Processed pianos: bug in dynamics?

This forum is for discussing Reason. Questions, answers, ideas, and opinions... all apply.
User avatar
moggadeet
Posts: 89
Joined: 27 Jun 2018
Location: Stuttgart, Germany

Post 06 Dec 2020

Hi everyone,

since I got Reason 11 suite I switched from Radical Piano to Processed Pianos for practicing, I love the sound of the Grand D Cocktail preset, but I noticed that the dynamics are anything but smooth in Processed Pianos, whereas in Radical Piano the transition to other dynamic layers are seamless. In Processed Pianos you can clearly hear the transitions as I marked them, you can even see it in the bounced audio. The velocities are increasing linearly of course.

I do not know much about virtual pianos, but this is clearly a bug?
You do not have the required permissions to view the files attached to this post.

User avatar
Loque
Moderator
Posts: 8634
Joined: 28 Dec 2015

Post 06 Dec 2020

I tried to repeat the test you are doing and i clearly could here 3 layers of samples which does not sound "hypersampled" to me. The switching between the node velocities is not smooth, maybe there are not enough samples. Maybe not a bug,but "just" bad or not enough samples?

Make the sound dry, chose only 1 piano, and hear your test sound. You can hear 3 layers, nothing more.
:reason: 11, Win10 64Bit.

User avatar
moggadeet
Posts: 89
Joined: 27 Jun 2018
Location: Stuttgart, Germany

Post 06 Dec 2020

Loque wrote:
06 Dec 2020
I tried to repeat the test you are doing and i clearly could here 3 layers of samples which does not sound "hypersampled" to me. The switching between the node velocities is not smooth, maybe there are not enough samples. Maybe not a bug,but "just" bad or not enough samples?

Make the sound dry, chose only 1 piano, and hear your test sound. You can hear 3 layers, nothing more.
I see, I supposed there would be some care taken in the transitions. I do not think the samples are that bad, because I checked out the same model, Yamaha Grand D, inside Reason Pianos and it has 4 layers in NNXT, sounds smooth in the transitions. I strongly guess they used the same samples and packed them in another device. I tested all other pianos in Processed Pianos, same problem in the transitions.
You do not have the required permissions to view the files attached to this post.

User avatar
mcatalao
Posts: 1398
Joined: 17 Jan 2015

Post 06 Dec 2020

For me the problem is not even the dynamics.

There's a bad bug with pedalling crossing too.

TBH, i still prefer to use the original refills. If you can grab it, it's a better option imho.

DecafDreams
Posts: 159
Joined: 07 Oct 2020

Post 07 Dec 2020

moggadeet wrote:
06 Dec 2020
Hi everyone,

since I got Reason 11 suite I switched from Radical Piano to Processed Pianos for practicing, I love the sound of the Grand D Cocktail preset, but I noticed that the dynamics are anything but smooth in Processed Pianos, whereas in Radical Piano the transition to other dynamic layers are seamless. In Processed Pianos you can clearly hear the transitions as I marked them, you can even see it in the bounced audio. The velocities are increasing linearly of course.

I do not know much about virtual pianos, but this is clearly a bug?
This definitely looks like a bug. If I remember rightly the RDK RE is perfect for velocities vs the ReFill, so there should be no reason why the Piano RE would do this and, as mentioned, the old Reason Pianos ReFill works perfectly.

You should raise it with Reason Studios as a bug in the RE.

User avatar
moggadeet
Posts: 89
Joined: 27 Jun 2018
Location: Stuttgart, Germany

Post 08 Dec 2020

DecafDreams wrote:
07 Dec 2020
moggadeet wrote:
06 Dec 2020
Hi everyone,

since I got Reason 11 suite I switched from Radical Piano to Processed Pianos for practicing, I love the sound of the Grand D Cocktail preset, but I noticed that the dynamics are anything but smooth in Processed Pianos, whereas in Radical Piano the transition to other dynamic layers are seamless. In Processed Pianos you can clearly hear the transitions as I marked them, you can even see it in the bounced audio. The velocities are increasing linearly of course.

