Topic: Preset demos sound way better than my own MIDI files

Hello everyone,

I just noticed something that I can’t find any reason for: The MIDI demo files by Modartt sound way better than my own MIDI files. I absolutely love the demo file for the Bechstein DG Warm preset but I can hardly stand the sound of my own recording of Chopin Op 69 No 2. And yes, the pitch of the melody in the Chopin piece is higher than in the demo file but I don’t think that this is the reason for the lower sound quality. I’m sure that it’s not about my playing either but the pianos themselves seem to sound different, no matter if a note is played with a high or low velocity. Would it be possible that my stage piano sends MIDI data that make the files different from those by Modartt? I use a Kawai MP10 which I connect to my iPhone via a powered USB hub. No interface involved. Any help is highly appreciated.

Re: Preset demos sound way better than my own MIDI files

Did your keys velocity is ok? When in the midi view, have you 127 when you hit hard the key and near 0 when you slowly touch the key?
Maybe it's a bad setting of yout keyboard or a bad velocity curve?
For example, I have an old Yamaha DX7 and it can't send a note harder than 100, so it's never FF without tweaking the velocity curve.

Re: Preset demos sound way better than my own MIDI files

Hansz wrote:

Did your keys velocity is ok? When in the midi view, have you 127 when you hit hard the key and near 0 when you slowly touch the key?
Maybe it's a bad setting of yout keyboard or a bad velocity curve?
For example, I have an old Yamaha DX7 and it can't send a note harder than 100, so it's never FF without tweaking the velocity curve.

I think this might be the case but something that I don’t understand is that even if I don’t play with the highest possible velocity some of the notes already start to sound rather harsh. It probably has nothing to do with the hammer hardness either because in the demo track the velocity is sometimes even higher and it doesn’t sound harsh at all. Nevertheless I will try to experiment with the velocity curve. Something I did notice about the demo file is that the variation in velocity is much higher than in my MIDI files. I might have to pay more attention to the very fine details of my playing because Pianoteq would react with more sensitivity then but I think it’s definitely worth trying. I will experiment with that tomorrow because it’s almost midnight here. Without changing the velocity curve it feels kind of similar to my acoustic upright. Anyway, I’ll try to make some adjustments to the velocity curve tomorrow and then I’ll see how it works. Thanks!

Re: Preset demos sound way better than my own MIDI files

Zeng Hua wrote:
Hansz wrote:

Did your keys velocity is ok? When in the midi view, have you 127 when you hit hard the key and near 0 when you slowly touch the key?
Maybe it's a bad setting of yout keyboard or a bad velocity curve?
For example, I have an old Yamaha DX7 and it can't send a note harder than 100, so it's never FF without tweaking the velocity curve.

I think this might be the case but something that I don’t understand is that even if I don’t play with the highest possible velocity some of the notes already start to sound rather harsh. It probably has nothing to do with the hammer hardness either because in the demo track the velocity is sometimes even higher and it doesn’t sound harsh at all. Nevertheless I will try to experiment with the velocity curve. Something I did notice about the demo file is that the variation in velocity is much higher than in my MIDI files. I might have to pay more attention to the very fine details of my playing because Pianoteq would react with more sensitivity then but I think it’s definitely worth trying. I will experiment with that tomorrow because it’s almost midnight here. Without changing the velocity curve it feels kind of similar to my acoustic upright. Anyway, I’ll try to make some adjustments to the velocity curve tomorrow and then I’ll see how it works. Thanks!

Changing the velocity curve didn’t work for me. However, I did find a preset now that I also like for my own recordings: The Bechstein DG Prelude. Thanks again for your help!

Re: Preset demos sound way better than my own MIDI files

The response of Pianoteq varies a lot depending on the keyboard and as correctly pointed out also depending on the velocity curve selected, that's why it's always best to do a couple of times the calibration process in Pianoteq which allows you to create a personal preset for your keyboard and for the style you're playing even if it is just for having an idea of what it should look like depending on the response.
Another thing that is crucial but often overlooked is the Dynamics slider which acts as a sort of natural compressor increasing or decreasing the db range between pp and ff. Whenever I wasn't able to actually craft a preset with velocity curves or calibration, using the slider has always "fixed" the problem somehow. It's worth giving it a try imho.

Last edited by Chopin87 (Yesterday 14:22)
"And live to be the show and gaze o' the time."  (William Shakespeare)

Re: Preset demos sound way better than my own MIDI files

"Un autre élément crucial, mais souvent négligé, est le curseur de dynamique, qui agit comme un compresseur naturel, augmentant ou diminuant la plage de décibels entre pp et ff"

je me permets de rebondir sur cette remarque qui me semble assez importante quant aux réglages ayant trait à la dynamique, en général, d'origine, le réglage est sur 40db,
certains d'entre vous utilisent-ils des réglages plus étendus?

