MIDI 2.0!?

[The Elf]

https://www.midi.org/articles-old/the-midi-manufacturers-association-mma-and-the-association-of-music-electronics-industry-amei-announce-midi-2-0tm-prototyping?fbclid=IwAR3yojtbqXc52uTwrBV4uaUV7JdsMHMKIXA2NudhUH4mw8uPlmbxAPoDW3Q

[Eddy Deegan]

The MIDI 2.0 initiative updates MIDI with auto-configuration, new DAW/web integrations, extended resolution, increased expressiveness, and tighter timing

I'm torn. On paper, some of those things look useful but the optimistic cynic in me hopes that they don't mess up a really solid protocol with decades of implementation under its belt and cause chaos as questionable support emerges on new hardware.

Not so sure about the DAW/Web integrations part either - that sounds like an unstable recipe.

That said, if it's backwards compatible... though I for one would be rather sorry to see the DINs go - I wonder if they'll adopt USB or something like it as 'the' physical interface.

Still, we'll see...

[Jumpeyspyder]

Great news!, It's a shame its taken so long!!

I can't help thinking this should have been happened 15 years ago!

Not so sure about the DAW/Web integrations part either - that sounds like an unstable recipe.

I think computer compatibility is vital as many musicians work mostly ITB

I for one would be rather sorry to see the DINs go - I wonder if they'll adopt USB or something like it as 'the' physical interface.

Still, we'll see...

Basing MIDI 2,0 on USB is a great suggestion and IMHO would be a very clever move !
Standard USB A-B cables are cheap, compact, reasonably strong and might remove the need for dedicated MIDI interfaces ?

[Sam Spoons]

How will it work without a host computer, USB/A to USB/A cables (which are rare as hens teeth)? Or would every device have USB/A and USB/B connectors?

[Folderol]

Hmmm,
Cautiously optimistic. The MMA have been very good 'gatekeepers' so far. Let's hope it stays that way.

I tend to regard USB connectivity with some reservation. It has been notoriously flaky (although it's getting better) and represents a colossal processing overhead - not exactly what you want it time-critical systems!

[The Elf]

USB seems, to me, to be the worst option. Time for MIDI to embrace the network, I'd say. As long as it does this then whatever network connection hardware (wired or wireless) arrives in the future, MIDI should happily tag long.

[wireman]

wonder if they'll adopt USB or something like it as 'the' physical interface.

Not thought about it much but how about a choice of protocols using UTP RJ45 sockets
There would be two variants, serial and Ethernet / TCP/IP defined by the standard.

[Folderol]

wonder if they'll adopt USB or something like it as 'the' physical interface.

Not thought about it much but how about a choice of protocols using UTP RJ45 sockets
There would be two variants, serial and Ethernet / TCP/IP defined by the standard.

You don't want TCP!
It guarantees that all packets will be delivered, but not when, nor what order - this could make for an 'interesting' performance
{Eric Morcambe}
"I'm playing all the right notes. Not necessarily in the right order"

[wireman]

wonder if they'll adopt USB or something like it as 'the' physical interface.

Not thought about it much but how about a choice of protocols using UTP RJ45 sockets
There would be two variants, serial and Ethernet / TCP/IP defined by the standard.

You don't want TCP!
It guarantees that all packets will be delivered, but not when, nor what order - this could make for an 'interesting' performance
{Eric Morcambe}
"I'm playing all the right notes. Not necessarily in the right order"

Like I said, I have not thought about it. Modern switched fabrics are pretty reliable and you would think that the packet rate is fast enough nowadays that TCP with any reordering/retransmission should work for MIDI without requiring some sort of real time solution (and these do exist).

[muzines]

Great! I look forward to reading a new SOS article all about MIDI - just like the good old days of the 80s..!

[Drew Stephenson]

Excellent, all this time not learning midi has been the right approach as I'll be able to learn about midi2.0 from scratch.

Alternatively they'll now be twice as much midi that I don't understand.

[Folderol]

Hate to upset you BD, but it's the latter.
Having watched a vid. on the matter, it seems the MMA have indeed done their homework, and the new system will be completely backward compatible.
P.S.
It is here

Also, it will be transport agnostic, so DIN, ETHERNET, USB whatever you like - although guy did point out the latency issues with Blutooth and wireless.

[Drew Stephenson]

Yeah, I thought that was going to be the case. So much to learn, so little time.

Because so much time is wasted on here....

[Drew Stephenson]

That is a fascinating video Folderol.

Just reminds you what we're capable of when people work together.

[The Elf]

Also, it will be transport agnostic, so DIN, ETHERNET, USB whatever you like

JUST the way it should be. Sounds like a very good start.

