How sx3 will handle external fx hardware

[annode]

I`ve been anxious to find out how version 3 will compensate for the timing delay,(latency),when sending an audio signal from sx to an external hardware effect and back in to sx while keeping it in phase with the source signal.

I looked on their site for the answer...no dice.
So I picked up the sx3/nuendo3 .pdf off their ftp server.
This is what they are doing;

This shows a dialog window for three post hookup ajustments.
1-delay=how much time do I want sx to wait after play/start button has occured.(this will allow for the external fx audio to get to your ears the same time fx source does.)
This is most important if you intend on mixing the dry source with the wet fx inside the host.
If these signals are off slightly,you`ll get a shift in phase alignment and frequency cancellations.

2-send gain
3-return gain
-no need to explain these.
-----------------------------------
In the steinberg document,it was implied that this new feature be used for playback...There was no mention of what might be expected if you want to, say,play a vsti while sending it`s audio signal through your hardware fx processor.
This is easy to see...BIG TIME LATENCY PROB!

Well,I can live with ext fx on playback only.
I`ll just use some similar vst plug so I can hear an effect while playing/recording.

Steinberg also made the assumption that only hardware with "lookahead"circuits,(sample hold),will need delay,(latency), compensation...taking no accounting for d/a a/d convertion processing at all!!(let alone that NO hardware fx will be without SOME delay between the in and outputs.

And I say again...if mixing dry and wet in the host...that`s enough to be out of phase.

I`m happy to see it in version 3 no matter what!

[stag]

The Latency is compensated automaticaly either by using a function called ping . a signal is sent to the hardware via the bus routing and show what the latency is.
Not bigger than Waves plugs or UAD ,IMO.
if during recording they introduce some serious latency there is a orange button on the top of track view that enables PDC to disable those plugs.
i can tell you i´m happy too for being a licensed user of SX.

[SJ_Digriz]

Steinberg also made the assumption that only hardware with "lookahead"circuits,(sample hold),will need delay,(latency), compensation...taking no accounting for d/a a/d convertion processing at all!!(let alone that NO hardware fx will be without SOME delay between the in and outputs.

YAY!! I get to play my favorite game of pointing out rediculous information about software that the poster doesn't know a damn thing about. And WOW, this post is really full of crap, I don't have time to deal with it all. Oh well, I'll just cover the quoted part for now.

The pdf says the "lookahead" FX are the ones that can have NOTICEABLE delays(and there are good reasons why).... Not that they are the only ones that have it. You just have to do some manual tweaking in the ping tool to get the phase resolved.

They DO consider the freakin D/A->A/D conversion in the compensation. It's just already accounted for by the PDC system. So you don't have to add it into the External FX calculation. How the hell are they going to avoid counting the D/A-A/D conversion time? CRIPES

It is the full chain that is pinged/compensated. So, the entire path is compensated. This includes I/O connection delay and such. They don't know if there is 1 compressor or a chain of 62 devices so how are they gonna remove delay that is I/O based? Answer: THEY DON'T...it's accounted for in the ping.

Finally, you can see the delay slider on the tool. It is very simple to print a copy of the return and adjust your latency until you get NO phase issues.

Just another attempt to bring a little sanity to the misinformation superhighway.

[C00kie]

... assumed you do a ping every time you have changed the hardware chain ofcourse...

You're absolutely right SJ_Digriz, Cubase doesn't care weather you send the signal to the moon & back, it just compensates.

If you do a ping once with input connected directly to output, then... well, you have to ping again

[SJ_Digriz]

... assumed you do a ping every time you have changed the hardware chain ofcourse...

If you do a ping once with input connected directly to output, then... well, you have to ping again

Yeah, I have a little sticky on my monitor with different chain values and I adjust manually each time I swap now. Supposedly they are adding presets in a patch. That will help a ton. But even without it is a really neat feature. I didn't think it was going to be all that big a deal, but I love it now.

[annode]

First off,let me get my dignity back.
stag sez;

i can tell you i´m happy too for being a licensed user of SX.

It seems your implying i`m NOT a licensed user.
If you are,i`m a licesed nuendo2 user that hasn`t decided to get the upgrade since nuendo 3 has been selected to showcase video purposes from now on,and i`m still undesided if I want to stay with nuendo.

As for SJ_Digriz,your just an asshole!

The pdf says the "lookahead" FX are the ones that can have NOTICEABLE delays

I said the same thing but with different words...almost quoting the doc.
As for the ping,the section I read said nothing about it...and if I had read about the ping,and it is clear I did not,I would have stated it.
Ping or not...if the ext fx introduces enough latency,keying a vsti through the fx will certainly be felt.

I said where I got the info for this.This was all I knew ...I don`t have or said I have sx3.
Again...ping or not,this doesn`t change any statement I made.

I stand by what I wrote ping or not.
Your must just be an asshole to attack me like you did.

