Jump to content


Please note: You can easily log in to MPN using your Facebook account!

Some Timely MIDI 2.0 News!


Recommended Posts

There's now a standard data transfer method over USB that combines MIDI 2.0 and MIDI 1.0 data. So, why does this matter?

 

· Better support for high-speed throughput (up to hundreds of times MIDI"s original speed)

· Old and new devices work with any operating system that supports the new USB Class Definition for MIDI Devices

· As with the previous version, no drivers needed for compliant devices

· Supports up to 256 MIDI Channels in 16 Groups

· Can provide more accurate timing for dense MIDI streams

· Over time, simpler to implement than USB MIDI 1.0

· Enhances using MIDI 2.0 mechanisms including MIDI Capability Inquiry (MIDI-CI)

· Added support for Interrupt transactions for better control over jitter and throughput

 

There's more info in the MIDI Association forum, as well as a link to download the spec if your hobby is MIDI rocket science. :)

Link to comment
Share on other sites



  • Replies 14
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Nice!

 

I am content to allow the geniuses to figure this all out. It sounds like there is a consistent paradigm guiding developement and a path moving forward without leaving the past behind.

That should engender full adaptation by all concerned.

It took a chunk of my life to get here and I am still not sure where "here" is.
Link to comment
Share on other sites

I am hopeful that somewhere we will see instruments designed with high-resolution controllers. My ownership of the VAX-77 and Non-Linear Labs C-15 confirms for me that the higher resolution is very helpful for nuanced playing.

 

You'll be happy to know that now there's a spec to accommodate them :)

Link to comment
Share on other sites

Years ago I read an interview of the musicians in Garbage. They had given up on MIDI recording because it was not tight enough. They built all song using audio recordings. I later found that a lot of musicians felt this way. Is this new spec getting us to the point that we can record MIDI parts with timing tight enough to satisfy professional recording musicians?

This post edited for speling.

My Sweetwater Gear Exchange Page

Link to comment
Share on other sites

Years ago I read an interview of the musicians in Garbage. They had given up on MIDI recording because it was not tight enough. They built all song using audio recordings. I later found that a lot of musicians felt this way. Is this new spec getting us to the point that we can record MIDI parts with timing tight enough to satisfy professional recording musicians?

 

I don't know if I agree that the original spec wasn't tight enough...I did a lot of research on the subject, and found that computer jitter was the main problem. With a Mac, you had the same timing discrepancies whether you had 240 pulses per quarter note or 960 ppq. The computer had to work four times harder with the higher PPQs, which led to four times the timing errors.

 

But that was then, and this is now. Computers are better, and MIDI 2.0 now provides time-stamping to virtually eliminate jitter. What's even cooler is that there's no reason why this can't also work with MIDI 1.0 gear. There are actually many aspects of MIDI 2.0 that can be added to MIDI 1.0 gear, it's more than just backward-compatible.

Link to comment
Share on other sites

I don't know if I agree that the original spec wasn't tight enough...I did a lot of research on the subject, and found that computer jitter was the main problem. With a Mac, you had the same timing discrepancies whether you had 240 pulses per quarter note or 960 ppq. The computer had to work four times harder with the higher PPQs, which led to four times the timing errors.

 

I remember reading in Roger Nichols' column in EQ magazine about when he was working with Steely Dan. After experimenting with everything they could find, they settled on the Atari ST (don't remember what program) as being the only one that had MIDI timing stable enough to make them happy.

 

We indeed have come a long way since then.

Link to comment
Share on other sites

If you're having problems with MIDI being too slow, not tight enough, it's not "MIDI" that's the problem. It's the latency caused by the round trip through the computer where your virtual instrument is probably too CPU-hungry for the computer you own. Or you have a really terrible and/or old MIDI interface.

 

Such an odd thing for Butch Vig to say about MIDI - he was one of the big MIDI freaks in the 90s.

 

I should know - I've had many problems with latency in VST instruments over the course of decades. There are solutions that don't involve ditching MIDI as a simple digitial communication protocol. It works great if you have enough computer horsepower and set it all up right.

 

Maybe there are people who's ears are so attuned to tightness that 2 ms bothers them. I don't think that's a normal person.

 

But what am I putting my 2 cents in here for when Craig is the MIDI guru of gurus. Listen to the man!

 

nat

Link to comment
Share on other sites

If you're having problems with MIDI being too slow, not tight enough, it's not "MIDI" that's the problem. It's the latency caused by the round trip through the computer where your virtual instrument is probably too CPU-hungry for the computer you own. Or you have a really terrible and/or old MIDI interface.

