[Csnd] GEN11 and the extended guard point
| Date | 2020-05-12 16:34 |
| From | Guillermo Senna |
| Subject | [Csnd] GEN11 and the extended guard point |
| Attachments | GEN11_gp.csd |
Hi,
I want to use a pulse made with GEN11 as an envelope. If I use a table
with a 2^n+1 size I get a notch in the middle of the envelope. Is this
the expected behaviour for the closed-form summation formula used inside
GEN11? The pulse looks fine with a 2^n size. I'm attaching a file that
shows the issue, but please don't run it! Render the code and import the
resulting audio with something like Audacity.
Cheers.
Csound mailing list
Csound@listserv.heanet.ie
https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
Send bugs reports to
https://github.com/csound/csound/issues
Discussions of bugs and features can be posted here
|
| Date | 2020-05-12 21:47 |
| From | Guillermo Senna |
| Subject | Re: [Csnd] GEN11 and the extended guard point |
I think I now what's happening. This conditional
(https://github.com/csound/csound/blob/a739e0b3c152aa5a34bd7407cc938c8908005db0/Engine/fgens.c#L865)
is not evaluating as true (as I think it should) when phs equals the
last index of the table (i.e., the guard point). This is because
sin(3.1415926535897931) isn't exactly 0, but 1.2246467991473532e-16.
I don't know if it's worth trying to fix this.
Cheers.
On 12/5/20 12:34, Guillermo Senna wrote:
> Hi,
>
> I want to use a pulse made with GEN11 as an envelope. If I use a table
> with a 2^n+1 size I get a notch in the middle of the envelope. Is this
> the expected behaviour for the closed-form summation formula used inside
> GEN11? The pulse looks fine with a 2^n size. I'm attaching a file that
> shows the issue, but please don't run it! Render the code and import the
> resulting audio with something like Audacity.
>
> Cheers.
>
>
Csound mailing list
Csound@listserv.heanet.ie
https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
Send bugs reports to
https://github.com/csound/csound/issues
Discussions of bugs and features can be posted here |
| Date | 2020-05-12 21:52 |
| From | john |
| Subject | Re: [Csnd] GEN11 and the extended guard point |
Thnks for the analysis. I will look at it to see if tereis a simple fix;
t too latetonight.
On Tue, 12 May 2020, Guillermo Senna wrote:
> I think I now what's happening. This conditional
> (https://github.com/csound/csound/blob/a739e0b3c152aa5a34bd7407cc938c8908005db0/Engine/fgens.c#L865)
> is not evaluating as true (as I think it should) when phs equals the
> last index of the table (i.e., the guard point). This is because
> sin(3.1415926535897931) isn't exactly 0, but 1.2246467991473532e-16.
>
> I don't know if it's worth trying to fix this.
>
> Cheers.
>
> On 12/5/20 12:34, Guillermo Senna wrote:
>> Hi,
>>
>> I want to use a pulse made with GEN11 as an envelope. If I use a table
>> with a 2^n+1 size I get a notch in the middle of the envelope. Is this
>> the expected behaviour for the closed-form summation formula used inside
>> GEN11? The pulse looks fine with a 2^n size. I'm attaching a file that
>> shows the issue, but please don't run it! Render the code and import the
>> resulting audio with something like Audacity.
>>
>> Cheers.
>>
>>
>
> Csound mailing list
> Csound@listserv.heanet.ie
> https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
> Send bugs reports to
> https://github.com/csound/csound/issues
> Discussions of bugs and features can be posted here
>
Csound mailing list
Csound@listserv.heanet.ie
https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
Send bugs reports to
https://github.com/csound/csound/issues
Discussions of bugs and features can be posted here |
| Date | 2020-05-13 21:01 |
| From | john |
| Subject | Re: [Csnd] GEN11 and the extended guard point |
I have committed what I think is a fix. Just hope it does not break
anything else
On Tue, 12 May 2020, john wrote:
> Thnks for the analysis. I will look at it to see if tereis a simple fix; t
> too latetonight.
>
>
>
> On Tue, 12 May 2020, Guillermo Senna wrote:
>
>> I think I now what's happening. This conditional
>> (https://github.com/csound/csound/blob/a739e0b3c152aa5a34bd7407cc938c8908005db0/Engine/fgens.c#L865)
>> is not evaluating as true (as I think it should) when phs equals the
>> last index of the table (i.e., the guard point). This is because
>> sin(3.1415926535897931) isn't exactly 0, but 1.2246467991473532e-16.
>>
>> I don't know if it's worth trying to fix this.
Csound mailing list
Csound@listserv.heanet.ie
https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
Send bugs reports to
https://github.com/csound/csound/issues
Discussions of bugs and features can be posted here |
| Date | 2020-05-13 22:40 |
| From | Guillermo Senna |
| Subject | Re: [Csnd] GEN11 and the extended guard point |
Thanks! Cheers.
On 13/5/20 17:01, john wrote:
> I have committed what I think is a fix. Just hope it does not break
> anything else
>
>
> On Tue, 12 May 2020, john wrote:
>
>> Thnks for the analysis. I will look at it to see if tereis a simple
>> fix; t too latetonight.
>>
>>
>>
>> On Tue, 12 May 2020, Guillermo Senna wrote:
>>
>>> I think I now what's happening. This conditional
>>> (https://github.com/csound/csound/blob/a739e0b3c152aa5a34bd7407cc938c8908005db0/Engine/fgens.c#L865)
>>>
>>> is not evaluating as true (as I think it should) when phs equals the
>>> last index of the table (i.e., the guard point). This is because
>>> sin(3.1415926535897931) isn't exactly 0, but 1.2246467991473532e-16.
>>>
>>> I don't know if it's worth trying to fix this.
>
> Csound mailing list
> Csound@listserv.heanet.ie
> https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
> Send bugs reports to
> https://github.com/csound/csound/issues
> Discussions of bugs and features can be posted here
Csound mailing list
Csound@listserv.heanet.ie
https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
Send bugs reports to
https://github.com/csound/csound/issues
Discussions of bugs and features can be posted here |