
Peter, could you make fb2k to accept a file with the extension .lst as m3u playlist? This way it's possible to use the %L parameter in Total Commander (TC) to feed foobar with the selected files in TC with a simple command. TC just creates a list file with the full path to all selected files and then starts foobar using "E:\Program Files\Foobar2000\foobar2000.exe list.lst".
why not just associate .lst files with foobar2000, shouldn't that work?
There's a Play/Pause hotkey, which frobs the playing state.
I think TMA meant to assign a the same key to "play/pause" and the shortcut to pop up the balloon, so that hitting the one key would both "frob" the playing state and show the balloon.
why not just associate .lst files with foobar2000, shouldn't that work?
Not if foobar doesn't know what to do with a lst file. Association jpegs with foobar won't turn it into an image viewer.
0.6 Beta 20
Saving m3u8, m3u, and pls playlists is broken for rar archived files (possibly others as well?).
Fpl playlists save and work fine.
Or was this intentional?
~Alfred Ska
why not just associate .lst files with foobar2000, shouldn't that work?
Nopez, (like Curious_George says) when using the command 'Foobar2000.exe list.lst', none of the files are added. When renaming the file to list.m3u the files are added. The problem is Total Commander makes this temporary list only as a file with the extension .lst.
Conclusion: fb2k must allow the extension and recognize it as a playlist. Kinda tricky, but Peter could also make a switch which enables me to force fb2k to see this file as a playlist. Somthin' like 'foobar2000.exe list.lst /forcepl' or '/forcem3u'.
or you could make a tclist.bat somewhere in your path, which contains simply
@Echo Off
ren *.lst *.m3u
and then associate tclist.bat with .lst files. Then whenever you have a new .lst file you simply double-click it and it gets renamed, along with any other .lst files in that directory.
Little bug:
Preferences\Title Formatting - Colors section says: Colors (playlist only), while it should say Colors (playlist and statusbar only).
I got a little bug with "sort by length" function (0.6b20 version)...
by example, I got :
1:00
10:00
2:00
3:00
instead of
1:00
2:00
3:00
10:00
Is it a known bug or not ??? It seemed to me it was known but I couldn't manage to find this anywhere on this forum.
I got a little bug with "sort by length" function (0.6b20 version)...
by example, I got :
1:00
10:00
2:00
3:00
instead of
1:00
2:00
3:00
10:00
Is it a known bug or not ??? It seemed to me it was known but I couldn't manage to find this anywhere on this forum.
you must be high on something, i don't see "sort by length" anywhere. if you use custom sort with "%_length%", big RTFM to you.
Heh heh, nice one Peter. Just checking, do you intend to implement fading on stop and seek or something similar? I'd much appreciate it if you do. No rush, don't get me wrong and I'm sure you've already read it somewhere around...
And I spotted a little typographical error in the "Help" for the Title Formatting:
- $progress2(a,b,c,d,e) - displays progress bar, eg. $progress(%_time_elapsed_seconds%,%_time_total_seconds%,20,'#','=') will display something like "#####==============="
Should say '$progress2(' in the example.
Don't work too hard, eh?
b17
pressing a hotkey assigned to "skip forward 4 minutes" (or skip back) actually skips by 4 minutes and 10 seconds. Not sure if this happened in previous betas, I only just noticed.
Just checked - this happens in b20 also.
It's not for me but a friend of mine needed to use this custom sort for some reason and he saw it gave strange results...
So he told me to repport it because he didn't own a forum account...
I asked a similar question to do with title formatting actually, try "sort by..." and insert
$num($div(%_length_seconds%,60),2):$num($mod(%_length_seconds%,60),2)
Should do the trick...
As for the intellimouse... Peter?
b17
pressing a hotkey assigned to "skip forward 4 minutes" (or skip back) actually skips by 4 minutes and 10 seconds. Not sure if this happened in previous betas, I only just noticed.
Just checked - this happens in b20 also.
You must have the same hotkey set to skip ahead 10 seconds as well as 4 minutes. That explains things well.
@Ghim: If they're bugs, they belong here. Say 'em, or you lose the right to complain once it's out of beta.
