FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 

Midi at 44.1 ES4, Report
MUFF WIGGLER Forum Index -> Expert Sleepers  
Author Midi at 44.1 ES4, Report
Just a little report for those interested.

Pretty good success here doing Midi in a 44.1/24bit session using the ES4 controller which isn't officially supported (48k and higher)

I've finally made up 8 cables and have successfully run 8 simultaneous midi outs from one gate expander.

I'm using an aurora 192 and PT9 with the plugin wrapped. I'm sending via AES and then the transformer and resistor pad as described by OS to the ES4. Tight sounding midi (much much tighter than the midi interfaces via the mac) - success with following midi keys Juno 60 kenton , Prophet 5 kenton, Jupiter 4 with IO midi, Midimoog, Polyevolver, Juno 106. Voyetra 8 worked but a bit loose, have to look into that as it's going via a filtering box, and the Sidstation didn't like it, maybe some powering issue with this - it has been flakey/loose when just run from the midi interface prior to the ES4. A few more keys to try. There's always the synths own midi latency/jitter to consider btw. I haven't tried the CV side of the ES4 at the same time as yet as i'm setup with ES3's.

As each synth has it's own cable i'm bypassing the channel output being stuck on one for the time being and have just set all to receive on ch 1. No daisy chaining.

Midi clock still a bit sloppy in PT, watching the machinedrum and evolver clocks wobbling a bit, i know here we have a possible Pro tools specific problem, and again 44.1 is not officially supported by oS.

Also success running the ES4 plug in VSTis Maschine within PT, setting up 8 plugins and routing all to a common AUX in tools then out to the ES4 as described above. Interestingly sending midi clock from within Maschine is tighter

Os - One note on the midi clock, is there anyway to prevent it from stopping when the DAW loops? In PT if i'm doing a 4 bar sequence when it loops round the midi device being clocked stops. It's fine if left to run unlooped. Only other prob is occasional hangs of notes when changing sequences as I mentioned before.

This is amazing for me, i'm keeping the es3's as well to run the CV stuff and just using the ES4 purely for midi now, probably get another expander. Being able to run at 44 means I don't have to convert all the sessions I recieve to get this to work. I wonder if the midi side of the ES4 could be spun off into a separate set of plugs? CC control and all that stuff, patch change, etc etc. ( For us PT users who can't use Midi arpeggiators and the like maybe a built in arpeggiator?) Good work again OS!
Great to hear someone's really using this feature in earnest.

actionmusic2 wrote:
Os - One note on the midi clock, is there anyway to prevent it from stopping when the DAW loops? In PT if i'm doing a 4 bar sequence when it loops round the midi device being clocked stops. It's fine if left to run unlooped.

Yes, should be OK - just an oversight on my part.
Please keep the ES4 add on within the SW voice controller though.
It's working great for me, and many of the older projects I'm still working on are 44.1Khz.
Hoping to switch to 96khz at some point.

I appreciate you posting this info. It helped me get a better grasp on it before I got my ES-4. I thought I would post some interesting results from the tests I've done on my midi gear.

For timing tests, I worked in Sonar 8.5, running SW inside Bidule. I did the majority at 44.1/24bit. I would just do two sequences of 16ths(50% length) separated by a few measures at 120bpm.

Overall, it's just nice to be able to confirm how much the receiving device contributes to the latency/jitter I'm used to. It's such a strange mixed bag, though. It has exposed things like the midi clock in Sonar might be questionable. Anyway, here's a breakdown by synth that really tells the story...

Nord 3 Rack

This is pretty much unusable at any sample rate. It was the first one I tried. I expected the second best results from this thing but it actually made me think the ES-4/GE weren't working right. It wouldn't even trigger notes at 44.1 or 48. The midi in light would show it was receiving data but nothing would happen. At 96, it would respond but if you fed it sequenced data, it would lock up the first notes it saw. Disappointing and a bit of a mystery.

UPDATE April 2014: The newly released ESX-8MD has the Nord working at 48kHz. It is amazingly tight and is basically in the same ballpark as the Akai below. Maybe a bit better on latency. Now, all my synths work with Expert Sleepers hardware/software. Disappointed no more. Quite the opposite.

UPDATE May 2015: The ESX-8MDmk2 has it working at 44.1kHz!


The ES-4 really can't help this thing too much. It's just sloppy. I can think of a few reasons why this is not a surprise.


For its age, it's pretty tight. Maybe around 2ms jitter - but it's a smooth jitter...if that makes sense. It moves like a wave. Really, the big thing is the reduced latency. This is true on most of the synths I checked. I saw as much as 4ms difference between the Motu Midi Express and the ES-4. It really makes things feel tighter with normal midi playback.


Considering the 106, this thing is pretty bad. It still has a considerable amount of jitter. Not as bad as the 600, though.


This one is super tight. I'd say around 1ms jitter. Nice. A solid third place.

K5000r and K-4r

These two I didn't test timing but I used them in existing projects and you could tell they were tighter. It made a huge difference in feel.


Tighter than the JP-8000. Not a surprise. It does not like the clock from Sonar. I actually thought it was a Mopho defect but Bidule standalone and Reaper are rock solid. Note timing was always tight, though. I expected it to be the best. I'm shocked how wrong I was.

