soundfont playback bug beeeeeeeep!!!!

fresh101 · Oct 20, 2003, 05:27 PM

Hi everyone,

Having a bit of a problem. When using soundfonts I noticed some of the samples gave off a continuos hi pitch tone (beeeeeeeeeeeeeeeep) after triggering them. Using Vienna I learned that it was the samples in which the local sample end had been edited that gave off the tone. I'm using the Audigy2 ex and latest drivers (this problem with 3534f also). Any help would be greatly appreciated.

Thanks in advance.

PS. These drivers are awesome , complicated , but still awesome thanks

**Eugene Gavrilov** · Oct 21, 2003, 12:28 AM

>> in which the local sample end had been edited that gave off the tone

could you pls give more details?

/E

fresh101 · Oct 21, 2003, 06:49 AM

Thanks for the response, here goes

In Vienna in the sample dialog box the first two parameter boxes are local sample start and local sample end. Lets say I have a snare drum with a sample length of 1000 (local sample start being 0 and sample end being 1000) and I edit the start to 125 and the end to 900 ( to remove silence, unwanted noise , ect ) I get the beeeeep tone during playback. If I return the local sample end to 1000 the beeeeep goes away.

This happens with any soundfont sample that has the local sample end parameter set to a lower value than the sample length value.

The beeeeep sound is a single high pitch tone which continues till you kill whatever soundfont playback application you have open.

**Eugene Gavrilov** · Oct 21, 2003, 11:26 AM

>> and the end to 900

ok, I'll check 'sample end' parameter

/E

Rox271 · Oct 21, 2003, 04:25 PM

Sounds like the famous bug.
Would you do some kind of masking or have granularity on some parameters?

fresh101 · Oct 21, 2003, 05:19 PM

>>The beeeeep sound is a single high pitch tone which continues till you kill whatever soundfont playback application you have open

Did some experimenting in Vienna and found out that the lower the sample end value the louder the beep. Also the length of the tone is relative to the release value in the volume envelope area.

**Eugene Gavrilov** · Oct 21, 2003, 05:21 PM

>> Sounds like the famous bug

not, it is not
its a simple error in SoundFont code -- perhaps, the negative values aren't interpreted correctly -- I'll check them soon

/E

**Eugene Gavrilov** · Oct 21, 2003, 08:15 PM

I was unable to reproduce this error - please create a (small!) soundfont that demonstrates the bug, upload it somewhere and post a link here

thanks

/E

fresh101 · Oct 21, 2003, 11:30 PM

sample files

Here is a sf2 and wav file demonstrating the bug.

**Eugene Gavrilov** · Oct 22, 2003, 09:59 PM

the link is invalid

/E

fresh101 · Oct 22, 2003, 11:08 PM

Sorry it should be fixed now.

sample files

fresh101 · Oct 28, 2003, 04:18 PM

Eugene

have you had a chance to look at the problem yet? If so any ideas?

HumanRes · Oct 28, 2003, 05:40 PM

I checked the provided sample sf2 with reason 2.5 (ASIO 5.33ms) but i did not get the beep. If i change the start / end of the sample in the reason NN-XT Sampler, it's still how it should be. I use many SF2's with presets that have samples with lower end parameter and i never expierienced such an issue before.

----Hardware / Software configuration---------------
Treiber Name:kX Audio Driver (Debug)
Treiber Datum: Oct 22 2003 22:40:38
Treiber Version: 5.10.00.3536 - debug
DB Name:non
CT4832 10k1 [d400]
PCI Information:
Device: 21102 Subsys: 80271102 ChipRev: 8
Card has MPU device
Card HAS AC97 codec
Codec name: Unknown AC97 codec [43525914
3D Extension: Crystal Semi 3D Stereo Enhancement
Codec is 2.0 compliant
Capabilities[1990] : -headphone out-
DAC resolutions : -16-bit- -20-bit-
ADC resolutions : -16-bit- -18-bit-
Ext Capabilities [200]: -slot/DAC mappings-
Port: d400 Irq: 0x11(17)
Playback buffer: 20a0
Record buffer: 4000
Number of AC3 buffers: 4
Tank memory: 256 kb
GSIF buffer: 256 samples
OS version: [2 5 1 Service Pack 1; 2 5 1 Service Pack 1; 156 148]
----------------------------------------------------

fresh101 · Oct 28, 2003, 08:24 PM

The beep does not occur when using software samplers to playback sf2 files. It only happens when using the creative hardware to playback the the soundfonts. Try opening them with vienna (which depends on creative hardware to function) and see if they still playback properly.