Re: Preset demos sound way better than my own MIDI files

bernard wrote:

"Un autre élément crucial, mais souvent négligé, est le curseur de dynamique, qui agit comme un compresseur naturel, augmentant ou diminuant la plage de décibels entre pp et ff"

je me permets de rebondir sur cette remarque qui me semble assez importante quant aux réglages ayant trait à la dynamique, en général, d'origine, le réglage est sur 40db,
certains d'entre vous utilisent-ils des réglages plus étendus?

Yes, for certain pieces I usually raise the level beyond the default values. Or if I want more control over the lower dynamics.

"And live to be the show and gaze o' the time."  (William Shakespeare)

Re: Preset demos sound way better than my own MIDI files

I have the same issue not only with the Pianoteq demos - especially the showcase recordings on modartt.com, but also with the inbuild demo songs of my digital piano.

Perhaps one has to be a really, really good pianist to achieve comparable results?

Be it as it may, regarding Pianoteq, imho it would help a lot if each recording setup and the preset parameters were laid open.
Or if the corresponding fxp-files would be made available (i don't mean the inbuild mini recordings of the presets).

Your thoughts?

Music was my first love. And it will be my last. Music of the future. And music of the past (John Miles)

Re: Preset demos sound way better than my own MIDI files

Chopin87 wrote:

The response of Pianoteq varies a lot depending on the keyboard and as correctly pointed out also depending on the velocity curve selected, that's why it's always best to do a couple of times the calibration process in Pianoteq which allows you to create a personal preset for your keyboard and for the style you're playing even if it is just for having an idea of what it should look like depending on the response.
Another thing that is crucial but often overlooked is the Dynamics slider which acts as a sort of natural compressor increasing or decreasing the db range between pp and ff. Whenever I wasn't able to actually craft a preset with velocity curves or calibration, using the slider has always "fixed" the problem somehow. It's worth giving it a try imho.

You may have a point there but I think that the demo MIDI files are played back with the default dynamic range. But I do agree – dynamics are very important and I sometimes increase the dynamic range up to 50 dB because I read somewhere that this corresponds with the dynamic range of an acoustic grand. Something that confuses me a lot is that on one hand the Chopin piece with a lot of high notes doesn’t sound that good but on the other hand I also tried another piece with a lot of midrange notes (so it’s kind of similar to the demo track, regarding the tonal range) and that one doesn’t sound good either. But the demo track does! My impression is that the tonal character is different, however, when I look at the MIDI data I can’t see anything that makes the MIDI tracks different from each other in the kind of data that are sent to Pianoteq.

Re: Preset demos sound way better than my own MIDI files

Pboy wrote:

I have the same issue not only with the Pianoteq demos - especially the showcase recordings on modartt.com, but also with the inbuild demo songs of my digital piano.

Perhaps one has to be a really, really good pianist to achieve comparable results?

Be it as it may, regarding Pianoteq, imho it would help a lot if each recording setup and the preset parameters were laid open.
Or if the corresponding fxp-files would be made available (i don't mean the inbuild mini recordings of the presets).

Your thoughts?

I actually suspect that they use modified presets for the MIDI demo files that can be played back inside Pianoteq (if they did that it should be visible easily on the UI) and I don’t know how the demo tracks on the Modartt website were made in terms of the setup etc. But since the MIDI files that come with Pianoteq sound amazing (at least for some of the presets) I can’t see any reason why my own MIDI files sound so different in terms of the piano sound. I double checked my Chopin file and noticed that I actually did cover quite a big velocity range but I also noticed that my midrange notes were relatively quiet. So I made a new recording focusing on playing the midrange notes a little louder. The sound did improve a little bit but I still wasn’t satisfied. And as I mentioned it doesn’t sound good if I play rather in the middle of the keyboard (stage piano with weighted keys) either. I do think that playing skills are crucial but nonetheless I’m quite sure that an acoustic instrument would behave differently – at least it would be more forgiving and it would never sound artificial in that way. That being said I had about six years of classical piano lessons and I played on a Yamaha grand and for concerts even on a Steinway. However, I have to admit that since I stopped taking lessons my technical playing skills got worse. So this will certainly have a negative impact on the sound but I’m sure that that’s not the only problem.

Re: Preset demos sound way better than my own MIDI files

Zeng Hua wrote:
Pboy wrote:

I have the same issue not only with the Pianoteq demos - especially the showcase recordings on modartt.com, but also with the inbuild demo songs of my digital piano.

Perhaps one has to be a really, really good pianist to achieve comparable results?

Be it as it may, regarding Pianoteq, imho it would help a lot if each recording setup and the preset parameters were laid open.
Or if the corresponding fxp-files would be made available (i don't mean the inbuild mini recordings of the presets).

