Korg Forums

Yes if I play the Kronos piano sounds via midi, directly from a Motif XF (ie. Motif OUT to Kronos IN - no interface), things feel obviously slow to me. Of course, I expect that - response is never as fast over midi as 'locally'... (one of my other boards is noticeably slow as a transmitter... the Motif is actually ok in this regard, though)

I do wonder if the Kronos piano samples have just a teeny bit of 'dead space' (silence!) at the beginning of them. That could also account for perceived latency.

Is it possible for a user to tweak the individual sample start points within the German and Japanese piano multisamples?

cosmicjazzer wrote:I do wonder if the Kronos piano samples have just a teeny bit of 'dead space' (silence!) at the beginning of them. That could also account for perceived latency.

For what possible reason would 'dead space' be included at the start of the piano samples?

cosmicjazzer wrote:Is it possible for a user to tweak the individual sample start points within the German and Japanese piano multisamples?

No.

dfahrner wrote: For what possible reason would 'dead space' be included at the start of the piano samples?

Well there's certainly no good reason! But the reality is that samples are not always trimmed as precisely as they might be... It's necessary to zoom in as much as possible to really get the sample start points perfect.. and this isn't always done!

I've quite often come across samples in keyboards and sample libraries that suffer from this...

I find it odd for keyboardists especially to not get accustomed to playing with various amounts of latency. Indeed, between the time your skin touches a piano's key, that key throws the hammer into the string and the sound reflecting off of the lid and walls to get to you, you're experiencing at least several milliseconds of "latency," and there's more as you play softer. Concert pianists seem to get by just fine with different pianos everywhere they go. And as has been mentioned, keyboards of the past are all over the map. For that matter, ever play a harpsichord? Mellotron?

Yes, it's different playing a Hammond with it's 9 contacts that complete circuits sequentially before you've even hit the bottom, and playing a piano, but I assure you that many people can play both, and not feel uncomfortable with either one.

At the end of the day, I don't think this is a Kronos problem.

Well I never experience any issue playing an acoustic piano, a Rhodes, Wurlitzer or indeed a Moog or DX7. Yes there must be some latency in all these instruments but in my perception it's just nothing compared with many current electronic keyboards. The Kronos is no worst than most in this regard, actually. In fact I'd say it fares fractionally better than a Motif XF, fractionally worse than a Motif ES and Nord Stage 1, and about the same as a Jupiter 80 and a Nord Stage 2!

I was saying previously that older boards usually have LESS latency, in my perception (obviously the Mellotron is an exception!)...

I would love to be able to 'adjust'! But my nervous system is just not wired that way. Most of my fellow pro-keyboard players are far less fussed than I about such stuff, that's clear to me! It's playing the really fast improvised lines that most exposes any latency to me. The Kronos, Motif etc are sometimes very obviously behind then... I can't deal with that!

JPWC wrote:while some people maybe able to hear a time shift of 3 mS to 7 mS. Most Keyboard players learn to compensate for this time shift, without even knowing most times.

Time shifts of 5 mS range are commonly refered to as the feel, rushing or dragging the beat.

Some keyboardists can actually accomodate much much larger latencies in the 100ms+ range. Ask any competent pipe organist.

Yep. Dunno how they manage it! Respect!

Strangely enough, I find that if I know what I'm playing, ie. if i'm playing a piece that I've learnt, then latency is not so important. When it becomes crucial to me is when I'm improvising...

cosmicjazzer wrote:Yep. Dunno how they manage it! Respect!

Strangely enough, I find that if I know what I'm playing, ie. if i'm playing a piece that I've learnt, then latency is not so important. When it becomes crucial to me is when I'm improvising...

Perhaps because you have more 'muscle memory' going on with the pieces you have learned. Many years ago when I played pipe organ it seemed like my immediate 'feedback' was more the touch of the keys than the sound of the pipes. After a time it seems the brain has some uncanny way of 'syncing' the real-time playing of the keyboard with the delayed sound.

I had a friend who was a concert pianist come visit me at the old Elks Lodge in Los Angeles. Robert Morton installed one of their largest pipe organs there in 1926. This beast was a 4 manual 61 rank instrument in a large hall.



She sat down and hardly got 2 or 3 notes played before stopping with a horrified look on her face. "I can't play on this!" she said.