I do not know much about virtual pianos, but this is clearly a bug?
This definitely looks like a bug. If I remember rightly the RDK RE is perfect for velocities vs the ReFill, so there should be no reason why the Piano RE would do this and, as mentioned, the old Reason Pianos ReFill works perfectly.

You should raise it with Reason Studios as a bug in the RE.
Thanks for the opinions, I opened a ticket.

User avatar
moggadeet
Posts: 89
Joined: 27 Jun 2018
Location: Stuttgart, Germany

Post 09 Dec 2020

So I got the answer from Support:
It's true that the dynamics is not as great as in Radical Piano (or Reason Pianos).
The main purpose of processed Pianos was to have a great "pop piano" instrument, i.e. piano sounds that fitted in the mix, rather than be played on their own. So if you want superior dynamics, go for Radical Piano.
So Processed Pianos are kind of a collection of piano sounds, but they are not intended to be played on their own ... go figure.

Steedus
Posts: 417
Joined: 31 Aug 2015
Location: Melbourne, AU

Post 09 Dec 2020

Interesting answer. I don't play piano enough to really have noticed this, but considering the source material, that's a bit disappointing honestly.

User avatar
Reasonable man
Posts: 589
Joined: 14 Jul 2016

Post 09 Dec 2020

Its great for ambient music i think . Freq shifter, chorus, comb filter and send those to the tap delay and reverb ..with the dry signal turned all he way down. I think it is possible to do and create similar effects with a chain of dedicated insert effects ...but its alot of work!

As a clean piano that plays clean piano'ry' sounds .. tis pretty hopeless imo.

User avatar
Boombastix
Posts: 1930
Joined: 18 May 2018
Location: Bay Area, CA

Post 09 Dec 2020

Midi velocity = 127 rules !
Welcome to hyposampled pianos! :roll:

Reminds me about the old discussion about setting up that "RE with known buggs" (and errors) pinned thread...

(apologies if my sarcastic joke didn't roll off like water on a goose)
10% off at Waves with link: https://www.waves.com/r/6gh2b0
Disclaimer - I get 10% as well.

DecafDreams
Posts: 159
Joined: 07 Oct 2020

Post 10 Dec 2020

I guess either support just fobbed you off and they don't know for sure, or Processed Pianos is designed to be used at a fixed velocity then!

Perhaps Buddard at Robotic Bean can chip in as he often posts on these forums and from the back of the Processed Pianos RE it states it was a Robotic Beans job.

Buddard: Is the behaviour exhibited here to be expected or has something gone wrong somehow with the device and the samples aren't being crossfaded in terms of velocities properly?

User avatar
buddard
RE Developer
Posts: 936
Joined: 17 Jan 2015
Location: Stockholm

Post 12 Dec 2020

DecafDreams wrote:
10 Dec 2020
Buddard: Is the behaviour exhibited here to be expected or has something gone wrong somehow with the device and the samples aren't being crossfaded in terms of velocities properly?
Sorry, hadn't seen this topic or your PM. :-)

Something might be off in the transitions, I'm not sure? It's been about 18 months since we worked on this one, but I believe we thought it sounded alright back then... But apparently not, I guess, haha.

There are 4 velocity layers for each piano just like on the NN-XT version, and each layer has its own volume and velocity sensitivity setting. I think those settings (along with the crossover velocities) also differ a bit between each piano.

Most settings were pulled directly from the original NN-XT patches, but I think there were some differences in the velocity response between NN-XT and Gorilla Engine that we had to adjust for. I don't remember it very clearly now, so I'd have to take a closer look at the sources to be sure!

DecafDreams
Posts: 159
Joined: 07 Oct 2020

Post 15 Dec 2020

buddard wrote:
12 Dec 2020
DecafDreams wrote:
10 Dec 2020
Buddard: Is the behaviour exhibited here to be expected or has something gone wrong somehow with the device and the samples aren't being crossfaded in terms of velocities properly?
Sorry, hadn't seen this topic or your PM. :-)

Something might be off in the transitions, I'm not sure? It's been about 18 months since we worked on this one, but I believe we thought it sounded alright back then... But apparently not, I guess, haha.

