• Register
  • Help
Results 1 to 9 of 9

Topic: Piano: Difference between hard strikes in hardware and soft samplers? My imagination?

Share/Bookmark
  1. #1

    Piano: Difference between hard strikes in hardware and soft samplers? My imagination?

    I was at Guitar Center today trying out both older keyboards (Yamaha P-250, Motif) and the new Roland 300SX (nice 10 piano sounds, but a loose feeling keyboard and few velocity layers). I couldn't help noticing a big difference between the sound of the upper velocity sounds on the hardware instruments and those in giga\kontakt libraries--there seems to be "bigger" wider sound in the hardware, with the notes more present, if not louder. The harmonic content is there in the softsamplers, but somehow, and this may be entirely a subjective experience, the softsample hard strikes seem a little thinner. There's more of a clear, pleasant jangle of strings in the hardware.

    Is this just me, or do Roland and Yamaha do something to their hardstrikes? (Their primary samples, I imagine, with filters to reduce the harmonic content for softer sounds.) Do they use a slight chorus or stereo widening or doubling of samples, or record them with a different mic set up? (For soft and middle velocity samples, the giga instruments sound better, for all the obvious reasons.)

  2. #2

    Re: Piano: Difference between hard strikes in hardware and soft samplers? My imagination?

    I haven't taken a look at the programing in any of the piano libs I have - but I know that almost all developers normalize their samples which may bring down the volume of the higher velocity samples drastically, especially if the high velocity has a heavy thud from the hammer hitting the string, these low frequencies would cause it to "over-normalize".

    Maybe nothing you can do if the dev did this to the raw wav data - if - however, they applied it to the patch as a filter triggered by the velocity to filter more the harder the key is pressed - than this process you can simply turn off in your sampler.
    Alan Lastufka | www.BelaDMedia.com
    Producer/Artistic Design | Content Producer

    20 Things

  3. #3

    Re: Piano: Difference between hard strikes in hardware and soft samplers? My imagination?

    I'm not sure if it's a different volume that I'm trying to describe. (I wish I owned one of the recent Rolands so I could record a few notes and post them.) It's more the presence of the harmonics. Not just that they're audible, but that I seem to hear the multiple strings vibrating more in the hardware samples.

    They may have just put mics over all the strings that would ring as harmonics. A time consuming, tedious effort, but it works, if that's what they did.

  4. #4

    Re: Piano: Difference between hard strikes in hardware and soft samplers? My imagination?

    It occurs to me that maybe developers are putting mutes on adjacent strings when they sample each note. (?)

  5. #5

    Re: Piano: Difference between hard strikes in hardware and soft samplers? My imagination?

    Hi,

    did you try the SRX-11 Piano Roland in a RD-700 or 700 sx ?

    http://www.rolandus.com/products/det...&prodid=SRX-11


    According to me the recording session is the most important.

    Size of samples doesn't mind quality.

    Regards

    Olivier

  6. #6

    Re: Piano: Difference between hard strikes in hardware and soft samplers? My imagination?

    I hope I'm making sense, here. I'm not trying to disparage the many great current libraries out there. I notice the difference most when playing certain types of music--rock, "gospel" chords with a big sound.

    One of the pianos that made me hear the difference was the main piano in the RD300SX. (I'm not sure what the SR11 is--an addon card?)

    Maybe what I'm noticing is some more programming in the hardware that emulates the difference between hard strikes and soft strikes: in a real piano, if you hit several notes at once, the harmonics\sympathetic vibrations are going be louder than if you hit one note, and there will be more excitation of all the surrounding strings, yes? Since the force of each hammer strike will be felt. In other words, the force of a three note chord will be closer to three times the force of a single note. (Disregarding harmonic cancellation...)

    In the soft libraries, I'm not sure that it's possible to program something like this--the more notes AND the higher the velocity, the more harmonics and the louder they play and the longer they decay\ring out (the more excitation, the longer the fade to a passive state.) With soft libraries, playing three notes gives you the sum of the three notes and whatever resonance the mics picked up as bleed, or other mics picked up, for each of these--there's no increase in the volume of the resonance that depends on the number of notes played. Maybe Kontakt 2 will help here, with its programming rules? (If number of notes played = A, play harmonics B-D at volume Z.)

    Or am I hearing things?

  7. #7

    Re: Piano: Difference between hard strikes in hardware and soft samplers? My imagination?

    Quote Originally Posted by Jake Johnson
    Maybe Kontakt 2 will help here, with its programming rules? (If number of notes played = A, play harmonics B-D at volume Z.)

    Or am I hearing things?
    I hope that Kontakt 2 can do this. I will spend several hours to program that.

    Regards,

    Olivier

  8. #8

    Re: Piano: Difference between hard strikes in hardware and soft samplers? My imagination?

    Come to think of it, the above formula doesn't take into consideration the velocity of the strikes and the decay of the harmonics. The correct formula would be closer to:

    If number of notes played = A AND velocity= Y, play harmonics B-D at volume/velocity level Z for N (time).

    I don't know what the increase in the volume of harmonics would be with the addition of each simultaneous note\hammer strike. Some trial and error ahead, I guess. (I really hope Kontakt 2 will allow this kind of formula, and that formulas won't cause latency while the engine calculates them.)

  9. #9

    Talking Re: Piano: Difference between hard strikes in hardware and soft samplers? My imagination?

    Jake,

    There are different paradigms when developing a library for hardware versus software.

    In software there is much more freedom for the developer than in hardware - he can use more space. There can be more samples, they can be longer, and they can be layered in more ways.

    The approach seems to be 'keep it as real and natural as possible'. Processing is often seen as the devil.

    I know Nemesys went to a lot of trouble to avoid using output compression in Gigasampler so that they could maximise dynamic range and polyphony. This is more difficult in hardware samplers where the output stage is sometimes not as good as a 'pro' audio card's output. In this case they need compression and gating to keep the signal level above the noise floor at low polyphony while avoiding output distortion at high polyphony.

    In hardware there are many more restrictions - lower CPU power, less polyphony and much less space to play with.

    I'm sure these hardware limitations give rise to a different approach to 'getting the job done'. You end up picking and choosing the best samples to stretch, the best samples to layer etc.,You do what HAS to be done to make it sound good, and making something 'sound good' doesn't always mean making it perfectly mimic the original (witness the inherent 'smiley' eq found on many domestic hifis)

    The makers of the GEM piano (the first one with the 'resonance model') once boasted to me that their piano was a combination of samples produced at three different recording sessions - by three different pianos! Their reasoning was that different pianos had strengths in different areas of their range, so they borrowed the best of each...

    Anyway, my guess is that what you're hearing is someone's heavily re-engineered idea of what a kickass top end on a piano should be - not necessarily what it was on the day they recorded it. That's not necessarily a bad thing - just different.

    As soft samplers continue to evolve, I'm sure the native processing options (like eq, multiband compression etc.,) available to developers will also continue to improve and they'll start generating patches which have a much wider range of variance than we're used to seeing.

Go Back to forum

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •