CLOCK! Why INNERCLOCK when EXPERT SLEEPERS?

[JSt0rm]

Originally Posted by elan

How could be sample accurate a software CV/gate like silent way that sends clock using the interface outputs?

There will be the driver lag and the latency will change depending on the buffer size, plus you have the jittering of the USB or the firewire port (depending on how many tracks you record at time, and in general, how much bandwidth you are using and if you have hubs, other things connected that influences the USB/firewire bus)

Thanks

Your system will all have the same latency. By tricking the daw to treat the sync as audio you are locking your machines to whatever latency the audio inside your daw is. Thats the point. If your machine were ahead of the audio in your daw you would have a problem instead of the initial problem of it being late and sloppy running off midi time from the daw.

As for jitter on the usb/firewire port get a pci-e sound card and external converters. When you playback audio do you actually notice jitter in your audio? I would think that would be a ore pressing issue then buying a sync box.

All of the problems presented have nothing to do with these products. I would say one should have a very tweaked rig and then grab one of these. There are many paths to the end result of producing some sounds that we feel have worth. Go with what you feel is important on your journey.

[chrisso]

I don't know the technical ins and outs, but my set up is current, also very simple and basic.
Latest DAW running on latest (music only) Mac.
Outboard sound modules, plus some softsynths. That's it.
Over traditional midi I was hearing extremely audible flamming. The flamming also moved erratically, so it wasn't possible to nudge my tracks forward or back by a set amount. One bar the flamming stopped, by the next bar it was back.
Once I started using ES4 and Innerclock IILS - no more flamming, and I can just get on with making music.

[Synth Buddha]

Originally Posted by expertsleepers

And this is why I don't visit GS much. Apparently a manufacturer isn't allowed to mention his support forums here.

That stuff is indeed weird as ****.

[JSt0rm]

walled garden.

[azone]

I haven't tried the Silent Way stuff but I can say that the Sync Gen Pro MKII just works. Yes, the website is a bit ADD if not schizophrenic but the product is solid. Install the software, hook up the box, get the signal level right for the sync/clk outputs, and you're off. I use it with Cubase on PC and Logic on Mac, and have also used it with Live in a live setup.

[elan]

Originally Posted by JSt0rm

Your system will all have the same latency. By tricking the daw to treat the sync as audio you are locking your machines to whatever latency the audio inside your daw is. Thats the point.

Yep, but only playback will be in sync, not the recording, because the latency is double, plus if you sequence VI inside the computer, they will have different latency then the audio tracks and then the instruments clocked with the software CV/gate

To me the point of having an external hardware sync is in that way you can avoid the computer latency and the latency differences depending on the buffer you are using (which probably will not be the same at the end of a project)

Example, if I track everything with an MPC, using CV and stuff, everything is lined up when I record, if I use a computer which already has its random latencies depending on processing and stuff, I think everything will be a little messed up

As for jitter on the usb/firewire port get a pci-e sound card and external converters. When you playback audio do you actually notice jitter in your audio? I would think that would be a ore pressing issue then buying a sync box.

Well, I think no one can actually hear jitter alone, because your brain get used to, but you can easily hear jitter if you hear two sources one with jitter and the other not, you hear jitter as a general blurriness (audio related jitter), but the most important, in that case, is timing related jitter

Have you ever tested USB and Firewire midi interfaces? I did and USB have a lot of jitter, meaning 1ms or more, while FW can deliver an almost sample accurate timing (I'm talking exclusively about RME in that case) but still there are many interfaces which are totally not accurate in timing even if they are FW and sometimes they "jump" some ms randomly, I know that because I've tested, and that depends mostly to the way the driver manages the flux of data, and to the way the interface is conceived, and that affects both audio and midi, and that's why I consider software CV/gate clock not reliable on paper

Considering very few interfaces are still PCI-e and considering my favorite converters are not PCI-e, I think that's a problem, also even if it was PCI-e there will still the problem that at different level of CPU usage the behavior of the computer changes and latencies change... so again it is not reliable on paper

It would be if you work almost only in the box and you sequence and record only using that CV/gate, but actually not reliable, simply you wouldn't hear any problem because everything has the same latencies and stuff, but that still means your recorded audio has these latencies and random latencies too, so when you'll change your settings and record other stuff you'll hear/see that the new tracks will not be perfectly aligned with the others

[JSt0rm]

The innerclock plug in is going to be delayed with everything else in the system via delay compensation. Its accurate. Most everything has automatic delay compensation and while yes things can go wrong here with plug ins not reporting correctly or something that isnt a problem with any of these systems we are talking about. And you would still have this problem working with usb midi.

