I'm getting the following message after I turnoff instrument 10, which  
is the instrument that calls the UDO:

PERF ERROR in instr 10: linearinterpolate: not initialised
kperturbation	linearinterpolate	0.5	0.0	1	0	ktemp	0
note aborted

The actual line of code referenced is:

kperturbation	 linearinterpolate	 0.5, 0.0, 1, 0, ktemp


note that there is no zero at the end of the line, yet a zero shows up  
at the end of the error message.

While instrument 10 is running, it executes the above line (calls this  
UDO) repeatedly with no errors. The error message consistently occurs  
only after instrument 10 has been turned off. If I continue running  
the csd and try to turn instrument 10 on again, csound (and  
QuteCsound) crash.

I'm running Csound version 5.12 (float samples) Jun 4 2010 with  
QuteCsound 0.6.0.

Can anyone explain what might be causing the error message?

many thanks!

=Lou







Send bugs reports to the Sourceforge bug tracker
            https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Hi Lou,

Could you paste the UDO? I'm wondering if it has an extra mandatory
argument in its definition that isn't being used.

Thanks!
Steven

On 8/29/10, Louis Cohen  wrote:
>
>
> I'm getting the following message after I turnoff instrument 10, which
> is the instrument that calls the UDO:
>
> PERF ERROR in instr 10: linearinterpolate: not initialised
> kperturbation	linearinterpolate	0.5	0.0	1	0	ktemp	0
> note aborted
>
> The actual line of code referenced is:
>
> kperturbation	 linearinterpolate	 0.5, 0.0, 1, 0, ktemp
>
>
> note that there is no zero at the end of the line, yet a zero shows up
> at the end of the error message.
>
> While instrument 10 is running, it executes the above line (calls this
> UDO) repeatedly with no errors. The error message consistently occurs
> only after instrument 10 has been turned off. If I continue running
> the csd and try to turn instrument 10 on again, csound (and
> QuteCsound) crash.
>
> I'm running Csound version 5.12 (float samples) Jun 4 2010 with
> QuteCsound 0.6.0.
>
> Can anyone explain what might be causing the error message?
>
> many thanks!
>
> =Lou
>
>
>
>
>
>
>
> Send bugs reports to the Sourceforge bug tracker
>             https://sourceforge.net/tracker/?group_id=81968&atid=564599
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>
>

Here's the UDO. Sorry I didn't think of including it.

As soon as I saw your message I took off a parameter from the UDO  
declaration without really checking. That's not the problem, it really  
needs 5 inputs. Here it is:


;----------------------------------------------------------------------------
;traces a straight line from ky0 to ky1 at time kt, over a period from  
kt0 to kt1
;syntax:
;ky	 linearinterpolate	 ky1, ky0, kt1, kt0, kt
opcode	 linearinterpolate, k, kkkkk	;traces a straight line from ky0  
to ky1 at time kt, over a period from kt0 to kt1
ky1, ky0, kt1, kt0, kt xin
kslope	 =	 (ky1 - ky0)/(kt1-kt0)
kintercept	 =	 ky0 - (kslope * kt0)
ky	 =	 (kslope * kt) + kintercept
xout	 ky
endop



On Aug 29, 2010, at 5:58 PM, Steven Yi wrote:

> Hi Lou,
>
> Could you paste the UDO? I'm wondering if it has an extra mandatory
> argument in its definition that isn't being used.
>
> Thanks!
> Steven
>
> On 8/29/10, Louis Cohen  wrote:
>>
>>
>> I'm getting the following message after I turnoff instrument 10,  
>> which
>> is the instrument that calls the UDO:
>>
>> PERF ERROR in instr 10: linearinterpolate: not initialised
>> kperturbation	linearinterpolate	0.5	0.0	1	0	ktemp	0
>> note aborted
>>
>> The actual line of code referenced is:
>>
>> kperturbation	 linearinterpolate	 0.5, 0.0, 1, 0, ktemp
>>
>>
>> note that there is no zero at the end of the line, yet a zero shows  
>> up
>> at the end of the error message.
>>
>> While instrument 10 is running, it executes the above line (calls  
>> this
>> UDO) repeatedly with no errors. The error message consistently occurs
>> only after instrument 10 has been turned off. If I continue running
>> the csd and try to turn instrument 10 on again, csound (and
>> QuteCsound) crash.
>>
>> I'm running Csound version 5.12 (float samples) Jun 4 2010 with
>> QuteCsound 0.6.0.
>>
>> Can anyone explain what might be causing the error message?
>>
>> many thanks!
>>
>> =Lou
>>
>>
>>
>>
>>
>>
>>
>> Send bugs reports to the Sourceforge bug tracker
>>            https://sourceforge.net/tracker/? 
>> group_id=81968&atid=564599
>> Discussions of bugs and features can be posted here
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>> "unsubscribe
>> csound"
>>
>>
>
> -- 
> Sent from my mobile device
>
>
> Send bugs reports to the Sourceforge bug tracker
>            https://sourceforge.net/tracker/?group_id=81968&atid=564599
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"
>



