Omniclock (MIDI sync Plugin)

Erik — to be honest I broke my golden rule about never engaging on public forums years ago. We could go back and forth on the 909 and its quirks forever and get nowhere. I’ve owned three over the years and run plenty more, every firmware revision along the way. TR-808’s too. But I didn’t come into this thread for that. Or to sell anything.

The rhythmic-feel debate is as old as time, and so are the myths riding along with it — jitter as “human feel,” some secret “magic” swing baked into the code of machines like the MPC-60 and MPC-3000. I can speak to that one from the inside: I was a beta tester on the Vailixi 3.50 update, approved by Roger Linn and written by Rohan Mansell, back in 2009. We spent months in the code and the behaviour. There was no magic in there — just very deliberate, super simple swing. Nothing mystical, nothing random.

So I’ll leave the feel debate where it is and address the one thing that isn’t a matter of taste or mythology: your main marketing claim is hardware-grade, sample-accurate tempo sync delivered with no hardware. That’s specific and measurable — and it’s just not true:

9 Likes

I can definitely see very good use cases for omniclock. But this quote

Omniclock delivers rock-solid MIDI clock to your hardware with sample-accurate precision.

from the website is completely dependent on the midi port hardware being used and cannot be solved in software alone.

1 Like

7 Likes

Only posting as a direct reply and to close off any further back and forth. Again — I don’t do forums as a general rule, and it was only about being very clear about the facts as stated. Discussions about feel and timing can quickly descend into endless wormholes that go nowhere in the end.

Hardly anyone posts actual timing specifications for tools we use every day — MIDI interfaces and hardware drum machines and sequencers, which, at the end of the day, are what our creative endeavours are all built on. They are our tools. Just like the audio interfaces we use every day.

Nobody questions the validity of audio interface specifications — THD / Latency / Channel Crosstalk — these matter because they are real. How important they are is always personal, but that doesn’t make these specifications irrelevant. In a final mix, what’s more important — a slightly wider stereo image, slightly better signal-to-noise ratio, slightly more dynamic range?

Or the rhythmic feel of the whole track itself?

If those audio interface numbers matter — which they obviously do, for companies like RME, SSL, Apogee, Lynx, Antelope and Burl — then MIDI interfaces, drum machine and sequencer timing specifications should, in an ideal world, be held to the same standard.

6 Likes

I’ll say that I like this type of discourse. Nothing wrong with having an intelligent and fact based , and courteous discussion stem from a product related post such as this, IMO.

While I appreciate the utility of a plugin like this, I am very much in the give me atomic clock tight timing, and let me decide how much slop I want to apply to my groove camp. Getting a Sync Gen II back in the day was the best thing that could have happened for my hybrid setup. Syncing PC’s and hardware with any sort of consistency sucks, or at least it used to really suck.
These days my Sync Gen is shelved in favor of CV clock to my hapax which is forwarded to all my gear. Can’t say for sure the timing is absolutely quite 100% as tight as the Sync gen but it allows me a much finer grained control over the integration between the gears. It makes me happy :slight_smile:
Carry on…

5 Likes

In an ideal world all routers ‘should’ be 100% transparent. There are a rare few, like the Hinton MIDIX (16 input / 24 output switchable MIDI matrix), that offer 200 ns (nanosecond) MIDI THRU. But as things became more complex over time — integrated MIDI interfaces / routers, and of course USB MIDI hosting and routing — all these ‘convenience’ features look good in marketing copy, but they require active processing to work, and you pay a price for that convenience.

Converting 5-pin MIDI to USB MIDI bi-directionally is never going to be transparent. USB was built as a general peripheral bus, never for real-time MIDI timing — and its polled, frame-based nature is exactly where a lot of that jitter comes from. 5-pin DIN is a dedicated serial line with none of that overhead, which is why it’s still the better path when you have it.

So in answer to your question — of course the mioXL will degrade the MIDI input signal timing as it travels through the system. How much isn’t stated, and like much in this thread, how much that matters is a personal thing. Knowing those numbers helps us make an informed choice rather than find out post-purchase.

To quote NRain’s post: “I am very much in the give me atomic clock tight timing, and let me decide how much slop I want to apply to my groove camp.”

4 Likes

That means using serial midi (5 pin or TRS) on hardware with serial midi in/outs (sequencers, synths, grooveboxes, midi thru boxes, midi routers etc.), as USB Midi interfaces (audio interfaces with built in midi in/out and dedicated usb midi interfaces) are also subject to the same limitations, right? Or can a dedicated midi interface be better than a direct usb midi connection straight to a computer?

Personally I try to avoid clock sync from/to my DAW and simply manually set the tempo on my hardware (using the most accurate sequencer I have as master) and then record into the DAW.
Using Overbridge to distribute midi clock from my DAW seems to work ok on my system, too.

I’ll run some tests (nothing special, just measuring how accurate recordings arrive in my DAW) with the Omniclock plugin vs clock sync via USB Midi and Overbridge later. If Omniclock can deliver more accurate timing, I’d probably buy it. It doesn’t have to beat dedicated hardware clocks, if it improves timing, that’s already a win I’d say. Good to know where it stands in regards to dedicated clock solutions, though.

Hi Innerclock, one of the things I most appreciate about Elektronauts is that discussions here tend to be courteous and fair, and I’d really like to keep it that way. I can absolutely relate to wanting to avoid forums sometimes. Things can get ugly on the web, but this place really is a great exception.

I was intrigued with your statement that our main claim is “hardware level precision”, which is not something we focus on in our language:

I’ve since gone back and checked the product page, the manual, the press release, and the tech specs PDFs, that language doesn’t appear in any of them. If I’ve missed something please do point me to it and I’ll address it. (The one place I did find it was in the FAQ, and I’ve already removed it.)

To address the “hardware level clocking” point directly: to my knowledge I used that language on YouTube and even there I wasn’t referencing Innerclock or any specific hardware system. What I meant, and probably didn’t express clearly enough, is that hardware timing is not one single standard. It spans a really wide range, and we were never claiming to be at the precision end of that spectrum.

However what I can say from using Omniclock in my own studio, is that it has replaced a hardware system I used previously. (This was not one of your systems to be clear.) We are getting similar feedback from OC users, so this is not a one-off “founder preference”.

The machines that inspired Omniclock, the 909s, the 808s, the MPCs, the stuff we all love, don’t produce notes with lab-grade precision either. That’s part of their character. Interestingly, your measurements put Omniclock solidly under 1 ms — for the producers using it, that’s a result we’re really proud of. The TR-909 recordings I shared earlier clearly show timing behavior with much wider tolerances.

I genuinely want to be fair here: your system is around 100x tighter on jitter than ours. That’s a real engineering achievement and I’m not trying to minimise it.

I beleive there’s genuinely room for both in this space. You’re chasing the absolute edge of timing precision, which is impressive work. We’re focused on something different: helping producers get their hardware and DAW truly in the pocket, with musical groove profiles and routing flexibility built in, at a price point that opens that up to a lot more people.

Hope this clarifies this and we can all get on with making music and / or audio tools.

4 Likes

Aren’t most sequencers using an incoming midi clock to set their own clock, afaik often averaging the incoming clock? Unlike analog sync that usually directly drives the sequencer. I can see how a more precise clock improves timing, but how does swing over midi clock work?
Couldn’t find anything on your website, so I thought I’d just ask…

1 Like

“Hi Innerclock, one of the things I most appreciate about Elektronauts is that discussions here tend to be courteous and fair, and I’d really like to keep it that way. I can absolutely relate to wanting to avoid forums sometimes. Things can get ugly on the web, but this place really is a great exception.”

Hi Erik, and thanks for the reply. I was hesitant to even draft a reply, but after reading your post a few times it still fundamentally takes the position that in sync and rhythm, loose = creative and tight = stiff/boring, and that people who work in this space are chasing something esoteric and counter to the creative process. It implies that these tools are only for lab experiments and, conversely, that your solution is designed for ‘real’ creatives. Which I disagree with. Music is a blend of maths and emotion — both have to be respected in equal measure. Forum discussions like this, even on Elektronauts, often get the two sides scrambled and it quickly can become personal. Facts and numbers keep the emotional/subjective heat out, which is really for another forum.

