could you post your csound code?
----- Original Message ----- 
From: "Mark Van Peteghem" 
To: 
Sent: Monday, July 21, 2008 9:54 PM
Subject: [Csnd] Deferred size not possible with GEN23?


> Hi,
>
> Now I ran into another problem with GEN23: the documentation says that if 
> the specified size is 0, the size is determined by the number of values in 
> the textfile. This would make what I want to do a lot easier, but when I 
> try it, I get the error "deferred-size ftable 1.000000 illegal here".
>
> I've had the same problem with GEN01, but then I could solve the problem 
> by first determining the size with the filelen opcode. I don't know of a 
> similar workaround for GEN23.
>
> -- 
>  Mark
>  _________________________________________
>  When you get lemons, you make lemonade.
>  When you get hardware, you make software.
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe 
> csound" 

Here is a minimal example in attachment, based on your code, but with 
ftgen, although that makes no difference. I noticed that if I delete the 
note in the score, there are no errors, so the error is most likely 
generated when the table is used with the oscil opcode. The file 
test.txt contains 17 values, so it includes a guard point as required by 
oscil. CSound actually prints out that the table contains 17 elements, 
so reading in the file seems ok.

victor schreef:
> could you post your csound code?
> ----- Original Message ----- From: "Mark Van Peteghem" 
> 
> To: 
> Sent: Monday, July 21, 2008 9:54 PM
> Subject: [Csnd] Deferred size not possible with GEN23?
>
>
>> Hi,
>>
>> Now I ran into another problem with GEN23: the documentation says 
>> that if the specified size is 0, the size is determined by the number 
>> of values in the textfile. This would make what I want to do a lot 
>> easier, but when I try it, I get the error "deferred-size ftable 
>> 1.000000 illegal here".
>>
>> I've had the same problem with GEN01, but then I could solve the 
>> problem by first determining the size with the filelen opcode. I 
>> don't know of a similar workaround for GEN23.
>>
>> -- 
>>  Mark
>>  _________________________________________
>>  When you get lemons, you make lemonade.
>>  When you get hardware, you make software.
>>
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body 
>> "unsubscribe csound" 
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body 
> "unsubscribe csound"
>
>

-- 
  Mark
  _________________________________________
  When you get lemons, you make lemonade.
  When you get hardware, you make software.

Hi Mark -

as you said, this error is due to oscil, not to gen23. When you use  
the statement

gi_test ftgen 0, 0, 17, -23, "text.txt" (instead of 0 as size)

it works. Size=0 or any table length (also not 2^n or 2^n+1) works  
with poscil, or with the tab opcodes.

Best -

	joachim



Am 22.07.2008 um 10:06 schrieb Mark Van Peteghem:

> Here is a minimal example in attachment, based on your code, but  
> with ftgen, although that makes no difference. I noticed that if I  
> delete the note in the score, there are no errors, so the error is  
> most likely generated when the table is used with the oscil opcode.  
> The file test.txt contains 17 values, so it includes a guard point  
> as required by oscil. CSound actually prints out that the table  
> contains 17 elements, so reading in the file seems ok.
>
> victor schreef:
>> could you post your csound code?
>> ----- Original Message ----- From: "Mark Van Peteghem"  
>> 
>> To: 
>> Sent: Monday, July 21, 2008 9:54 PM
>> Subject: [Csnd] Deferred size not possible with GEN23?
>>
>>
>>> Hi,
>>>
>>> Now I ran into another problem with GEN23: the documentation says  
>>> that if the specified size is 0, the size is determined by the  
>>> number of values in the textfile. This would make what I want to  
>>> do a lot easier, but when I try it, I get the error "deferred- 
>>> size ftable 1.000000 illegal here".
>>>
>>> I've had the same problem with GEN01, but then I could solve the  
>>> problem by first determining the size with the filelen opcode. I  
>>> don't know of a similar workaround for GEN23.
>>>
>>> -- 
>>>  Mark
>>>  _________________________________________
>>>  When you get lemons, you make lemonade.
>>>  When you get hardware, you make software.
>>>
>>>
>>>
>>> Send bugs reports to this list.
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>> "unsubscribe csound"
>>
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>> "unsubscribe csound"
>>
>>
>
> -- 
>  Mark
>  _________________________________________
>  When you get lemons, you make lemonade.
>  When you get hardware, you make software.
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"
>
> 
> sr=44100
> ksmps=1
> nchnls=2
>
> gi_test ftgen 0, 0, 0, -23, "test.txt"
>
>     instr 1
> asig oscil 10000, 200, gi_test
>
> outs asig, asig
>
>     endin
> 
>
> 
> i1 0 5
>
> e
> 
> 
> 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
>