Send bugs reports to the Sourceforge bug tracker
            https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Maybe the full CSD would be more helpful. I tried your UDO here and  
got no such PERF errors.

Victor
On 29 Aug 2010, at 23:36, Louis Cohen wrote:

> Here's the UDO. Sorry I didn't think of including it.
>
> As soon as I saw your message I took off a parameter from the UDO  
> declaration without really checking. That's not the problem, it  
> really needs 5 inputs. Here it is:
>
>
> ;----------------------------------------------------------------------------
> ;traces a straight line from ky0 to ky1 at time kt, over a period  
> from kt0 to kt1
> ;syntax:
> ;ky	 linearinterpolate	 ky1, ky0, kt1, kt0, kt
> opcode	 linearinterpolate, k, kkkkk	;traces a straight line from ky0  
> to ky1 at time kt, over a period from kt0 to kt1
> ky1, ky0, kt1, kt0, kt xin
> kslope	 =	 (ky1 - ky0)/(kt1-kt0)
> kintercept	 =	 ky0 - (kslope * kt0)
> ky	 =	 (kslope * kt) + kintercept
> xout	 ky
> endop
>
>
>
> On Aug 29, 2010, at 5:58 PM, Steven Yi wrote:
>
>> Hi Lou,
>>
>> Could you paste the UDO? I'm wondering if it has an extra mandatory
>> argument in its definition that isn't being used.
>>
>> Thanks!
>> Steven
>>
>> On 8/29/10, Louis Cohen  wrote:
>>>
>>>
>>> I'm getting the following message after I turnoff instrument 10,  
>>> which
>>> is the instrument that calls the UDO:
>>>
>>> PERF ERROR in instr 10: linearinterpolate: not initialised
>>> kperturbation	linearinterpolate	0.5	0.0	1	0	ktemp	0
>>> note aborted
>>>
>>> The actual line of code referenced is:
>>>
>>> kperturbation	 linearinterpolate	 0.5, 0.0, 1, 0, ktemp
>>>
>>>
>>> note that there is no zero at the end of the line, yet a zero  
>>> shows up
>>> at the end of the error message.
>>>
>>> While instrument 10 is running, it executes the above line (calls  
>>> this
>>> UDO) repeatedly with no errors. The error message consistently  
>>> occurs
>>> only after instrument 10 has been turned off. If I continue running
>>> the csd and try to turn instrument 10 on again, csound (and
>>> QuteCsound) crash.
>>>
>>> I'm running Csound version 5.12 (float samples) Jun 4 2010 with
>>> QuteCsound 0.6.0.
>>>
>>> Can anyone explain what might be causing the error message?
>>>
>>> many thanks!
>>>
>>> =Lou
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> Send bugs reports to the Sourceforge bug tracker
>>>           https://sourceforge.net/tracker/? 
>>> group_id=81968&atid=564599
>>> Discussions of bugs and features can be posted here
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>> "unsubscribe
>>> csound"
>>>
>>>
>>
>> -- 
>> Sent from my mobile device
>>
>>
>> Send bugs reports to the Sourceforge bug tracker
>>           https://sourceforge.net/tracker/?group_id=81968&atid=564599
>> Discussions of bugs and features can be posted here
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>> "unsubscribe csound"
>>
>
>
>
> Send bugs reports to the Sourceforge bug tracker
>           https://sourceforge.net/tracker/?group_id=81968&atid=564599
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"
>



Send bugs reports to the Sourceforge bug tracker
            https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

I am thinking that the use of the opcode in context would be useful.
My guess at this point is that the UDO is used within an if-then-else
clause or some other if-goto and that it is being skipped at init time
but not at perf-time.