 

That's why time-stamping is a big step forward, even with stressed computer systems, they'll know at what precise time to spew out data. As long as they have a stable clock (and that's pretty much a given), even those who do find 2 ms a problem should be okay IF the MIDI gear itself doesn't add delay.

 

A good analog is your brain knows when it wants your fingers to play a note, the problem isn't with your brain...it's with your fingers executing the command.

Link to comment
Share on other sites

That's why I use an array of hardware synthesizers. The latency is so slight, for all practical purposes it isn't there.

 

Plus since they all have about the same minuscule latency I can mix and match from as many synths and synth modules as I like to get the best sound for each voice.

 

And when the computer OS abandons software synths, those 5 pin DIN synths still work.

 

Until computers get strong enough and soft-synths get light enough, I'll watch and wait.

 

But there is more than one right way to make music. My way isn't best for everyone.

 

Insights and incites by Notes

Bob "Notes" Norton

Owner, Norton Music http://www.nortonmusic.com

Style and Fake disks for Band-in-a-Box

The Sophisticats http://www.s-cats.com >^. .^< >^. .^<

Link to comment
Share on other sites

That's why I use an array of hardware synthesizers. The latency is so slight, for all practical purposes it isn't there.

 

You might be surprised...measure the delay on scanning keyboards between hitting the lowest and highest notes of complex chords. In reality, many of the processors in older synths are underpowered chips compared to what's available today. If you don't hear latency with hardware synths, that probably means 10 - 20 ms of delay isn't really objectionable. Computers+virtual instruments can hit that level too, especially if you're running the synth stand-alone.

Link to comment
Share on other sites

That's why I use an array of hardware synthesizers. The latency is so slight, for all practical purposes it isn't there.

 

You might be surprised...measure the delay on scanning keyboards between hitting the lowest and highest notes of complex chords. In reality, many of the processors in older synths are underpowered chips compared to what's available today. If you don't hear latency with hardware synths, that probably means 10 - 20 ms of delay isn't really objectionable. Computers+virtual instruments can hit that level too, especially if you're running the synth stand-alone.

Thanks!

 

That's nice to know. How about softsynth latency when running 5 to 10 different synths at the same time?

 

When I make a backing track for my duo, I often mix 6 or more. Some of the old synths like the TX81z, DS8 or MT32 have dated sounds, but some of those dated sounds work perfectly, especially when covering an 1980s - 1990s song (we take requests). And some of those sounds just sound good, while others sound cheesy - but I can use the good and avoid the cheese.

 

10 ms is about like a musician 10 feet away. I recall reading that one of my synths (I think it was the Ketron SD-2 but not sure) being 6ms and I can't hear any of them 'flamming' when using them, so I suspect they are as close as musicians in a room --- with one exception: The physical modeling Yamaha VL70m. But it expresses itself so much like an acoustic instrument it's worth it. I only use that with my wind MIDI controller, which I've learned to play a tad ahead of the beat.

 

About the only time I use software synths is when writing new styles for Band-in-a-Box. I use the soft synth (VST) provided with BiaB because I know many of my customers will use that synth. I often have to adjust the BiaB display to compensate for the synth. I remember the first one being over 400ms, and the newest one around 100ms.

 

If I can run over a half dozen soft synths at the same time with no latency and no adjusting the tracks to compensate for the latency of different synths it might be time for me to move on.

 

Notes

Bob "Notes" Norton

Owner, Norton Music http://www.nortonmusic.com

Style and Fake disks for Band-in-a-Box

The Sophisticats http://www.s-cats.com >^. .^< >^. .^<

Link to comment
Share on other sites

Well...the more soft synths, the harder the computer has to work, and that can increase latency. With your hardware synths, the latency is what it is - the synths won't affect each other.

 

But also bear in mind that different soft synths and plug-ins stress out a computer to varying degrees. In Studio One, it's interesting to see the Performance view, and find out just how much CPU each plug-in draws, and how much latency it adds.

 

Also note that you can render the synths as audio tracks, which means negligible latency. I don't know how much synth tweaking you do during a performance, though. When I've had to load multiple songs without knowing the order in which I'd need them, I put all the audio tracks for the songs in Folder tracks, so there was a folder for each song. I could solo a Folder track, and voila - one click, and good to go.

 

Craig

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Unfortunately, your content contains terms that we do not allow. Please edit your content to remove the highlighted words below.
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...