velocity curve doesn't reach 127

Ten2One · Post by **Ten2One** » Sat Jul 14, 2012 11:41 pm

Yes, the only reason I took it in was because Dan is a standup guy and tries to help here from time to time and he suggested that I do so.

My Kronos was checked out by KORG before they shipped it to me so I'm skeptical as to whether there are any adjustments that can be made. My hope is that there is a setting that will open the Kronos to the last sample set if there is one, i.e. 118-127, and not lose any dynamics. I will be very disappointed if I go through the expense and delay and then have a board that has no dynamics.

I could get a little more volume using velocity curves 6,7,8 (I think) but there were no dynamics (who wants that?).

It shouldn't be that big of a deal. My Kurzweil PC2X could produce the full range and its not near the board the Kronos is.

There are some great people at Korg who really try to make a difference but sometimes management can make decisions that ties the hands of those who care.

Ten2One

MarPabl · Post by **MarPabl** » Sun Jul 15, 2012 7:50 am

The fact I just can't understand is why this can't be easily fixed with a software update. The velocity curves, after all, define a mapping of the velocity detected on the keybed. Even the 1.5.2 OS added a special hidden parameter to "adjust" the feel of the fixed keybed. So this one is surely fixed with software.

The fact of changing the velocity curve will not modify (move gears or anything mechanical) the keybed: just the mapping will be modified.

On Kurzweil, there's even an option to change the sensitivity for adapting a totally different physical keybed. Is like saying: treat my (synth action) keybed as if it is a fully weighted keybed.

IMHO we're again with the same issue we faced with the faulty RH3 keybeds (cutoff notes): we needed pages and more pages and more threads (and letting know on different forums how bad playing a Kronos was) talking about the actual failure and Korg took so long to "believe us" (less than 1%, etc. etc.) as well as so many fanboys around saying crazy things like the "failure" was because the player had bad playing technique.

I'm seeing here the same thing... How long will Korg take to accept the issue and provide a fix for everyone? BTW I disagree with those thinking that is actually desirable not being able to get 127 because this way, making practically impossible to get 127, you don't "waste" dynamics. IMHO this sounds like the "incorrect playing technique" that was commented regarding the cutoff notes.

I can understand it's "desirable" that there are velocity curves where someone just can't get 127 because the dynamics (strength) of some players "isn't enough" while or others it's easy to go 127. But I think there should be some velocity curve where this (weak) player can reach 127 and at the same time the player feels comfortable with the dynamics he can get.

As I can see, there are many reports regarding this. So I think we have here other issue...

michelkeijzers · Post by **michelkeijzers** » Sun Jul 15, 2012 11:40 am

@MarParbl: do you mean you would like some kind of 'conversion factor', so that 1.00 means the velocities stay as they are and e.g. 1.1 would increase the velocity range from [0..100] to [0..110] ? So anybody can define their own 'weight'? And after/before that the velocity curves are taken into account.

Might be a good idea, very simple to implement, the only downside is that using a conversion factor is that some velocities cannot be reached (step of 2 instead of 1) ... if this is bad? probably less bad than not reaching a large range of velocities (high or low) at all.

huibn · Post by **huibn** » Sun Jul 15, 2012 1:09 pm

MarPabl wrote:The fact I just can't understand is why this can't be easily fixed with a software update. The velocity curves, after all, define a mapping of the velocity detected on the keybed. Even the 1.5.2 OS added a special hidden parameter to "adjust" the feel of the fixed keybed. So this one is surely fixed with software.

The fact of changing the velocity curve will not modify (move gears or anything mechanical) the keybed: just the mapping will be modified.

On Kurzweil, there's even an option to change the sensitivity for adapting a totally different physical keybed. Is like saying: treat my (synth action) keybed as if it is a fully weighted keybed.

IMHO we're again with the same issue we faced with the faulty RH3 keybeds (cutoff notes): we needed pages and more pages and more threads (and letting know on different forums how bad playing a Kronos was) talking about the actual failure and Korg took so long to "believe us" (less than 1%, etc. etc.) as well as so many fanboys around saying crazy things like the "failure" was because the player had bad playing technique.

