need for larger velocity scales, 0-128 to small

jemkeys25 · Post by **jemkeys25** » Mon Jan 31, 2011 3:05 am

here we are talking about all this new technology of the kronos, yet keyboards will struggle to advance if commom ideas used for years by all keyboard companies don't change.
take key velocity for instance, it's set at 0-128, yet theoretically you can have an infinite number of velocity values from a players softest playing force to thier hardest. todays keyboard sounds are set at a threshold where a player is limited in thier expression. say it takes 20 pounds of finger force to reach a key velocity of 128. playing with a velocity over twenty pounds won't change the sound one ioda. this doesn't happen in the natural world. hit a piano key harder, the sound changes. even changing from 0-128 to 0-1280 would give a ten fold increase, where the hardest velocity value couldn't be reached by the human hand and the softest would become inaudible to the human ear, just like acoustic instruments. simple little ideas like this would do more for keyboards than a new sound engine or longer samples or what have you.
not to sound negative, I really like the kronos and will probably pick one up, but i've heard all the sounds before.

CfNorENa · Post by **CfNorENa** » Mon Jan 31, 2011 3:07 am

jemkeys25 wrote:here we are talking about all this new technology of the kronos, yet keyboards will struggle to advance if commom ideas used for years by all keyboard companies don't change.
take key velocity for instance, it's set at 0-128, yet theoretically you can have an infinite number of velocity values from a players softest playing force to thier hardest. todays keyboard sounds are set at a threshold where a player is limited in thier expression. say it takes 20 pounds of finger force to reach a key velocity of 128. playing with a velocity over twenty pounds won't change the sound one ioda. this doesn't happen in the natural world. hit a piano key harder, the sound changes. even changing from 0-128 to 0-1280 would give a ten fold increase, where the hardest velocity value couldn't be reached by the human hand and the softest would become inaudible to the human ear, just like acoustic instruments.

0-127.

I think your general point is well taken -- why stay within the (arbitrary) constraints of old technology? But I'm not sure velocity is the best example. For that particular parameter, a range of 128 individual values seems more than sufficient...

billysynth1 · Post by **billysynth1** » Mon Jan 31, 2011 3:16 am

Hi

That's strange. About 2 months ago i was playing Rachmaninoff Etude op 39 no5, i hit a five note chord ( at the piano, not a synth), i assume i used a force above 20lbs...and bang, i broke 3 strings. Actaully, 2 strings broke but three keys were not sounding, or just making a clunk noise. One of the non working keys shared the same string as one of the other keys.

How does your premise regarding higher velocity levels work regarding string instruments, whereby, if you strike the strings harder on a real instrument they just break?

Regards
Billy

jemkeys25 · Post by **jemkeys25** » Mon Jan 31, 2011 3:25 am

this is just one example,i'm sure there are more, i'm just trying to covey the idea that old ideas of programming values seem to be stuck in place.
it would work fine if sound could only be broken up into 128 parts, yet physics shows us different. the human eye can detect millions of colors, and the ear, millions of shades of sound. you gave me the sound engines,now give me more controll over them.

jemkeys25 · Post by **jemkeys25** » Mon Jan 31, 2011 3:38 am

hey billy,
thats the beauty of synths, you wouldn't have broken the string on a synth, there aren't any to break, where acoustic instruments have a physical breaking point, synths dont. pianos and guitars, any stringed instrument can break strings, but it doesn't apply to say a trumpet, you can't blow so hard you'll break it.
(love that rachmaninoff piece, one of my favorites)

StephenKay · Post by **StephenKay** » Mon Jan 31, 2011 5:40 am

0-127 is a technological limitation of the current MIDI spec. It is not possible to send a MIDI note with a velocity larger than 127. It is not possible to divide the full range into more than 128 steps, for higher-resolution.

Currently, although I don't know all that much about it, the MMA (MIDI Manufacturers Association) is supposedly finalizing an update to the MIDI Spec, after many years of discussion, that would change this (the new spec is often referred to as HD MIDI) - and allow higher-resolution for controllers, velocity, and much more.

The whole problem with this is that MIDI 1.0 has been in use for 20 or 25 years, and most of the manufacturers don't really want to change it because of the huge work involved. So even when it passes, I would imagine it will take some time for it to be accepted and utilized by many of the large manufacturers.

billysynth1 · Post by **billysynth1** » Mon Jan 31, 2011 9:18 am

So i guess with MIDI HD all current stock of musical instruments wont be able to be updated with it...it will only be applicable ( if accepted ) to new generation Musical Instruments.

Oh goodness, 25yrs of buying synths all gone down the drain over night.
I'm sure I've spent over 50, 60, maybe 70 thousand bucks on synths, and keys/pianos over the past 25yrs

I should of learnt to knit

Billy

ozy · Post by **ozy** » Mon Jan 31, 2011 9:26 am

I think there's a bit of misunderstanding here.

1) first of all, keyboard feeling is not related to ABSOLUTE resolution, but to velocity curve [see point 3)].

