Artifacts with sinc interpolation

Artifacts with sinc interpolation

Date	2015-09-09 00:07
From	Alex Weiss
Subject	Artifacts with sinc interpolation
Attachments	None None
	Hi, could somebody kindly double-check the code for tablexkt? I'm getting bad distortion and/or high frequency ringing whenever I try to slow a sound down by a factor of 10 or so when using sinc interpolation. I was under the impression that sinc interpolation is supposed to be higher quality than cubic interpolation -- but here, it seems to be significantly worse and introduces more artifacts than even linear interpolation. Here is the snippet I'm using: aindex phasor sr/(ftlen(1)*10) asigL tablexkt aindex, 1, 0, 1024, 1 outs asigL/1000, asigL/1000 As you can see from the example, I also have to scale the output considerably -- not sure if that's normal. Thanks, Alex

Date	2015-09-09 04:19
From	Alex Weiss
Subject	Re: Artifacts with sinc interpolation
Attachments	None None
	I did some further testing and the same files are slowed down flawlessly using diskin2 with sinc interpolation. No output scaling required either. But unfortunately, I do need table access since I want to jump around and play forwards and backwards. Would it be hard to take the interpolation code from diskin2 and plop it into tablexkt? Thanks, Alex On Tue, Sep 8, 2015 at 4:07 PM Alex Weiss <alexweiss86@gmail.com> wrote: Hi, could somebody kindly double-check the code for tablexkt? I'm getting bad distortion and/or high frequency ringing whenever I try to slow a sound down by a factor of 10 or so when using sinc interpolation. I was under the impression that sinc interpolation is supposed to be higher quality than cubic interpolation -- but here, it seems to be significantly worse and introduces more artifacts than even linear interpolation. Here is the snippet I'm using: aindex phasor sr/(ftlen(1)*10) asigL tablexkt aindex, 1, 0, 1024, 1 outs asigL/1000, asigL/1000 As you can see from the example, I also have to scale the output considerably -- not sure if that's normal. Thanks, Alex

Date	2015-09-09 06:24
From	Victor Lazzarini
Subject	Re: Artifacts with sinc interpolation
Attachments	None None
	Looks like a bug. Can you file an issue at github? Victor Lazzarini Dean of Arts, Celtic Studies, and Philosophy Maynooth University Ireland On 9 Sep 2015, at 04:19, Alex Weiss <alexweiss86@gmail.com> wrote: I did some further testing and the same files are slowed down flawlessly using diskin2 with sinc interpolation. No output scaling required either. But unfortunately, I do need table access since I want to jump around and play forwards and backwards. Would it be hard to take the interpolation code from diskin2 and plop it into tablexkt? Thanks, Alex On Tue, Sep 8, 2015 at 4:07 PM Alex Weiss <alexweiss86@gmail.com> wrote: Hi, could somebody kindly double-check the code for tablexkt? I'm getting bad distortion and/or high frequency ringing whenever I try to slow a sound down by a factor of 10 or so when using sinc interpolation. I was under the impression that sinc interpolation is supposed to be higher quality than cubic interpolation -- but here, it seems to be significantly worse and introduces more artifacts than even linear interpolation. Here is the snippet I'm using: aindex phasor sr/(ftlen(1)*10) asigL tablexkt aindex, 1, 0, 1024, 1 outs asigL/1000, asigL/1000 As you can see from the example, I also have to scale the output considerably -- not sure if that's normal. Thanks, Alex ------------------------------------------------------------------------------ Monitor Your Dynamic Infrastructure at Any Scale With Datadog! Get real-time metrics from all of your servers, apps and tools in one place. SourceForge users - Click here to start your Free Trial of Datadog now! http://pubads.g.doubleclick.net/gampad/clk?id=241902991&iu=/4140 _______________________________________________ Csound-users mailing list Csound-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/csound-users Send bugs reports to https://github.com/csound/csound/issues Discussions of bugs and features can be posted here

Date	2015-09-09 09:01
From	Alex Weiss
Subject	Re: Artifacts with sinc interpolation
Attachments	None None