With precedent version, my previous and next buttons woked very well and it doesn't work anymore with 0.6b20 version... do you know why ?
This sorting does not work because if you use Sort by ... --> %_length%, fb2k
alphabetically sorts by the length string of each file, not by its numerical value.
To make it work, use Sort by ... --> $num(%_length_seconds%,6)
(assuming that your longest track does not exceed 999999 seconds...)
As far as your IntelliMouse is concerned - I have no idea. I have one as well, and for me the previous/next buttons do still work like a charm...
superdumprob & ssamadhi97 > Thx for the sorting trick...
For the Intellimouse, just forget about it... I think it's still a prob from my side and not from foobar's side... sorry ^^;;;; (I'm gonna try tonight... if it doesn't work tomorrow, i'll say it...)
So don't mind about my problem...
b17
pressing a hotkey assigned to "skip forward 4 minutes" (or skip back) actually skips by 4 minutes and 10 seconds. Not sure if this happened in previous betas, I only just noticed.
Just checked - this happens in b20 also.
You must have the same hotkey set to skip ahead 10 seconds as well as 4 minutes. That explains things well.
I just checked, and this is certainly not the case. I have no overlapping keybindings.
b17
pressing a hotkey assigned to "skip forward 4 minutes" (or skip back) actually skips by 4 minutes and 10 seconds. Not sure if this happened in previous betas, I only just noticed.
Just checked - this happens in b20 also.
Yeah, confirmed here.
QUOTE (rejj @ Apr 16 2003 - 01:58 PM)
QUOTE (rejj @ Apr 15 2003 - 11:23 AM)
b17
pressing a hotkey assigned to "skip forward 4 minutes" (or skip back) actually skips by 4 minutes and 10 seconds. Not sure if this happened in previous betas, I only just noticed.
Just checked - this happens in b20 also.
Yeah, confirmed here.
And here, not that I ever used it before.
Just noticed a bug in 0.6b20. It's reproducible every time. If I load a PLS playlist in foobar2000, quit, and reload the player then none of the files loaded from that PLS file are playable in the current foobar playlist. Trying to play any of the files gives this console message:
INFO (CORE) : opening file for playback : file://<filename>
ERROR (CORE) : error opening file for playback : file://<filename>
If I manually reload the PLS playlist all the files are playable again. Attempting to load a PLS file from Windows Explorer into foobar also gives the same error messages. Any files in the current playlist that weren't initially loaded from a PLS file do not have this problem and play fine.
This occurs on every PLS file I've tried.
This only appears to affect PLS not M3U playlists.
or you could make a tclist.bat somewhere in your path, which contains simply
@Echo Off
ren *.lst *.m3u
and then associate tclist.bat with .lst files. Then whenever you have a new .lst file you simply double-click it and it gets renamed, along with any other .lst files in that directory.
No, it doesn't work that way with Total Commander. TC makes a temporary .lst file and then starts foobar immediately. Making a batch file that renames the list file (filename is unknown to me and could very well be variable) and after that runs foobar can't be done. Besides that, I would have a command prompt visible until Foobar is closed, hideous...
I found that fb2k play *.cue fine if cuesheet and image file store in the same directory,but otherwise it don't work.
I have revised the third line of cuesheet(FILE "xxxxxxx" WAVE) to specify the path.
The cue file can be add to playlist,but can't play.
INFO (CORE) : opening file for playback : file://D:\temp\CDImage.cue
ERROR (CORE) : error opening file for playback : file://D:\temp\CDImage.cue
It seem to be a bug.If I missed something,please tell me.
Please don't mind my bad English
Bug in playlist_oper::add_location() in 0.6_beta20/0.6_beta14_sdk:
No matter what |url| is passed, add_location() always adds the string in the Add Location dialog, or nothing if there is no string available.
Edit: BTW, is there any difference between add_location() and add_items(), other than add_location() takes a const char*, and add_items() takes a list of metadb_handles?
Maybe it would be better to depreciate add_location() and/or make it a helper function that just calls add_items().