On Sun, Aug 29, 2010 at 6:53 PM, Victor Lazzarini
 wrote:
> Maybe the full CSD would be more helpful. I tried your UDO here and got no
> such PERF errors.
>
> Victor
> On 29 Aug 2010, at 23:36, Louis Cohen wrote:
>
>> Here's the UDO. Sorry I didn't think of including it.
>>
>> As soon as I saw your message I took off a parameter from the UDO
>> declaration without really checking. That's not the problem, it really needs
>> 5 inputs. Here it is:
>>
>>
>>
>> ;----------------------------------------------------------------------------
>> ;traces a straight line from ky0 to ky1 at time kt, over a period from kt0
>> to kt1
>> ;syntax:
>> ;ky      linearinterpolate       ky1, ky0, kt1, kt0, kt
>> opcode   linearinterpolate, k, kkkkk    ;traces a straight line from ky0
>> to ky1 at time kt, over a period from kt0 to kt1
>> ky1, ky0, kt1, kt0, kt xin
>> kslope   =       (ky1 - ky0)/(kt1-kt0)
>> kintercept       =       ky0 - (kslope * kt0)
>> ky       =       (kslope * kt) + kintercept
>> xout     ky
>> endop
>>
>>
>>
>> On Aug 29, 2010, at 5:58 PM, Steven Yi wrote:
>>
>>> Hi Lou,
>>>
>>> Could you paste the UDO? I'm wondering if it has an extra mandatory
>>> argument in its definition that isn't being used.
>>>
>>> Thanks!
>>> Steven
>>>
>>> On 8/29/10, Louis Cohen  wrote:
>>>>
>>>>
>>>> I'm getting the following message after I turnoff instrument 10, which
>>>> is the instrument that calls the UDO:
>>>>
>>>> PERF ERROR in instr 10: linearinterpolate: not initialised
>>>> kperturbation   linearinterpolate       0.5     0.0     1       0
>>>> ktemp   0
>>>> note aborted
>>>>
>>>> The actual line of code referenced is:
>>>>
>>>> kperturbation    linearinterpolate       0.5, 0.0, 1, 0, ktemp
>>>>
>>>>
>>>> note that there is no zero at the end of the line, yet a zero shows up
>>>> at the end of the error message.
>>>>
>>>> While instrument 10 is running, it executes the above line (calls this
>>>> UDO) repeatedly with no errors. The error message consistently occurs
>>>> only after instrument 10 has been turned off. If I continue running
>>>> the csd and try to turn instrument 10 on again, csound (and
>>>> QuteCsound) crash.
>>>>
>>>> I'm running Csound version 5.12 (float samples) Jun 4 2010 with
>>>> QuteCsound 0.6.0.
>>>>
>>>> Can anyone explain what might be causing the error message?
>>>>
>>>> many thanks!
>>>>
>>>> =Lou
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> Send bugs reports to the Sourceforge bug tracker
>>>>          https://sourceforge.net/tracker/?group_id=81968&atid=564599
>>>> Discussions of bugs and features can be posted here
>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>>>> csound"
>>>>
>>>>
>>>
>>> --
>>> Sent from my mobile device
>>>
>>>
>>> Send bugs reports to the Sourceforge bug tracker
>>>          https://sourceforge.net/tracker/?group_id=81968&atid=564599
>>> Discussions of bugs and features can be posted here
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>>> csound"
>>>
>>
>>
>>
>> Send bugs reports to the Sourceforge bug tracker
>>          https://sourceforge.net/tracker/?group_id=81968&atid=564599
>> Discussions of bugs and features can be posted here
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>> csound"
>>
>
>
>
> Send bugs reports to the Sourceforge bug tracker
>           https://sourceforge.net/tracker/?group_id=81968&atid=564599
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>
>


Send bugs reports to the Sourceforge bug tracker
            https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Thanks for helping me with this problem.