Im confused as to your argument as of course midi clock over usb is sloppy as all hell. I thought you were referring to your audio being jitttery. Thats what these products solve - sloppy usb/firewire midi. They solve it by using a audio signal, something that daws are very good at keeping in sync.

Have you used one of these systems?

[elan]

Originally Posted by JSt0rm

The innerclock plug in is going to be delayed with everything else in the system via delay compensation. Its accurate. Most everything has automatic delay compensation and while yes things can go wrong here with plug ins not reporting correctly or something that isnt a problem with any of these systems we are talking about. And you would still have this problem working with usb midi.

Im confused as to your argument as of course midi clock over usb is sloppy as all hell. I thought you were referring to your audio being jitttery. Thats what these products solve - sloppy usb/firewire midi. They solve it by using a audio signal, something that daws are very good at keeping in sync.

Have you used one of these systems?

No I haven't used any of these software CV that's why I asked because I could be interested

Well, yes midi is very sloppy using USB and even certain firewire interfaces, my argument was because I noticed random delays even in audio in certain cases and even with very expensive interfaces, have you ever tried looping back stuff, using like 8 channels at time and then trying to null? Doing that I found my converter is very silent (meaning it nulls a lot) but after a while boom! I had the volume quite loud and I found a part of the audio file was not nulled at all... I zoom to see the waveform and I've seen it didn't null because the audio was shifted compared to the other track... both tracks tracked simultaneously..

Since then I've done that more frequently and I've seen that it happens... not that often to be noticed, but happens... so, considering when I do tests I'm just using 8 tracks and no processing at all... I imagine these things happen much more frequently when you are producing or mixing... and probably we don't notice that because we are making music not tests, but still happen and if that audio clocks stuff could be a problem

But, well... I'll try it out one day

[kpsiegel]

Originally Posted by JSt0rm

The innerclock plug in is going to be delayed with everything else in the system via delay compensation. Its accurate. Most everything has automatic delay compensation and while yes things can go wrong here with plug ins not reporting correctly or something that isnt a problem with any of these systems we are talking about.

It goes without saying that you have to turn off plugin delay compensation for this in your DAW. If you don't do that it won't work right but it seems sort of obvious.

[JSt0rm]

Originally Posted by elan

No I haven't used any of these software CV that's why I asked because I could be interested

Well, yes midi is very sloppy using USB and even certain firewire interfaces, my argument was because I noticed random delays even in audio in certain cases and even with very expensive interfaces, have you ever tried looping back stuff, using like 8 channels at time and then trying to null? Doing that I found my converter is very silent (meaning it nulls a lot) but after a while boom! I had the volume quite loud and I found a part of the audio file was not nulled at all... I zoom to see the waveform and I've seen it didn't null because the audio was shifted compared to the other track... both tracks tracked simultaneously..

Since then I've done that more frequently and I've seen that it happens... not that often to be noticed, but happens... so, considering when I do tests I'm just using 8 tracks and no processing at all... I imagine these things happen much more frequently when you are producing or mixing... and probably we don't notice that because we are making music not tests, but still happen and if that audio clocks stuff could be a problem

But, well... I'll try it out one day

I suspect what you are seeing is converter latency.

[JSt0rm]

Originally Posted by kpsiegel

It goes without saying that you have to turn off plugin delay compensation for this in your DAW. If you don't do that it won't work right but it seems sort of obvious.

Im not having any problems with delay comp on.

[chrisso]

I'm using Ableton Live with my Innerclock.
I'm not sure if Ableton automatically uses delay compensation, but my innerclock clocked recordings are PERFECTLY in time.

I think we are trying to find problems where problems don't exist here.
The Innerclock sync is plug and play. It just works.