**Eugene Gavrilov** · Oct 28, 2003, 10:00 PM

>> have you had a chance to look at the problem yet? If so any ideas?

yep -- sorry, haven't posted a message

it seems I've fixed the bug -- the next driver version (probably, 3536b) will include the fix

btw, concerning the reason of the bug: by default, kX driver adds special 'zero' samples at the end of each soundfont sample
if the sample isn't looped, the loop-start/loop-end positions are set to this 'zero' sample to ensure normal -note-off- event
however, if the sample start/end positions are changed by soundfont parameters, the 'zero' sample is shifted, thus, creating an invalid loop with non-zero audio data -- thus, producing a beeeep

/Eugene

fresh101 · Oct 29, 2003, 03:33 AM

THANKS

AWSOME!!!

**Eugene Gavrilov** · Sep 18, 2006, 03:33 AM

is the link still available?
pls let me know if the bug is still present with 3538l

E.

Oct 20, 2003, 05:27 PM	#1
fresh101 DriverHeaven Newbie Join Date: Oct 2003 Posts: 11 Rep Power: 0	soundfont playback bug beeeeeeeep!!!! Hi everyone, Having a bit of a problem. When using soundfonts I noticed some of the samples gave off a continuos hi pitch tone (beeeeeeeeeeeeeeeep) after triggering them. Using Vienna I learned that it was the samples in which the local sample end had been edited that gave off the tone. I'm using the Audigy2 ex and latest drivers (this problem with 3534f also). Any help would be greatly appreciated. Thanks in advance. PS. These drivers are awesome , complicated , but still awesome thanks

Oct 21, 2003, 11:30 PM	Thread Starter #9
fresh101 DriverHeaven Newbie Join Date: Oct 2003 Posts: 11 Rep Power: 0	sample files Here is a sf2 and wav file demonstrating the bug. Last edited by fresh101; Oct 22, 2003 at 11:04 PM.

Oct 22, 2003, 11:08 PM	Thread Starter #11
fresh101 DriverHeaven Newbie Join Date: Oct 2003 Posts: 11 Rep Power: 0	Sorry it should be fixed now. sample files Last edited by fresh101; Oct 28, 2003 at 04:15 PM.

Oct 21, 2003, 12:28 AM	#2
Eugene Gavrilov kX Project Lead Programmer and Coordinator Join Date: Dec 2002 Posts: 3,106 Rep Power: 73	>> in which the local sample end had been edited that gave off the tone could you pls give more details? /E

Oct 21, 2003, 06:49 AM	Thread Starter #3
fresh101 DriverHeaven Newbie Join Date: Oct 2003 Posts: 11 Rep Power: 0	Thanks for the response, here goes In Vienna in the sample dialog box the first two parameter boxes are local sample start and local sample end. Lets say I have a snare drum with a sample length of 1000 (local sample start being 0 and sample end being 1000) and I edit the start to 125 and the end to 900 ( to remove silence, unwanted noise , ect ) I get the beeeeep tone during playback. If I return the local sample end to 1000 the beeeeep goes away. This happens with any soundfont sample that has the local sample end parameter set to a lower value than the sample length value. The beeeeep sound is a single high pitch tone which continues till you kill whatever soundfont playback application you have open.

Oct 21, 2003, 11:26 AM	#4
Eugene Gavrilov kX Project Lead Programmer and Coordinator Join Date: Dec 2002 Posts: 3,106 Rep Power: 73	>> and the end to 900 ok, I'll check 'sample end' parameter /E

Oct 21, 2003, 04:25 PM	#5
Rox271 DriverHeaven Addict Join Date: Apr 2003 Location: Here Posts: 251 Rep Power: 0	Sounds like the famous bug. Would you do some kind of masking or have granularity on some parameters?

Oct 21, 2003, 05:19 PM	Thread Starter #6
fresh101 DriverHeaven Newbie Join Date: Oct 2003 Posts: 11 Rep Power: 0	>>The beeeeep sound is a single high pitch tone which continues till you kill whatever soundfont playback application you have open Did some experimenting in Vienna and found out that the lower the sample end value the louder the beep. Also the length of the tone is relative to the release value in the volume envelope area.