Attached is the csd, which must be run under QTCsound. You will see  
two buttons labeled "start note" and "stop note".

First press "start note" -- no problems, and the UDO is being used  
every k-cycle in instrument 10 (and some other places I think.)

Then press "stop note" and the error message appears. If you press  
"start note" again, csound will crash (at least it does for me.)

best,
Lou Cohen


On Aug 29, 2010, at 7:50 PM, Steven Yi wrote:

> I am thinking that the use of the opcode in context would be useful.
> My guess at this point is that the UDO is used within an if-then-else
> clause or some other if-goto and that it is being skipped at init time
> but not at perf-time.
>
> On Sun, Aug 29, 2010 at 6:53 PM, Victor Lazzarini
>  wrote:
>> Maybe the full CSD would be more helpful. I tried your UDO here and  
>> got no
>> such PERF errors.
>>
>> Victor
>> On 29 Aug 2010, at 23:36, Louis Cohen wrote:
>>
>>> Here's the UDO. Sorry I didn't think of including it.
>>>
>>> As soon as I saw your message I took off a parameter from the UDO
>>> declaration without really checking. That's not the problem, it  
>>> really needs
>>> 5 inputs. Here it is:
>>>
>>>
>>>
>>> ;----------------------------------------------------------------------------
>>> ;traces a straight line from ky0 to ky1 at time kt, over a period  
>>> from kt0
>>> to kt1
>>> ;syntax:
>>> ;ky      linearinterpolate       ky1, ky0, kt1, kt0, kt
>>> opcode   linearinterpolate, k, kkkkk    ;traces a straight line  
>>> from ky0
>>> to ky1 at time kt, over a period from kt0 to kt1
>>> ky1, ky0, kt1, kt0, kt xin
>>> kslope   =       (ky1 - ky0)/(kt1-kt0)
>>> kintercept       =       ky0 - (kslope * kt0)
>>> ky       =       (kslope * kt) + kintercept
>>> xout     ky
>>> endop
>>>
>>>
>>>
>>> On Aug 29, 2010, at 5:58 PM, Steven Yi wrote:
>>>
>>>> Hi Lou,
>>>>
>>>> Could you paste the UDO? I'm wondering if it has an extra mandatory
>>>> argument in its definition that isn't being used.
>>>>
>>>> Thanks!
>>>> Steven
>>>>
>>>> On 8/29/10, Louis Cohen  wrote:
>>>>>
>>>>>
>>>>> I'm getting the following message after I turnoff instrument 10,  
>>>>> which
>>>>> is the instrument that calls the UDO:
>>>>>
>>>>> PERF ERROR in instr 10: linearinterpolate: not initialised
>>>>> kperturbation   linearinterpolate       0.5     0.0     1       0
>>>>> ktemp   0
>>>>> note aborted
>>>>>
>>>>> The actual line of code referenced is:
>>>>>
>>>>> kperturbation    linearinterpolate       0.5, 0.0, 1, 0, ktemp
>>>>>
>>>>>
>>>>> note that there is no zero at the end of the line, yet a zero  
>>>>> shows up
>>>>> at the end of the error message.
>>>>>
>>>>> While instrument 10 is running, it executes the above line  
>>>>> (calls this
>>>>> UDO) repeatedly with no errors. The error message consistently  
>>>>> occurs
>>>>> only after instrument 10 has been turned off. If I continue  
>>>>> running
>>>>> the csd and try to turn instrument 10 on again, csound (and
>>>>> QuteCsound) crash.
>>>>>
>>>>> I'm running Csound version 5.12 (float samples) Jun 4 2010 with
>>>>> QuteCsound 0.6.0.
>>>>>
>>>>> Can anyone explain what might be causing the error message?
>>>>>
>>>>> many thanks!
>>>>>
>>>>> =Lou
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> Send bugs reports to the Sourceforge bug tracker
>>>>>          https://sourceforge.net/tracker/?group_id=81968&atid=564599
>>>>> Discussions of bugs and features can be posted here
>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>>>> "unsubscribe
>>>>> csound"
>>>>>
>>>>>
>>>>
>>>> --
>>>> Sent from my mobile device
>>>>
>>>>
>>>> Send bugs reports to the Sourceforge bug tracker
>>>>          https://sourceforge.net/tracker/? 
>>>> group_id=81968&atid=564599
>>>> Discussions of bugs and features can be posted here
>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>>> "unsubscribe
>>>> csound"
>>>>
>>>
>>>
>>>
>>> Send bugs reports to the Sourceforge bug tracker
>>>          https://sourceforge.net/tracker/?group_id=81968&atid=564599
>>> Discussions of bugs and features can be posted here
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>> "unsubscribe
>>> csound"
>>>
>>
>>
>>
>> Send bugs reports to the Sourceforge bug tracker
>>           https://sourceforge.net/tracker/?group_id=81968&atid=564599
>> Discussions of bugs and features can be posted here
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>> "unsubscribe
>> csound"
>>
>>
>
>
> Send bugs reports to the Sourceforge bug tracker
>            https://sourceforge.net/tracker/?group_id=81968&atid=564599
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"
>

