You are here

MIDI Channel 10 is bad for drums - fact or fiction?

For fans of synths, pianos or keyboard instruments of any sort.

MIDI Channel 10 is bad for drums - fact or fiction?

Postby BillB » Tue Oct 22, 2019 11:42 pm

I am starting to set up an Akai MPC1000 as the MIDI sequencer for my hardware-based studio. I plan to set up templates for MIDI recording, which will be consistent across the various synths, modules and drum machines.

I have read that, despite it being the General MIDI standard, Ch10 is not good for drums, with Channel 1 providing ‘tighter’ timing. As it is just a matter of the order of 4 bits in the MIDI message, it is not obvious why it should make any perceptible difference - unless, in a multi-Channel data stream, the drum voices ‘come tenth’.

On the other hand, with Ch 10 being the ‘standard’, sticking to it will offer higher levels of compatibility with a wider range of hardware/software. Purely as an example, the Akai Tomcat / Rhythm Wolf drum machines have fixed Ch10 assignment for the drum voices.

So, Channel 01 for tightness (or does it not matter?) or Channel 10 for GM etc compatibility?

Opinions please....
BillB
Frequent Poster
Posts: 757
Joined: Mon Apr 07, 2003 12:00 am
Location: East Yorkshire

Re: MIDI Channel 10 is bad for drums - fact or fiction?

Postby desmond » Wed Oct 23, 2019 12:13 am

BillB wrote:So, Channel 01 for tightness (or does it not matter?) or Channel 10 for GM etc compatibility?

It probably doesn't matter.

You can run your own tests if you like that sort of thing. The channel 1 thing came from early MIDI/software sequencers that processed events by track/channel, so when they would output MIDI notes on a given clock, they would run through in channel order - so any events on channel 1 would get sent first, then channel 2 etc.

I think (I should check) it was early Pro24 or Cubase that specified for the tightest timing, use tracks at the top of the track list, and events on MIDI channel 1, as they would always go "first" - and the more MIDI data in general, the more likely events on lower tracks and channels will get pushed back a little due to higher priority events getting sent first.

In your case, I don't think it will matter, so stick to 10 if it makes logistical sense. :thumbup:
User avatar
desmond
Jedi Poster
Posts: 8992
Joined: Tue Jan 10, 2006 1:00 am
mu:zines | music magazine archive | difficultAudio

Re: MIDI Channel 10 is bad for drums - fact or fiction?

Postby Ben Asaro » Wed Oct 23, 2019 1:26 am

I've been using channel 10 since the mid 80s. I think even if channel 1 is tighter, I'm so used to it I wouldn't be able to handle the change! :tongue:
Ben Asaro
Frequent Poster
Posts: 573
Joined: Thu Feb 26, 2009 1:00 am
Location: NYC

Re: MIDI Channel 10 is bad for drums - fact or fiction?

Postby resistorman » Wed Oct 23, 2019 5:03 am

Ben Asaro wrote:I've been using channel 10 since the mid 80s. I think even if channel 1 is tighter, I'm so used to it I wouldn't be able to handle the change! :tongue:
Ha!

Never heard of this, and I’ve been using midi since its beginning.
User avatar
resistorman
Frequent Poster
Posts: 678
Joined: Sun Nov 22, 2015 1:00 am
Location: Asheville NC

Re: MIDI Channel 10 is bad for drums - fact or fiction?

Postby Folderol » Wed Oct 23, 2019 7:07 am

It depends...
If you have a number of events happening at exactly the same time, channels are transmitted in sequence, but there is no such thing as the 'first' one in absolute time. However, ones carrying sharp attack instruments should ideally be as close together as possible, so leave drums where they are and put (say) a track with piano at 9 and brass at 11.

If they are a millisecond or more apart, it won't matter anyway, as channels are only transmitted when there is something on them to send.

Probably more relevant is don't use MIDI THRU if you can possibly avoid it - especially not on kit that uses active sensing.
User avatar
Folderol
Jedi Poster
Posts: 9195
Joined: Sat Nov 15, 2008 1:00 am
Location: The Mudway Towns, UK
Yes. I am that Linux nut.
Onwards and... err... sideways!

Re: MIDI Channel 10 is bad for drums - fact or fiction?

Postby BJG145 » Wed Oct 23, 2019 8:06 am

All I would say is, don't use channel 13. Your MIDI channels should skip from 12 to 14, like US hotel floors. :beamup:
User avatar
BJG145
Frequent Poster (Level2)
Posts: 3916
Joined: Sat Aug 06, 2005 12:00 am

Re: MIDI Channel 10 is bad for drums - fact or fiction?

Postby Wonks » Wed Oct 23, 2019 8:42 am

Avoid it like the plague for drums if using a hardware MIDI sequencer to play everything from hardware MIDI modules. Zuke will tell you the first thing he and his other programming mates did back in the 80's and 90's was set the drums up on channel 1. The drum machines often had their own timing foibles, so you didn't want extra timing issues being added in. Channel 10 drums were noticeably laggy.

But for modern software MIDI within DAWs? I doubt it makes any difference at all now.

If you're not sure, then just try recording a part on channel 1 and then channel 10 and seeing if you can see any timing differences on the resulting waveforms.
User avatar
Wonks
Jedi Poster
Posts: 10147
Joined: Thu May 29, 2003 12:00 am
Location: Reading, UK
Correcting mistakes on the internet since 1853

Re: MIDI Channel 10 is bad for drums - fact or fiction?

Postby BillB » Wed Oct 23, 2019 8:51 am

Thanks for all the comments, folks.

Resistorman, it’s definitely ‘a thing’, although I can’t recall the source(s) - seen it several times over the decades.

Desmond, I’m sure you are right that it is something archaic from the early days of multi-Channel sequencing - it would be interesting to know what / when. Not sure what search term to use on Muzines.... :D

Will, good point about using softer sounds with higher channels, although assuming that synths have been pre-allocated channels (just to maintain the sanity of a fixed system) no one synth will always be on, say, string synth duties. So that one may have to slide a little.

As for avoiding Channel 13... I shall be scientific, like NASA, and boldly jump regardless into one of the most unlucky missions ever. There’s no room for superstition in MIDI, or Apollo missions :headbang:

Another thought is that the MPC has two MIDI outs, so it would be possible to dedicate one to the ‘drum modules’ (Volca Beats, TR-626, EMX1, Alesis DM5) although these would have to go via a Thru box or switch. Other possible percussion sources (eg JV-2080) would just have to carry on via the other outputs with the rest of the synths. Notwithstanding Will’s advice about avoiding Thru, it is not possible to avoid having one/some somewhere, so maybe a single Thru box feeding the drum units.... I think you can get away with driving two MIDI inputs from a single MIDI out via splitter lead, but four might be pushing it!
just when I thought I had finished wiring MIDI :headbang:

Thanks for you thoughts, any others are welcome :D
BillB
Frequent Poster
Posts: 757
Joined: Mon Apr 07, 2003 12:00 am
Location: East Yorkshire

Re: MIDI Channel 10 is bad for drums - fact or fiction?

Postby BillB » Wed Oct 23, 2019 8:55 am

Thanks, Wonks, so it is ‘a thing’, I didn’t just make it up...

I guess you are right, I should really do some scientific tests :crazy:

But maybe, if drums are run from their own port and not in the main 16-channel stream, the ‘drum’ channel wouldn’t make any difference.
BillB
Frequent Poster
Posts: 757
Joined: Mon Apr 07, 2003 12:00 am
Location: East Yorkshire

Re: MIDI Channel 10 is bad for drums - fact or fiction?

Postby The Elf » Wed Oct 23, 2019 9:17 am

This all comes from the early days of MIDI. Nowadays I doubt it's an issue at all.
User avatar
The Elf
Jedi Poster
Posts: 13062
Joined: Tue Aug 14, 2001 12:00 am
Location: Sheffield, UK
An Eagle for an Emperor, A Kestrel for a Knave.

