Successfully using external inserts?

So, I’ve searched…I’ve seen some folks bring up issues before…but, I am getting ready to invest in a big bank of converters to connect all my external gear to integrate with Cubase, and I ran a simple test (below) and has cause me alarm.

So, I takes a mono track…send it prefader to a mono aux channel. I insert an external unit (for testing purposes, this is a bass, and a SansAmpRBI unit)…the unit is bypassed (for now)-so it’s a straight loop. I flip the phase on the aux channel–almost NO phase cancelation. Some, mind you, but not much. And then as I click the “ping” test button on the external instrument, it changes the reported need for compensation…it starts with .11ms, which is absurd, because the converters them selves take longer, without even considering the buffer size and active PDC…when I calculate it based on reported IO latency…still nada cancellation.

Tell me what I’m doing wrong.

Side note that’s hilarious now–I recently sent back a Little Labs IBP, because I felt like I was getting better phase correction in the DAW…of course, I just assumed the actual compensation was working correctly. Otherwise, all you’re doing is throwing the track MORE out of phase with the other mics on the kit and then aligning pieces of it with the IBP.

So, I guess I’m wondering aloud if there’s anyone actually using the external effects as INSERTS…for reverb/time FX, a few MS isn’t really significant…but, if I got this bank of converters only to find that when I insert my graphicEQ and La3a on the kick, Speck EQ on the snare, and say my ProVLA on the overheads, and wound up with a phasey mess…I’ll be out thousands of dollars without a system that’s useable as advertised.

I REALLY hadn’t budgeted for a Mac and HDX/HDio system for this project, you know? If this is a fail, I need to rethink how I’m morphing my studio. I know ProTools SOFTWARE doesn’t work. I’ve read that Reaper’s insert compensation doesn’t work at all. But, Cubase’s has been solid for me in the past…I will go download the lastest 6.05 or whatever now and see if that was a glitch reported and fixed…

So,I tracked down the majority of the issue to two components:

“adjust for record latency” needs to be OFF…and my Echo interface was somehow inverting the phase, making it actually work the opposite of an internal plug in (which works as expected and fully cancels).

But, this has gotten me a little suspicious…and is still an open question…are there people out there using a rack of external gear integrated intothe cubase mixer?

The test I was running was running an analog TRS cable from output to input…setting up that IO as a compensated External FX…and inserting it on an aux channel. So, bass track sending pre fader to the auxchannel, where the insert took it out of the machine and back in. real world use, I was trying to use it to add a SansAmpRBI channel to the bassDI for a little dirt and girth…so parallel processing.

the above Echo phase inversion explains why I wasn’t successful using the IBP to correct phase issues…it was 180 deg out PLUS whatever the IBP changed.

Ok, I think I get what you are doing and what you are trying to accomplish. But spell it out even more? Are you trying to place the ext efx on the insert of a track within the DAW as you record? Sorry to be so thick here but it seems this is what you are saying. Is it?

I don’t see you using the word ‘mixdown’, that’s why I’m confused.

And then as I click the “ping” test button on the external instrument, it changes the reported need for compensation…it starts with .11ms, which is absurd, because the converters them selves take longer, > without even considering the buffer size and active PDC> …when I calculate it based on reported IO latency…still nada cancellation.

That sounds about right (0.11ms) the other delays are known are should be taken into account.

Well, I disagree with split here, .11ms does not seem like something reported by C6 when the ext efx bus is connected correctly. From my experience, using an ext bus, if I have set things up INCORRECTLY I will get 0ms. This seems likely in your case as well, just as you suspect.

OK, real world report on my system using FireWire, setting up the ext efx bus correctly, my best reading for the C6.5 ping is 17ms. …yeah, that’s a far cry away from that .11ms you reported. :sunglasses:

Do you, or would you EVER get true cancellation with any other system doing the equivalent operations as described?
Please point me to an article which says that this should happen 100% everywhere.
I’m confused. :mrgreen:

Might this help?

Very professional support forum at RME. Turned out to be his cables. Informative.

Well using a PCI card I get around 0.1ms for a loop through.

17ms is way too big!!!

Ha! Split. OK, Like I said, I can do a loop thru and get 0ms but it doesn’t mean the function of the ext efx bus is working correctly.

So to you I ask, have you set up an ext efx bus and done the ping test and next finished the process of recording a new EFX’d track, placed it back in your project and compared it to your original track? And found the two tracks, new and old, time aligned perfectly? For this to happen on my system, like I said, the C6.5 ping feature show 17ms in my system. My point being that 0.1ms number means nothing in relation to the actual use of the ping feature in C6.5 as it applies to using an ext efx bus.

Man, I don’t really understand why, when I’m sitting here with working results, that you would say that my 17ms delay time is too big.

If I’m going to move to a software mixer, I HAVE to be able to leverage my racks of external gear. Otherwise, I’ll put the money into an analog mixer and give up on software for mixing.

This specific example was simple…adding amp sim to a DI bass track in parallel.

And as pointed out, I tracked down a majority of the issue…Echo, despite having no phase switches anywhere in their control panel, was inverting phase of the loop. And I had to DISABLE “adjust for record latency”. It now works “OK”…it’s canceling down to where I might consider the difference Echo’s subpar conversion.