	Done. Alex On Tue, Sep 8, 2015, 22:25 Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote: Looks like a bug. Can you file an issue at github? Victor Lazzarini Dean of Arts, Celtic Studies, and Philosophy Maynooth University Ireland On 9 Sep 2015, at 04:19, Alex Weiss <alexweiss86@gmail.com> wrote: I did some further testing and the same files are slowed down flawlessly using diskin2 with sinc interpolation. No output scaling required either. But unfortunately, I do need table access since I want to jump around and play forwards and backwards. Would it be hard to take the interpolation code from diskin2 and plop it into tablexkt? Thanks, Alex On Tue, Sep 8, 2015 at 4:07 PM Alex Weiss <alexweiss86@gmail.com> wrote: Hi, could somebody kindly double-check the code for tablexkt? I'm getting bad distortion and/or high frequency ringing whenever I try to slow a sound down by a factor of 10 or so when using sinc interpolation. I was under the impression that sinc interpolation is supposed to be higher quality than cubic interpolation -- but here, it seems to be significantly worse and introduces more artifacts than even linear interpolation. Here is the snippet I'm using: aindex phasor sr/(ftlen(1)*10) asigL tablexkt aindex, 1, 0, 1024, 1 outs asigL/1000, asigL/1000 As you can see from the example, I also have to scale the output considerably -- not sure if that's normal. Thanks, Alex ------------------------------------------------------------------------------ Monitor Your Dynamic Infrastructure at Any Scale With Datadog! Get real-time metrics from all of your servers, apps and tools in one place. SourceForge users - Click here to start your Free Trial of Datadog now! http://pubads.g.doubleclick.net/gampad/clk?id=241902991&iu=/4140 _______________________________________________ Csound-users mailing list Csound-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/csound-users Send bugs reports to https://github.com/csound/csound/issues Discussions of bugs and features can be posted here ------------------------------------------------------------------------------ Monitor Your Dynamic Infrastructure at Any Scale With Datadog! Get real-time metrics from all of your servers, apps and tools in one place. SourceForge users - Click here to start your Free Trial of Datadog now! http://pubads.g.doubleclick.net/gampad/clk?id=241902991&iu=/4140_______________________________________________ Csound-users mailing list Csound-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/csound-users Send bugs reports to https://github.com/csound/csound/issues Discussions of bugs and features can be posted here

Date	2015-09-10 19:44
From	Alex Weiss
Subject	Re: Artifacts with sinc interpolation
Attachments	None None
	Just saw that the bug is already fixed in git. And I got variable table length, too! Thanks, Victor. Will this fix make it into the next release that's due at the end of September? Out of curiosity: from a sound quality point of view, is there any point in allowing window sizes bigger than 1024 for interpolation? Alex On Wed, Sep 9, 2015 at 1:01 AM Alex Weiss <alexweiss86@gmail.com> wrote: Done. Alex On Tue, Sep 8, 2015, 22:25 Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote: Looks like a bug. Can you file an issue at github? Victor Lazzarini Dean of Arts, Celtic Studies, and Philosophy Maynooth University Ireland On 9 Sep 2015, at 04:19, Alex Weiss <alexweiss86@gmail.com> wrote: I did some further testing and the same files are slowed down flawlessly using diskin2 with sinc interpolation. No output scaling required either. But unfortunately, I do need table access since I want to jump around and play forwards and backwards. Would it be hard to take the interpolation code from diskin2 and plop it into tablexkt? Thanks, Alex On Tue, Sep 8, 2015 at 4:07 PM Alex Weiss <alexweiss86@gmail.com> wrote: Hi, could somebody kindly double-check the code for tablexkt? I'm getting bad distortion and/or high frequency ringing whenever I try to slow a sound down by a factor of 10 or so when using sinc interpolation. I was under the impression that sinc interpolation is supposed to be higher quality than cubic interpolation -- but here, it seems to be significantly worse and introduces more artifacts than even linear interpolation. Here is the snippet I'm using: aindex phasor sr/(ftlen(1)*10) asigL tablexkt aindex, 1, 0, 1024, 1 outs asigL/1000, asigL/1000 As you can see from the example, I also have to scale the output considerably -- not