Re: MIDI Channel 10 is bad for drums - fact or fiction?

Postby BillB » Wed Oct 23, 2019 9:26 am

The Elf wrote:This all comes from the early days of MIDI. Nowadays I doubt it's an issue at all.

Even in a hardware-based system?
BillB
Frequent Poster
Posts: 757
Joined: Mon Apr 07, 2003 12:00 am
Location: East Yorkshire

Re: MIDI Channel 10 is bad for drums - fact or fiction?

Postby Moroccomoose » Wed Oct 23, 2019 9:33 am

I guess most modern midi is now either in the box or goes through USB, so most instruments have their own port, and unless you are really hammering multi timberal, I'd wager there is rarely much more than one channel in use most of the time anyway. So I'd be interested to know more about how the timing between ports is managed.

That said I have just set up my Cubase template for a Proteus Virtuoso for orchestration. Its an old school rompler, with 32 MIDI channels over 2 ports and 128 voice multi timbral. Interestingly, I have set the channels up to reflect the orchestra pit, left to right and front to back. This means that all the slow attach stuff (Strings) are using the lower channels and the fast attack stuff (Brass and percussion) is all in the higher channels.

While I am convinced that I will never be able to hear any discernible timing discrepancies - especially after they have gone through Elf's reverb delay technique (from another recent thread), I wondered if there is a MIDI standard or protocol which defines how orchestral instruments' channels are arranged, beyond channel 10 for percussion. And if there is, whether it is sensible given the points raised in this thread.

Stu.
Moroccomoose
Regular
Posts: 192
Joined: Tue Apr 12, 2011 12:00 am
Location: Staffordshire
 

Re: MIDI Channel 10 is bad for drums - fact or fiction?

Postby BillB » Wed Oct 23, 2019 9:56 am

Moroccomoose wrote:I guess most modern midi is now either in the box or goes through USB, so most instruments have their own port.

Yes, that's how I have it when driving from a DAW. I'm lucky enough to have a couple of UM-880s and a UM-550, so most things have their own port.

But when you are looking at driving from a hardware sequencer (albeit one with two ports), channels are the only way to direct parts to devices, thus the question.

Apart from GM drums on Ch10, I am not aware of any MIDI protocols which propose a relationship between instrumentation and channels. But who knows: someone, somewhere, probably has! GM is mainly concerned with having program change numbers per instrument, plus various standard controller messages.

Having said that, and based on some of the comments above, I think I will be setting more percussive instruments (eg monosynths likely to take bass duties) to lower channels). Apart from that (and drums) it is likely that any system that makes sense to you is a good one.
BillB
Frequent Poster
Posts: 757
Joined: Mon Apr 07, 2003 12:00 am
Location: East Yorkshire

Re: MIDI Channel 10 is bad for drums - fact or fiction?

Postby Martin Walker » Wed Oct 23, 2019 10:05 am

desmond wrote:I think (I should check) it was early Pro24 or Cubase that specified for the tightest timing, use tracks at the top of the track list, and events on MIDI channel 1, as they would always go "first" - and the more MIDI data in general, the more likely events on lower tracks and channels will get pushed back a little due to higher priority events getting sent first.

This - I certainly remember reading it about happening with Cubase (and may even have written about it at the time as well ;) )

As Desmond says, I doubt that it would make any difference timing-wise nowadays, although bets might be off if you overlay loads of controller & pitch bend messages on your drum MIDI note data stream.


Martin
User avatar
Martin Walker
Moderator
Posts: 14602
Joined: Wed Jan 13, 2010 9:44 am
Location: Cornwall, UK

Re: MIDI Channel 10 is bad for drums - fact or fiction?