I just found that the documentation of GEN01 says you can only use 
deferred sizes with loscil units (probably just loscil and loscil3). I 
suppose it also applies to GEN23, although it is not mentioned there.

It's not intuitive. You'd think that the tables are read before the 
sound of any note is created, so the tablesizes should be known when 
they are needed.

joachim heintz schreef:
> Hi Mark -
>
> as you said, this error is due to oscil, not to gen23. When you use 
> the statement
>
> gi_test ftgen 0, 0, 17, -23, "text.txt" (instead of 0 as size)
>
> it works. Size=0 or any table length (also not 2^n or 2^n+1) works 
> with poscil, or with the tab opcodes.
>
> Best -
>
>     joachim
>
>
>
> Am 22.07.2008 um 10:06 schrieb Mark Van Peteghem:
>
>> Here is a minimal example in attachment, based on your code, but with 
>> ftgen, although that makes no difference. I noticed that if I delete 
>> the note in the score, there are no errors, so the error is most 
>> likely generated when the table is used with the oscil opcode. The 
>> file test.txt contains 17 values, so it includes a guard point as 
>> required by oscil. CSound actually prints out that the table contains 
>> 17 elements, so reading in the file seems ok.
>>
>> victor schreef:
>>> could you post your csound code?
>>> ----- Original Message ----- From: "Mark Van Peteghem" 
>>> 
>>> To: 
>>> Sent: Monday, July 21, 2008 9:54 PM
>>> Subject: [Csnd] Deferred size not possible with GEN23?
>>>
>>>
>>>> Hi,
>>>>
>>>> Now I ran into another problem with GEN23: the documentation says 
>>>> that if the specified size is 0, the size is determined by the 
>>>> number of values in the textfile. This would make what I want to do 
>>>> a lot easier, but when I try it, I get the error "deferred-size 
>>>> ftable 1.000000 illegal here".
>>>>
>>>> I've had the same problem with GEN01, but then I could solve the 
>>>> problem by first determining the size with the filelen opcode. I 
>>>> don't know of a similar workaround for GEN23.
>>>>
>>>> -- 
>>>>  Mark
>>>>  _________________________________________
>>>>  When you get lemons, you make lemonade.
>>>>  When you get hardware, you make software.
>>>>
>>>>
>>>>
>>>> Send bugs reports to this list.
>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body 
>>>> "unsubscribe csound"
>>>
>>>
>>>
>>> Send bugs reports to this list.
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body 
>>> "unsubscribe csound"
>>>
>>>
>>
>> -- 
>>  Mark
>>  _________________________________________
>>  When you get lemons, you make lemonade.
>>  When you get hardware, you make software.
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body 
>> "unsubscribe csound"
>>
>> 
>> sr=44100
>> ksmps=1
>> nchnls=2
>>
>> gi_test ftgen 0, 0, 0, -23, "test.txt"
>>
>>     instr 1
>> asig oscil 10000, 200, gi_test
>>
>> outs asig, asig
>>
>>     endin
>> 
>>
>> 
>> i1 0 5
>>
>> e
>> 
>> 
>> 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
>>
>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body 
> "unsubscribe csound"
>
>

-- 
  Mark
  _________________________________________
  When you get lemons, you make lemonade.
  When you get hardware, you make software.

It's not about not knowing the size, but the fact that the tables are
non-power-of-two (deferred tables are always like that, even if they
are power of two in the end).