“I was intrigued with your statement that our main claim is “hardware level precision”, which is not something we focus on in our language.”

Maybe not in those exact words, but your marketing positioning leans hard on it: “hardware level synchronisation box precision,” no external hardware needed. That’s the hardware-grade-without-the-hardware comparison, baked into how it’s presented — and the Gearnews piece runs the same angle straight off your press material, right down to a headline asking whether Omniclock can “replace expensive hardware clock boxes.” So even if the phrase isn’t on the spec page, the framing of your marketing pitch is all around it.

“I’ve since gone back and checked the product page, the manual, the press release, and the tech specs PDFs, that language doesn’t appear in any of them. If I’ve missed something please do point me to it and I’ll address it. (The one place I did find it was in the FAQ, and I’ve already removed it.)”

When making public claims on the maths behind the music, clarity and transparency are important, and credit for doing it. If I sound pedantic, it’s because sync is a serious business and we treat it that way. That’s why we lean on specs and stay away from spin.

“To address the “hardware level clocking” point directly: to my knowledge I used that language on YouTube and even there I wasn’t referencing Innerclock or any specific hardware system. What I meant, and probably didn’t express clearly enough, is that hardware timing is not one single standard. It spans a really wide range, and we were never claiming to be at the precision end of that spectrum.”

*This is where we genuinely part ways, Erik. “Hardware timing is not one single standard” — yes, there’s no industry spec for rhythmic timing, not because it can’t be measured or doesn’t matter, but because the industry never had to. That gap is where decades of sloppy drum machines, sequencers and MIDI interfaces have sat with random rhythmic jitter reframed as “human feel” or deliberate, baked-in “character” — which it isn’t.

Time is the standard — fixed since 1967, not elastic. It’s the canvas everything else in music production gets painted on.

Two drummers, same straight 100 BPM 4/4 — kick, snare, eighth hats.

A kid sitting down for the first time can play the pattern OK, but the motor skills and internal clock aren’t developed yet — it’s a rough listen, the hits landing with random timing errors. Those errors aren’t deliberate feel or repeatable character. They’re just slop.

Put any great player on the same pattern — Ferrone, Carlock, Gadd, Colaiuta — and the tight internal clock shows up instantly: listenable, infectious. That’s a stable clock with deliberate micro-timing placed on top — swing, late, early — placements that only read as intentional because the clock underneath is solid. Take the core stability away and any micro-timing expression collapses back into slop.

Electronic music and the tools we use are no different.

“However what I can say from using Omniclock in my own studio, is that it has replaced a hardware system I used previously. (This was not one of your systems to be clear.) We are getting similar feedback from OC users, so this is not a one-off “founder preference”.”

This heads into serious subjective territory again, Erik — what works for you works for you, and if it works for others, that’s brilliant too. It’s totally subjective on all counts. My Dad can’t dance to save his life but he still likes music.

But your marketing pitch isn’t presented as subjective. It’s framed as: you don’t need expensive hardware to get the same results — and that’s clearly not the case. “We were never at the precision end” sits oddly next to the marketing that calls it a “precision-driven timing engine” with “sub-millisecond accurate precision,” built on profiles “developed through mathematical analysis.” That’s the science-and-precision angle front and centre.

And to be clear, this isn’t me defending my own position exclusively — “hardware-grade sync without the hardware” takes a swing at everyone doing serious sync, including Multiclock (E-RM, now Floatingpoint), SND, Expert Sleepers and Sim’n’ Tonic. Whether a plugin-only solution suits your studio is entirely your call. But “hardware-grade” used as a comparison to dedicated, tested, repeatable solutions is a timing claim — and timing is measurable.

“The machines that inspired Omniclock, the 909s, the 808s, the MPCs, the stuff we all love, don’t produce notes with lab-grade precision either. That’s part of their character. Interestingly, your measurements put Omniclock solidly under 1 ms — for the producers using it, that’s a result we’re really proud of. The TR-909 recordings I shared earlier clearly show timing behaviour with much wider tolerances.”

The TR-909 / MPC point I’ve already covered — for decades. Roger Linn said as much himself. Systematic, repeatable timing is experienced as rhythmic feel; random rhythmic jitter isn’t. The tighter the better. I won’t run it again.