Hi Lou,

I took a bit of a look and it's a subtle issue; I'm not sure if it's a
bug or how we want to address it.  The root cause is that your
instrument that calls turnoff2 has a lower number than than the one it
is turning off (i3 turning off i10).  Csound processes down the
instrument in numerical order, but I think it doesn't clean up
deactivated instruments until the next kpass. When i3 processes, it
deactivates the instrument instances in i10, tries to run i10 with
deactivated UDO's (causing the initialise error messages, for UDO's,
when it deacts, it sets its performance function pointer to one that
just reports that message, which normally isn't an issue), then at the
end of kpass tries to cleanup the instrument instances that were
deactivated.

I was able to fix it here by renumbering instrument 3 to instr 13 and
changing the stop button to create an i13 0 1 note instead of i3.
Doing that, I was able to repeatedly use the start/stop note buttons
in your project.

I think we need to change how this works for deactivation or just note
carefully that if using turnoff2 it should only be called on
instruments with numbers less than the one it is called from.
(Thinking about how turnoff works, the behavior works fine in that
context, but if turnoff2 was based on turnoff, then this situation may
not have come into play during testing.)

So, is this a serious enough issue to merit change or is documentation
sufficient?

Thanks!
steven

On Sun, Aug 29, 2010 at 8:54 PM, Louis Cohen  wrote:
>
>
> Thanks for helping me with this problem.
>
> Attached is the csd, which must be run under QTCsound. You will see two
> buttons labeled "start note" and "stop note".
>
> First press "start note" -- no problems, and the UDO is being used every
> k-cycle in instrument 10 (and some other places I think.)
>
> Then press "stop note" and the error message appears. If you press "start
> note" again, csound will crash (at least it does for me.)
>
> best,
> Lou Cohen
>
>
> On Aug 29, 2010, at 7:50 PM, Steven Yi wrote:
>
>> I am thinking that the use of the opcode in context would be useful.
>> My guess at this point is that the UDO is used within an if-then-else
>> clause or some other if-goto and that it is being skipped at init time
>> but not at perf-time.
>>
>> On Sun, Aug 29, 2010 at 6:53 PM, Victor Lazzarini
>>  wrote:
>>>
>>> Maybe the full CSD would be more helpful. I tried your UDO here and got
>>> no
>>> such PERF errors.
>>>
>>> Victor
>>> On 29 Aug 2010, at 23:36, Louis Cohen wrote:
>>>
>>>> Here's the UDO. Sorry I didn't think of including it.
>>>>
>>>> As soon as I saw your message I took off a parameter from the UDO
>>>> declaration without really checking. That's not the problem, it really
>>>> needs
>>>> 5 inputs. Here it is:
>>>>
>>>>
>>>>
>>>>
>>>> ;----------------------------------------------------------------------------
>>>> ;traces a straight line from ky0 to ky1 at time kt, over a period from
>>>> kt0
>>>> to kt1
>>>> ;syntax:
>>>> ;ky      linearinterpolate       ky1, ky0, kt1, kt0, kt
>>>> opcode   linearinterpolate, k, kkkkk    ;traces a straight line from ky0
>>>> to ky1 at time kt, over a period from kt0 to kt1
>>>> ky1, ky0, kt1, kt0, kt xin
>>>> kslope   =       (ky1 - ky0)/(kt1-kt0)
>>>> kintercept       =       ky0 - (kslope * kt0)
>>>> ky       =       (kslope * kt) + kintercept
>>>> xout     ky
>>>> endop
>>>>
>>>>
>>>>
>>>> On Aug 29, 2010, at 5:58 PM, Steven Yi wrote:
>>>>
>>>>> Hi Lou,
>>>>>
