Piano World Home Page Forums Our Most Popular Forums Digital Pianos - Electronic Pianos - Synths & Keyboards Piano software post-recording editing

ChrisA,

I did a pretty bad job explaining what I meant because everything you said I agree with. I guess I was a little too concise.

I've heard this question asked once or twice before so here's a website I'm just finishing up.

BlackGrand.com

I thought it would be better if, rather than just throw up another personal blog, I make a site that others can actually benefit from. It is an experiment, to a certain degree, since I realize that there are a lot of different ways to share your music already. I just thought it would be helpful to people who are looking into a certain piece of equipment or software to be able to here and see what others are already doing.

This is sort of an unofficial announcement since I'm planing on posting a new topic to make it more widely known but I thought it was appropriate to mention here since it goes along with what Morodiene was asking.

Certainly, setchman...thanks for posting this!

Blackgrand is interesting.

Have you considered providing your midis so we could create wave/mp3 files using our sources of sound?

Glenn

That's a cool idea Glenn! And setchman, I love the song Breaking Out...it's gorgeous! I also love the sound of the piano there.

Thanks for the feedback. I didn't intend on hijacking this thread so I just posted a new one.

Glenn,

I hadn't thought of that but I don't have a problem sharing my midis. I'll get them posted when I have a chance.

You copy a section of the music. If you've got a midi capture then you can select (with a mouse) just one voice. and copy it to a parallel track. Then the trick is to make it slightly different and mix to the final stereo mix

Some drastic things you can do might be to transpose the pitch and assign the track to (say) a string section and mix that sound under the piano.

Once I made a string bass line from the roots of the left hand chords and then moved it down an octave.

A more subtle trick is to copy a track, delay it by a few milliseconds (3 to 12) and add it back. this adds "space" to the sound and such a small delay does not sound like an echo.

You can also do both. Once for fun I recorded a four hands piano plus bass guitar, woodwind and drum version of a theme from Greig's "morning mood" It was grossly over produced but fun to try. Back then I could only play "one finger piano" so it took hours to "play" all the parts.

To make the DP sound "acoustic" you really do have to put it in a mix. But mix it with what Well an altered copy of itself

Last night I think I decided that DPs sound and play different than acoustics because the DPs are near perfect while the acoustics not. (I had signed out a practice room at the college and inside was a 20+ year old Kawai piano that is literally played 12 hours a day, 6 days a week and gets tuning about every second year.)

To get that sound, you are going to have to "adjust" some pitches to make them "off" slightly. I'm serious, de-tune the DP using the software. Perfection is a dead give away that the piano is digital. I actually got this idea from the service manual for a Hammond organ I have, I cautions NOT to attempt "perfect" tune because the slight key to key "error" is what gives the Hammond it's characteristic sound.

So maybe flatten and sharpen a few keys by one or two "cents"

Morodiene,

There are some walk throughs on Mahlzeit's site.

www.pianoclues.com

Look under MIDI and Recording in the categories and you will find plenty of information.

Rich

Interesting idea, Chris, but it sounds right, acoustics aren't perfect and so these imperfections make it work. It sounds a bit sophisticated for me, but I'll have to give it a try. I can't wait to be well again so I can go into my studio and play with this stuff. Of course, I'll have to teach then .

That looks like a great resource, Rich. thanks!

OK, so here's what I've been able to figure out so far. Below are two samples of the same music, one a recording directly from my FP-7 as a wav file via an Edirol UA-1ex usb interface, then converted to mp3. The second is as a midi file into Pianoteq and exported as mp3 file using their C3 Recording sound.

FP-7 recording: http://www.box.net/shared/a3by9ycnd2

Pianoteq C3: http://www.box.net/shared/xp9uy3sj5k

What do you think? I really like the Pianoteq version much better, of course. Any suggestions as to what else I could do to make it sound more realistic/acoustic?

Both are good recordings, but I agree, the Pianoteq version is much better to my ears.

I would play around with all the various settings in PT, mic placement, and all the other settings, EQ, reverb, etc.

Damn! I just posted elsewhere, that while I love PT, it it not quite 'there' for me, but your recording is starting to change my mind again!!! :-)

I just have the trial version of PT. How do I change the mic settings? Or is that just the different "voice" selection?

edited to add: I think perhaps because I was playing Bach, which doesn't rely on a lot of low notes that it sounded pretty good. I was also going to play something Romantic to see how it held up.

Just click on the microphone icon ( near the centre of the screen) It will open up a screen with a graphic of the piano, and choices for placing microphones. Just click and drag.

Oh! Cool!! OK, now time to fiddle some more

The Pianoteq version is noticeably better sounding to me, as well. I've only played with the demo myself so I can't offer any suggestions as to improving on what you've done but it sure sounds like you're on the right track.

So what piano brand does PT use for these? Steinways? Yamahas? I couldn't find it on their site anywhere.

I always thought the C3 and M3 were supposed to be Steinways, while the YC5 is meant to be a Yamaha. But I don't think the modeling is good enough that you can really distinguish the brand based on sound alone, it's more a mellow vs bright thing...

So how exactly is modeling different than sampling?

Sampling means recording a real, physical, acoustic piano.

Modeling instead simulates the whole physical system (hammers, strings, soundboard, case, air around the piano, microphones) to create the sound, which is very complicated to do in real-time, so the Pianoteq developers are supposedly using all kinds of mathematical tricks to make it feasible.

The advantage of modeling is that things like resonance between strings or a hammer striking a string that is still vibrating can be accurately taken into account, in a way that wouldn't even be possible with thousands of samples using the "normal", sample-based approach. The drawback is that modeled pianos still sound a little fake sometimes.