Spidericemidas needs your help with Kronecker in AUM and other hosts

reasOne

wow that recording you did really does sound crazy different

brambos

@Spidericemidas said:

@brambos said:
Probably something samplerate-related. Try it with NS2: by playing and exporting. If the exported version sounds very different then the AU has some things hardcoded which should be variable (such as filter sample frequencies, etc).

Hey, thanks for dropping in. With Kronecker loaded into Cubasis2 on a totally dry channel, I get a sound which is very close to the standalone sound, played live and mixed down. So different from the sound in a dry channel in AUM. I will also try it in NS2. Cheers.

The difference in Ns2 should be very pronounced, because it plays in 48Khz and renders for exports in 192khz (off the top of my head).

Spidericemidas

@Soundscaper said:
Kronecker is a pretty weird synth ... particles driving emitters, resonators, FM synths popping up in odd places - all with filters in them and the like.

I've found a few resonator and string modelling synths to be highly variable when it comes to presets ... often don't "play true" from one session to the next to my ear... play back a bit dead and flat. The Karplus-Strong string synthesis models work on a filtered delay line which could point to where the gremlins are too - filters all over the place .

There's a lot of odd stuff going on in these Icegear gadgets under the hood - deceptively normal ...

Yeah, my gut feeling is it’s something to do with the Particle Emitter thing. It’s the only thing I can see that is unique to Kronecker, seeing as I have never ever experienced such a wild and obvious difference between any preset in AUM, standalone and Cubasis with any of my other 70+ synths on my iPad. It is a very unusual synth compared to all my other ones. I do like it a lot. But I need it to behave consistently in all my hosts to be useful.

Spidericemidas

I have emailed IceGear a link to this thread in hope they could shed some light on what is happening. I would love to make a soundbank for Kronecker, it’s an interesting synth, but at the moment it is too variable in different hosts.

Spidericemidas

@reasOne said:
wow that recording you did really does sound crazy different

Ikr!! 😅😅

espiegel123

@Spidericemidas said:
@Philandering_Bastard @Fruitbat1919 @Carnbot @craftycurate

Here’s a link to a short audio recording. One made in AUM. The other made in Cubasis2. For an A-B comparison. The first muffled sound is in AUM. The second sharper sound with attack is in Cubasis2. Both recordings completely dry, only sound coming from Kronecker. Channel strip and EQ bypassed in Cubasis2 so as not to influence the sound. You wouldn’t believe these two short sequences are the same patch! 😲

The second part of the recording, from Cubasis2, sounds closer to what I hear in standalone and is how I designed the sound to be...plucky and snappy. The first part of the recording is from AUM and sounds absolutely nothing like the behaviour of the same patch in Cubasis2. It’s lost all its attack and snappiness and sounds muffled and soft. A completely different patch. This A-B comparison is pretty crazy! 🤔

Thanks for all your thoughts and ideas on this, guys! Much appreciated!

https://www.dropbox.com/s/3a80skyam2v0qin/ - Kronecker AB compare (1).wav?dl=0

The two versions sound radically different.

Spidericemidas

Have just received very quick reply from IceGear.

“Hello,

I apologise for the inconvenience.

I have confirmed the problem you pointed out. The problem is probably in the audio buffer size. Could you set the buffer size in the standalone app to the same value as specified in AUM and try again?

Best regards,
Satoshi Taneda”

Philandering_Bastard

@Spidericemidas said:
@Philandering_Bastard @Fruitbat1919 @Carnbot @craftycurate

Here’s a link to a short audio recording. One made in AUM. The other made in Cubasis2. For an A-B comparison. The first muffled sound is in AUM. The second sharper sound with attack is in Cubasis2. Both recordings completely dry, only sound coming from Kronecker. Channel strip and EQ bypassed in Cubasis2 so as not to influence the sound. You wouldn’t believe these two short sequences are the same patch! 😲

The second part of the recording, from Cubasis2, sounds closer to what I hear in standalone and is how I designed the sound to be...plucky and snappy. The first part of the recording is from AUM and sounds absolutely nothing like the behaviour of the same patch in Cubasis2. It’s lost all its attack and snappiness and sounds muffled and soft. A completely different patch. This A-B comparison is pretty crazy! 🤔

Thanks for all your thoughts and ideas on this, guys! Much appreciated!

https://www.dropbox.com/s/3a80skyam2v0qin/ - Kronecker AB compare (1).wav?dl=0

Holy mackerel, that is a drastic difference. Just went back to Kronecker in AUM and it sounds like the stand alone or your Cubasis bit.