[perky]

I have had a SYNC LOCK II now for some time, and i can only say that its completely awesome. stays perfectly in sync, and what a relief, after years and years of nightmarish slightly unstable "normal" midi-clock. I do however have a good eye to the SND ACME, because of the shift possibilities, but its more a future project. I also had my SYNC LOCK sent to Australia to get upgraded (from vers. I) and can inform that all went smooth and i got it back really fast, so thumbs up for a great product and good service. On top of that i think the Innerclock HP is very informative and have been a big help for me, cause at some points i thought i was going crazy from dealing with the "normal" midi clock. just my 2 cents. YO

[jessem]

Originally Posted by clusterchord

@Jessem:

so you're saying Dark Time is tight/accurate when being slaved to ES clock?


i am interested as ive bought the ES3+ES5 system few days ago, but always planned to add some analog step-sequencing to this system, for some immediate hands-on operation.

and i've narrowed down to Dark Time, or Dot Com 960. i like that DT is completely standalone, and its got quantizing on board and its small. but 960 is classic moog setup with 3x8.. decisions


thanks

Sure, Anything that takes a pulse usually syncs well. In my experience it's once you start dealing with midi sync that you may (or may not) have issues. It's like clocking a SH-101,J60, or Pro-one in that regard.

That being said, the dark time has midi sync as well of course so you could use that with something like Innerclock or use the pulse outs.

[kpsiegel]

Originally Posted by chrisso

I'm using Ableton Live with my Innerclock.
I'm not sure if Ableton automatically uses delay compensation, but my innerclock clocked recordings are PERFECTLY in time.

I think we are trying to find problems where problems don't exist here.
The Innerclock sync is plug and play. It just works.

It definitely is plug and play but on Sonar X1/X2 I had to turn off delay compensation. Not really a problem at all.

[elan]

Originally Posted by JSt0rm

I suspect what you are seeing is converter latency.

No because this happened in the middle of the audio file, and I haven't done anything while recording

Also it happened the other day, I was sampling from battery and I've done a midi file that contained every note I needed to sample, then rerouted internally in PT and recorded every note in one pass, I had to do two tracks because it was stereo, but some notes where mono, so, just for the sake of it I inverted the phase while listening, the stereo notes obviously didn't null, the mono yes, but one didn't null at all, I checked and one note was shifted and that was the reason it didn't null, both tracks were recorded at the same time because as I said there were stereo notes too...so I just did one pass in stereo and deleted on track for the mono notes... recording everything in one pass means that if there was an error or something it should have been happened to both tracks or at least to one track but to all the track, not just one note...

then I did another pass and everything was lined up

I'm not saying that these things happen all the time, I'm saying that it is hard to notice, but it means it is not that reliable

[JSt0rm]

Originally Posted by elan

I'm not saying that these things happen all the time, I'm saying that it is hard to notice, but it means it is not that reliable

so because of your tests you think usb midi is more reliable then audio? I dont understand what your point is.

And I deal with pro tools everyday and the audio is in sync, always unless it isnt because of humans. Sometimes 6 hours of audio is in sync to picture across 13 hours of timeline. Does it null test back onto itself every time? I dunno. I bet I could alternate black and white frames for 24 hours and put a 1khz -20dbfs pop only on the white frames and it would be in sync for 24 hours.

Im not going to confirm that though

[clusterchord]

Originally Posted by jessem

Sure, Anything that takes a pulse usually syncs well. In my experience it's once you start dealing with midi sync that you may (or may not) have issues. It's like clocking a SH-101,J60, or Pro-one in that regard.

That being said, the dark time has midi sync as well of course so you could use that with something like Innerclock or use the pulse outs.

thanks, that's good to know.

yeah, i intend to use strictly pulse clock for cv synths/seqs. no midi at all.

however, when needed i will dedicate one ES3 channel to midi clk - for modern instruments where its the only possible option. for example, my Andromeda's arp and seq.

[spaceacademy]

Originally Posted by kpsiegel

It goes without saying that you have to turn off plugin delay compensation for this in your DAW. If you don't do that it won't work right but it seems sort of obvious.