[C00kie]

Again...ping or not,this doesn`t change any statement I made.

Yes, it does matter, Annode. Lets go through it from start to finish.

Suppose you use Cubase (or any other host for that matter) for the first time, and you want it to compensate for any latency in a hardware loopback. So you connect output to input directly and let it measure the total latency.

Now you are saying "yes, but what if I plug in this hardware pitch shifter (or whatever) that adds 20ms latency?"
Answer: hook it up, and ping again.

You don't have to specify it is actually 20ms. Simple as that. Do it every time you change hardware (or presets, they might differ!) and you can't go wrong. Never any phasing issues, always sample accurate.

[SJ_Digriz]

As for SJ_Digriz,your just an asshole!

I can't argue with that. You are not the first to make the claim. Fortunately for me my wife doesn't think so and my kids are still young enough that they haven't discovered it yet.

I said the same thing but with different words...almost quoting the doc

No sir, you took the words from the manual completely out of context and here is what you wrote

Steinberg also made the assumption that only hardware with "lookahead"circuits,(sample hold),will need delay,(latency), compensation

This clearly indicates to me that you are saying that Steinberg made a huge mental error in dealing with hardware. You are the one that made a bad assumption. Yes you "ALMOST" quoted the book. You just missed enough to try and make Steinberg seem the fools.

As for the ping,the section I read said nothing about it...and if I had read about the ping,and it is clear I did not,I would have stated it.

You have a picture of the ping tool in your post. How could you not read about it?

Ping or not...if the ext fx introduces enough latency,keying a vsti through the fx will certainly be felt.

And again, it is compensated FULLY

I said where I got the info for this.This was all I knew ...I don`t have or said I have sx3.
Again...ping or not,this doesn`t change any statement I made. I stand by what I wrote ping or not.

Yes, like I said, you almost read something that you almost understood and felt obligated to share your misunderstanding with the world. You can stand by it all you want, it's still wrong.

Your must just be an asshole to attack me like you did.

Again, can't argue the asshole part. Truthfully, it my reply was half in fun and half serious. I'm sorry I hurt your feelings. I just hate it when people who don't know what they are writing about put it as gospel. Sorry that you got caught up in it.

[annode]

I believe your confusing what i`m saying.
All of this is true where the fx latency is of no consequence when playing back audio.This I said above.
But i`m also saying,that if after the ext fx in pinged and set,and you are playing a vsti through that ext fx and monitoring it through the host...this is where a longer latency can be felt.

I don`t have any experience with sx3,but this must be a fact...delay/latency and real time keyboard playing happens at different times.That`s all i`m saying.

[C00kie]

But i`m also saying,that if after the ext fx in pinged and set,and you are playing a vsti through that ext fx and monitoring it through the host...this is where a longer latency can be felt.

Indeed. So you ping again, and the added latency is compensated. No problemo!

edit: oh you also play & monitor the VSTi? Well you can't travel in time can you? Most hardware fx have neglible latency. Just don't record midi input while having a whole chain of effects with added latencies, and again you should have nothing to worry about.

[SJ_Digriz]

I don`t have any experience with sx3,but this must be a fact...delay/latency and real time keyboard playing happens at different times.That`s all i`m saying.

As you said in the first post, external FX are not for live tracking. Nobody made the claim that it is. In fact the manual and the literature all say specifically that it ISN'T for that.

And that's NOT all you were saying. You said and implied a LOT more than that.

[SJ_Digriz]

oh, and btw I did test this out just for fun. You can set the latency(from the plugin list value) for any given VSTi and adjust the latency ahead of time for that device. You have to make a lot of adjustments for it though, it won't be accurate on the fly. But, once you get it right, you can actually live track a VSTi with a send going to the external FX fairly accurately.

[SJ_Digriz]

just to be extra clear:

MIDI track plays the VSTi with send to external FX is fully compensated.

Playing the VSTi live is not, but you can fiddle with it and get it close on a case by case basis. I would also remind you that analog gear has innate latency that is NOT compensated for and never has been. Having "sample accurate" effects is almost an oxymoron.

[annode]

This is as far as i`m going with this non-sense.

[SJ_Digriz]

I don't know what nonsense you are speaking of. Thanks for posting the doc.

Steinberg says "Effect devices that may have noticeable latency include digital dynamics processors with look-ahead........"

You say "Steinberg also made the assumption that only hardware with "lookahead"circuits,(sample hold),will need delay,(latency), compensation"

Nowhere does Steinberg say that the ONLY types of devices that get delay compensation are the "look-ahead" kind. They just mention that these are going to have extra latency because of the way they work.

In most "analog" device situations, the value can be zero because the D/A-A/D is already being calculated by the PDC system. This is no different than tracking on a board to tape with a bunch of FX on the aux send.

Again, all I'm saying is that you ran off with info that is flat wrong with added inuendo that Steinberg are a bunch of dumbasses for even comeing up with this.