Wouldn’t an issue that is related to the sample rate primarily affect the pitch? Your example sounds more as if it were heavily EQd.

Spidericemidas

@espiegel123 @brambos @Carnbot @craftycurate @reasOne @Philandering_Bastard @Fruitbat1919 @Soundscaper

Ok! Problem solved with a very quick response from IceGear! The buffer size in Kronecker and AUM simply needed to be the same. Kronecker was initially in 512 and AUM was left in 1024 from a previous project. I changed AUM to 512 to match Kronecker, and I now have EXACTLY the same sound and behaviour from this patch in AUM as I designed it in standalone mode.

Now I can finally sleep! (1am UK time!) and continue my soundbank for Kronecker tomorrow. Relief!! Thanks for all your time and participation in this thread, guys! Much appreciated!

I have learned something this evening! 😅👍

Philandering_Bastard

@Spidericemidas said:
Have just received very quick reply from IceGear.

“Hello,

I apologise for the inconvenience.

I have confirmed the problem you pointed out. The problem is probably in the audio buffer size. Could you set the buffer size in the standalone app to the same value as specified in AUM and try again?

Best regards,
Satoshi Taneda”

Buffer at 256 both stand alone and AUM. Little if any difference.

espiegel123

@Spidericemidas said:
@espiegel123 @brambos @Carnbot @craftycurate @reasOne @Philandering_Bastard @Fruitbat1919 @Soundscaper

Ok! Problem solved with a very quick response from IceGear! The buffer size in Kronecker and AUM simply needed to be the same. Kronecker was initially in 512 and AUM was left in 1024 from a previous project. I changed AUM to 512 to match Kronecker, and I now have EXACTLY the same sound and behaviour from this patch in AUM as I designed it in standalone mode.

Now I can finally sleep! (1am UK time!) and continue my soundbank for Kronecker tomorrow. Relief!! Thanks for all your time and participation in this thread, guys! Much appreciated!

I have learned something this evening! 😅👍

Nice when a mystery is solved!

craftycurate

@Spidericemidas said:
@espiegel123 @brambos @Carnbot @craftycurate @reasOne @Philandering_Bastard @Fruitbat1919 @Soundscaper

Ok! Problem solved with a very quick response from IceGear! The buffer size in Kronecker and AUM simply needed to be the same. Kronecker was initially in 512 and AUM was left in 1024 from a previous project. I changed AUM to 512 to match Kronecker, and I now have EXACTLY the same sound and behaviour from this patch in AUM as I designed it in standalone mode.

Now I can finally sleep! (1am UK time!) and continue my soundbank for Kronecker tomorrow. Relief!! Thanks for all your time and participation in this thread, guys! Much appreciated!

I have learned something this evening! 😅👍

Get iiiiinnnn!! Other apps cause problems with mismatched buffer sizes (e.g. Wotja crackles like crazy).

Philandering_Bastard

Really interesting. Mismatched buffer would have to been one of the last places I would have looked.

[Deleted User]

And I would have got away with it too if it wasn’t for you pesky kids!

brambos

Errmmm.. sound should be the same regardless of buffersize. Or are you going to ask all your users to conform to a fixed buffersize? Some would say this is a bug in the AU

Spidericemidas

@brambos said:
Errmmm.. sound should be the same regardless of buffersize. Or are you going to ask all your users to conform to a fixed buffersize? Some would say this is a bug in the AU

It is a very interesting outcome. Something I have never come across so far with any of the other 70+ synths on my iPad. Obviously I will suggest to users that for best results it is necessary to have matching buffer sizes. In this situation, the synth is still very useable in a host for sampling or running in smaller AUM projects. In Cubasis, tracks can be rendered to audio for freeing up space for more AU instances as a workaround. However, the good news is, Satoshi from IceGear got back to me very quickly last night and said that he will look into this further and see if he can rectify the issue in Kronecker. Interesting stuff! Many thanks for your input here. 🙏

Philandering_Bastard

@brambos said:
Errmmm.. sound should be the same regardless of buffersize. Or are you going to ask all your users to conform to a fixed buffersize? Some would say this is a bug in the AU

It is strange. From 64 to 1024 buffer sizes the sounds becomes increasingly dull.