Comparing Omniclock to a TR-909 and concluding it’s “tighter than the TR-909, so it’s enough” compares two different things. A TR-909 is an instrument — a sound source with its own timing signature. A master clock is the reference that instrument gets slaved to. They do opposite jobs: the reference has to be steadier than anything hanging off it, because everything is measured against it. “Tighter than a 909” is the wrong yardstick.

One correction on the numbers. “Your measurements put Omniclock solidly under 1 ms” — that’s only the one path I tested: a clean DIN output through one interface. But there’s no absolute figure here, and that’s the point. Omniclock’s clock precision ends at the computer’s MIDI layer — the interface, the path, the host all sit downstream, and that’s where the timing is delivered. Final output precision isn’t Omniclock’s to own; it’s inherited from whatever it’s plugged into. A hardware sync box owns its output stage and can put a number on it. A plugin can’t — which is exactly why “sub-millisecond accurate precision” isn’t a spec it can stand behind.

“I genuinely want to be fair here: your system is around 100x tighter on jitter than ours. That’s a real engineering achievement and I’m not trying to minimise it.”

Appreciated . We’ve been refining this since 2004; they’re simply the numbers the system was designed to deliver. And again, you might not minimise the achievement itself, but you minimise its importance in the creative process.

“I believe there’s genuinely room for both in this space. You’re chasing the absolute edge of timing precision, which is impressive work. We’re focused on something different: helping producers get their hardware and DAW truly in the pocket, with musical groove profiles and routing flexibility built in, at a price point that opens that up to a lot more people. Hope this clarifies this and we can all get on with making music and / or audio tools.”

Room for both — of course — if a clear statement of the differences is front and centre. They are not the same.

You said “truly in the pocket.”

We’re firmly back in subjective land with that, Erik.

What a good pocket feels like to you is personal — to everyone — and it’s not something you can measure or promise. But where the pocket sits depends entirely on the stability of the timing canvas underneath it. That part you can measure.

And you are still driving home the myth that Loose = Human Feel/Soul and Tight = Sterile/Boring.

So go back to the drummers. Plenty of people would happily listen to a school-band kid on a good day. But the implication that wanting that same part played by a Ferrone or a Carlock is just “chasing the absolute edge of timing precision” — for no real creative or expressive gain — doesn’t fly. Everyone hears that difference, instantly. It’s the whole reason those players get the call.

The fundamentals haven’t changed — acoustic or electronic, it makes no difference.

3 Likes

We’ve been living in times of the marketing blubb for quite some time now. I’m pretty sure potential customers reading the website will take the claims with a big grain of salt, because what sounds too good to be true, usually is.

Does the Ommiclock plugin only send clock when the DAWs transport is running? I read that in the Gearspace thread. If that’s the case, why? It would make integrating any gear with a sequencer more complicated, because you’d go back and forth between DAW clock and Omniclock all the time.

Edit, or no, forgot you have to turn off the DAW clock for the ports that use Omniclock, you’d be going back and forth between the hardware clock and Omniclock.

Hi Innerclock, thanks for the response. I think at this point readers have enough information to draw their own conclusions, but I’d like to clarify a few final things.

First, I don’t agree that this discussion is somehow irrelevant or unhelpful. The TR-909 data I posted was not a distraction from the topic - it was directly relevant. People are using Omniclock to synchronize real instruments, and the reality is that many of those instruments, including revered classics like the TR-909, do not themselves operate at the level of absolute timing precision being presented here as the only meaningful benchmark. Selectively bypassing that context entirely makes it difficult to have a discussion about what musicians actually experience in practice.

I also think it’s important to separate my own claims from the independent feedback others have shared. The Gearnews article was not written by me, but it offers a great real-world perspective. The reviewer specifically noted that Omniclock worked perfectly well even alongside a dedicated hardware clock (a Multiclock), allowing him to sync additional devices without the usual headaches. That is exactly the point I have been making: software and hardware solutions aren’t mutually exclusive, and there is clear value in a tool that solves these problems for users in actual studios.