I just noticed a problem with play_control::play_item (using 0.6_beta20/0.6_beta14_sdk):
Calling this function does not propagate the new track to the playlist, i.e. titlebar still shows the previously playing track. play_callback::on_playback_new_track and play_control::get_now_playing both pass/return the old track.
why not just associate .lst files with foobar2000, shouldn't that work?
Nopez, (like Curious_George says) when using the command 'Foobar2000.exe list.lst', none of the files are added. When renaming the file to list.m3u the files are added. The problem is Total Commander makes this temporary list only as a file with the extension .lst.
Conclusion: fb2k must allow the extension and recognize it as a playlist. Kinda tricky, but Peter could also make a switch which enables me to force fb2k to see this file as a playlist. Somthin' like 'foobar2000.exe list.lst /forcepl' or '/forcem3u'.
make a bat file with this in it:
CALL REN %~n1.lst %~n1.m3u
CALL START foobar2000.exe %~n1.m3u
That will rename the file to .m3u, then start foobar in its own window using the .m3u file, then the dos box will close.
why not just associate .lst files with foobar2000, shouldn't that work?
Nopez, (like Curious_George says) when using the command 'Foobar2000.exe list.lst', none of the files are added. When renaming the file to list.m3u the files are added. The problem is Total Commander makes this temporary list only as a file with the extension .lst.
Conclusion: fb2k must allow the extension and recognize it as a playlist. Kinda tricky, but Peter could also make a switch which enables me to force fb2k to see this file as a playlist. Somthin' like 'foobar2000.exe list.lst /forcepl' or '/forcem3u'.
make a bat file with this in it:
CALL REN %~n1.lst %~n1.m3u
CALL START foobar2000.exe %~n1.m3u
That will rename the file to .m3u, then start foobar in its own window using the .m3u file, then the dos box will close.
Yeah, that works nicely, however, still a DOS box
1)I use following formating string in playlist:
$if(%_isplaying%,FFFFFF,)$num(%_playlist_number%,3).
[E1FFA1%artist% - ]$if(%title%,['['%album%[ #[%disc%/]
$num(%tracknumber%,2)]'] ']E1FFA1%title%,%_filename_ext%)
[ | 00DDD2%date%][ | B388FC%genre%][ | %comment%][ FFFF00%__bitrate%kbps]
$if(%_isplaying%,FFFFFF $repeat(*PLAYING*,100),)
$if(%title%,,40FFFF No Tags )
$if(%__replaygain_track_gain%,, 4040FFNO GAIN! )$if(%_isplaying%,FFFFFF,)[%_length%]
and if I have 2 or more entries of one songs, than ALL entries have
state 'PLAYING' and I don't known, what exactly entry is actually playing !
Because %_isplaying% use only filename, but not playlist number
2)Is there no way to change background color of item ? will be better
if I can see what track is played by changing its background color
3)feature request - make 'presets'(save and load) for many objects wich can changes -
each formating strings, all formating strings, sorting strings,
configs...
4) make options: "autoreset 'stop after current' state" to config.
5) make some description of standart dsp plugins - sometimes i don't
here changes in sound when I activate some of dsps
p.s. Each new sdk version will not be compatible with previous ? Or
until 1.0 release ?
Minor bug/unexpected feature in 0.6b22/0.6b14 SDK: play_control::VOLUME_SET has no lower bounds (i.e. -100db / -10000). I turned the volume down to -1000dB and everything seemed to work fine, though.
Also a feature request - add a method to access the Title Format strings via the SDK. Actually, some sort of generic access to the preferences would be nice!
or you could make a tclist.bat somewhere in your path, which contains simply
@Echo Off
ren *.lst *.m3u
and then associate tclist.bat with .lst files. Then whenever you have a new .lst file you simply double-click it and it gets renamed, along with any other .lst files in that directory.
No, it doesn't work that way with Total Commander. TC makes a temporary .lst file and then starts foobar immediately. Making a batch file that renames the list file (filename is unknown to me and could very well be variable) and after that runs foobar can't be done. Besides that, I would have a command prompt visible until Foobar is closed, hideous...
well, the actual name of the .lst file would be irrelevant anyways because my batch file would rename all. But if that won't work for you, it won't work. How about hexediting TC so it makes .m3u files instead of .lst
or you could make a tclist.bat somewhere in your path, which contains simply
@Echo Off
ren *.lst *.m3u
and then associate tclist.bat with .lst files. Then whenever you have a new .lst file you simply double-click it and it gets renamed, along with any other .lst files in that directory.