Oct 21, 2003, 05:21 PM	#7
Eugene Gavrilov kX Project Lead Programmer and Coordinator Join Date: Dec 2002 Posts: 3,106 Rep Power: 73	>> Sounds like the famous bug not, it is not its a simple error in SoundFont code -- perhaps, the negative values aren't interpreted correctly -- I'll check them soon /E

Oct 21, 2003, 08:15 PM	#8
Eugene Gavrilov kX Project Lead Programmer and Coordinator Join Date: Dec 2002 Posts: 3,106 Rep Power: 73	I was unable to reproduce this error - please create a (small!) soundfont that demonstrates the bug, upload it somewhere and post a link here thanks /E

Oct 22, 2003, 09:59 PM	#10
Eugene Gavrilov kX Project Lead Programmer and Coordinator Join Date: Dec 2002 Posts: 3,106 Rep Power: 73	the link is invalid /E

Oct 28, 2003, 04:18 PM	Thread Starter #12
fresh101 DriverHeaven Newbie Join Date: Oct 2003 Posts: 11 Rep Power: 0	Eugene have you had a chance to look at the problem yet? If so any ideas?

Oct 28, 2003, 05:40 PM	#13
HumanRes DriverHeaven Junior Member Join Date: Mar 2003 Location: Germany / Duesseldorf Posts: 26 Rep Power: 0	I checked the provided sample sf2 with reason 2.5 (ASIO 5.33ms) but i did not get the beep. If i change the start / end of the sample in the reason NN-XT Sampler, it's still how it should be. I use many SF2's with presets that have samples with lower end parameter and i never expierienced such an issue before. ----Hardware / Software configuration--------------- Treiber Name:kX Audio Driver (Debug) Treiber Datum: Oct 22 2003 22:40:38 Treiber Version: 5.10.00.3536 - debug DB Name:non CT4832 10k1 [d400] PCI Information: Device: 21102 Subsys: 80271102 ChipRev: 8 Card has MPU device Card HAS AC97 codec Codec name: Unknown AC97 codec [43525914 3D Extension: Crystal Semi 3D Stereo Enhancement Codec is 2.0 compliant Capabilities[1990] : -headphone out- DAC resolutions : -16-bit- -20-bit- ADC resolutions : -16-bit- -18-bit- Ext Capabilities [200]: -slot/DAC mappings- Port: d400 Irq: 0x11(17) Playback buffer: 20a0 Record buffer: 4000 Number of AC3 buffers: 4 Tank memory: 256 kb GSIF buffer: 256 samples OS version: [2 5 1 Service Pack 1; 2 5 1 Service Pack 1; 156 148] ----------------------------------------------------

Oct 28, 2003, 08:24 PM	Thread Starter #14
fresh101 DriverHeaven Newbie Join Date: Oct 2003 Posts: 11 Rep Power: 0	The beep does not occur when using software samplers to playback sf2 files. It only happens when using the creative hardware to playback the the soundfonts. Try opening them with vienna (which depends on creative hardware to function) and see if they still playback properly.

Oct 28, 2003, 10:00 PM	#15
Eugene Gavrilov kX Project Lead Programmer and Coordinator Join Date: Dec 2002 Posts: 3,106 Rep Power: 73	>> have you had a chance to look at the problem yet? If so any ideas? yep -- sorry, haven't posted a message it seems I've fixed the bug -- the next driver version (probably, 3536b) will include the fix btw, concerning the reason of the bug: by default, kX driver adds special 'zero' samples at the end of each soundfont sample if the sample isn't looped, the loop-start/loop-end positions are set to this 'zero' sample to ensure normal -note-off- event however, if the sample start/end positions are changed by soundfont parameters, the 'zero' sample is shifted, thus, creating an invalid loop with non-zero audio data -- thus, producing a beeeep /Eugene

Oct 29, 2003, 03:33 AM	Thread Starter #16
fresh101 DriverHeaven Newbie Join Date: Oct 2003 Posts: 11 Rep Power: 0	THANKS AWSOME!!!

Sep 18, 2006, 03:33 AM	#17
Eugene Gavrilov kX Project Lead Programmer and Coordinator Join Date: Dec 2002 Posts: 3,106 Rep Power: 73	is the link still available? pls let me know if the bug is still present with 3538l E.