sure if that's normal. Thanks, Alex ------------------------------------------------------------------------------ Monitor Your Dynamic Infrastructure at Any Scale With Datadog! Get real-time metrics from all of your servers, apps and tools in one place. SourceForge users - Click here to start your Free Trial of Datadog now! http://pubads.g.doubleclick.net/gampad/clk?id=241902991&iu=/4140 _______________________________________________ Csound-users mailing list Csound-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/csound-users Send bugs reports to https://github.com/csound/csound/issues Discussions of bugs and features can be posted here ------------------------------------------------------------------------------ Monitor Your Dynamic Infrastructure at Any Scale With Datadog! Get real-time metrics from all of your servers, apps and tools in one place. SourceForge users - Click here to start your Free Trial of Datadog now! http://pubads.g.doubleclick.net/gampad/clk?id=241902991&iu=/4140_______________________________________________ Csound-users mailing list Csound-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/csound-users Send bugs reports to https://github.com/csound/csound/issues Discussions of bugs and features can be posted here

Date	2015-09-10 20:11
From	Victor Lazzarini
Subject	Re: Artifacts with sinc interpolation
Attachments	None None
	yes, it will be there in the next release. Not sure about the merits of higher orders, never thought too much about it. It was simple to fix, but awfully hard to spot, this one. Victor Lazzarini Dean of Arts, Celtic Studies, and Philosophy Maynooth University Ireland On 10 Sep 2015, at 19:44, Alex Weiss <alexweiss86@gmail.com> wrote: Just saw that the bug is already fixed in git. And I got variable table length, too! Thanks, Victor. Will this fix make it into the next release that's due at the end of September? Out of curiosity: from a sound quality point of view, is there any point in allowing window sizes bigger than 1024 for interpolation? Alex On Wed, Sep 9, 2015 at 1:01 AM Alex Weiss <alexweiss86@gmail.com> wrote: Done. Alex On Tue, Sep 8, 2015, 22:25 Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote: Looks like a bug. Can you file an issue at github? Victor Lazzarini Dean of Arts, Celtic Studies, and Philosophy Maynooth University Ireland On 9 Sep 2015, at 04:19, Alex Weiss <alexweiss86@gmail.com> wrote: I did some further testing and the same files are slowed down flawlessly using diskin2 with sinc interpolation. No output scaling required either. But unfortunately, I do need table access since I want to jump around and play forwards and backwards. Would it be hard to take the interpolation code from diskin2 and plop it into tablexkt? Thanks, Alex On Tue, Sep 8, 2015 at 4:07 PM Alex Weiss <alexweiss86@gmail.com> wrote: Hi, could somebody kindly double-check the code for tablexkt? I'm getting bad distortion and/or high frequency ringing whenever I try to slow a sound down by a factor of 10 or so when using sinc interpolation. I was under the impression that sinc interpolation is supposed to be higher quality than cubic interpolation -- but here, it seems to be significantly worse and introduces more artifacts than even linear interpolation. Here is the snippet I'm using: aindex phasor sr/(ftlen(1)*10) asigL tablexkt aindex, 1, 0, 1024, 1 outs asigL/1000, asigL/1000 As you can see from the example, I also have to scale the output considerably -- not sure if that's normal. Thanks, Alex ------------------------------------------------------------------------------ Monitor Your Dynamic Infrastructure at Any Scale With Datadog! Get real-time metrics from all of your servers, apps and tools in one place. SourceForge users - Click here to start your Free Trial of Datadog now! http://pubads.g.doubleclick.net/gampad/clk?id=241902991&iu=/4140 _______________________________________________ Csound-users mailing list Csound-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/csound-users Send bugs reports to https://github.com/csound/csound/issues Discussions of bugs and features can be posted here ------------------------------------------------------------------------------ Monitor Your Dynamic Infrastructure at Any Scale With Datadog! Get real-time metrics from all of your servers, apps and tools in one place. SourceForge users - Click here to start your Free Trial of Datadog now! http://pubads.g.doubleclick.net/gampad/clk?id=241902991&iu=/4140_______________________________________________ Csound-users mailing list Csound-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/csound-users Send bugs reports to https://github.com/csound/csound/issues Discussions of bugs and features can be posted here ------------------------------------------------------------------------------ Monitor Your Dynamic Infrastructure at Any Scale With Datadog! Get real-time metrics from all of your servers, apps and tools in one place. SourceForge users - Click here to start your Free Trial of Datadog now! http://pubads.g.doubleclick.net/gampad/clk?id=241902991&iu=/4140 _______________________________________________ Csound-users mailing list Csound-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/csound-users Send bugs reports to https://github.com/csound/csound/issues Discussions of bugs and features can be posted here

Date	2015-09-10 20:16
From	Victor Lazzarini
Subject	Re: Artifacts with sinc interpolation
Attachments	None None
	I'm enquiring about the interpolation length Victor Lazzarini Dean of Arts, Celtic Studies, and Philosophy Maynooth University Ireland On 10 Sep 2015, at 20:11, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote: yes, it will be there in the next release. Not sure about the merits of higher orders, never thought too much about it. It was simple to fix, but awfully hard to spot, this one. Victor Lazzarini Dean of Arts, Celtic Studies, and Philosophy Maynooth University Ireland On 10 Sep 2015, at 19:44, Alex Weiss <alexweiss86@gmail.com> wrote: Just saw that the bug is already fixed in git. And I got variable table length, too! Thanks, Victor. Will this fix make it into the next release that's due at the end of September? Out of curiosity: from a sound quality point of view, is there any point in allowing window sizes bigger than 1024 for interpolation? Alex On Wed, Sep 9, 2015 at 1:01 AM Alex Weiss <alexweiss86@gmail.com> wrote: Done. Alex On Tue, Sep 8, 2015, 22:25 Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote: Looks like a bug. Can you file an issue at github? Victor Lazzarini Dean of Arts, Celtic Studies, and Philosophy Maynooth University Ireland On 9 Sep 2015, at 04:19, Alex Weiss <alexweiss86@gmail.com> wrote: I did some further testing and the same files are slowed down flawlessly using diskin2 with sinc interpolation. No output scaling required either. But unfortunately, I do need table access since I want to jump around and play forwards and backwards. Would it be hard to take the interpolation code from diskin2 and plop it into tablexkt? Thanks, Alex On Tue, Sep 8, 2015 at 4:07 PM Alex Weiss <alexweiss86@gmail.com> wrote: Hi, could somebody kindly double-check the code for tablexkt? I'm getting bad distortion and/or high frequency ringing whenever I try to slow a sound down by a factor of 10 or so when using sinc interpolation. I was under the impression that sinc interpolation is supposed to be higher quality than cubic interpolation -- but here, it seems to be significantly worse and introduces more artifacts than even linear interpolation. Here is the snippet I'm using: aindex phasor sr/(ftlen(1)*10) asigL tablexkt aindex, 1, 0, 1024, 1 outs asigL/1000, asigL/1000 As you can see from the example, I also have to scale the output considerably -- not sure if that's normal. Thanks, Alex ------------------------------------------------------------------------------ Monitor Your Dynamic Infrastructure at Any Scale With Datadog! Get real-time metrics from all of your servers, apps and tools in one place. SourceForge users - Click here to start your Free Trial of Datadog now! http://pubads.g.doubleclick.net/gampad/clk?id=241902991&iu=/4140 _______________________________________________ Csound-users mailing list Csound-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/csound-users Send bugs reports to https://github.com/csound/csound/issues Discussions of bugs and features can be posted here ------------------------------------------------------------------------------ Monitor Your Dynamic Infrastructure at Any Scale With Datadog! Get real-time metrics from all of your servers, apps and tools in one place. SourceForge users - Click here to start