Akai S-3000(non-XL)

Average jitter: 2 samples


I had to leave the room after the first test. Once I composed myself and tested again - same results. There is some latency from the input but it is easy to adjust for. I've had this since 94 and I've been moving more to Kontakt over the last half-decade but I think I might be reconsidering. At least for drums. I still can't believe this. And this is 44.1/24bit!

I do get a hanging note issue on many synths if I stop playback when notes are sustained. As has been mentioned, a panic button would be nice. But this is my only real gripe at this point. Other than that, this is great!


Amazing work, once again! Thank you!

Thanks for posting that - great results. Glad to hear that MIDI at 44.1kHz seems to be working more often than not - I think I need to revise the user manual on that point.

Panic button will be added.
da Vinci
Hi, want to settle the usb midi output crap timing once and for all. I have most of my qubase projects in 44.1 and I have gathered from various threads that it should work fine in most cases with the es-4 and esx-8gt. But does the es-3 and es-5 combo officially support 44.1? My only intended use at present is to send midi output to my yamaha s90 and escape the awful usb midi timing. Thanks for this product :-)

ES-3 and ES-5 behave exactly as ES-4/8GT as far as MIDI at 44.1kHz goes - it mostly works, but there are some exceptions.
da Vinci
Ok, thanks :-) Will put an order in for the ES-4/8GT from Swedish online store "Escape from noise" tomorrow then.
This will be my next purchase as midi timing is my biggest hate.

Now I notice (as I am tending not to quantise absolutely everything) that usb midi input kills the timing recorded.

Wonder if the reverse could be done?
Just got the ES4, and it seems it was an excellent choice and an excellent product!

I'd really appreciate I've got you correctly here:

So, you can take a midi out from ES4 or the ES4 gate expander and it has much faster timing than regular midi, and you do that using a cable that you made yourself with general midi on one side and a eurorack cable on the other, with separate cables for each device taken direct out for least latency, less than standard midi cables like we said... is this all correct?

If so seems pretty awesome.

EDIT ...just saw the "jitter free midi video ... which kind of answered my question ...

Wow, I'm really impressed by this.

Was just wondering how to make those adapters or if there is a place to buy them?

Oh and it seems you need the expander to achieve this right? Just wondering if you can do this with just the ES4, thats all i have for now!

You need the ESX-8GT for midi out. The ES-4 outs alone won't accomplish this. If you want to make your own cv to midi cables, the instructions can be found here. If you want to purchase them you can check whoever you normally get modules from. A list of resellers is here. If you are in the States, it looks like Control still has some in stock. They're pretty simple to make. I ended up getting a panel for the ones I made(far right).


I'm pretty sure midi input isn't an option...yet. It would be interesting to see how much a difference it could make, considering the external midi device would be creating the midi stream. It would be cool if the ES-6/7 could do this.

There was someone who was working on a device for midi in but I cannot remember who or if it has been released.
It was this: e-120-no-jitter-midi-recording-interface-17694

Never materialised as far as I know, and the company website doesn't exist anymore.

It's on my radar, but low priority as I've never really heard anyone make a compelling case for it. @Analog>Plugins, what are you recording via MIDI that you're really noticing the off timing?
os wrote:
It was this:

I figured you would know. Not surprised it never came about.

It's on my radar, but low priority as I've never really heard anyone make a compelling case for it.

That's how I feel about it, as well. While it would be cool, it still feels like a situation where the damage is already done, or at least the worst of it. I can't articulate it, myself. It just feels like midi in would be more of a crap-shoot and you would just be transferring flawed data into the computer sample-accurately. There's also the penchant for quantizing in the computer. Kinda makes it pointless if you do that. Maybe, there would be a small advantage for playing softsynths live but that just doesn't seem like a compelling argument.

The only case for it I can think of is just unification - having everything on one hardware system. That's not really a strong case, though.
But recording a performance to hear it played back NOT as you played it in is a bummer.
Indeed. Just curious what you're recording. A keyboard part? Drums? Some kind of MPC sample triggering thing?

And are you playing it along with the DAW to which you're recording?
Analog>Plugins wrote:
But recording a performance to hear it played back NOT as you played it in is a bummer.

Absolutely and I'm not trying to ignore or diminish the obvious but is this the midi interface or the device itself or something else or all of them together? Will it really help is what I wonder. It would be interesting to know more and that may be the reason Os is asking what you are using.

I'd still buy an ES-6 or 7 for this and use it constantly. Then again, I'm not the best to be talking about this. Interested to hear your experiences.
I was following that MIDI Bridge quite eagerly and was sorely disappointed when it didn't materialize.

I use the ES5 to sync my Cirklon to Cubase, and I often use the Cirklon to trigger VSTs via the Cirklon's USB-MIDI interface. It works quite well, even for fairly intricate percussion. There is a hint of jitter, and I'd be curious to try MIDI input via Silent Way to see if there is a perceptible difference.

At this point the issue is largely academic. I don't know if I could detect any rhythmic difference, but even a small amount of jitter becomes audible during rapid tracker-style rolls and fills.
MUFF WIGGLER Forum Index -> Expert Sleepers  
Page 1 of 1
Powered by phpBB © phpBB Group