Your thoughts?

I actually suspect that they use modified presets for the MIDI demo files that can be played back inside Pianoteq (if they did that it should be visible easily on the UI) and I don’t know how the demo tracks on the Modartt website were made in terms of the setup etc. But since the MIDI files that come with Pianoteq sound amazing (at least for some of the presets) I can’t see any reason why my own MIDI files sound so different in terms of the piano sound. I double checked my Chopin file and noticed that I actually did cover quite a big velocity range but I also noticed that my midrange notes were relatively quiet. So I made a new recording focusing on playing the midrange notes a little louder. The sound did improve a little bit but I still wasn’t satisfied. And as I mentioned it doesn’t sound good if I play rather in the middle of the keyboard (stage piano with weighted keys) either. I do think that playing skills are crucial but nonetheless I’m quite sure that an acoustic instrument would behave differently – at least it would be more forgiving and it would never sound artificial in that way. That being said I had about six years of classical piano lessons and I played on a Yamaha grand and for concerts even on a Steinway. However, I have to admit that since I stopped taking lessons my technical playing skills got worse. So this will certainly have a negative impact on the sound but I’m sure that that’s not the only problem.

It’s in Modartt’s best interests to tell everyone how to make their software sound its best.

Why would they hide that?

Re: Preset demos sound way better than my own MIDI files

dikrek wrote:
Zeng Hua wrote:
Pboy wrote:

I have the same issue not only with the Pianoteq demos - especially the showcase recordings on modartt.com, but also with the inbuild demo songs of my digital piano.

Perhaps one has to be a really, really good pianist to achieve comparable results?

Be it as it may, regarding Pianoteq, imho it would help a lot if each recording setup and the preset parameters were laid open.
Or if the corresponding fxp-files would be made available (i don't mean the inbuild mini recordings of the presets).

Your thoughts?

I actually suspect that they use modified presets for the MIDI demo files that can be played back inside Pianoteq (if they did that it should be visible easily on the UI) and I don’t know how the demo tracks on the Modartt website were made in terms of the setup etc. But since the MIDI files that come with Pianoteq sound amazing (at least for some of the presets) I can’t see any reason why my own MIDI files sound so different in terms of the piano sound. I double checked my Chopin file and noticed that I actually did cover quite a big velocity range but I also noticed that my midrange notes were relatively quiet. So I made a new recording focusing on playing the midrange notes a little louder. The sound did improve a little bit but I still wasn’t satisfied. And as I mentioned it doesn’t sound good if I play rather in the middle of the keyboard (stage piano with weighted keys) either. I do think that playing skills are crucial but nonetheless I’m quite sure that an acoustic instrument would behave differently – at least it would be more forgiving and it would never sound artificial in that way. That being said I had about six years of classical piano lessons and I played on a Yamaha grand and for concerts even on a Steinway. However, I have to admit that since I stopped taking lessons my technical playing skills got worse. So this will certainly have a negative impact on the sound but I’m sure that that’s not the only problem.

It’s in Modartt’s best interests to tell everyone how to make their software sound its best.

Why would they hide that?

That’s true…

Re: Preset demos sound way better than my own MIDI files

Just to clarify, there are two kind of Pianoteq demos:

- those hosted on our website, for example https://www.modartt.com/modeld. The preset used for recording the demos is displayed on the player, and unless it is specified "(customized)", the demos were recorded with the built in presets without any alteration. For example, among the 7 classical demos of the New York Steinway Model D, 6 are using the default presets, and one (the second demo) is using a customized preset;

- those embedded in Pianoteq, accessible in the presets manager. In that case, as you are playing live the embedded MIDI file, you are hearing the result produced by the preset which is loaded in Pianoteq, with possibly the changes you made yourself to the preset (experiment by changing for example the diapason). If you did not modify the preset, than you will hear the default preset without any alteration.

Re: Preset demos sound way better than my own MIDI files

Philippe Guillaume wrote:

Just to clarify, there are two kind of Pianoteq demos:

- those hosted on our website, for example https://www.modartt.com/modeld. The preset used for recording the demos is displayed on the player, and unless it is specified "(customized)", the demos were recorded with the built in presets without any alteration. For example, among the 7 classical demos of the New York Steinway Model D, 6 are using the default presets, and one (the second demo) is using a customized preset;

- those embedded in Pianoteq, accessible in the presets manager. In that case, as you are playing live the embedded MIDI file, you are hearing the result produced by the preset which is loaded in Pianoteq, with possibly the changes you made yourself to the preset (experiment by changing for example the diapason). If you did not modify the preset, than you will hear the default preset without any alteration.

Thank you very much for clarifying that. In this case it must be my playing style. Seems I’ve unlearned a lot… I hope I haven’t caused too much confusion now…