Clicks in 2 Bechstein notes

Hi all,

I'm new to the forum and relatively new to Akoustik Piano. Overall I'm really happy with it, but I've noticed a problem that's preventing me from finishing a piece.

2 notes in the Bechstein piano have clicks in them: C1 (MIDI note number 36) at velocities of 45 to 57, and C#1 (MIDI note number 37) at velocities of 45 to 71. There's a single click in both of these notes at approx. the 0.8 second mark (only guessing - I haven't measured it).

Does it behave this way for everyone else, or is it just on my system?

Thanks in advance, Bill

 

I just installed Akoustik Piano and have the identical issue with those two Bechstein notes! It must be bad samples on the install DVDs... not sure how to resolve this issue though? I installed all of the available upgrades, but apparently these bad samples were not replaced by the upgrades. I also replaced the installed Bechstein samples by re-dragging the sample directories from the source DVD, but this had no effect, so obviously the problem is on the DVD source files.

I have not tried installing the 16bit files as I would like to use 24bit (so I dont know if this issue is only for the 24bit files or not).

 

Thanks for the confirmation, ZFire. Now I know I don't have to spend days troubleshooting.

But, ummm, how do we get NI to fix it?

 

Well these two waves have glitches in them on my A.P. installation. The first one is a pair of softer clicks around the 5 second mark, and the second one has the big obvious click in the first second and another softer one around 4.5 seconds.

Bech-R(C1)-dB(3155)-T(ML)-M(64)-P(659-04-01).wav
Bech-R(C1)-dB(3316)-T(ML)-M(52)-P(695-02-04).wav

If someone else could give these a listen on their installation it would be nice to know if they are glitched too. They are in the directory:
...\Akoustic Piano Library \Samples \Bechstein \Samples

 

Yes the clicks are there, i can confirm. The Bechstein preset are my favorite of the A.P. Library. I hope there will be a fix some day. Or OlivierFRAPPIER can do a new preset, like his great work with the Steinway preset :-D

 

I submitted a support request to NI, so we'll see what they say.

 

Just to update the status of this: NI has confirmed that the issue is a bug and will be fixed in a future software update.

There is no immediate fix, though I imagine that a person could get some click-removal software/plugins to clean up the wave files.

It makes me wonder how many other of the samples might have small glitches in them? I certainly didn't audition the hundreds of samples there to look for problems.. only the C1 samples, of which those two had problems.

 

Hi, I looked at these wav files using Adobe Audition and found (most) of the clicks.

In Bech-R(C1)-dB(3155)-T(ML)-M(64)-P(659-04-01).wav I found a small click at sample number 223364 and another small one at 243328 (both about 4 samples wide).

In Bech-R(C1)-dB(3316)-T(ML)-M(52)-P(695-02-04).wav there was a large click at sample number 35582. Softer clicks were found at samples 216966, 217085 and 217203.

All of these clicks were short enough to correct by hand in the Adobe Audition waveform edit window.

 

Just found this thread, and also posted about other issues with the Bechstein model: clicks in the attack portion of many notes (sounds like the sample start times are jacked up or start in the middle of the waveform).

This is a shame, as this is an incredible piano otherwise! Most of the problems with this tend to be in the middle range (C1-C4) when playing at light velocities. As it stands, this model is totally useless for an exposed solo piano passage...

 

So, it's about 10 months since I first reported this as a bug AND IT'S *STILL* NOT FIXED! >:-(

So much for all the hype about the quality of NI products!

I sorta kinda expected something like "Oh sh*t! How the hell did we
let that happen? Sorry - we'll fix it a.s.a.p.!"

Anyone from the NI team reading this? Markus? Hellooooooooooo! Mr Product Manager Instruments Division Native Instruments, Berlin?

It's not like it's some tricky little bug that could take months to track down! It's just a couple of fscked samples! No, I'm not very good at fixing things like that, and I shouldn't have to be!

Yep, definitely regretting buying this instead of Ivory... >:-(

To finish on a friendly note, has anyone actually managed to fix this problem for themselves? I've spent a few hours on it but they're still not perfect and I'm just tired and frustrated.

