Home | ER-101 | ER-102 | ER-301 | Wiki | Contact

V0.3.x Firmware Workout: Better late than never



What about if you want no slice and no loop ?


use the manual grains unit instead?


Yes, of course. But still, it doesn’t feel right to me.


im not in front of it, but check in the slice menu - there may be a playback behaviour mode. I remember seeing this in the sample splicer menu earlier today. Maybe it’s the same .


With the new playback behaviors (12 modes in all!) for sample player units, there is definitely possibility for some weirdness. I would like to address ASAP but it would help if you could give details like:

  • How many slices?
  • If you had slices, what slice was active at the time the trig was not working?
  • What were the values of: How often? How far? and Which side?

Or you could save a preset and drop it here. :vulcan_salute:


How many slices?
If you had slices, what slice was active at the time the trig was not working?

What were the values of: How often? How far? and Which side?
How often : once
How far ? well, are we talking of the ER 301 :grin: ??
How Much : all
Which Side : both

(sorry for the formatting weirdness. No idea why)


Trig doesn’t work for me either in grain stretch even with slices defined. Will test out some other samples


Haha. Oops. It used to be called “How far?” I guess I changed it to How much?

This helps. I think I know what the problem is. It’s basically a case of the reset point and the stop point being on top of each other so that it is triggering but playing only like 0 or 1 sample.

If it is not triggering in “both” try changing it to “right”. In the meantime, I will have a think about how to fix it. I agree that the goal is to have no situation where a trig causes nothing to happen.

Also, I’ll make a diagram showing how the 3 playback behaviors interact.


I bet all of you guys having trouble are working with samples where all of the good stuff is right at the beginning (like single hits). For some reason, it never occurs to me to test with these “left-justified” samples :sweat_smile: All of my samples are field recordings with all of the interesting stuff in the middle.

I’ll try to make a point to use more sample kits. :space_invader:


This probably has nothing to do with my issue, but what does “which side” mean ??? i assumed it was referring to audio channels. If not, this is quite confusing.
More generally (and don’t get me wrong, i love jokes, humor and poetry), it’s maybe worth considering that not anybody reads english fluently. Some of the displayed messages, even though quite pleasant, can be a little disturbing to non-english natives. We are used to a more “common tongue” when it comes to music machines, manuals, etc.

i’m personally ok with it, but i thought it’s worth mentioning.


But I am trying to use common language…what examples are you referring to? Definitely, welcome any language suggestions.


By the way, the How often? How much? and Which side? are not attempts to be cute. I really haven’t been able to come up with better terminology.

How often?
As in, how often does the buffer/slice play back when triggered? Just once or loop forever?

How much?
As in, how much of the buffer should be played? The entire buffer (all) or just a slice?

Which side?
This is the most complicated (and probably unusual) one and really needs a diagram. So don’t feel bad if it is not obvious to you. It has nothing to with English ability. In a nutshell,

  • left: use only the section to the left of the active slice.
  • both: use both the sections to the left and right of the active slice.
  • right: use only the section to the right of the active slice.


Well, i need to apologize because i realize it’s far from being true. I don’t know why i got this impression. Maybe, as i left my hands far from the ER 301 for quite a long time (just got back to test the new firmware), i was left was a false impression.

Period-o-meter comes to my mind but it’s been renamed.
Bump Map is not obvious to me.

It’s more in the “Admin” section that, from my point of view, there is a bit too much literature.
For instance “Bypassing disables unit” instead of "Does bypassing a unit also disable it ?"
or “GUI Animation” simply instead of “Set the GUI animation speed or disable it”.
This sort of things. i mean, the less grammar the better.

But really, it’s no big deal.


No need to apologize for your impressions! I am doing this exercise in public exactly for feedback such as you are giving.

Those both do not have common terms or at least not ones that I familiar with. So I do my best which admittedly probably comes off a bit nerdy and arcane sounding. Also, let’s remember that I’m not trying to do “obvious” things. :sweat_smile: I’m Orthogonal Devices. Orthogonal means different (in the extreme sense) from something else. Naming something that is new and not in the common consciousness of the users is difficult and I need help!

That’s really interesting :astonished: because I thought I was improving the situation for non-native speakers by trying to use complete sentences and not any shortened jargon. These issues are important.


Makes sense.

Instead of “How often”, why not

Play Mode : Once / Loop

Instead of “How Much”,

Play Region : all / slice

Instead of “Which side”,

Well, i’m not sure, something like “Relative Region” : left to slice / around slice / right to slice

Instead of “Address Mode”,

Slice Access : nearest / index / chromatic


Oh and in general, option values and unit names need to be made up of words that are each at most 6 or 7 characters so that they will fit in one of the 6 columns. The new menu system is alleviating that somewhat but still we need to stick to short words.


I really like the new view on menus and the terminology made sense at least for me =)
This is not related on the new firmware, but what kind of filters are the ladders? I’m working on multiband resonator (I think I share this after I’m done, if anyone have any suggestions what there should be tell me) and I’m just trying to decide should the filters crossover area be on the same frequency or should they override each other a bit to get more rid of the original content.

Anyways after update everything seems to be a bit faster, really good job workflow wise =)


Oh my word, all my dreams are even more true!


The answer is complicated due to fragmentation of the sample memory. In general, these destructive edits cause the sample to be cloned and the actual edit is performed there. So assuming no fragmentation, the sample needs to be less than the free sample memory available. Probably less when there is fragmentation.

This is not the final word though. One of these days I will get the virtual memory sub-system up and running (the CPU has the hardware to support it). This is one of those things where you have to roll your own when you shirked Linux and went bare-metal. That and dynamic linking. That one is going to be a doozie.


The Ladder filter units are all 24dB/octave (4th-order) filters. The slope will naturally change though when you crank up the resonance.