At 10:38 22/07/2008, you wrote:
>I just found that the documentation of GEN01 says you can only use 
>deferred sizes with loscil units (probably just loscil and loscil3). 
>I suppose it also applies to GEN23, although it is not mentioned there.
>
>It's not intuitive. You'd think that the tables are read before the 
>sound of any note is created, so the tablesizes should be known when 
>they are needed.
>
>joachim heintz schreef:
>>Hi Mark -
>>
>>as you said, this error is due to oscil, not to gen23. When you use 
>>the statement
>>
>>gi_test ftgen 0, 0, 17, -23, "text.txt" (instead of 0 as size)
>>
>>it works. Size=0 or any table length (also not 2^n or 2^n+1) works 
>>with poscil, or with the tab opcodes.
>>
>>Best -
>>
>>     joachim
>>
>>
>>
>>Am 22.07.2008 um 10:06 schrieb Mark Van Peteghem:
>>
>>>Here is a minimal example in attachment, based on your code, but 
>>>with ftgen, although that makes no difference. I noticed that if I 
>>>delete the note in the score, there are no errors, so the error is 
>>>most likely generated when the table is used with the oscil 
>>>opcode. The file test.txt contains 17 values, so it includes a 
>>>guard point as required by oscil. CSound actually prints out that 
>>>the table contains 17 elements, so reading in the file seems ok.
>>>
>>>victor schreef:
>>>>could you post your csound code?
>>>>----- Original Message ----- From: "Mark Van Peteghem" 
>>>>
>>>>To: 
>>>>Sent: Monday, July 21, 2008 9:54 PM
>>>>Subject: [Csnd] Deferred size not possible with GEN23?
>>>>
>>>>
>>>>>Hi,
>>>>>
>>>>>Now I ran into another problem with GEN23: the documentation 
>>>>>says that if the specified size is 0, the size is determined by 
>>>>>the number of values in the textfile. This would make what I 
>>>>>want to do a lot easier, but when I try it, I get the error 
>>>>>"deferred-size ftable 1.000000 illegal here".
>>>>>
>>>>>I've had the same problem with GEN01, but then I could solve the 
>>>>>problem by first determining the size with the filelen opcode. I 
>>>>>don't know of a similar workaround for GEN23.
>>>>>
>>>>>--
>>>>>  Mark
>>>>>  _________________________________________
>>>>>  When you get lemons, you make lemonade.
>>>>>  When you get hardware, you make software.
>>>>>
>>>>>
>>>>>
>>>>>Send bugs reports to this list.
>>>>>To unsubscribe, send email sympa@lists.bath.ac.uk with body 
>>>>>"unsubscribe csound"
>>>>
>>>>
>>>>
>>>>Send bugs reports to this list.
>>>>To unsubscribe, send email sympa@lists.bath.ac.uk with body 
>>>>"unsubscribe csound"
>>>>
>>>
>>>--
>>>  Mark
>>>  _________________________________________
>>>  When you get lemons, you make lemonade.
>>>  When you get hardware, you make software.
>>>
>>>
>>>Send bugs reports to this list.
>>>To unsubscribe, send email sympa@lists.bath.ac.uk with body 
>>>"unsubscribe csound"
>>>
>>>
>>>sr=44100
>>>ksmps=1
>>>nchnls=2
>>>
>>>gi_test ftgen 0, 0, 0, -23, "test.txt"
>>>
>>>     instr 1
>>>asig oscil 10000, 200, gi_test
>>>
>>>outs asig, asig
>>>
>>>     endin
>>>
>>>
>>>
>>>i1 0 5
>>>
>>>e
>>>
>>>
>>>1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
>>
>>
>>
>>Send bugs reports to this list.
>>To unsubscribe, send email sympa@lists.bath.ac.uk with body 
>>"unsubscribe csound"
>>
>
>--
>  Mark
>  _________________________________________
>  When you get lemons, you make lemonade.
>  When you get hardware, you make software.
>
>
>
>Send bugs reports to this list.
>To unsubscribe, send email sympa@lists.bath.ac.uk with body 
>"unsubscribe csound"

Victor Lazzarini
Music Technology Laboratory
Music Department
National University of Ireland, Maynooth