Thats not true.
The Sync Gen plugin has a +/- delay function for the midi clock signal,.
You may have to adjust depending on your buffer settings.
I set it up by recording 1/4 or 1/8 notes using a click from my MPC and then make sure they line up exactly on the grid. You could do this once using various buffer settings and take a note of the plugins delay setting for each one. Once you have done that you wouldn't have to set it up again.
Mind you, were talking almost undetectable delay at low buffer settings such as 32 and 64 which many can work at these days.

[bil_g]

Originally Posted by clusterchord

thanks, that's good to know.

yeah, i intend to use strictly pulse clock for cv synths/seqs. no midi at all.

however, when needed i will dedicate one ES3 channel to midi clk - for modern instruments where its the only possible option. for example, my Andromeda's arp and seq.

Just to add on to what jessem said, there is a sync plug-in in the Silent Way suite that can take care of all the normal cv and dinsync clock duties. I use it to sync a Linndrum.

All midi functions, including midi clock, would be handled by the ES-5 via the ES-5 Controller plug-in. So, you'll be able to play the Andromeda(via controller) and sync the seq/arp with a single ES-5 output. The ES-3 outs, alone, can't do midi.

The ES-5 requires two channels of the ES-3(default - channels 7/8), but if I understand the manual(see-Connecting Expansion Modules) correctly, you could attach two more Gate Expanders to ES-5 expansion connectors 2 and 3 and still only be using ES-3 channel 7 for all 24 Expander outs. You could still use channel 8 on the ES-3, which is big since those channels are valuable for so many duties.

[elan]

Originally Posted by JSt0rm

so because of your tests you think usb midi is more reliable then audio? I dont understand what your point is.

where have I wrote that?

I say USB MIDI is unreliable as hell, but audio can have random latencies that's why I think is better sending clock with something external, independent to the CPU... like an MPC, but hey, no prob, I'll check these problem out and I'll see.. what I think is then when things become complex, like sending clock and automations, things will become a bit of a mess.. but maybe I'm wrong (actually I hope so)

[expertsleepers]

Originally Posted by Entrainer

However, I have found the ES-3 to be a suitable and reliable source for clocking analog sequencers as well as reseting with triggers. It's just the calibration and tuning I have an issue with. It works, just a massive PITA.

Maybe add an ESX-4CV?

Expert Sleepers ESX-4CV CV Expander

The outputs are hardware calibrated and so do not require the use of e.g. Silent Way's software calibration when used with standard V/octave synths/oscillators.

[carfian]

Originally Posted by elan

Have you ever tested USB and Firewire midi interfaces? I did and USB have a lot of jitter, meaning 1ms or more, while FW can deliver an almost sample accurate timing (I'm talking exclusively about RME in that case).....

Hi elan, please advise which RME Firewire interface you have used that provides accurate MIDI timing.

[Morechips]

Originally Posted by carfian

Hi elan, please advise which RME Firewire interface you have used that provides accurate MIDI timing.

I can tell you one that is/was an improvement over my previous midi interface (a midiman 4x4 usb device), the RME Multiface II, they come up relatively cheap second hand but you need a PCI slot, not PCIe. RME probably do a modern equivalent on PCIe.

[Morechips]

Originally Posted by elan

where have I wrote that?

I say USB MIDI is unreliable as hell, but audio can have random latencies that's why I think is better sending clock with something external, independent to the CPU... like an MPC, but hey, no prob, I'll check these problem out and I'll see.. what I think is then when things become complex, like sending clock and automations, things will become a bit of a mess.. but maybe I'm wrong (actually I hope so)

You took up a fair amount of thread space for someone who hasn't used either of these systems, please report back when you have done some tests. ES is free to download and evaluate and if you have the right hardware you can test at least a CV pulse clock output, maybe even dinsync if you make up the cable, midi you will need dedicated hardware for.

[schmuck]

Originally Posted by Morechips

I can tell you one that is/was an improvement over my previous midi interface (a midiman 4x4 usb device), the RME Multiface II, they come up relatively cheap second hand but you need a PCI slot, not PCIe. RME probably do a modern equivalent on PCIe.

I had bad experience with midi out from RME PCI cards (=even "directer" than firewire). My point is: I think the midi interface/audio interface is only one element, and if someone has good results with X does not mean that it is the same for others. It seems to really heavily depend on the entire system as well. If you want to be sure to be tight, one of the dedicated solutions discussed here seem to be the only way (why it is not possible to create a tight usb-midi solution I do not understand tbh).

