Welcome to the Piano World Piano Forums Over 3 million posts about pianos, digital pianos, and all types of keyboard instruments. Over 100,000 members from around the world.
Join the World's Largest Community of Piano Lovers
(it's free)
It's Fun to Play the Piano ... Please Pass It On!
It sounds good to me too. I might just be able to hear the effect of looping in the final bar, but it's not distracting in any way. I also think the overall effect might not be that you would hear anything wrong with the individual notes, but rather that the sound is less spacious, or perhaps a bit fatiguing. Maybe the Cathedral reverb setting in the recording helps?
It will be interesting to try this out for myself, to see if it bothers me in any way while actually playing something.
Originally Posted by Kawai James
Originally Posted by bitnick
If I had a small gripe (with the CA 67 action) it would be that the feel of the let-off point was a bit uneven, but that was only noticeable when I depressed the keys slowly enough not to make a sound (while checking for string resonance).
This sensitivity can be adjusted using the 'Minimum Touch' parameter in the Virtual Technician menu. The default setting is 1 (e.g. notes with a velocity > 1 will produce a sound), however perhaps 5+ would be preferable for some players.
Well, I'm not sure when I would play silent notes anyway, so I don't think it's an issue for me, but it's nice to hear that it can be affected if it turns out that I'm wrong.
BTW, do you know if the CA17 uses the whole range of midi velocity values (Off, 1-127)? If you set Minimum Touch to 5, does that then compress the rest of the velocity range into 127-5=122 values (less resolution), or does it shift the range (possibly cutting off the top)? Or does the CA17 support MIDI "High resolution velocity prefix" for >16000 possible velocities?
Originally Posted by Kawai James
The CA97/CA67 and CA17 share the same HI-XL sound, however the CA97/CA67 feature additional 'Undamped String Resonance' and 'Cabinet Resonance' modelling that is not present on the CA17. There are also additional Virtual Technician parameters on the CA97/CA67 that are not present on the CA17.
What physical mechanism is modelled by 'Cabinet Resonance'?
Interesting about the Undamped String Resonance. I guess this is the always-on resonance of the higher notes that are undamped on a real piano? But the CA17 does support string resonance (if I'm not mistaken). Am I correct in that one should be able to get that Undamped String Resonance by just putting a weight on the top keys, i.e. the CA17 is deliberately modelled as having dampers on all keys (to make the CA67/CA97 more attractive, perhaps)?
It sounds good to me too. I might just be able to hear the effect of looping in the final bar, but it's not distracting in any way. I also think the overall effect might not be that you would hear anything wrong with the individual notes, but rather that the sound is less spacious, or perhaps a bit fatiguing. Maybe the Cathedral reverb setting in the recording helps?
It will be interesting to try this out for myself, to see if it bothers me in any way while actually playing something.
Yes, playing actual music is a good way to check the sound of any piano - be it acoustic or digital.
Originally Posted by bitnick
BTW, do you know if the CA17 uses the whole range of midi velocity values (Off, 1-127)?
Yes, I believe all Kawai instruments utilise the full range of MIDI velocity values.
Originally Posted by bitnick
If you set Minimum Touch to 5, does that then compress the rest of the velocity range into 127-5=122 values (less resolution), or does it shift the range (possibly cutting off the top)?
I'm afraid I do not know, however I expect the former is true - I will have to check this point.
Originally Posted by bitnick
Or does the CA17 support MIDI "High resolution velocity prefix" for >16000 possible velocities?
No, the CA17 does not support this prefix.
Originally Posted by bitnick
What physical mechanism is modelled by 'Cabinet Resonance'?
As explained on page 90 of the CA97/CA67 owner's manual:
Originally Posted by bitnick
Interesting about the Undamped String Resonance. I guess this is the always-on resonance of the higher notes that are undamped on a real piano?
Correct, as explained on page 89 of the CA97/CA67 owner's manual:
Originally Posted by bitnick
But the CA17 does support string resonance (if I'm not mistaken).
Yes, as does the CA97/CA67, and a number of other models, but this is separate from the 'Undamped String Resonance' modelling unique to the higher specification CA97/CA67.
Originally Posted by bitnick
Am I correct in that one should be able to get that Undamped String Resonance by just putting a weight on the top keys, i.e. the CA17 is deliberately modelled as having dampers on all keys (to make the CA67/CA97 more attractive, perhaps)?
Interesting assumption, but I don't believe this is the case.
For your information, here is a slide from the CA97/CA67 sales manual prepared for dealers. I've mosaiced the 'sales points', but kept the blue part below, as I rather like the analogy.
First, I'd like to say thank you for taking time to answer these questions.
I can't say I really understand why the cabinet resonances (vibrations) should be modelled separately - shouldn't this sound be recorded in the samples? Can you say anything on what this actually entails (is it panning, equalizing, extra sound added, ...?).
Originally Posted by "Kawai James"
Interesting assumption, but I don't believe this is the case.
