It sounds a lot more realistic than Ivory 2 with its 18 velocity levels and anyway than the ridiculous Ivory 3. Over 100 velocity levels - impressive. But: How can it be that there are even fewer GB per stereo microphone position (Ivory 2 German D: 27 GB, VSL Steinway D-274 SY-Concert-D_06_Main.syvolume: 25 GB). Were all 88 keys sampled?

Ok, is it lossless compression like flac? Is it more than 100 max.? I think the higher keys don't need 100...

Hmm... So the installed library files aren't compressed if I understand you right. Sounds logical, otherwise the software would also have to decompress while playing... Ivory 18 vel. levels 27 GB vs. VSL >50 (?) vel. levels 25 GB. I can not comprehend that..

One more thing, could you please give me the preset used in the Tchaikovsky March? Thanks!

I'm really sorry, but I found another bug: The release sample of many notes, e.g. 64 E4, are way too loud (RS Level 0). It happens when the key is pressed at low volume (approx velocity 35-45) and for about half a second and the sustain pedal is pressed before the key is pressed and released before the key is released (1.wav and 4.wav). You can hear it without the sustain pedal, but much less. I used the main microphone.

In general, I am bothered by the fact that a note with the same velocity sounds exaggeratedly louder with the sustain pedal than without the sustain pedal (2.wav). It should be almost no difference, since just some other strings will resonate. Compare 8.wav and 9.wav. The only (!) difference: At 8.wav I deleted the sustain pedal, which is pressed after the first chord but before the 2nd chord. This means that whenever the sustain pedal is pressed earlier than the key, it is much too loud. With chords, this adds up.

File Attachment(s):

1.wav (440kb) downloaded 6 time(s).

2.wav (537kb) downloaded 3 time(s).

4.wav (612kb) downloaded 2 time(s).

8.wav (1,423kb) downloaded 3 time(s).

9.wav (1,376kb) downloaded 3 time(s).

You cannot view/download attachments. Try to login or register.

Hello Frederic, 

Thanks for all your input!

Any chance that you can send your sequencer project with those findings, so we can check in the software directly and are definitely using the same material?

Also, which OS are you working on, and which version of the Synchron Pianos Player?

Best, 
Paul

I'm now sure what happened:

- The keys are a little smoother when the sustain pedal is pressed because the dampers are already up.

- This has been ignored.

Proof 1:

A Steinway D-274 has dampers up to and including g''' (dreigestrichenes g). The problem occurs exactly up to this key (see 11.wav). Compare the first tones of the two pitches: there is no difference in volume, although the first has a damper and the second does not.

Proof 2:

The error does not occur with the Vienna Imperial without robot finger (see 12.wav).

Proff 3:

The error increases towards the lower tones, consistent with the fact that the dampers become progressively heavier towards the lower tones.

Best regards

Frederic Chopin

File Attachment(s):

11.wav (797kb) downloaded 2 time(s).

12.wav (797kb) downloaded 1 time(s).

You cannot view/download attachments. Try to login or register.

Originally Posted by: FredericChopin

The error does not occur with the Vienna Imperial without robot finger (see 12.wav).

I wonder if it's actually an error or the nature of the pianos. I notice this behavior to a greater/lesser degree on the Bosendorfer Imperial, 280VC, and Steinway, but I don't notice it on the F308 which is also recorded with the robot finger.

It could be that the robot finger is able to capture pedal-up velocities that are lower than what a human finger can consistently play. That makes the lowest pedal-up/down velocity contrast more discernible. 

It could be that the F308 doesn't have a discernible pedal-up/down low velocity contrast because the piano itself is more capable of accommodating the extreme end of the robot finger movement. 

God Bless,

David

I've been reading this post and you are not only right but your opinion is exactly the same that the technician who tunes and maintains my Yamaha U3. The cause was clear for him, as I showed and demoed him the D274 library: the piano that was sampled was not perfectly tuned and very badly toned in some keys.

The sound is delicious but, as a matter of fact, the playability of the CFX is slightly better.

I consider that, provided that the price paid for this library is quite high, VSL should rework and correct the wrong notes in the D274 library and make a new release.

After extensive testing, I can at least add that the sustain pedal problem is not noticeable above velocity 50 (and as I said with keys without a damper, i.e. g#''' and higher). With the key speed there, the percentage difference is negligible, which makes sense.

I haven't received an answer about the Tchaikovsky preset either. I think I ask too much...

1. Die früher erwähnten Klickgeräusche wurden nun doch korrigiert.
2. Bei den Tönen 70 und 71 mit Haltepedal und Velocity 107 ist auch noch ein Klickgeräusch. Für mich sieht das nach Sampling jedes zweiten Tones aus. Das legt auch der GB-Vergleich mit Ivory German D (s.o.) nahe, welches nur über 18 Velocity Layers und über keine Haltepedal-Samples verfügt und dabei etwa gleich viele installierte GB hat wie eine VSL-Stereo-Mikrofonposition (Komprimierung hin oder her).
3. Auch in anderen Foren weist man darauf hin, dass der Haltepedal-Klangeffekt wieder verschwinden müsste, wenn man das Pedal loslässt. Dafür müsste wohl mit einem künstlichen Effekt Vorlieb genommen werden, obwohl jeglicher künstliche Anteil grundsätzlich abzulehnen ist. Dies erscheint mir der bessere Weg, da der Unterschied sowieso marginal sein sollte und ganz sicher nicht den oben erwähnten massiven Lautstärkeunterschied bei eher leisen Noten aufweisen sollte. Dies und der Fakt, dass die Lautstärke der Releasesamples relativ oft nicht passt, sind für mich die entscheidenden Punkte bei einer Neuauflage. Die Neuauflage sollte auch gleich kompromissloser sein. 200 GB pro Mikrofonposition dürfen's schon sein.

1. The clicking noises mentioned earlier have now been corrected.
2. Notes 70 and 71 with sustain pedal and velocity 107 also have a clicking noise. To me it looks like sampling every other note. This is also suggested by the GB comparison with Ivory German D (see above), which only has 18 velocity layers and no damper pedal samples and has about the same installed GB as a VSL stereo microphone position (compression or not) .
3. Other forums also point out that the sustain pedal sound effect should disappear again when the pedal is released. An artificial effect would probably have to be preferred for this, although any artificial component should be rejected as a matter of principle. This seems the better way to me as the difference should be marginal anyway and certainly shouldn't have the massive difference in volume mentioned above on rather soft notes. This and the fact that the volume of the release samples often doesn't fit are the decisive points for a new edition for me. The new edition should also be more uncompromising. It can be 200 GB per microphone position.