>>>>> Could you paste the UDO? I'm wondering if it has an extra mandatory
>>>>> argument in its definition that isn't being used.
>>>>>
>>>>> Thanks!
>>>>> Steven
>>>>>
>>>>> On 8/29/10, Louis Cohen  wrote:
>>>>>>
>>>>>>
>>>>>> I'm getting the following message after I turnoff instrument 10, which
>>>>>> is the instrument that calls the UDO:
>>>>>>
>>>>>> PERF ERROR in instr 10: linearinterpolate: not initialised
>>>>>> kperturbation   linearinterpolate       0.5     0.0     1       0
>>>>>> ktemp   0
>>>>>> note aborted
>>>>>>
>>>>>> The actual line of code referenced is:
>>>>>>
>>>>>> kperturbation    linearinterpolate       0.5, 0.0, 1, 0, ktemp
>>>>>>
>>>>>>
>>>>>> note that there is no zero at the end of the line, yet a zero shows up
>>>>>> at the end of the error message.
>>>>>>
>>>>>> While instrument 10 is running, it executes the above line (calls this
>>>>>> UDO) repeatedly with no errors. The error message consistently occurs
>>>>>> only after instrument 10 has been turned off. If I continue running
>>>>>> the csd and try to turn instrument 10 on again, csound (and
>>>>>> QuteCsound) crash.
>>>>>>
>>>>>> I'm running Csound version 5.12 (float samples) Jun 4 2010 with
>>>>>> QuteCsound 0.6.0.
>>>>>>
>>>>>> Can anyone explain what might be causing the error message?
>>>>>>
>>>>>> many thanks!
>>>>>>
>>>>>> =Lou
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> Send bugs reports to the Sourceforge bug tracker
>>>>>>         https://sourceforge.net/tracker/?group_id=81968&atid=564599
>>>>>> Discussions of bugs and features can be posted here
>>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>>>>>> "unsubscribe
>>>>>> csound"
>>>>>>
>>>>>>
>>>>>
>>>>> --
>>>>> Sent from my mobile device
>>>>>
>>>>>
>>>>> Send bugs reports to the Sourceforge bug tracker
>>>>>         https://sourceforge.net/tracker/?group_id=81968&atid=564599
>>>>> Discussions of bugs and features can be posted here
>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>>>>> "unsubscribe
>>>>> csound"
>>>>>
>>>>
>>>>
>>>>
>>>> Send bugs reports to the Sourceforge bug tracker
>>>>         https://sourceforge.net/tracker/?group_id=81968&atid=564599
>>>> Discussions of bugs and features can be posted here
>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>>>> csound"
>>>>
>>>
>>>
>>>
>>> Send bugs reports to the Sourceforge bug tracker
>>>          https://sourceforge.net/tracker/?group_id=81968&atid=564599
>>> Discussions of bugs and features can be posted here
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>>> csound"
>>>
>>>
>>
>>
>> Send bugs reports to the Sourceforge bug tracker
>>           https://sourceforge.net/tracker/?group_id=81968&atid=564599
>> Discussions of bugs and features can be posted here
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>> csound"
>>
>
>


Send bugs reports to the Sourceforge bug tracker
            https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Steven,

Many thanks for figuring this out.

I don't know if the following would provide any guidance as to whether  
the csound design needs to be changed:

As a relatively naive user, I would use the following rule to avoid  
trouble:

"Only turnoff instruments whose numbers are lower than the instrument  
executing the turnoff opcode." I would treat turnoff, turnoff2 and  
turnon as a family of opcodes for which this rule applies.

I can't think of a situation where this rule would cause me any  
difficulties (but it's early in the morning here!)