Yours disappointedly, Bill

 

Yes its pretty disappointing alright. Especially for something that should be relatively straight-forward for them to fix.

I think NI is crazy if they think that not supporting their products doesn't cost them customers. Who else is a more likely customer than someone who had a good experience with one of their products? I think they need to get some more realistic priorities.

 

Native where are you?. I have same problem. It would be polite to answer this tread. You can not use it in your music production with the clicks.
Thanks
Michael

 

Hey guys

Seeing this thread reminded me of a post I made a few months back when I first installed AP. It's more related to the trueness of the notes themselves, rather than sample 'quality', but I would interested in your thoughts... and of course the possibility of a fix from NI.

Full text of my original post is below.

Cheers



Sorry to dig this up, but I've just installed AP today. Can others confirm that they also have what I can only describe as non-pure samples on G1, G#1, C2, C#2 and E3 on the Bechstein?

I'm not talking about interference (clicks) etc, simply that the notes don't seem to ring as true on these two keys. I'm sure there's a more technical/musical term for this, but it just sounds like the notes aren't ringing quite right, as if the piano hammer has mis-struck slightly. The samples are mostly underpinned by a shrill, bell-like sound, compared to the other notes, and are most noticable when played between other 'true' notes as part of a scale.

Any ideas?

 

Well it has now been ONE YEAR and these obvious glitches in samples have not been fixed by NI. For a product that touts itself as having super fidelity, it is just amazing to me that they wouldn't care.

Apparently supporting Akoustik Piano is a very low priority for Native Instruments. To me this seems incredibly foolish, as the most likely purchasers of NI products are those who have made past purchases and are happy!

I am currently in the market for an orchestral plugin... and to be honest I won't even audition NI's plugin. Why should I? If there are issues with it, then those issues will very likely be completely ignored.

 

I just checked the first sample you mention and can hear the clicks clearly. Well it seems the de-clicking with soundforge or audition or with some plugins isn't the most elegant method, otherwise NI would have fixed that in the past year or so. The golden ears of course would find you lose someting while reducing the clicks.
There are, surely, some bugs in some products that get fixed VERY very slow. True. But well, we're not owners of fine plugins of a single developer who always listens to the mails he/she gets from the paying customers , this you can say without wanting to be unfair against Native Instruments. They have a whole lot of people they sell their products to, and apparently Norah Jones or Chris Martin didn't knock at their door until today to say "I want this sample replaced" .
All in all I am very happy NI included a Bechstein into their Akoustik piano, but the bug should be fixed for all klavier-wizards now. I can live with that, but well, I don't play the Svjatoslav-Richter-like solos anyway )).
C'mon, NI, I did beta-tests for 3 weeks every night last February after a strange and foolish Vista-bug that constantly deactivated all of my NI-software, and got nothing in return after I solved the problem (or, err, after 3 idiotic weeks a microsoft hotfix and sp1 of Vista helped me to hunt down the challenge-response-bug that drove me crazy before) ! Now you could at least fix this bug for us.
(Maybe they plan a new recordings for the mighty Bechstein, and we just should be a bit patient...let's hope so )

 

Hi people,
I fixed the two samples in question. You can download them at
http://www.kim-bostroem.de/Stuff/Bech-R(C1)-dB(3155)-T(ML)-M(64)-P(659-04-01).wav
and
http://www.kim-bostroem.de/Stuff/Bech-R(C1)-dB(3316)-T(ML)-M(52)-P(695-02-04).wav
It was really no big deal and I find no quality diminuishing. And even if there was some (for the Super-ear-o), it's still dramatically better than having clicks.

 

Thanks a lot. Very generous of you to do NI's job and share these files.

 

Thanks for the files. I have manually corrected them myself in Soundforge, but was not sure if that would have made the sound worse for our golden-ear-masterpianists and did not post it. Glad you did
Myself I am not such a grreeat pianist, so to be honest I could live with the few mistakes in the old files, but I know it is important for others. Still glad we have a Bechstein too in the AP, I love this instrument.