Besides the size, I am sad for Spitfire sampler guys, sampling those 4 billion layers.

 

Don't know if this is DAW-dependent (and I don't think so) but last time I checked some pitch values in an editor, there were + - values and a 0 for no change...
AFAIK pitch is transmitted with MSB and LSB and therefore has 127*127 values of which one half is positive and the other negative. So there is a standard zero.

 

This is due to a bad controller or some concussions while playing.

 

:D I think it has to do w/ pitch bend range and separate pitch bend sensitivity. Maybe?

Spoiler

You can have PB that will let you bend (usually) +- a semitone (+- 8192 @ 14 bits) or something else entirely depending on random chance. It seems. I dunno. I don't wanna get into a whole MMJ thingy about it, but I hope they standardize it.

 

Yeah... Every library will be ended in 2119 or later. Not coming soon! Good news for our grandsons' grandsons! It is so because of midi 2.0

 

Ha, this is correct, 127*127 was wrong. Thanks.

That's a good thing.

Well, don't know, I played keys for years and still use a master keyboard to play everything and never had any big issues with faulty PB values...
Let's wait and see how they mess it up this time.

 

i wait for midi3

 

I mean, I don't rly know much about MIDI, my lovely tigery friend, but I know that PB (sometimes) & automation (all times) rly are the two big things that screw up. I just hope it gets fixed! It's weird and spooky how consistently and unconciously I can hit a 64 velocity all the damn time, which feels rly wrong given my ineptitude! There needs to be more accuracy than that!

 

Ever saw a tiger blushing?

You should see how consistently and unconciously I can easily hit a different velocity all the time, that's really impressive.

But you're right, 127 steps are just not enough for an expressive performance.

 

MIDI 2.0 development discontinued? Or it will be announced to be released in 2029?
)

 

They already did it!!
We're prophets!!!

"Get BBC Symphony Orchestra delivered to you on our new 1TB SSD for fast, instant playability."

 

Dave Smith of Instruments fame. One of the creators of MIDI. Thirty seven years later, we're still at Version 1.0.

While you're waiting for the next version, don't hold your breath.

 

Hoping that it will be really backward compatible, I think it's a good thing. Improved velocity dynamics is great for pianos, and it gets nearer to advanced instrument standards like the AvantGrand.

Also, while I understand that anyone is rightly worried about the huge sample libraries that could follow to this new protocol (they're already too fat for my taste), I believe that physical modeling is going to take over samples (it will take a while, though).

 

I'm afraid when they finally implement MIDI 2.0 this fine gentleman will be holding breath... eternally...

 

So this thread is 6 years old now. Did anything change? As far as I know there are at least 2 controllers that support Midi 2.0 now? What an achievement.

I am not too sure about 32 bit controllers though. How do you want to utilize them if we have music that is still rendered to (even mostly below) 96 kHz? Not that there is anything wrong with 96 kHz though.
But my point is, there are a max of 96000 value changes that can be recorded in one second. This means even to be able to record the 4.294.967.296‬ possible values the controller should be able to represent you would need roughly 44740 Seconds if you would record one sample per value, which would be the absolute minimum. That would be roughly 746 minutes or about 12.5 hours for one complete sweep. Using 48 kHz, you would already need more than one day.

Then again, I guess 16/32 bit are the most obvious choices for the controllers?

 

MIDI & timing – we're all familiar with these problems. The MIDI protocol operates at 31,250 bps and uses serial data transmission. For 16 channels, that can be quite tight. The protocol update promises a significant improvement. Parameter resolution is generally expected to be improved:

The MIDI 2.0 protocol brings us a new Universal MIDI Packet. Now, up to 128 bits are possible. This Universal MIDI Packet is only intended for high-speed transmission via (e.g.) USB and is not transmitted via the 5-pin DIN connector; at least, there are currently no plans to do so. Therefore, using a 5-pin DIN connector does not allow (full-fledged) MIDI 2.0.

Previously, most MIDI values were transmitted in 7 bits. A 7-bit resolution allows for a gradation of 128 individual steps. This might be sufficient for velocity sensitivity, but it could be far too coarse for a fader progression. Although some parameters are transmitted with two data words (in 14 bits), this only applies to selected values and not to all MIDI devices. At least this enabled increments beyond 16,000. Velocity will now be transmitted at a generous 16 bits, providing more than 65,000 values for increments. Pitch bend will also run at 16 bits in the future.

I don't think this is all that important for velocity. Most sampling instruments only partially utilize the old capabilities. The resolution of controllers has been increased to 32 bits. This increases the increment to over 4 billion, and nothing burns out at this point – extremely smooth filter transitions are no longer a problem. Any slow parameter changes will sound smoother this way, and parameter jumps will no longer be audible.

EDIT Source: www.amazona.de/alle-informationen-zum-midi-2-0-standard/

 

Last edited: Mar 16, 2025

Currently only these 3 devices support MIID 2.0: Roland A-88 MKII, Yamaha MONTAGE M and Korg Keystage.

 

Even if you just copied (and auto-translated) it from Amazona, this relates to bandwidth, not timing.

Microsoft already released a new MIDI stack with MIDI 2.0 support to Canary last month.
https://devblogs.microsoft.com/wind...ing-windows-midi-services-customer-preview-1/
Maybe it will be shipped with 25H2, but I would guess later.