I don’t want people who are guessing what the issue is…simple as my asking, is there ANYONE using a rack of external gear integrated I to the Cubase mixer, day in and day out. I’ve asked at GS, too, and there have been some responses that it seems dependent on the hardware’s mixer and how it reports to Cubase. Ironically–that’s actually the thing the “ping” is supposed to make up for-incorrect reporting. I think Lynx Auroras were successful…but, GS loves those converters…kinda creepy, how they seem to be the single source of great reviews of them.

As to other systems? Yes…it’s why TDM has been the industry standard for more than a decade. Studios can leverage all their hardware because TDM IS a hardware mixer. The feature (full io compensation) is supposed to be Steiny’s workaround for not having a hardware mixer with 100% predictable latency (ie simple compensation). Also, of course any analog mixer that has been used on 98% of the mixes commercially released…ever.

I would prefer a system that reports actual latency. .11ms might be “in addition to everything else already compensated for”–but, it’s obviously NOT the actual latency, as it will take 1.5-2ms just to be converted out and back in. Let alone my 1024 buffer. But, I get that if that reading is “I’m addition”. 0ms is what you get when the loop is hooked up incorrectly. .11ms means it came back.

Well if it works then it works. When I put an outboard compressor on my system and ping it, I get about the .11ms mark, the same as a loop back, and it’s all fine. What more can I say?

I do a realtime mixdown and it works.

I would post this exact question on the Sound on Sound site. You’re much more likely to get a good technical answer.
As far as I can tell you’re referencing two bass signals and inverting one to get phase cancellation to test how any inserts are likely to behave?
This is from one of their articles, which you might have read and where I have indicated in bold what might be happening to you as I understand what you’ve written in the initial post:

You should also take care when layering sounds with prominent low frequencies (such as basses and kick drums), because it can really suck the power out of the track if the combination cancels out even a single powerful low-frequency sine-wave component. > (In fact, this is as much a concern with live instruments, and accounts for the comparative rarity of layered bass sounds on record.)

From this article:

As these writers also work with many sequencers including Cubase and post in the forum quite often the chances of getting an interesting reply from them are pretty good.

Split - so you are doing a ‘realtime’ mixdown? This means you are going out of your DAW to an external mixer? If so, well geez, I hope this would be 0ms! I do not believe the OP has a mixer to do the ‘realtime’ mixdown, nor do I think this is why he was questioning the ping feature in Cubase.

Maybe I don’t understand your use of the word ‘realtime’ here. :sunglasses:

It’s not complicated.

Realtime… as in Cubase realtime. using external plugin routing as that is the only mixdown that is allowed when using external plugins.

Of course you use realtime mixdown. It’s mandated by the app when using external gear of any kind.

And, yes…to the SOS post…maybe I should back up. I’ve mixed records for decades now. Yes, the problem is phase cancellation–which is CAUSED by the timing shift of the computer IO not being properly compensated for. It doesn’t ONLY effect low end…at all…but, that just happened to be a great real world acid test…something I needed for a specific mix. I ended up using a plug in amp sim…that’s one thing I’ll give Cubase–if you stay 100% ITB using plug ins, it’s PDC compensation is relatively bulletproof. I’ve got one plug that doesn’t report correctly…but, grand scheme, that’s got to be on them, since it also drops settings sometimes.

I guess I’m just gun shy about getting a bank of Apogee/Lynx level converters to integrate all this gear (previously used on analog and digital hardware mixers)…if I am going to be constantly testing and retesting the phase. Or doing what I did with the Little Labs IBP I demo’d–assuming it was working and just “not liking” the external gear.

Split, OK. So you are doing a mixdown of prerecorded tracks in real time using an external analog efx unit. OK, this would be using the insert of the prerecorded track? Hm, then this sounds like what the OP wants to accomplish. This is not what I do or want to do so I will bail here. Sorry for the misdirected responses, OP, Split, is your guy here. My comments have regarding using the ext efx bus to record a new time aligned efx’d track and bring it back into the project to do a mixdown later.

It may be a moot point, but I want to end up with an original track and and efx’d track that are time aligned so I can bring one up in volume, eq, add vst efx, etc.,against the other and then do a final mixdown. Really, as I think about it, how can you compare the alignment of the original track using your method? In your scenario how would you ever be able to compare the time alignment to the original track? Well, I suppose here you could just mixdown the one track and import it into the project. Have you made this comparison? That would be something to share. :slight_smile:

It’s pretty easy. In the scenario I described, you flip the phase on the Aux track. If it’s aligned, it will (almost) disappear, as it’s the same waveform with it’s phase inverted. I say almost, because converters add/lose some info…so those will remain as the differential between the two channels. If you do this with a plug in inserted on the Aux, it will 100% go bye bye. With the plug bypassed, of course…

Yeah, this is not rocket science, this I’m doing here. It’s how records are mixed. And likely why they’re still all mixed on hardware mixers.

The reason I ask is because I feel like…informally, around town…there are three camps of mix engineers:

-still mix on a desk
-Mix in a TDM environment with external gear (and plug ins)
-Mix in software with 95-100% plug ins

Cubase’s PDC is incredibly good. Literally, one bad plug in is all I have…I should look for an update to it, FWIW…but, the external seems to be more hit or miss. I lost half of a stereo aux this morning. Why? Not sure. I’d say cable, except I swapped it…level on the KSP8 in and out was fine…control panel of the Echo, the left was down like 90%. WTF?

Maybe I just buy the bank of Apogees…AES card…permanent cabling…test tone align all the levels…and see if the problem persists. Everything has tended to be traced back to the Echo in one way or another at this point, rather than Cubase.