I'm seeing here the same thing... How long will Korg take to accept the issue and provide a fix for everyone? BTW I disagree with those thinking that is actually desirable not being able to get 127 because this way, making practically impossible to get 127, you don't "waste" dynamics. IMHO this sounds like the "incorrect playing technique" that was commented regarding the cutoff notes.

I can understand it's "desirable" that there are velocity curves where someone just can't get 127 because the dynamics (strength) of some players "isn't enough" while or others it's easy to go 127. But I think there should be some velocity curve where this (weak) player can reach 127 and at the same time the player feels comfortable with the dynamics he can get.

As I can see, there are many reports regarding this. So I think we have here other issue...

AMEN!

I'm totally with you on this one!

I think Korg has a 2nd chance with this issue to show that they are capable and willing to take their custumors serious.. from what i know in my country, they destroyed their reputation on how they handled the note cutoff issue..

QuiRobinez · Post by **QuiRobinez** » Sun Jul 15, 2012 6:17 pm

just tried in by playing my Kronos 88 keys in cubase.

Without any effort i could play notes with velocities ranging from 1 - 127 on curve 4 and 9 (didn't tried the other ones).

So no problem over here.

MarPabl · Post by **MarPabl** » Sun Jul 15, 2012 6:22 pm

michelkeijzers wrote:@MarParbl: do you mean you would like some kind of 'conversion factor', so that 1.00 means the velocities stay as they are and e.g. 1.1 would increase the velocity range from [0..100] to [0..110] ? So anybody can define their own 'weight'? And after/before that the velocity curves are taken into account.

Might be a good idea, very simple to implement, the only downside is that using a conversion factor is that some velocities cannot be reached (step of 2 instead of 1) ... if this is bad? probably less bad than not reaching a large range of velocities (high or low) at all.

This is one way to modify the velocity curves. If you see the horizontal axis (x) in Parameter guide (PreMIDI), it's named "soft/strong" which clearly is a subjective definition.

One way to fix this issue is to change the definition of "strong", which means that the signal received from the keybed will be modified because evidence shows the current definition of strong is (practically speaking) impossible to achieve. This way, the curves don't change at all and the overall horizontal axis is stretched. If the resolution of the hardware is currently adjusted to deliver 127 degrees, when you stretch the horizontal axis, you lose certain degrees. In our case, we're talking between 7 and 13 and this implies that you lose one step every 10 or 20 degrees, but you can access the whole range. However, if the resolution of the hardware can still take this stretching, we don't lose resolution and we also get the whole range.

Other way to solve this is effectively adding, modifying or allowing the player to tweak the curves. Almost all the PreMIDI curves make it hard to play hard. Take the PreMIDI graph and add a straigth line going from 0 (at soft) to 127 (at strong). You'll see that this line is above many curves as you approach to strong. Now check the graph for PostMIDI, when you draw the straight line, you can see there are curves above the line, meaning it will be easier for us to achieve stronger speeds. With PreMIDI graphs, there are some curves which cause lose of resolution: when you see that at certain point the curve will deliver the same value. I don't like those curves. This solution can get us to velocities nearest to 127, but in reality we still would not reach the full range because, after all, currently all the curves end at 127 which right now is not possible to get.

MarPabl · Post by **MarPabl** » Sun Jul 15, 2012 6:25 pm

qrobinez wrote:just tried in by playing my Kronos 88 keys in cubase.

Without any effort i could play notes with velocities ranging from 1 - 127 on curve 4 and 9 (didn't tried the other ones).

So no problem over here.

The real test is to achieve those velocities directly on the Kronos. This may be an incorrect test because some gear provides extra velocity maps to adjust incoming velocities.

Even Kronos has this feature when you set the Convert Position parameter to PostMIDI

QuiRobinez · Post by **QuiRobinez** » Sun Jul 15, 2012 6:29 pm

MarPabl wrote:
qrobinez wrote:just tried in by playing my Kronos 88 keys in cubase.

Without any effort i could play notes with velocities ranging from 1 - 127 on curve 4 and 9 (didn't tried the other ones).

So no problem over here.
The real test is to achieve those velocities directly on the Kronos. This may be an incorrect test because some gear provides extra velocity maps to adjust incoming velocities.

Even Kronos has this feature when you set the Convert Position parameter to PostMIDI

i don't agree

i's used the kronos as a master keyboard and recorded this midi information directly in cubase, so the values i played were registered correctly. The velocity information was recorded by using the usb midi connection from the kronos. So no external gear involved.

MarPabl · Post by **MarPabl** » Sun Jul 15, 2012 7:16 pm

qrobinez wrote:i don't agree

i's used the kronos as a master keyboard and recorded this midi information directly in cubase, so the values i played were registered correctly. The velocity information was recorded by using the usb midi connection from the kronos. So no external gear involved.

However it would be great if you can repeat the test on the Kronos, just to confirm

You don't need to go Seq mode to check. You can use a parameter receiving velocity and use Enter + smashing key to check the velocity you're getting.

huibn · Post by **huibn** » Mon Jul 16, 2012 1:05 am

More and more cases of the velocity issue are starting tot pop the surface on this forum.. Also in other topics.. Seems to be a design flaw after all just like the double trigger problemm...

QuiRobinez · Post by **QuiRobinez** » Mon Jul 16, 2012 7:54 am

MarPabl wrote:
qrobinez wrote:i don't agree

i's used the kronos as a master keyboard and recorded this midi information directly in cubase, so the values i played were registered correctly. The velocity information was recorded by using the usb midi connection from the kronos. So no external gear involved.
However it would be great if you can repeat the test on the Kronos, just to confirm

You don't need to go Seq mode to check. You can use a parameter receiving velocity and use Enter + smashing key to check the velocity you're getting.

ok, although i still believe that it shouldn't matter if you record it in a DAW or do it internally in the kronos i did the check in the sequencer.

- selected the default piano
- played notes in the sequencer on the kronos ranging from 1 - 127
- selected the midi event filter and checked the recorded velocities on the Kronos itself.

Everything was recorded perfectly!

I saw notes ranging from 1 -127, so it works fine on my Kronos.

So in my opinion this is not a design flaw. Everything works on my kronos, I have the keyboard replaced earlier this year because of the cutoff notes problem, maybe the new keyboard has a better response?

I have no idea why some of you have problems with it, but it isn't in the firmware because otherwise i would have this problem too.

michelkeijzers · Post by **michelkeijzers** » Mon Jul 16, 2012 8:40 am

As said before, I can go up to 123 with velocity curve 3, didn't try the other ones. For me 123 is good enough and it was night, so I didn't want to make too much noise hitting the keys.

For me those last few promille of range does not matter much.

Davidb · Post by **Davidb** » Mon Jul 16, 2012 11:44 am

huibn wrote:More and more cases of the velocity issue are starting tot pop the surface on this forum.. Also in other topics.. Seems to be a design flaw after all just like the double trigger problemm...

True.

michelkeijzers · Post by **michelkeijzers** » Mon Jul 16, 2012 1:01 pm

Davidb wrote:
huibn wrote:More and more cases of the velocity issue are starting tot pop the surface on this forum.. Also in other topics.. Seems to be a design flaw after all just like the double trigger problemm...

True.

I'm not really sure. Maybe on other keyboards 127 can be reached to fast, meaning that you have the same "problem" as described above but reversed.

I rather have a keyboard that goes "only" to 124 instead of one that reaches 127 way too quickly.

felsineus · Post by **felsineus** » Mon Jul 16, 2012 1:07 pm

No more velocity issues to me.
Now I have no need to change the standard (4) velocity curve.
My new Kronos 88 is delightful to play.
The old one has been correctly substituted by korg's italian distributor, and I am a satisfied costumer...
I must thank all of you for the accurate information.