However, Csound is actually giving me some conflicting signals:

In instr 3 (the one you changed to 13) I use "active 10" to test  
whether instr 10 is running. I then have a loop that turns off instr  
10 until "active 10" returns 0. This is because a few years ago, I  
think there was a bug in turnoff2 -- it could not turn off multiple  
instances of an instrument, so a loop was necessary.

 From what you say, it would seem that even though "active 10" is  
returning a zero, instr 10 is not completely turned off. That  
possibility creates uncertainty which perhaps would be best cleared up.

I hope these thoughts help.

Again, thanks for spending time on this problem! and for providing a  
workaround.

best,
Lou Cohen




On Aug 29, 2010, at 9:51 PM, Steven Yi wrote:

> Hi Lou,
>
> I took a bit of a look and it's a subtle issue; I'm not sure if it's a
> bug or how we want to address it.  The root cause is that your
> instrument that calls turnoff2 has a lower number than than the one it
> is turning off (i3 turning off i10).  Csound processes down the
> instrument in numerical order, but I think it doesn't clean up
> deactivated instruments until the next kpass. When i3 processes, it
> deactivates the instrument instances in i10, tries to run i10 with
> deactivated UDO's (causing the initialise error messages, for UDO's,
> when it deacts, it sets its performance function pointer to one that
> just reports that message, which normally isn't an issue), then at the
> end of kpass tries to cleanup the instrument instances that were
> deactivated.
>
> I was able to fix it here by renumbering instrument 3 to instr 13 and
> changing the stop button to create an i13 0 1 note instead of i3.
> Doing that, I was able to repeatedly use the start/stop note buttons
> in your project.
>
> I think we need to change how this works for deactivation or just note
> carefully that if using turnoff2 it should only be called on
> instruments with numbers less than the one it is called from.
> (Thinking about how turnoff works, the behavior works fine in that
> context, but if turnoff2 was based on turnoff, then this situation may
> not have come into play during testing.)
>
> So, is this a serious enough issue to merit change or is documentation
> sufficient?
>
> Thanks!
> steven
>
> On Sun, Aug 29, 2010 at 8:54 PM, Louis Cohen   
> wrote:
>>
>>
>> Thanks for helping me with this problem.
>>
>> Attached is the csd, which must be run under QTCsound. You will see  
>> two
>> buttons labeled "start note" and "stop note".
>>
>> First press "start note" -- no problems, and the UDO is being used  
>> every
>> k-cycle in instrument 10 (and some other places I think.)
>>
>> Then press "stop note" and the error message appears. If you press  
>> "start
>> note" again, csound will crash (at least it does for me.)
>>
>> best,
>> Lou Cohen
>>
>>
>> On Aug 29, 2010, at 7:50 PM, Steven Yi wrote:
>>
>>> I am thinking that the use of the opcode in context would be useful.
>>> My guess at this point is that the UDO is used within an if-then- 
>>> else
>>> clause or some other if-goto and that it is being skipped at init  
>>> time
>>> but not at perf-time.
>>>
>>> On Sun, Aug 29, 2010 at 6:53 PM, Victor Lazzarini
>>>  wrote:
>>>>
>>>> Maybe the full CSD would be more helpful. I tried your UDO here  
>>>> and got
>>>> no
>>>> such PERF errors.
>>>>
>>>> Victor
>>>> On 29 Aug 2010, at 23:36, Louis Cohen wrote:
>>>>
>>>>> Here's the UDO. Sorry I didn't think of including it.
>>>>>
>>>>> As soon as I saw your message I took off a parameter from the UDO
>>>>> declaration without really checking. That's not the problem, it  
>>>>> really
>>>>> needs
>>>>> 5 inputs. Here it is:
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> ;----------------------------------------------------------------------------
>>>>> ;traces a straight line from ky0 to ky1 at time kt, over a  
>>>>> period from
>>>>> kt0