Here's a look at the front half of the hall ...



Huge amounts of latency going on in there! The 'echo' pipe chamber on the back wall was about 150' from the console. That's about 150ms acoustic delay not counting the significant latency of key press to when the pipe speaks.

7ms latency is a piece of cake!

Latency on physical grand pianos varies by the force of the key press (aka velocity) - and, according to this study, the amount of variation between pp and ff is about <b>110 ms</b> (!!!).

Thoraldus wrote:

cosmicjazzer wrote:Yep. Dunno how they manage it! Respect!

Strangely enough, I find that if I know what I'm playing, ie. if i'm playing a piece that I've learnt, then latency is not so important. When it becomes crucial to me is when I'm improvising...

Perhaps because you have more 'muscle memory' going on with the pieces you have learned. Many years ago when I played pipe organ it seemed like my immediate 'feedback' was more the touch of the keys than the sound of the pipes. After a time it seems the brain has some uncanny way of 'syncing' the real-time playing of the keyboard with the delayed sound.

I had a friend who was a concert pianist come visit me at the old Elks Lodge in Los Angeles. Robert Morton installed one of their largest pipe organs there in 1926. This beast was a 4 manual 61 rank instrument in a large hall.



She sat down and hardly got 2 or 3 notes played before stopping with a horrified look on her face. "I can't play on this!" she said.

Here's a look at the front half of the hall ...



Huge amounts of latency going on in there! The 'echo' pipe chamber on the back wall was about 150' from the console. That's about 150ms acoustic delay not counting the significant latency of key press to when the pipe speaks.

7ms latency is a piece of cake!

That's a brilliant example.

I think I play reasonably in time (most of the time), but whenever I record into a DAW, I always appear to be ahead of the beat, but it's quite consistent, not much, but noticeable, and I'm convinced it's latency through my PC system that I am unconsciously compensating for.

danatkorg wrote:Latency on physical grand pianos varies by the force of the key press (aka velocity) - and, according to this study, the amount of variation between pp and ff is about <b>110 ms</b> (!!!).

That will be the same on a modern keyboard as well, as the time taken to measure the velocity from key on detection to key fully depressed detection will vary according to how hard you hit it.

Derek Cook wrote:

danatkorg wrote:Latency on physical grand pianos varies by the force of the key press (aka velocity) - and, according to this study, the amount of variation between pp and ff is about <b>110 ms</b> (!!!).

That will be the same on a modern keyboard as well, as the time taken to measure the velocity from key on detection to key fully depressed detection will vary according to how hard you hit it.

That's part of it - but not all of it. Read the paper.

- Dan

danatkorg wrote:Latency on physical grand pianos varies by the force of the key press (aka velocity) - and, according to this study, the amount of variation between pp and ff is about <b>110 ms</b> (!!!).

Well that's kinda right i guess. But - and this is crucial! - by latency I'm meaning the time between the key at the very bottom of its travel and the sound occuring. How long the key takes to get to the bottom is not experienced as part of the latency by my nervous system!

On an electronic keyboard with a piano like keyboard, any latency imposed by the electronics is of course in addition to whatever might be thought of as being imposed by a key's travel and to me, at least, prompts a very different feeling from anything mechanical...

cosmicjazzer wrote:

danatkorg wrote:Latency on physical grand pianos varies by the force of the key press (aka velocity) - and, according to this study, the amount of variation between pp and ff is about <b>110 ms</b> (!!!).

Well that's kinda right i guess. But - and this is crucial! - by latency I'm meaning the time between the key at the very bottom of its travel and the sound occuring.

According to that paper, on an acoustic piano, the the time between the key being at the bottom of its throw and the sound occuring is also different depending on force. They found that the difference between pp and ff - which I would characterize as jitter - is 13ms.

Just to share ...

I had an issue last month when I was organizing a setlist. I experienced a bit of latency when I was playing from setlist in edit mode. Probably I copied a slot and didn't paste, then I started to play without exit from edit mode.

After closing the edit mode of the setlist and accessing the play mode the latency was gone. I couldn't replicate it and i didn't care because editing a setlist is not a live condition, so to me was acceptable that maybe there was a different CPU/ sys resources load...