No, it doesn't work that way with Total Commander. TC makes a temporary .lst file and then starts foobar immediately. Making a batch file that renames the list file (filename is unknown to me and could very well be variable) and after that runs foobar can't be done. Besides that, I would have a command prompt visible until Foobar is closed, hideous...
well, the actual name of the .lst file would be irrelevant anyways because my batch file would rename all. But if that won't work for you, it won't work. How about hexediting TC so it makes .m3u files instead of .lst
The batch file zanson proposed works, however I get an annoying Dos box for some seconds. Really annoying...
Create a shortcut to the .bat file. Under Properties/Shortcut/Run set it to minimized. That should hide the DOS window.
This is a very minor nitpick and it might be intentional, but play_control::WINDOW_HIDE isn't quite the same as minimizing Foobar 0.6b22 via the minimize button/menu item. The big difference is that the window state isn't serialized. So if a component calls WINDOW_HIDE, Foobar will reappear at next launch.
In order to serialize the window state, I had to do this:
ShowWindow(service_factory_base::get_main_window(), SW_MINIMIZE);
play_control::command(play_control::WINDOW_HIDE, 0);
It would also be nice if WINDOW_HIDE could be set, with a -1 accessor to get the current state (ala PLAYBACK_SHUFFLE). Not a big deal, though, since you can get the window style and figure it out for yourself.
BTW, Foobar and especially the SDK rock! B)
3)feature request - make 'presets'(save and load) for many objects wich can changes -
each formating strings, all formating strings, sorting strings,
configs...
You can do this for all formatting strings like this:
$puts(string1,<your first string here>)
$puts(string2,<your second string here>)
$get(string1)
$get(string2)
Then just comment the string you DON'T want to show. I myself use this way. Presets would be better, but it works like this also.
4) make options: "autoreset 'stop after current' state" to config.
Isn't there one yet? Properties -> Core -> "Reset stop-after-current when stopping"
Support for m3u files with URLs using extended characters (question marks, ampersands, etc...) broke at some point between .586 and .6
An example would be:
http://domain.com/artist/album/filename.mp...?stream=1&log=0 (http://domain.com/artist/album/filename.mp3?stream=1&log=0)
Also support for choosing to display #EXTINF from m3u playlists would also be cool.
why not just associate .lst files with foobar2000, shouldn't that work?
Nopez, (like Curious_George says) when using the command 'Foobar2000.exe list.lst', none of the files are added. When renaming the file to list.m3u the files are added. The problem is Total Commander makes this temporary list only as a file with the extension .lst.
Conclusion: fb2k must allow the extension and recognize it as a playlist. Kinda tricky, but Peter could also make a switch which enables me to force fb2k to see this file as a playlist. Somthin' like 'foobar2000.exe list.lst /forcepl' or '/forcem3u'.
make a bat file with this in it:
CALL REN %~n1.lst %~n1.m3u
CALL START foobar2000.exe %~n1.m3u
That will rename the file to .m3u, then start foobar in its own window using the .m3u file, then the dos box will close.
Yeah, that works nicely, however, still a DOS box
Yeah there's no way to get around the dos box coming up for a few seconds. You might be able to right click on the bat file and tell it to start the dos box minimized so atleast it would just go into the task bar.
Support for m3u files with URLs using extended characters (question marks, ampersands, etc...) broke at some point between .586 and .6
An example would be:
http://domain.com/artist/album/filename.mp...?stream=1&log=0 (http://domain.com/artist/album/filename.mp3?stream=1&log=0)
Also support for choosing to display #EXTINF from m3u playlists would also be cool.
that url seems broken, returns some HTML crap.
whoever maintains that site should be told to obey some F. standards (most of shoutcast clients out there just assume all http:// urls to be mp3 thanks to nullsoft and their l33t softw34r des1gn sk1llz)
This is a very minor nitpick and it might be intentional, but play_control::WINDOW_HIDE isn't quite the same as minimizing Foobar 0.6b22 via the minimize button/menu item. The big difference is that the window state isn't serialized. So if a component calls WINDOW_HIDE, Foobar will reappear at next launch.