2) keyboards (better: keybeds) and their interfaces have a FAR better resolution that 0/127.

it is the RECEIVING end of the chain from keys to sound engine, and above all the processing protocol (midi), which is constrained by the 0/127 scale.

We don't need better keybeds [ok, we need, but not under this profile],

we need a better protocol [which would made everything we own unusable etc etc].

And more refined sound engines.

Finally.

3) I'd ask jemkeys25 to perform the following exercise, if it pleases him.

take a keyboard with an embedded midi tracking screen [pc3, e.g.].

Play it.

Try playing a scale at velocity 1/2. I mean, Try getting a 1/64 resolution from your FINGERS, not from the keybed.

Then a scale at velocity 3/4

Etc etc, up to the last scale, played at 127/127, fixed.

Can you?

Can you master that?

No?

Ok, try this,easier:

try playing a scale at 120/127,

then a scale at 120/115
then a scale at 115/110,

meaning that your fingers can master a resolution of TWENTYFIVE steps along the keys' run.

Can you? Can you master 25 different grades of touch on the keys?

Music notation has a resolution of... let me count them:

ppp
pp
p
mp
mf
f
ff
fff

That makes... er... eight.

Ok, add silence, or the legendary strawinskij's ffff.

It will take some theoretical effort to come to 10 or 12 steps.

If you can play 25 steps, you are right to complain about the keybeds' THEORETICAL resolution.

And boy, I'd like to share you command of the keyboard!

Otherwise,

you probably have got a BAD keybed, with a bad curve (meaning that it will JUMP, e.g., from 90 to 110 then stay flat up to 127 then stay flat again).

X-Trade · Post by **X-Trade** » Mon Jan 31, 2011 10:20 am

Should note, that 0-127 is a limit in resolution, not range.

0 could be wherever the manufacturer wants, and 127 likewise. Perhaps you just want a keyboard that you can hit harder to reach 127?

I think 127 steps of expression (for velocity anyway) is more than enough.

Controller resolution on the other hand (and potentially number of controllers) is another matter entirely.

jg:: · Post by **jg::** » Mon Jan 31, 2011 10:47 am

I want one that goes to 128....

jg::

cello · Post by **cello** » Mon Jan 31, 2011 10:47 am

ozy wrote:Music notation has a resolution of... let me count them:

ppp
pp
p
mp
mf
f
ff
fff

That makes... er... eight.

Absolutely spot on ozy! Sometimes technological capability takes focus away from musical reality. There are principally 8 dynamics in music (excuding silence unless you call John Cage silence as music

) so the other 120 enabled by technology are fairly meaningless (musically speaking).

dokido · Post by **dokido** » Mon Jan 31, 2011 11:19 am

if I had a synth that the strings blow when I push harder I would be very happy, this is really a concept for me, well, perhaps I can make it with sampling

you gave me an idea anyway, thanks

Melodialworks Music · Mon Jan 31, 2011 11:37 am

StephenKay wrote:It is not possible to divide the full range into more than 128 steps, for higher-resolution.

Sure it is. The Avant Grand N3 sitting in my studio is proof!

synthjoe · Post by **synthjoe** » Mon Jan 31, 2011 12:02 pm

ozy wrote:...it is the RECEIVING end of the chain from keys to sound engine, and above all the processing protocol (midi), which is constrained by the 0/127 scale.

I was always wondering whether modern digital synths work with greater internal resolution and limit the MIDI output to the specs, or generate values within MIDI limitations even internally? I think we see both of them on the marked but I'd be inclined to think that most serious keyboards/workstations (especially those with velocity curve control) work with a higher internal resolution than present on their MIDI OUT. Anyone capable of sharing some facts on this?

synthjoe · Post by **synthjoe** » Mon Jan 31, 2011 12:12 pm

cello wrote:
ozy wrote:Music notation has a resolution of... let me count them:

ppp
pp
p
mp
mf
f
ff
fff

That makes... er... eight.
Absolutely spot on ozy! Sometimes technological capability takes focus away from musical reality. There are principally 8 dynamics in music (excuding silence unless you call John Cage silence as music ) so the other 120 enabled by technology are fairly meaningless (musically speaking).

Those are the overall 'volume' indications. I sure hope I can play further nuances within those - how 'equally' I could play a scale I don't know. But I know when I play a musical phrase where I want to put the emphasis and where not, within (say) PP.

I'm not trying to say that 128 steps are not enough in most cases and not trying to say that I have the command of 127 absolute levels in my fingers under all circumstances. But I do believe having more than 127 relative expression levels in my play and I sure notice when a keyboard does not react as I intend - and interestingly that did not happen to me very often on a real piano (other than my bad command of the fingers

)

And some modern electronic instruments are better than others in this respect, hence my previous question about internal resolution... Sometimes I seem to notice a difference between my live play and the MIDI recorded version of it (other than tempo quantization), so while I agree that 'sometimes technological capability takes focus away from musical reality', I think that reassessing axioms in light of new advancements is not a bad thing.

Korg Forums

need for larger velocity scales, 0-128 to small

need for larger velocity scales, 0-128 to small

Re: need for larger velocity scales, 0-128 to small