your Free Trial of Datadog now! http://pubads.g.doubleclick.net/gampad/clk?id=241902991&iu=/4140_______________________________________________ Csound-users mailing list Csound-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/csound-users Send bugs reports to https://github.com/csound/csound/issues Discussions of bugs and features can be posted here ------------------------------------------------------------------------------ Monitor Your Dynamic Infrastructure at Any Scale With Datadog! Get real-time metrics from all of your servers, apps and tools in one place. SourceForge users - Click here to start your Free Trial of Datadog now! http://pubads.g.doubleclick.net/gampad/clk?id=241902991&iu=/4140 _______________________________________________ Csound-users mailing list Csound-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/csound-users Send bugs reports to https://github.com/csound/csound/issues Discussions of bugs and features can be posted here ------------------------------------------------------------------------------ Monitor Your Dynamic Infrastructure at Any Scale With Datadog! Get real-time metrics from all of your servers, apps and tools in one place. SourceForge users - Click here to start your Free Trial of Datadog now! http://pubads.g.doubleclick.net/gampad/clk?id=241902991&iu=/4140 _______________________________________________ Csound-users mailing list Csound-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/csound-users Send bugs reports to https://github.com/csound/csound/issues Discussions of bugs and features can be posted here

Date	2015-09-11 16:18
From	Victor Lazzarini
Subject	Re: Artifacts with sinc interpolation
	Apparently, there is nothing to gain from increasing the window size beyond 1024. ======================== Dr Victor Lazzarini Dean of Arts, Celtic Studies and Philosophy, Maynooth University, Maynooth, Co Kildare, Ireland Tel: 00 353 7086936 Fax: 00 353 1 7086952 > On 10 Sep 2015, at 19:44, Alex Weiss wrote: > > Just saw that the bug is already fixed in git. And I got variable table length, too! Thanks, Victor. Will this fix make it into the next release that's due at the end of September? > > Out of curiosity: from a sound quality point of view, is there any point in allowing window sizes bigger than 1024 for interpolation? > > Alex > > On Wed, Sep 9, 2015 at 1:01 AM Alex Weiss wrote: > Done. > > Alex > > On Tue, Sep 8, 2015, 22:25 Victor Lazzarini wrote: > Looks like a bug. Can you file an issue at github? > > Victor Lazzarini > Dean of Arts, Celtic Studies, and Philosophy > Maynooth University > Ireland > > On 9 Sep 2015, at 04:19, Alex Weiss wrote: > >> I did some further testing and the same files are slowed down flawlessly using diskin2 with sinc interpolation. No output scaling required either. But unfortunately, I do need table access since I want to jump around and play forwards and backwards. >> >> Would it be hard to take the interpolation code from diskin2 and plop it into tablexkt? >> >> Thanks, >> Alex >> >> On Tue, Sep 8, 2015 at 4:07 PM Alex Weiss wrote: >> Hi, >> >> could somebody kindly double-check the code for tablexkt? I'm getting bad distortion and/or high frequency ringing whenever I try to slow a sound down by a factor of 10 or so when using sinc interpolation. I was under the impression that sinc interpolation is supposed to be higher quality than cubic interpolation -- but here, it seems to be significantly worse and introduces more artifacts than even linear interpolation. Here is the snippet I'm using: >> >> aindex phasor sr/(ftlen(1)*10) >> asigL tablexkt aindex, 1, 0, 1024, 1 >> outs asigL/1000, asigL/1000 >> >> As you can see from the example, I also have to scale the output considerably -- not sure if that's normal. >> >> Thanks, >> Alex >> ------------------------------------------------------------------------------ >> Monitor Your Dynamic Infrastructure at Any Scale With Datadog! >> Get real-time metrics from all of your servers, apps and tools >> in one place. >> SourceForge users - Click here to start your Free Trial of Datadog now! >> http://pubads.g.doubleclick.net/gampad/clk?id=241902991&iu=/4140 >> _______________________________________________ >> Csound-users mailing list >> Csound-users@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/csound-users >> Send bugs reports to >> https://github.com/csound/csound/issues >> Discussions of bugs and features can be posted here > ------------------------------------------------------------------------------ > Monitor Your Dynamic Infrastructure at Any Scale With Datadog! > Get real-time metrics from all of your servers, apps and tools > in one place. > SourceForge users - Click here to start your Free Trial of Datadog now! > http://pubads.g.doubleclick.net/gampad/clk?id=241902991&iu=/4140_______________________________________________ > Csound-users mailing list > Csound-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/csound-users > Send bugs reports to > https://github.com/csound/csound/issues > Discussions of bugs and features can be posted here > ------------------------------------------------------------------------------ > Monitor Your Dynamic Infrastructure at Any Scale With Datadog! > Get real-time metrics from all of your servers, apps and tools > in one place. > SourceForge users - Click here to start your Free Trial of Datadog now! > http://pubads.g.doubleclick.net/gampad/clk?id=241902991&iu=/4140_______________________________________________ > Csound-users mailing list > Csound-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/csound-users > Send bugs reports to > https://github.com/csound/csound/issues > Discussions of bugs and features can be posted here ------------------------------------------------------------------------------ _______________________________________________ Csound-users mailing list Csound-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/csound-users Send bugs reports to https://github.com/csound/csound/issues Discussions of bugs and features can be posted here

Date	2015-09-11 17:30
From	Joel Ross
Subject	Re: Artifacts with sinc interpolation
	Sorry to jump in on this but I saw the link in [music-dsp] to Julius O. Smith's walkthrough of Windowed Sinc Interpolation. Interestingly, following the rule of thumb there gives a figure of 2^17 or 2^16 for 32-bit table values. 1024 would be good for a ~20-bit signal according to this rule. What actually sounds good is certainly a different question and I'm inclined to believe higher values aren't worth it, but it's nice to have a guideline. https://ccrma.stanford.edu/~jos/pasp/Choice_Table_Size_Word.html Joel On 11 September 2015 at 16:18, Victor Lazzarini wrote: > Apparently, there is nothing to gain from increasing the window size beyond 1024. > ======================== > Dr Victor Lazzarini > Dean of Arts, Celtic Studies and Philosophy, > Maynooth University, > Maynooth, Co Kildare, Ireland > Tel: 00 353 7086936 > Fax: 00 353 1 7086952 > >> On 10 Sep 2015, at 19:44, Alex Weiss wrote: >> >> Just saw that the bug is already fixed in git. And I got variable table length, too! Thanks, Victor. Will this fix make it into the next release that's due at the end of September? >> >> Out of curiosity: from a sound quality point of view, is there any point in allowing window sizes bigger than 1024 for interpolation? >> >> Alex >> >> On Wed, Sep 9, 2015 at 1:01 AM Alex Weiss wrote: >> Done. >> >> Alex >> >> On Tue, Sep 8, 2015, 22:25 Victor Lazzarini wrote: >> Looks like a bug. Can you file an issue at github? >> >> Victor Lazzarini >> Dean of Arts, Celtic Studies, and Philosophy >> Maynooth University >> Ireland >> >> On 9 Sep 2015, at 04:19, Alex Weiss wrote: >> >>> I did some further testing and the same files are slowed down flawlessly using diskin2 with sinc interpolation. No output scaling required either. But unfortunately, I do need table access since I want to jump around and play forwards and backwards. >>> >>> Would it be hard to take the interpolation code from diskin2 and plop it into tablexkt? >>> >>> Thanks, >>> Alex >>> >>> On Tue, Sep 8, 2015 at 4:07 PM Alex Weiss wrote: >>> Hi, >>> >>> could somebody kindly double-check the code for tablexkt? I'm getting bad distortion and/or high frequency ringing whenever I try to slow a sound down by a factor of 10 or so when using sinc interpolation. I was under the impression that sinc interpolation is supposed to be higher quality than cubic interpolation -- but here, it seems to be significantly worse and introduces more artifacts than even linear interpolation. Here is the snippet I'm using: >>> >>> aindex phasor sr/(ftlen(1)*10) >>> asigL tablexkt aindex, 1, 0, 