In order to serialize the window state, I had to do this:
ShowWindow(service_factory_base::get_main_window(), SW_MINIMIZE);
play_control::command(play_control::WINDOW_HIDE, 0);
It would also be nice if WINDOW_HIDE could be set, with a -1 accessor to get the current state (ala PLAYBACK_SHUFFLE). Not a big deal, though, since you can get the window style and figure it out for yourself.
i think it is OK to use raw win32 api calls on main window handle for querying status (at least for now).
no idea whats wrong with window state being serialized, all "remember window position" stuff uses GetWindowPlacement()/SetWindowPlacement() in win32 API (called on startup/shutdown, so no extra state change handlers needed).
Do you think that "%_isplaying%" has a chance of becoming available to plugins? I was thinking it would be nice to be able add "$if(%_isplaying%,'*',)" to some plugin display strings. (namely, mine )
no, you have to code it yourself (compare to get_now_playing() and set proper extra params for title formatting)
I figured as much, but thought I might as well ask.
Whats the 4/18/2003 version of foo_input strd on Case's site all about? What's different?
Bug in album-replaygain scanning 48Khz mp3's?I have an album that scans correctly except for one track, but only on the album scan.. if i scan it for per-track gain it finishes and writes the values to the file, but if i select 'scan as album' it encounters an error on that file and stops.
The problem file is a 48kHz 128kbps MP3, joint-stereo, padding size 1 byte, frame size 1 152 samples (if that's any help; I think that's the only 48kHz file I've tried to replaygain yet). There appear to be two synch errors at the very end of the file (MP3Utility [http://www.geocities.com/mp3utility/] reports one at 99%, which could actually be the APE tag, mp3utility is oldish; the MAD winamp plugin reports 2 of them just before playback stops; foobar's console doesn't report anything. There is no audible defect in the file in any player.).
Upon trying this with a different file, also 48kHz but 160kbps this time, the same thing happened: the file scans properly in per-track scan, but the scan process encounters an undescribed error in the last part of the file if I try to scan it as part of an album, which leads me to wonder if it's a generic problem with 48kHz files... this second file also had a synch error at the very end, so in the interests of controlled experiments I went & found a utility for fixing such errors. I found something called MP3Fixer [http://www.nomadness.feck.net/MP3FIXER.zip] and ran it on the file.. it said that there were no synch problems but chucked some bytes out of the file anyway. Tried the replaygain scan again, same "An error occurred while processing files", so it would seem to be 48kHz that sets it off (I've known it to be ok with synch errors in files of other samplerates).
here's what the console has to say about it:
INFO (CORE) : attempting to edit file info : file://M:\The Doors - (Self-titled) The Doors - Take It As It Comes.mp3
INFO (CORE) : file info update successful on : file://M:\The Doors - (Self-titled) The Doors - Take It As It Comes.mp3
WARNING (foo_input_std) : mpeg stream error at 2762710 bytes, attempting resync
INFO (foo_input_std) : mpeg stream resync successful
(the total filesize is only 2 152 308 bytes, so that console output is kind of weirdish... also, that warning doesn't show up when I play the file, only when I'm replaygaining it)
Does anyone know the exact calculation foobar uses for the album replaygain scan? I exported the replaygain numbers for some other albums (ones that scanned correctly) into a spreadsheet to see if I could find a calculation for the album gain based on the individual track gains (and fill in the values for that pesky file myself), with no luck yet. The closest I've gotten is to take the average [mean] of the track gain values and subtract half a standard deviation, but that doesn't seem to be quite it :P
Note: scanning as album
does seem to work when I only scan the one 48kHz file -- but not much point to that :)
ach.. sorry this is so long... I'm trying to be thorough :P
Minor bug with beta 22:
When fb2k is minimized and the track changes to an off-screen entry, the playlist won't scroll to make it visible. It works when the window is visible.
Does anyone know the exact calculation foobar uses for the album replaygain scan?