>>>>> to kt1
>>>>> ;syntax:
>>>>> ;ky      linearinterpolate       ky1, ky0, kt1, kt0, kt
>>>>> opcode   linearinterpolate, k, kkkkk    ;traces a straight line  
>>>>> from ky0
>>>>> to ky1 at time kt, over a period from kt0 to kt1
>>>>> ky1, ky0, kt1, kt0, kt xin
>>>>> kslope   =       (ky1 - ky0)/(kt1-kt0)
>>>>> kintercept       =       ky0 - (kslope * kt0)
>>>>> ky       =       (kslope * kt) + kintercept
>>>>> xout     ky
>>>>> endop
>>>>>
>>>>>
>>>>>
>>>>> On Aug 29, 2010, at 5:58 PM, Steven Yi wrote:
>>>>>
>>>>>> Hi Lou,
>>>>>>
>>>>>> Could you paste the UDO? I'm wondering if it has an extra  
>>>>>> mandatory
>>>>>> argument in its definition that isn't being used.
>>>>>>
>>>>>> Thanks!
>>>>>> Steven
>>>>>>
>>>>>> On 8/29/10, Louis Cohen  wrote:
>>>>>>>
>>>>>>>
>>>>>>> I'm getting the following message after I turnoff instrument  
>>>>>>> 10, which
>>>>>>> is the instrument that calls the UDO:
>>>>>>>
>>>>>>> PERF ERROR in instr 10: linearinterpolate: not initialised
>>>>>>> kperturbation   linearinterpolate       0.5     0.0      
>>>>>>> 1       0
>>>>>>> ktemp   0
>>>>>>> note aborted
>>>>>>>
>>>>>>> The actual line of code referenced is:
>>>>>>>
>>>>>>> kperturbation    linearinterpolate       0.5, 0.0, 1, 0, ktemp
>>>>>>>
>>>>>>>
>>>>>>> note that there is no zero at the end of the line, yet a zero  
>>>>>>> shows up
>>>>>>> at the end of the error message.
>>>>>>>
>>>>>>> While instrument 10 is running, it executes the above line  
>>>>>>> (calls this
>>>>>>> UDO) repeatedly with no errors. The error message consistently  
>>>>>>> occurs
>>>>>>> only after instrument 10 has been turned off. If I continue  
>>>>>>> running
>>>>>>> the csd and try to turn instrument 10 on again, csound (and
>>>>>>> QuteCsound) crash.
>>>>>>>
>>>>>>> I'm running Csound version 5.12 (float samples) Jun 4 2010 with
>>>>>>> QuteCsound 0.6.0.
>>>>>>>
>>>>>>> Can anyone explain what might be causing the error message?
>>>>>>>
>>>>>>> many thanks!
>>>>>>>
>>>>>>> =Lou
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> Send bugs reports to the Sourceforge bug tracker
>>>>>>>         https://sourceforge.net/tracker/?group_id=81968&atid=564599
>>>>>>> Discussions of bugs and features can be posted here
>>>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>>>>>>> "unsubscribe
>>>>>>> csound"
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> --
>>>>>> Sent from my mobile device
>>>>>>
>>>>>>
>>>>>> Send bugs reports to the Sourceforge bug tracker
>>>>>>         https://sourceforge.net/tracker/?group_id=81968&atid=564599
>>>>>> Discussions of bugs and features can be posted here
>>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>>>>>> "unsubscribe
>>>>>> csound"
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>> Send bugs reports to the Sourceforge bug tracker
>>>>>         https://sourceforge.net/tracker/? 
>>>>> group_id=81968&atid=564599
>>>>> Discussions of bugs and features can be posted here
>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>>>> "unsubscribe
>>>>> csound"
>>>>>
>>>>
>>>>
>>>>
>>>> Send bugs reports to the Sourceforge bug tracker
>>>>          https://sourceforge.net/tracker/? 
>>>> group_id=81968&atid=564599
>>>> Discussions of bugs and features can be posted here
>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>>> "unsubscribe
>>>> csound"
>>>>
>>>>
>>>
>>>
>>> Send bugs reports to the Sourceforge bug tracker
>>>           https://sourceforge.net/tracker/? 
>>> group_id=81968&atid=564599
>>> Discussions of bugs and features can be posted here
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>> "unsubscribe
>>> csound"
>>>
>>
>>
>
>
> Send bugs reports to the Sourceforge bug tracker
>            https://sourceforge.net/tracker/?group_id=81968&atid=564599
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"
>



Send bugs reports to the Sourceforge bug tracker
            https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"