To be honest, it’s been hard not to notice that this thread—which was started to discuss a software plugin—has gradually shifted into a space for detailed support and promotion of a competing hardware product. I’ll let readers draw their own conclusions about that. To me, it mainly confirms that affordable software-only solutions are now being taken seriously enough to be seen as part of the same conversation.

I’ve never claimed that software somehow bypasses every limitation of an OS or USB interface. My position has always been consistent: meaningful improvements can be achieved in software, and those improvements are valuable to musicians.

If someone needs the absolute most deterministic timing possible at the final physical output, dedicated hardware has its place. But for many others, the priority is affordability, port count, and ease of use. Not everyone can justify the cost or the extra desk clutter of a hardware box. For those users, a flexible, accessible software solution isn’t some “lesser” category—it is the only practical one.

Ultimately, most musicians just want to know if their gear feels tighter and their workflow is easier. For a lot of people, Omniclock does exactly that. I’m happy for people to choose the solution that best fits their music and their budget.

6 Likes

I think this quote about the TR909, TR808 and MPCs timing tolerances (=jitter) is a little misleading.

If you clock your 909 from a clock that has its own „wider timing tolerance“, the clock sources jitter won’t hide behind the 909s own jitter. It will accumulate and the 909s timing precision will be worse.
You need a stable clock to get the best out of your 909 (or any other machine for that matter). To claim clock precision doesn’t matter here because the 909 isn’t precise anyway is very misleading tbh

2 Likes

Gearnews “reviews” are always presenting any new gear or software as an absolute game changer and alternative to more expensive options. It’s like an extension of the Thomann catalogue. They don’t even know where the “n” for nope is on a keyboard.

As an Ableton LIve user, i have a general question about the omniclock shift function. Is there any difference in shifting the midiclock +/- ms in the Ableton Live midi port settings? It seems, with the omniclock it now only gets a dedicated dial with the UI of the plugin. The Shuffle function is a unique feature indeed, in comparison with live.

Thanks!

Has anyone had a chance to compare Clock V3 and Omniclock?

So from my, admittedly not very thorough tests, the recordings with Omniclock aren’t really tighter than if I clock my gear from USB Midi or from Din Midi (all are what I´d call ok, non is particularly bad or smt - once dialed in, they all drift a bit over time, but that´s alright for me), but the swing is soo nice. Especially if combined with the internal swing of my Analog Rytm, I could get some really sweet swung patterns.

Din Midi routed via Mio4, always using direct monitoring in Ableton Live.
The latency knob in the plugin is pretty useful, too! Don’t have to go to Live’s settings all the time.

And since non of you midi wizards answered my question how swing over midi clock works…I guess it just delays the odd pulses and the receiving device averages the incoming clock over such a small time that a swung midi clock simply delays the odd steps and does not produce a chaotic tempo. Something like that.

1 Like

Gearnews is the worst. There is zero effort to accurately describe stuff, the main goal is driving Thomann sales. And that is because Gearnews IS Thomann. Or to be specific:

GEARNEWS is […] owned by Remise 3 Medienservice GmbH, a daughter company of Thomann

source

No idea if it’s just bad writers, or mostly AI, and I don’t care. Don’t read more out of Gearnews than “there is this new thing”.

Anyways, back on topic:

Accuracy is good, but it is IMHO not an absolute ideal, simply because there are limits to human perception. Any system needs to be accurate enough, and how much accuracy that is would at least partially depend on context.

Of course people pay all kinds of money for things they believe makes a difference. That is fine, even though blind tests often reveal that “difference” does not manifest when tested rigorously.

So for me and my tiny small hybrid setup the real question is not even "is 100 times more accuracy worth 50x more money (Gridlock II vs. Omniclock), but what defines the context where Omniclock can make a meaningful difference (if that context exists at all).

Yeah, I know they are basically owned by Thomann. Lot´s of other gear zines and forums, too! Seems unlike other websites/forums owned by Thomann, Gearnews did not retain their autonomy.

1 Like

So Midi clock is just 24PPQN ticks. Omniclock is presumably moving (not just delay but could also push forward) the ticks based on templates from iconic hardware swing patterns. This might work fine for altering the midi clock sent to most hardware. But some hardware ‘regenerates’ it’s own internal tempo clock based on averaging of incoming midi clock ticks.

1 Like