The gain scanner collects the gain information for the entire album as if it were a single file. Obviously, this cannot work unless all files are the same sample rate. Now, if these are all from a CD album, why are they not 44.1KHz? Or is it a live recording amongst a set of rips? Or do you even know? ;P
The gain scanner collects the gain information for the entire album as if it were a single file. [...] Now, if these are all from a CD album, why are they not 44.1KHz? Or is it a live recording amongst a set of rips? Or do you even know? ;P
Cuz I'm a pirate and I downloaded them off a p2p network; there, are ya happy? :P
I have no idea how that came to be 48kHz... maybe some person had a DAT source or something... or maybe they were just being stupid and did an analogue rip at 48kHz.. in any case, I still think it shouldn't trip up the scanner :P
as to the value produced for the album gain (and I realise I'm probably overlooking something important and making myself look foolish), why isn't it equal to the weighted mean of the individual track gains? i.e. for each track value multiply by the track length (say, in seconds), then add up all those products and divide by the total length of all the tracks? or is there something with logorithms I'm missing (cuz the gain values in are decibels)?
The scanner must be reset for different sample rates.
And as for the weighted average .... I dunno. Might have something to do with the scale not being linear, but I really have no idea, sorry.
as to the value produced for the album gain (and I realise I'm probably overlooking something important and making myself look foolish), why isn't it equal to the weighted mean of the individual track gains?
http://replaygain.hydrogenaudio.org//stati...al_process.html (http://replaygain.hydrogenaudio.org//statistical_process.html)
Having calculated RMS signal levels every 50ms through the file, a single value must be calculated to represent the perceived loudness of the entire file. The above histograms show how many times each RMS value occurred in each file.
The most common RMS value in the speech track was -45dB (background noise) - so the most common RMS value is clearly NOT a good indicator of perceived loudness! The average RMS value is similarly misleading with the speech sample, and also with classical music.
A good method to determine the overall perceived loudness is to sort the RMS energy values into numerical order, and then pick a value near the top of the list.
Moneo, I believe you misunderstood what he meant. Goes back to another situation too. Perhaps someone could explain why you can't simply take existing track replaygain values and derive the album value from those values? Foobar rescans every track even if every single one has track values already when you do an album scan. Surely the track values in combination with the track lengths is all you need to calculate the album value, no? If not, why not?
<edit> ok, I tried an album I have with 16 tracks.
-9.75000 is album value according to foobar's album scan.
-9.33125 is straight average of the track gain values.
229 7.34 1680.86
382 9.3 3552.6
236 8.96 2114.56
223 7.38 1645.74
322 9.81 3158.82
242 10.16 2458.72
336 9.83 3302.88
380 9.62 3655.6
195 10.33 2014.35
198 9.64 1908.72
237 9.1 2156.7
325 10.9 3542.5
266 8.28 2202.48
309 11.02 3405.18
373 7.99 2980.27
202 9.64 1947.28
4455 secs total
41727.26 total
9.3663883277216610549943883277217 weighted average.
hmm, so much for that theory. I read that link before I tried this, but now after doing this and reading that page again I understand why it's calculated again. Because all rms samples get used, not just a track's track-gain. Nevermind
foobar2000 0.6 beta 22, after loading close to 4,000 files and adding them to the database, beta 22 has around 469MB of virtual memory used up, (maybe this is a memory leak?), either way it shouldn't have a footprint of that size at any time. foobar2000 remained open overnight after I loaded the files (it wasn't playing though).
- Mig
foobar2000 0.6 beta 22, after loading close to 4,000 files and adding them to the database, beta 22 has around 469MB of virtual memory used up, (maybe this is a memory leak?), either way it shouldn't have a footprint of that size at any time. foobar2000 remained open overnight after I loaded the files (it wasn't playing though).
- Mig
cant reproduce (loaded ~1000 files, VM size stays 5meg). what file types did you load ? do you have any third-party components installed ?
I only have 2 third party components installed, I loaded mostly mp3s, Maybe a few wavs.
The third party components i've got installed are:
foo_focus and foo_shizzle
Btw, I just re-loaded the files and my VM size becomes about 250mb.