Can you elaborate on this? I am a software developer, and one basic thing in software is that code should not be duplicated - i.e., if you have a way of adding string resonance that works well for multiple notes (as I guess it must if it is to be of any use), it would be beneficial to use that same code for all string resonance (this gives less bugs, less development cost, and much easier code maintenance). One reason to have a separate model for the top strings might be to save polyphony slots - I guess each resonating string takes up one of these? However, with 192 note polyphony available, that seems rather pointless(?).
I hope I'm not being a PITA - I actually think this is very interesting!
Edit: I did look at/read the excerpt from the sales manual, and found the information about resonances all being modelled rather than sampled interesting. But it doesn't really answer my questions about these resonances.
I can't say I really understand why the cabinet resonances (vibrations) should be modelled separately - shouldn't this sound be recorded in the samples?
Yes and no.
Sampling will capture a certain degree of the piano's natural resonance (depending on the micing positions used), however we're primarily interested in capturing the main 'voice' of the instrument. If we were to play a piece of music using just these individual samples alone, the sound would be an accurate representation of the grand piano's main 'voice', however we may feel that something is 'missing', and that the piano sound is static or lifeless. This is because sampling single notes does not capture the rich variety of resonances produced by an acoustic piano. In order to overcome this, the 'missing' resonances are calculated by the instrument dynamically (i.e. depending on the notes played), and added back to the captured main 'voice'. This combination of sampling and modelling is arguably the best way to capture the true 'voice' of a piano, while also giving the captured sound a living, breathing, acoustic-like quality.
Originally Posted by bitnick
Can you say anything on what this actually entails (is it panning, equalizing, extra sound added, ...?).
As noted above, the resonances are additional sounds, generated dynamically.
Originally Posted by bitnick
I am a software developer, and one basic thing in software is that code should not be duplicated - i.e., if you have a way of adding string resonance that works well for multiple notes (as I guess it must if it is to be of any use), it would be beneficial to use that same code for all string resonance (this gives less bugs, less development cost, and much easier code maintenance).
Unfortunately I am not a software developer, however I expect my colleagues responsible for software development share your philosophies regarding efficient, easily maintainable code, and have their reasons for creating separate 'String Resonance' and 'Undamped String Resonance' parameters.
Originally Posted by bitnick
One reason to have a separate model for the top strings might be to save polyphony slots - I guess each resonating string takes up one of these? However, with 192 note polyphony available, that seems rather pointless(?).
I believe these resonances are generated by the instrument's DSP and do not impact polyphony.
James, say, looking at the CA17 user manual, I see no instructions for firmware updates. For CA67 and 97 there have been firmware updates already. Also, there are no updates listed in the download section for CA17.
Is this not possible for the CA17 or only by a Kawai technician?
Kawai CN35. Daughter wanted a piano, so we got one. Now who'll learn faster? ;-)
Thanks for the video-link, James! The CA 17 really does sound fantastic (in the video - I haven't tried it myself yet)! I did try a Yamaha CLP-535 today, which costs about the same as the CA 17, and the Yamaha really sucks when it comes to both action and sound IMO (not even close to Kawai CN-24 even).
Also, I realise that I might be better served by some scientific text about the physics of piano sound rather than asking you about the inner implementations of the CA 17. Thanks for trying though!
BTW, does anyone know what the name of the piece is that's played starting at 5:00?
James, say, looking at the CA17 user manual, I see no instructions for firmware updates. For CA67 and 97 there have been firmware updates already. Also, there are no updates listed in the download section for CA17.
Is this not possible for the CA17 or only by a Kawai technician?
While it's technically possible to update the firmware of the CA17 (and other models), the procedure is rather less straight forward than with 'USB to Device' equipped instruments such as the CA97/CA67. The firmware must to be transmitted to the instrument via MIDI using playback software that does not strip-away important SysEx data. The process takes up to 10 minutes, and - unlike the USB stick method - can result in a non-functioning instrument if the firmware transfer is interrupted prior to completion.
For this reason, we do not tend to make such firmware updates publicly available, however technicians may be supplied with this data if an issue reported by a customer can be resolved by a firmware update.
Does anyone know if the HI-XL sampling is used for all the sounds ons the CA-17, or just the EX and SK-EX? In other words: does the "mellow grand" for example have lower quality samples than te EX of SK-EX?
My first footage (and audio) from my CA 17. These where the songs for my first piano exam. I'm in my first year of piano, so these are very simple pieces, but anyway here it goes:
Assuming that this was recorded from the CA17's headphone jacks, am I correct in thinking that SHS (Spatial Headphone Sound) is enabled? Did you make any other adjustments, other than selecting a bright voicing? The sound is very natural to my ears.
Yes, you are correct that this is recorded through the headphone outputs of the CA17. The SHS setting is set to the default setting: "Normal". I have the voicing set to bright and the reverb to "lounge" or "small hall", I'm not sure.
![[Linked Image]](http://i60.tinypic.com/rc1tu1.png)
![[Linked Image]](http://i61.tinypic.com/2zs7m2r.png)