[Eddy Deegan]

All very encouraging!

I think my initial reservations were broadly based around the feeling that MIDI 1.0 is one of the most 'wholesome' technical innovations around, and if they can do anything like as good a job with 2.0 then I'll be a happy bunny.

The transition period should be interesting ...!

[ef37a]

Too late and I am too knrd to read the links but to my mind there is only one logical connector for "new MIDI" ? The RJ45.
The cables are cheap and you can get them at Rock Botton type shops. Almost certainly well below CAT5e spec but easily good enough for this application. Patch bays can be very compact and wired with an IDC tool.

TRS is an audio connector (as was DIN) RJ45 IS data.

Dave.

[Eddy Deegan]

Too late and I am too knrd to read the links but to my mind there is only one logical connector for "new MIDI" ? The RJ45.
The cables are cheap and you can get them at Rock Botton type shops. Almost certainly well below CAT5e spec but easily good enough for this application. Patch bays can be very compact and wired with an IDC tool.

TRS is an audio connector (as was DIN) RJ45 IS data.

Dave.

Well, if it's physical link independent then it's up to the manufacturers of course. I would expect they'll standardise on something.

STP ethernet cables are great from a cabling point of view, and RJ45 is a compact and efficient connector but I'm not a fan of the RJ45 in many implementations because I've encountered a number of circumstances where a slightly dodgy plug (or more specifically, the little plastic lever you press to release it) causes no end of problems trying to extract the connector from the socket, and also if you trip over a cable terminated with a plugged-in RJ45 you can be in for a world of hurt trying to decouple it afterwards.

We need better RJ45 terminators basically. Maybe Neutrik or someone could come up with something a little more robust than the injection-moulded levers on conventional CAT-x cables if STP is adopted widely for MIDI 2.0!

[ef37a]

Hi Eddy, one of those situations I think where, to misquote Flog It, "quality matters".

In my last "proper" job I was responsible for building test beds for CATx modules and these had thousands of RJ45S plugged in and out of them weekly. Jammed plugs were very rare but then these were good quality plugs fitted with an AMP crimp tool. Often the even more rugged shielded plug.

If you do get a jam the latch can be released with a small screwey, about 2mm AF. The jacks are a "blind" hole there so no danger of electrical zapping.

There has always been a sprinkling of posts re dodgy connectors. I had some XLR 3's jam really well..Who said "There is very little in this world that cannot be made a little cheaper and a little worse"?

Just a thought. Where you have an expensive bit of kit it is worth making up "sacrificial" stubs so the onboard connectors never get strained. I did this at Blackstar for the AP test rig. Even relativley cheap kit like a cable tester wil benefit because it is a PITA if the sockets break.

Dave.

[The Elf]

I never had a problem with the 5-pin DIN, but why on earth didn't they utilise all five connections so we didn't have to use two cables all the time? MIDI in on the inner pins and out on the outer pins, for example. That always seemed like a missed opportunity to me...

[muzines]

No they needed to be 8ndependant, you’re not always connecting the same devices bidirectionally. Or else you’d need breakout cables, but it would be a confusing mess...

[The Elf]

No they needed to be 8ndependant, you’re not always connecting the same devices bidirectionally. Or else you’d need breakout cables, but it would be a confusing mess...

A receiving instrument that didn't want to receive would just ignore the incoming stream. Works for USB!

[Dave B]

Dave Smith always pushed for midi to be a 'star' topology, but the Roland bods kept pushing back that they wanted the 'thru'. They got their way which is why we have all these din cables everywhere.

One of the interesting thins about midi over usb is that it finally brings the concept back to Dave's original idea - a star topology.

[Folderol]

When the MIDI spec. was set there was no such thing as USB. Also the instrument manufacturers were thinking in terms of controllers direct to synths without any form of hub - for which separate connectors make sense. Computers were expensive, unreliable and barely entered into their thoughts. Dave Smith was significantly ahead of his time in that respect.

[The Elf]

They didn't need to time travel to see the benefit of a bi-directional cable. MIDI was based on RS232, which was itself bi-directional. The pins were sitting there waiting to be used. It could have solved us all this two-cable mess we ended up in - and the difference between 'in' and 'out' seems to baffle many people - myself included on occasions!

Atari used the spare pins as a MIDI Thru on the Atari ST, which caused a lot of trouble, since by then many cables had the spare pins shorted to avoid confusion over which pins were the correct ones! The number of times I saw that one...

All water under the bridge and out to sea, though. Let's hope the backwards compatibility works as seamlessly as it always does...