[elan]

Originally Posted by carfian

Hi elan, please advise which RME Firewire interface you have used that provides accurate MIDI timing.

Fireface 800, bought in 2006

In Pro Tools gives me a almost sample accurate MIDI timing, meaning that most of the notes (16th at 160bpm) are in a range of 4/5 samples of random latency, and many of them have 0 or 1 samples and very few of them are max 10 or so samples (I did the test a couple of years ago)

That happens in Ableton live too, but not in Logic, I don't know why... seems it drifts of its own

Unfortunately now I haven't the Fireface 800, but I've seen even a Focusrite Saffire 24 pro (which is firewire) is much better than usual USB interfaces and it is the one I'm using now

Originally Posted by Morechips

You took up a fair amount of thread space for someone who hasn't used either of these systems, please report back when you have done some tests. ES is free to download and evaluate and if you have the right hardware you can test at least a CV pulse clock output, maybe even dinsync if you make up the cable, midi you will need dedicated hardware for.

Sorry for the intrusion... I was interested in knowing more about software CV/gate, because in a future I would probably need it, and considering in the past years I've noticed strange DAW behaviors I wanted to have an idea about it, now I just have a Yamaha CS15 (which I love), but I've seen it isn't fully supported

I think you'll see if it is reliable while producing, to see how it works controlling many synths and with the CPU at 50% and higher, I bet it works right with one synth and one track :P

[elan]

[QUOTE=elan;8364294]Fireface 800, bought in 2006

In Pro Tools gives me a almost sample accurate MIDI timing, meaning that most of the notes (16th at 160bpm) are in a range of 4/5 samples of random latency, and many of them have 0 or 1 samples and very few of them are max 10 or so samples (I did the test a couple of years ago)

That happens in Ableton live too, but not in Logic, I don't know why... seems it drifts of its own

Unfortunately now I haven't the Fireface 800, but I've seen even a Focusrite Saffire 24 pro (which is firewire) is much better than usual USB interfaces and it is the one I'm using now



Sorry for the intrusion... I was interested in knowing more about software CV/gate, because in a future I would probably need it, and considering in the past years I've noticed strange DAW behaviors I wanted to have an idea about it, now I just have a Yamaha CS15 (which I love), but I've seen it isn't fully supported

I think you'll see if it is reliable while producing, to see how it works controlling many synths and with the CPU at 50% and higher, I bet it works right with one synth and one track :P

Originally Posted by schmuck

I had bad experience with midi out from RME PCI cards (=even "directer" than firewire). My point is: I think the midi interface/audio interface is only one element, and if someone has good results with X does not mean that it is the same for others. It seems to really heavily depend on the entire system as well. If you want to be sure to be tight, one of the dedicated solutions discussed here seem to be the only way (why it is not possible to create a tight usb-midi solution I do not understand tbh).

I don't know your experience but the Fireface 800 has been great and tight in every situations and computer I've used with, I think there are two things, one is the driver and how well has been written for the system you are using (like Mac or PC or both), and second there's a limit with USB, it isn't very good for sending real time messages, it is even worst then the good old MIDI, because it has delays, "sometimes it stops to think" it is much more capable then midi in term of quantity and speed but not good for real time stuff, firewire is much more capable on this, almost like a PCI

[jessem]

I think this is getting a bit off topic but midi is serial and DAWs prioritize audio, which makes it more likely to be tighter for sync.

Expert Sleepers even offer ADAT compatible units, which is great if you don't want to use up analog outs.

USB Midi is really completely irrelevant to any conversation on either of these products. It's like going into a TR-808 vs AcidLab Miami thread and trying to tell people that a drummer in your old band had bad timing and lived at 808 Miami Blvd so there is no way you can trust either of those units.

[carfian]

Originally Posted by elan

Fireface 800, bought in 2006

In Pro Tools gives me a almost sample accurate MIDI timing, meaning that most of the notes (16th at 160bpm) are in a range of 4/5 samples of random latency, and many of them have 0 or 1 samples and very few of them are max 10 or so samples (I did the test a couple of years ago)

Thanks, thats useful information: good to know.

Presumably the Fireface 400 will have similar performance.