There are 4 velocity layers for each piano just like on the NN-XT version, and each layer has its own volume and velocity sensitivity setting. I think those settings (along with the crossover velocities) also differ a bit between each piano.

Most settings were pulled directly from the original NN-XT patches, but I think there were some differences in the velocity response between NN-XT and Gorilla Engine that we had to adjust for. I don't remember it very clearly now, so I'd have to take a closer look at the sources to be sure!
It's no worries, thanks for taking the time to reply!

Does the Gorilla Engine allow for as seamless a layer transition as NN-XT? On the screenshot of the waveform for Reason Pianos ReFill, posted by the op, it's actually impossible to see where one of the 4 layers ends and the next one begins.

Just wondering if the fact the layers show more obviously in the waveform on Processed Pianos, is because of something more inherent in the way the Gorilla Engine works when handling sample layers, rather than a programming problem in the patch?

Anything you can do to look into this and let us know if it's a limitation rather than a bug, would be much appreciated!

User avatar
moggadeet
Posts: 89
Joined: 27 Jun 2018
Location: Stuttgart, Germany

Post 15 Dec 2020

DecafDreams wrote:
15 Dec 2020
buddard wrote:
12 Dec 2020


Sorry, hadn't seen this topic or your PM. :-)

Something might be off in the transitions, I'm not sure? It's been about 18 months since we worked on this one, but I believe we thought it sounded alright back then... But apparently not, I guess, haha.

There are 4 velocity layers for each piano just like on the NN-XT version, and each layer has its own volume and velocity sensitivity setting. I think those settings (along with the crossover velocities) also differ a bit between each piano.

Most settings were pulled directly from the original NN-XT patches, but I think there were some differences in the velocity response between NN-XT and Gorilla Engine that we had to adjust for. I don't remember it very clearly now, so I'd have to take a closer look at the sources to be sure!
It's no worries, thanks for taking the time to reply!

Does the Gorilla Engine allow for as seamless a layer transition as NN-XT? On the screenshot of the waveform for Reason Pianos ReFill, posted by the op, it's actually impossible to see where one of the 4 layers ends and the next one begins.

Just wondering if the fact the layers show more obviously in the waveform on Processed Pianos, is because of something more inherent in the way the Gorilla Engine works when handling sample layers, rather than a programming problem in the patch?

Anything you can do to look into this and let us know if it's a limitation rather than a bug, would be much appreciated!
Yes, it would be interesting to know how does the engine mix between the layers.

DecafDreams
Posts: 159
Joined: 07 Oct 2020

Post 21 Dec 2020

buddard wrote:
12 Dec 2020
I don't remember it very clearly now, so I'd have to take a closer look at the sources to be sure!
Hey Buddard, did you get chance to look into this and if so is there anything to report?

User avatar
buddard
RE Developer
Posts: 936
Joined: 17 Jan 2015
Location: Stockholm

Post 21 Dec 2020

DecafDreams wrote:
21 Dec 2020
buddard wrote:
12 Dec 2020
I don't remember it very clearly now, so I'd have to take a closer look at the sources to be sure!
Hey Buddard, did you get chance to look into this and if so is there anything to report?
No, I haven’t had time to do that yet, I’ve been busy with other work.
Hopefully I’ll find some time during the holidays!

DecafDreams
Posts: 159
Joined: 07 Oct 2020

Post 21 Jan 2021

Just noticed that Reason Pianos never did get pulled from the shop and is still available: https://www.reasonstudios.com/shop/prod ... on-pianos/

It seems to me like Processed Pianos isn't seen as a straight replacement like the Bass and Drums REs are for their ReFills counterparts...

DecafDreams
Posts: 159
Joined: 07 Oct 2020

Post 28 Jan 2021

DecafDreams wrote:
21 Jan 2021
Just noticed that Reason Pianos never did get pulled from the shop and is still available: https://www.reasonstudios.com/shop/prod ... on-pianos/

It seems to me like Processed Pianos isn't seen as a straight replacement like the Bass and Drums REs are for their ReFills counterparts...
Ah, it's gone now!

  • Information
  • Who is online

    Users browsing this forum: CommonCrawl [Bot], Jabba Bakes, selig, Trendiction [Bot] and 5 guests