1024, 1 >>> outs asigL/1000, asigL/1000 >>> >>> As you can see from the example, I also have to scale the output considerably -- not sure if that's normal. >>> >>> Thanks, >>> Alex >>> ------------------------------------------------------------------------------ >>> Monitor Your Dynamic Infrastructure at Any Scale With Datadog! >>> Get real-time metrics from all of your servers, apps and tools >>> in one place. >>> SourceForge users - Click here to start your Free Trial of Datadog now! >>> http://pubads.g.doubleclick.net/gampad/clk?id=241902991&iu=/4140 >>> _______________________________________________ >>> Csound-users mailing list >>> Csound-users@lists.sourceforge.net >>> https://lists.sourceforge.net/lists/listinfo/csound-users >>> Send bugs reports to >>> https://github.com/csound/csound/issues >>> Discussions of bugs and features can be posted here >> ------------------------------------------------------------------------------ >> Monitor Your Dynamic Infrastructure at Any Scale With Datadog! >> Get real-time metrics from all of your servers, apps and tools >> in one place. >> SourceForge users - Click here to start your Free Trial of Datadog now! >> http://pubads.g.doubleclick.net/gampad/clk?id=241902991&iu=/4140_______________________________________________ >> Csound-users mailing list >> Csound-users@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/csound-users >> Send bugs reports to >> https://github.com/csound/csound/issues >> Discussions of bugs and features can be posted here >> ------------------------------------------------------------------------------ >> Monitor Your Dynamic Infrastructure at Any Scale With Datadog! >> Get real-time metrics from all of your servers, apps and tools >> in one place. >> SourceForge users - Click here to start your Free Trial of Datadog now! >> http://pubads.g.doubleclick.net/gampad/clk?id=241902991&iu=/4140_______________________________________________ >> Csound-users mailing list >> Csound-users@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/csound-users >> Send bugs reports to >> https://github.com/csound/csound/issues >> Discussions of bugs and features can be posted here > > > ------------------------------------------------------------------------------ > _______________________________________________ > Csound-users mailing list > Csound-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/csound-users > Send bugs reports to > https://github.com/csound/csound/issues > Discussions of bugs and features can be posted here ------------------------------------------------------------------------------ _______________________________________________ Csound-users mailing list Csound-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/csound-users Send bugs reports to https://github.com/csound/csound/issues Discussions of bugs and features can be posted here

Date	2015-09-11 19:09
From	Alex Weiss
Subject	Re: Artifacts with sinc interpolation
Attachments	None None
	Thanks, Joel and Victor. That's what I thought. Alex On Fri, Sep 11, 2015 at 9:32 AM Joel Ross <joel.binarybrain@gmail.com> wrote: Sorry to jump in on this but I saw the link in [music-dsp] to Julius O. Smith's walkthrough of Windowed Sinc Interpolation. Interestingly, following the rule of thumb there gives a figure of 2^17 or 2^16 for 32-bit table values. 1024 would be good for a ~20-bit signal according to this rule. What actually sounds good is certainly a different question and I'm inclined to believe higher values aren't worth it, but it's nice to have a guideline. https://ccrma.stanford.edu/~jos/pasp/Choice_Table_Size_Word.html Joel On 11 September 2015 at 16:18, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote: > Apparently, there is nothing to gain from increasing the window size beyond 1024. > ======================== > Dr Victor Lazzarini > Dean of Arts, Celtic Studies and Philosophy, > Maynooth University, > Maynooth, Co Kildare, Ireland > Tel: 00 353 7086936 > Fax: 00 353 1 7086952 > >> On 10 Sep 2015, at 19:44, Alex Weiss <alexweiss86@gmail.com> wrote: >> >> Just saw that the bug is already fixed in git. And I got variable table length, too! Thanks, Victor. Will this fix make it into the next release that's due at the end of September? >> >> Out of curiosity: from a sound quality point of view, is there any point in allowing window