Postby The Elf » Wed Oct 23, 2019 10:07 am

BillB wrote:
The Elf wrote:This all comes from the early days of MIDI. Nowadays I doubt it's an issue at all.
Even in a hardware-based system?
Buyer beware. I doubt that all hardware is made equal.
User avatar
The Elf
Jedi Poster
Posts: 13062
Joined: Tue Aug 14, 2001 12:00 am
Location: Sheffield, UK
An Eagle for an Emperor, A Kestrel for a Knave.

Re: MIDI Channel 10 is bad for drums - fact or fiction?

Postby BillB » Wed Oct 23, 2019 10:16 am

The Elf wrote:
BillB wrote:
The Elf wrote:This all comes from the early days of MIDI. Nowadays I doubt it's an issue at all.
Even in a hardware-based system?
Buyer beware. I doubt that all hardware is made equal.
So... it may be an issue...

I'm not trying to make a big deal about it, just to understand whether it should be a consideration in designing a channel-based template for the MPC1000.
I have everything from 'probably not' to 'yes definitely' at the moment :headbang:
... which is fine - all interesting comments.
BillB
Frequent Poster
Posts: 757
Joined: Mon Apr 07, 2003 12:00 am
Location: East Yorkshire

Re: MIDI Channel 10 is bad for drums - fact or fiction?

Postby BillB » Wed Oct 23, 2019 10:18 am

Martin Walker wrote: bets might be off if you overlay loads of controller & pitch bend messages on your drum MIDI note data stream.

Good point, Martin. It would be sensible to keep percussion channels (ports!) clear of controller info.
BillB
Frequent Poster
Posts: 757
Joined: Mon Apr 07, 2003 12:00 am
Location: East Yorkshire

Re: MIDI Channel 10 is bad for drums - fact or fiction?

Postby Wonks » Wed Oct 23, 2019 10:21 am

I think you've misread Martin's comment. The data will be on the drum stream as it's affecting the drum sounds.

If it's a hardware unit, just make the drums channel 1 and stop worrying about it.
User avatar
Wonks
Jedi Poster
Posts: 10147
Joined: Thu May 29, 2003 12:00 am
Location: Reading, UK
Correcting mistakes on the internet since 1853

Re: MIDI Channel 10 is bad for drums - fact or fiction?

Postby The Elf » Wed Oct 23, 2019 10:23 am

BillB wrote:
The Elf wrote:
BillB wrote:
The Elf wrote:This all comes from the early days of MIDI. Nowadays I doubt it's an issue at all.
Even in a hardware-based system?
Buyer beware. I doubt that all hardware is made equal.
So... it may be an issue...
Well the problem is that unless you know how the software within the hardware is functioning it really is impossible to advise with certainty. When we were all running Commodore 64s and Atari STs then I can tell you that there definitely *was* an issue using higher channel numbers. These days, with ultra-fast processing, there's really no reason it should be an issue. But when cost-cutting and human error enter the fray...

I'm with Wonks - if the potential troubles you then use Channel 1 and end the speculation.
User avatar
The Elf
Jedi Poster
Posts: 13062
Joined: Tue Aug 14, 2001 12:00 am
Location: Sheffield, UK
An Eagle for an Emperor, A Kestrel for a Knave.

Re: MIDI Channel 10 is bad for drums - fact or fiction?

Postby BillB » Wed Oct 23, 2019 10:43 am

The Elf wrote:unless you know how the software within the hardware is functioning it really is impossible to advise with certainty.

Accepted, in the end all I can expect is people's experience and practice...

The Elf wrote:I'm with Wonks - if the potential troubles you then use Channel 1 and end the speculation.

Hah - it troubles me if it is a problem... thus the reason for asking. The disadvantage of setting drums to Ch1 is that you lose out-of-the-box compatibility with the majority of new hardware or software introduced into the studio (like the Tomcat example I gave - HARDWIRED to Ch10!). That's the conundrum. If it was an entirely isolated system and I was never likely to introduce or connect to anything new :lol: then yes, Ch1.
BillB
Frequent Poster
Posts: 757
Joined: Mon Apr 07, 2003 12:00 am
Location: East Yorkshire

Next

Who is online

Users browsing this forum: No registered users