working in IRs
-
- KVRian
- 539 posts since 25 Aug, 2006
Hi!
I'm just Working on IRs for acoustic instruments in a very experimental way.
I could need some Automation / Batchprocessing for comparing IRs:
I have made hundreds of experimental IRs from acoustic instruments but I'd just like to compare their 'naked' Room(reflections, body Vibration, damping...) aspect, not the overall frequency response .
I'd just like to (at least narrowing, not copying) the frequency response to one reference file, so that the Rest is EQ work on the same Basis, so that when switching IRs I do not necessarily have to change EQ settings when zapping through possibilities.
Is there a batch processing possibility to load one File and let freeform EQ analyse overall Frequency response and then adopt a whole bunch of files with different frequency responses to that Prototype as batchprocessing, so that every single file will be analyzed and then the overall frequency response made to fit (not perfectly, just the right direction) the Freqresponse of the reference file?
Thanks
Martin
I'm just Working on IRs for acoustic instruments in a very experimental way.
I could need some Automation / Batchprocessing for comparing IRs:
I have made hundreds of experimental IRs from acoustic instruments but I'd just like to compare their 'naked' Room(reflections, body Vibration, damping...) aspect, not the overall frequency response .
I'd just like to (at least narrowing, not copying) the frequency response to one reference file, so that the Rest is EQ work on the same Basis, so that when switching IRs I do not necessarily have to change EQ settings when zapping through possibilities.
Is there a batch processing possibility to load one File and let freeform EQ analyse overall Frequency response and then adopt a whole bunch of files with different frequency responses to that Prototype as batchprocessing, so that every single file will be analyzed and then the overall frequency response made to fit (not perfectly, just the right direction) the Freqresponse of the reference file?
Thanks
Martin
-
- KVRian
- Topic Starter
- 539 posts since 25 Aug, 2006
Trust me, I know what I'm doing. No, don't trust, that I know WHAT I'm doing, but trust me that it sounds great
I said "experimental", so I do not generate these impulses in a classic way and need to unify them afterwards. Same Instrument = same frequency response (which is not there from the beginning in my method). From there on I'll work with EQ presets, but it would be interesting to have them about compatible, not really, but in the right direction.
What I described to be usefull for me as a batch processing is what I am doing manually with LOTs of time invested, but the results are great. It doesn't even count for me, if it relys on corret theory. They just sound great.
What I described to be usefull for me as a batch processing is what I am doing manually with LOTs of time invested, but the results are great. It doesn't even count for me, if it relys on corret theory. They just sound great.
-
- KVRian
- Topic Starter
- 539 posts since 25 Aug, 2006
... but could you explain, why it sounds different, when 'baking' EQ settings into an impulse compared to using that EQ settings after the signal has been processed by the IR?
Do I have to take another IR-'shot' with Mconvolution/EZ, to get the correct sound when I EQ a Signal processed by IR in Mconvolution? Is there another possibility? I used deconvolver by voxengo, but that does not lead to good results with my IR sources, it would only work if I used a dirac or maby a gunshot...
It's ok to use EQ, but I could use higher quality settings to bake EQ settings into the impulses (because I need convolution in realtime with picked up acoustic instruments)... better than realtime. What' sthe trick to change impulses to sound like being equed after processing the signal?
Do I have to take another IR-'shot' with Mconvolution/EZ, to get the correct sound when I EQ a Signal processed by IR in Mconvolution? Is there another possibility? I used deconvolver by voxengo, but that does not lead to good results with my IR sources, it would only work if I used a dirac or maby a gunshot...
It's ok to use EQ, but I could use higher quality settings to bake EQ settings into the impulses (because I need convolution in realtime with picked up acoustic instruments)... better than realtime. What' sthe trick to change impulses to sound like being equed after processing the signal?
-
MeldaProduction MeldaProduction https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=176122
- KVRAF
- 14339 posts since 15 Mar, 2008 from Czech republic
Too many questions, some of which I'm not sure if are questions
. Anyways there's currently no tool that would batch process files by analyzing them, changing the EQ curves and processing the output, then saving it. You'd need to do that manually. But to be honest, I don't think that's what you want - you said you want that to compare the IRs, but that wouldn't really help you too much. There a whole lot of stuff the IR contains besides that... like stereo image, transient modification and general phase alteration etc... Just making the same overall frequency levels would get you much closer...
-
- KVRian
- Topic Starter
- 539 posts since 25 Aug, 2006
I do it manually now, it would just save a lot of time with some batch-possibilities... Right now I don't want to open my little processing secrets, because for me it really sounds like a live-Cello/Violin... Revolution. I have some sounds here on my Cello / Violin which completely blew my mind and I want to go even deeper.
- KVRist
- 495 posts since 18 Aug, 2006 from Italy
"Cello/Violin"… So, if I understand correctly, you are trying to make instrumental hybrids by using a source-filter model, representing the filter part (the body vibration) with an IR and producing the sounds live, by using MConvolution? Do I understand correctly?
If so, I like this kind of experiments. I used to do such things with MarcoHack, a very old and now abandoned Mac software (Mac OS 8 or earlier!) and now I have a project to go on with such experiments with another program (IRCAM Trax SF, which means "Source/Filter"). I had never thought to do so with IRs and MConvolution and I am interested in the results.
Of course, if I understand correctly what you are doing (which is not granted)…
On your question: an EQ setting is usually static, while IRs are not static, they have a time dimension, so this is (mainly) why they sound different.
If so, I like this kind of experiments. I used to do such things with MarcoHack, a very old and now abandoned Mac software (Mac OS 8 or earlier!) and now I have a project to go on with such experiments with another program (IRCAM Trax SF, which means "Source/Filter"). I had never thought to do so with IRs and MConvolution and I am interested in the results.
Of course, if I understand correctly what you are doing (which is not granted)…
On your question: an EQ setting is usually static, while IRs are not static, they have a time dimension, so this is (mainly) why they sound different.
-
MeldaProduction MeldaProduction https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=176122
- KVRAF
- 14339 posts since 15 Mar, 2008 from Czech republic
Ha! Like a year or 2 ago a guy came to me telling me he did exactly that!mccy wrote: Sat Dec 01, 2018 8:55 pm I do it manually now, it would just save a lot of time with some batch-possibilities... Right now I don't want to open my little processing secrets, because for me it really sounds like a live-Cello/Violin... Revolution. I have some sounds here on my Cello / Violin which completely blew my mind and I want to go even deeper.
Anyways if it is so awesome, let me know, perhaps we can turn that into a plugin
-
Chandlerhimself Chandlerhimself https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=318799
- KVRAF
- 1823 posts since 19 Dec, 2013 from Japan
I’m interested in this too. Would you mind sharing the IRs. I have an idea for MSF, but I don’t have any violin, viola, etc IRs.
My Youtube page https://www.youtube.com/user/GuitarChandler
-
- KVRian
- Topic Starter
- 539 posts since 25 Aug, 2006
These IRs are so highly individual, because of the pickupposition & sound of the cello itself...
I had a serious desease so all I was doing was a bit slower in the last months. I will keep on going with this stuff and let you know then...
I had a serious desease so all I was doing was a bit slower in the last months. I will keep on going with this stuff and let you know then...