I removed the 3rd party components and loaded up a directory of files that weren't already
in the database (which is enabled), they weren't all media files however, some were zip/rar/exe
and the vm size again rose past 200mb, (though only the media files were added (and for some reason .fla files)
- Mig
Well, foo_focus doesn't allocate any memory (if I recall correctly), and foo_shizzle is only allocating memory on open or refresh (and deallocating on close/refresh).
rar/zip readers eat a lot of memory for temporary data but i cant find any leaks, at least not with rar (working on it)
Adding arbitrary directories (e.g. Visual Studio; Microsoft.NET) seems to cause some memory leakage. Adding the VS.net directory caused the memory use to rise by several megs (from 7.3 to ~12) and only drop part of the way back down on completion (to 8.6). Adding other arbitrary directories seemed to have the same effect (i.e. an additional rise). These numbers are from the task manager, though, and I know the numbers it reports can be misleading.
Quick question (suggestion?)
Do you think you might change the playlist dragging behavior anytime soon? Selecting and dragging tracks doesn't work quite as well as one might hope. Dragging upward causes the dragging action to stop as soon as the selected track(s) reach the top of the list "window". Same thing when dragging down, although depending on how it's sized, there might be a "magic spot" where jiggling the mouse will cause the list to scroll, thus allowing the selections to move down a bit more.
beta23 posted, one confirmed memory leak in rar reader fixed
Support for m3u files with URLs using extended characters (question marks, ampersands, etc...) broke at some point between .586 and .6
An example would be:
http://domain.com/artist/album/filename.mp...?stream=1&log=0 (http://domain.com/artist/album/filename.mp3?stream=1&log=0)
Also support for choosing to display #EXTINF from m3u playlists would also be cool.
that url seems broken, returns some HTML crap.
whoever maintains that site should be told to obey some F. standards (most of shoutcast clients out there just assume all http:// urls to be mp3 thanks to nullsoft and their l33t softw34r des1gn sk1llz)
domain.com/artist/... was just some bullshit example I used instead of my real website. It was an example of the syntax of the kind of link I was talking about, not an actual web-based mp3-via-cgi url.
if you want me to research the problem, then at least post an example that is supposed to work.
anyway, this kind of shit has no chance of working correctly if eg. user decides to limit added files to *.mp3
With latest betas when saving playlist as .m3u8, result is utf-8 but has no '#subsong' and 'file://' stuff.
Has it been dropped recently?
yes, m3u8 format will probably go away, #subsong stuff got removed for compatibility with other players.
Just noticed a minor "bug". First, put transparency on and disable buttons. Then, if you enable buttons, they get shown pretty ugly at first:
(http://koti.mbnet.fi/antsau/foobar/pic/buttons.png)
This gets fixed as soon as you select foobar's main window as the active window.
The input tab on the configuration page has <empty page> written on it. This is with beta 24.
It's cosmetical at best though.
Also, when alt is pressed so the accelerator keys get underlined in the menu, the control button for open ([^]) gets underlined. Don't know if this is intended or not, because it also happens in 0.586.
Again, a minor cosmetical thing. I just can't find big bugs .
Those are both intended behaviors. The <empty page> thing was added fairly recently (presumably because people kept asking why the page was empty). The open button is underlined, because that's the standard symbol for "open". e.g. Look at Winamp's open button.
I'm not sure if you would consider this a bug or not..
I am using diskwriter and I chose a particular output directory (i.e. change it from the C:\ default) and I subsequently deleted this directory.
When I next went to use diskwriter I got the error:
ERROR (foo_diskwriter) : error writing file
It took me quite a while to discover that the problem was that I had deleted the output directory. I think it would be more informative if the error said "output directory does not exist" as well as the writing error.
It isn't an uncommon situation to create a temporary directory to diskwrite to, delete it when you've finished with it and then not use diskwriter for long enough to forget that you need to change the output directory.
Those are both intended behaviors. The <empty page> thing was added fairly recently (presumably because people kept asking why the page was empty). The open button is underlined, because that's the standard symbol for "open". e.g. Look at Winamp's open button.
Ah, right. Figures.
Thanks for explaining.