sizes bigger than 1024 for interpolation? >> >> Alex >> >> On Wed, Sep 9, 2015 at 1:01 AM Alex Weiss <alexweiss86@gmail.com> wrote: >> Done. >> >> Alex >> >> On Tue, Sep 8, 2015, 22:25 Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote: >> Looks like a bug. Can you file an issue at github? >> >> Victor Lazzarini >> Dean of Arts, Celtic Studies, and Philosophy >> Maynooth University >> Ireland >> >> On 9 Sep 2015, at 04:19, Alex Weiss <alexweiss86@gmail.com> wrote: >> >>> I did some further testing and the same files are slowed down flawlessly using diskin2 with sinc interpolation. No output scaling required either. But unfortunately, I do need table access since I want to jump around and play forwards and backwards. >>> >>> Would it be hard to take the interpolation code from diskin2 and plop it into tablexkt? >>> >>> Thanks, >>> Alex >>> >>> On Tue, Sep 8, 2015 at 4:07 PM Alex Weiss <alexweiss86@gmail.com> wrote: >>> Hi, >>> >>> could somebody kindly double-check the code for tablexkt? I'm getting bad distortion and/or high frequency ringing whenever I try to slow a sound down by a factor of 10 or so when using sinc interpolation. I was under the impression that sinc interpolation is supposed to be higher quality than cubic interpolation -- but here, it seems to be significantly worse and introduces more artifacts than even linear interpolation. Here is the snippet I'm using: >>> >>> aindex phasor sr/(ftlen(1)*10) >>> asigL tablexkt aindex, 1, 0, 1024, 1 >>> outs asigL/1000, asigL/1000 >>> >>> As you can see from the example, I also have to scale the output considerably -- not sure if that's normal. >>> >>> Thanks, >>> Alex >>> ------------------------------------------------------------------------------ >>> Monitor Your Dynamic Infrastructure at Any Scale With Datadog! >>> Get real-time metrics from all of your servers, apps and tools >>> in one place. >>> SourceForge users - Click here to start your Free Trial of Datadog now! >>> http://pubads.g.doubleclick.net/gampad/clk?id=241902991&iu=/4140 >>> _______________________________________________ >>> Csound-users mailing list >>> Csound-users@lists.sourceforge.net >>> https://lists.sourceforge.net/lists/listinfo/csound-users >>> Send bugs reports to >>> https://github.com/csound/csound/issues >>> Discussions of bugs and features can be posted here >> ------------------------------------------------------------------------------ >> Monitor Your Dynamic Infrastructure at Any Scale With Datadog! >> Get real-time metrics from all of your servers, apps and tools >> in one place. >> SourceForge users - Click here to start your Free Trial of Datadog now! >> http://pubads.g.doubleclick.net/gampad/clk?id=241902991&iu=/4140_______________________________________________ >> Csound-users mailing list >> Csound-users@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/csound-users >> Send bugs reports to >> https://github.com/csound/csound/issues >> Discussions of bugs and features can be posted here >> ------------------------------------------------------------------------------ >> Monitor Your Dynamic Infrastructure at Any Scale With Datadog! >> Get real-time metrics from all of your servers, apps and tools >> in one place. >> SourceForge users - Click here to start your Free Trial of Datadog now! >> http://pubads.g.doubleclick.net/gampad/clk?id=241902991&iu=/4140_______________________________________________ >> Csound-users mailing list >> Csound-users@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/csound-users >> Send bugs reports to >> https://github.com/csound/csound/issues >> Discussions of bugs and features can be posted here > > > ------------------------------------------------------------------------------ > _______________________________________________ > Csound-users mailing list > Csound-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/csound-users > Send bugs reports to > https://github.com/csound/csound/issues > Discussions of bugs and features can be posted here ------------------------------------------------------------------------------ _______________________________________________ Csound-users mailing list Csound-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/csound-users Send bugs reports to https://github.com/csound/csound/issues Discussions of bugs and features can be posted here