[PD] simultaneous read and write in tables

[matthew jones]

aha I get you.  Well to be honest I'm not entirely sure how to ensure 
tabwrite~ writes before read.  Things happen in BLOCKS of data, not sample 
by sample, so for example 64 numbers get written all at once and 64 samples 
get READ all at once.  This however shouldn't mean you can't get what you're 
trying to do to work, you just have to take care.  Maybe creation order 
would dictate this? ie. make the read happen before the write? somebody?
A workaround may be to make the table larger and read ahead a few samples to 
give you a bit more leeway.

The way I did loopsampling with tables is somewhat different tho.  First 
fill the table completely, then just repeatedly read with tabplay~s 
connected to a metro.  Guess you have a special reason for doing it your way 
though.
Regards,

Matt

-=-=-=-=-=-=-=-=-=-=-=-
http://www.loopit.org
-=-=-=-=-=-=-=-=-=-=-=-
----- Original Message ----- 
From: "Peter Plessas" <plessas at sbox.tugraz.at>
To: "matthew jones" <mj at isvr.soton.ac.uk>
Cc: <pd-list at iem.at>
Sent: Thursday, July 14, 2005 9:00 AM
Subject: Re: [PD] simultaneous read and write in tables


> Hi,
>
> Well i want to read first at position N, and then write into position N at 
> the same time. So as i read i get the audio signal from some seconds ago, 
> just like a delay.
>
> I hope my description is not too complicated, hasn't anyone got this 
> before? LiveLooping with feedbacks in tables?
>
> regards, Peter
>
> On Mon, 11 Jul 2005, matthew jones wrote:
>
>> But reading from the same point you are writing to (after having written) 
>> would be equivalent to just connecting it up straight with cables (ie. no 
>> delay)...?
>> Maybe I've got the wrong end of the stick here.  If tabwrite~ is 5 
>> samples BEHIND tabread~ then this would give you a delay of (TABLENGTH-4) 
>> samples, I think.  But for a loopsampler you would not need to use a 
>> table to output the CURRENT audio, only the repetitions, so there should 
>> be no issue with reading the table WHEN it is being written to.  Then 
>> again if I have misunderstood your problem, can you provide some more 
>> info on what you need to do?
>> Cheers,
>>
>> Matt
>>
>>> Hi List,
>>>
>>> I need some ideas for reading and writing to a table at the same 
>>> position with audio signals.
>>> I need to create a looping sampler which is based on tables, not delays.
>>> So i need to write to a table with tabwrite~ and read from the exactly 
>>> same position with a tabplay~, though this gives me strange behavior 
>>> (weird audio, distortion, etc). So i figure it is a problem between 
>>> tabwrite~ and tabplay~ accessing the same sample in the table...
>>> Sadly, it has to be exactly the same, as i then feed my playback signal 
>>> into the tabwrite~ again, and a minimum lag would add up with each 
>>> repetition.
>>> As i have tabwrite~ write 5 samples behind tabplay~, audio is fine, but 
>>> it's not the same position (see above).
>>> I know that xsample from Thomas Grill has some of this functionality, 
>>> but it's not quite what i need.
>>>
>>> Any hints?
>>>
>>> regards, Peter
>>
>>
>
> _______________________________________________
> PD-list at iem.at mailing list
> UNSUBSCRIBE and account-management -> 
> http://lists.puredata.info/listinfo/pd-list