I found a bug/side-effect with the 'friendly' KS device names. They don't work unless you have Admin priviledges under xp: you just get blank combo-box entries otherwise.
Happens under Windows XP SP1 with KX drivers for SB Live! 5.1 and for Aureal Vortex 2 (original XP drivers from MS).
edit: In fact, the entries aren't always blank - sometimes they're random garbage, or even segments of song paths
very small aesthetic bug (hope it hasn't already been addressed). in .6, if you drop a playlist onto the window, it shows up as the default formatting (black text on a white background) regardless of what you have your formatting set to. as soon as you move the highlight or click a track, it updates the formatting, but it won't otherwise. it does this even if there already is a playlist loaded ... the old playlist is the correct formatting, but the new one isn't.
here's an example ... my normal background is grey, but when you drop a playlist on, it's white. (and nothing is highlighted, my highlighted color isn't black on white either)
(http://www.adhesionband.com/foo/foobug.jpg)
it doesn't happen when you go playlist->open however
also, it doesn't happen on .586 ... but that might just be because that version would automatically highlight the new selection, so when you unhighlight it it changes, just like in .6
not a big deal, just thought you'd like to know. oh, and i'm using the win9x ver
I'm not sure if this has been discussed, but I found a weird bug when going to the display settings.
Right click on desktop >> properties >> settings >> advanced >> card name tab causes the currently playing song to pause. This occurs both when foobar is open and when it's minimized to tray.
I don't know if it's Nvidia specific or not, or if it's just something weird on my comp. I'm using the 43.45 drivers BTW with beta 26.
Also, I haven't changed any of the global keyboard settings or anything like that.
I'm not sure if this has been discussed, but I found a weird bug when going to the display settings.
Right click on desktop >> properties >> settings >> advanced >> card name tab causes the currently playing song to pause. This occurs both when foobar is open and when it's minimized to tray.
I don't know if it's Nvidia specific or not, or if it's just something weird on my comp. I'm using the 43.45 drivers BTW with beta 26.
Also, I haven't changed any of the global keyboard settings or anything like that.
Doesn't happen to me (with 43.45 too). I'm running WinXP.
very small aesthetic bug (hope it hasn't already been addressed). in .6, if you drop a playlist onto the window, it shows up as the default formatting (black text on a white background) regardless of what you have your formatting set to. as soon as you move the highlight or click a track, it updates the formatting, but it won't otherwise. it does this even if there already is a playlist loaded ... the old playlist is the correct formatting, but the new one isn't.
here's an example ... my normal background is grey, but when you drop a playlist on, it's white. (and nothing is highlighted, my highlighted color isn't black on white either)
This is beacuse when you drag a playlist or any files on to the window they are "selected" and white is likely your selected item color.. by clicking on another item it becomes selected, and the others will deselect. boom bam. word.
wes.
(and nothing is highlighted, my highlighted color isn't black on white either)
well ... that quote was inaccurate ... my highlighted color
wasn't black on white, but i changed it when i was goofing around with some different playlist formatting ideas .. ... i tell ya ... sorry about that.
I notice the crossfeed link no longer works, was there something wrong with it? or does it just not work with the latest beta/sdk?
Well, just found one small thing, it has to do with the keyboard shortcuts...
for example, i have pause assigned to ctrl+shift+c (globally) and when i try to use ctrl+shift to edit thing in e.x a normal text editor those keys don't work normally (prolly because of foobar has them locked them or something) usually u can use them to mark/jump between words/text and other stuff, usefull stuff imo.
might not qualify as a bug, rather a side effect, nevertheless it's still annoing
ah, if only i could bind to those pesky windows keys ....
oh yeah, i should work on my spelling as well B)
EDIT: YES, i know i can bind to shift+alt for example, but i just wanted to mention it...
hence the defintion of global hotkeys
for example, i have pause assigned to ctrl+shift+c (globally) and when i try to use ctrl+shift to edit thing in e.x a normal text editor those keys don't work normally
Can't reproduce.
With Shift+Control+C as a global hotkey for "Pause", e.g., shift+control+arrows work normally in other applications.
0.6 final posted, topic closed (since all recent posts are uninteresting)