Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Glide questions
#1
Hi, I have read every mention of Glide in the manual and the userforum.  I still have a lot of questions about glide.  I tried to do it just by listening, but I haven't had much luck in cracking the code yet.

I understand that there are two parameters that control glide - time & steps.

I guess I don't understand how the glide time maps to actual time, and what the defaults are for glide steps.  If I just want a smooth glide between notes, do I have to manipulate the steps value or leave on default 0? Being that FX1 is executed before FX2, which setting should I manipulate first? Also, I assume if I turn glide on in a row (by setting time >0) then the glide will occur between that row and the next row (and every subsequent row until I set time back to 0).  So if I set glide time in row 3 >0, and set it to 0 in row 4, I will only glide between the notes one 3 & 4, correct? A pictogram similar to the one for the trigger value may be warranted here.

Can this be explained in more detail?  Is glide steps default 0 maximum steps or minimum steps?  Is the number of steps related to bit-depth?  If steps value is high but time is low, what happens?

Also as an advanced topic, I find that logarithmic glide shapes work best (at least for traditional berlin or acid style).  I read about using add/mul in another topic and I can see potentially how you cold possibly do that in a table but even with a table speed of 1 it doesn't seem like you would get it very smooth, but either way I don't see how to start messing with the curve if I don't know how time & steps work and are inter-related!

Thanks.
Reply
#2
Interesting question Smile

I've just played with it a bit and my impression is that Glide Speed is actually some form of delay between steps and Glide Steps is the amount of "micro voltage steps" (for lack of a better word) the voltage is changed (glides) per iteration. Higher values for Steps yield faster glide and higher values for Speed yield slower glide. However even Steps == 0 yields some glide.

While it certainly is possible to adjust Glide by setting Steps and Speed by listening to it I clearly would like to see a table or formula for the various values and the resulting glide time. Such would be especially helpful when trying to create non linear glides by means of ADD/SUB/MULT/DIV (not sure such is needed though...)

In other words:
I second geremy's request Smile

Kind regards,
Michael
Reply
#3
Glide speed is just the value how fast the glide function is being updated. (Like a samplerate)
Glide steps is indeed the voltage change.

If a new pitch (or note) is detected, then the voltages changes from the old pitch/note with the glide speed in the amount of volatge (glide steps).
Glide steps to 0 is the finest resolution but since the voltage updates are so small, it might not be 'snappy' enough to glide to the desired note/pitch.
A bit higher, makes it snappier. Much higher makes it go in steps to the desired voltage which is an additional effect.
If I read this now it might confuse people to think it is all about sequencer steps.

It doesn't matter in which order you change the values.
The result is linear and there is no plan to make it logarithmic. You probably use an external slew or so for that.
Also making a table to show glide times would be a bit of overkill. They are too many parameters for that. (Old pitch/note, New Pitch/Note, Glide Speed, Glide Steps).

But in one scentence: The Pitch/Note is changing to a new detected Pitch/Note in the Glide Speed (sample rate) in the glide steps (voltage steps).
PLEASE use the search function if something have been asked or discussed before.
Every (unnessesary) forum support means less time to develop! But of course, i am here to help!  Smile
Reply
#4
(07-22-2021, 08:04 AM)XORadmin Wrote: Glide speed is just the value how fast the glide function is being updated. (Like a samplerate)
Glide steps is indeed the voltage change.

Glide steps I do understand.

Glide speed I'm not so sure. When it behaves like a samplerate my expectation would be that higher values yield faster glide as it is sampled more often. However my tests seem to indicate that higher values in speed do actually yield a slower glide. When I doubled the value for speed and halved the value for glide my listening impression was, the glide was happening at the same pace.

That's what gave me the impression that speed is actually some form of delay that happens between applying steps.

Could you elaborate a bit more how speed influences the glide?
Reply
#5
(07-22-2021, 10:42 AM)mgd Wrote:
(07-22-2021, 08:04 AM)XORadmin Wrote: Glide speed is just the value how fast the glide function is being updated. (Like a samplerate)
Glide steps is indeed the voltage change.

Glide steps I do understand.

Glide speed I'm not so sure. When it behaves like a samplerate my expectation would be that higher values yield faster glide as it is sampled more often. However my tests seem to indicate that higher values in speed do actually yield a slower glide. When I doubled the value for speed and halved the value for glide my listening impression was, the glide was happening at the same pace.

That's what gave me the impression that speed is actually some form of delay that happens between applying steps.

Could you elaborate a bit more how speed influences the glide?
Yes, 1 is the fastest speed(sample rate) and FFF the slowest. Else you would get 0 is off and 1 would be  the slowest possible which wouldn’t be naturally.
PLEASE use the search function if something have been asked or discussed before.
Every (unnessesary) forum support means less time to develop! But of course, i am here to help!  Smile
Reply
#6
Thanks for the reply. So for the slowest, smoothest glide, the answer is to leave steps at default 000, and change speed to simply 001.

Would you be open to a feature suggestion of constant-time glide option, or would that be too difficult to calculate between pattern rows? In constant time glide (slide on the 303, otherwise known as constant time portamento), regardless whether the notes are 1 semitone apart or 3 octaves apart, the glide takes the same amount of time.
Reply
#7
(07-22-2021, 01:25 PM)geremy Wrote: Thanks for the reply.  So for the slowest, smoothest glide, the answer is to leave steps at default 000, and change speed to simply 001.  

Would you be open to a feature suggestion of constant-time glide option, or would that be too difficult to calculate between pattern rows?  In constant time glide (slide on the 303, otherwise known as constant time portamento), regardless whether the notes are 1 semitone apart or 3 octaves apart, the glide takes the same amount of time.

The slowest glide you get with speed FFF, 000 is no glide and 001 is fastest glide.
Smoothest glide is as you write with steps to 0.
Reply
#8
Oops yes mis-typed. Thanks.
Reply
#9
You really don't need the smoothest glide for snappy fast and good glides. In my experience some glide steps between 5 and 15 are mostly perfect for nice fast glides and still very fine with their 'quantization'.

I did not plan to implement a constant glide time function (and I don't really give it a chance). A slide on the 303 is from one to the other step, and here with the NerdSEQ to limit the glide to one step would not cover all possibilities. Glide through multiple steps couldn't be done with it. And even if you would want to add a function to glide to the next note...you don't know what the next note is...can be in a pattern that hasn't been called yet, so this wouldn't work.
PLEASE use the search function if something have been asked or discussed before.
Every (unnessesary) forum support means less time to develop! But of course, i am here to help!  Smile
Reply
#10
Hi.  That makes sense.   I just want to point out that it could be handled another way...it could be another FX distinct from GLIDE, only valid within a pattern.  I have seen other sequencers (like metropolix/metropolis) handle the 'not knowing what the next note is' by actually making it work slightly backward.  Example:

Note 1 - Slide 0
Note 2 - Slide >0
Note 3 - Slide 0

In the glide function this would cause a glide from note 2 to note 3.  However, in the slide function this often created the slide from note 1 to note 2.

Thanks!
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)