external fx with spdf

[jb hoste]

hi every one,
sorry about my poor english, I'm French

because of the lack of CPU power with heavy sessions (G4 400 with 448 M ram), I plan to get an external device for reverbs such as Lexicon / TC M-one etc ... Is it possible to keep the reverb in a final mix when I bounce to two traks ? Is the FX return should be on aux trak in analog inputs or spdif input ? Is anyone use this configuration ? Thanks for your answers

[bassmac]

Yes, but there is some latency. Some things sound fine, but if the latency is too obvious, you can record the verb aux to an audio track and nudge it back - this will add clarity.

[Dirren]

i'm putting both my TC reverb and my dbx DDP on aux channels just like you mentioned to keep the CPU at a reasonable leve.

works great!

[accession]

I use a Lexicon MPX 500 on every session as an Insert on an Aux Track connected via S/PDIF with auto preset recall over MIDI.

Works an absolute charm, as convenient as any plug-in, but without the CPU load.

If I was to do it all over again, I'd go the M•One over the MPX 500 for it's ability to run two stereo reverbs at once.

Phil

[Chompers]

WARNING:
MAny claim that the spdif fx method works great.
However there is latency, that will sound like crap. Some dont notice it. IF you are using the fx as a "send" then you will get phasing and other bad results. IF you are using it as an insert, then you are safe.
This topic has raised a few debates. None of the oppinions can dismiss the fact that a reverb send with the SPDIF can not do real time send/reverbing without the latency creating a noticeable sound. Most dont notice this out of lack of experience. The sound of a drum kit getting a reverb send without latency is pure solid effect...the sound of drum kit getting a reverb send with latency is a tinny reverb sound that introduces phase problems and screws up a mix.
IF you disagree, you are simply not educated on the matter. Recording the verb and nudging the results will fix the issue, but at that point..the only advantage is the quality of the reverb, not the real time cpu power savings.

Some will argue that if a reverb setting is only the WET signal, then the latency will not be an issue. This is true in theory, but after dealing with these methods for quite a while, I have found that the almost all the best reverb sounds that you desire require a phase acurrate relationship with the source. And in many cases the desired reverb setting does in fact include a portion of the dry signal in the return channel.
THe method can be used to gain extra cpu power and new reverb sounds. But dont expect to get the same sounds as a real world reverb send can get.

[unclemurray]

For reverb, I route sends via optical to my Kork Oasys and back. There is latency and I do notice it. I find thought that if i adjust the predelay setting carefully I can minimize, but not eliminate, the muck. It's a pain.

[mario gabrielli]

Acession,
I use a MPX 500 as well as auside effects whith a M-Box . I use the S/PDIF from the MPX to the M-Box but the sound is not as good as when I use in analog. Is this part of the latency that has been discuss in here. I have set PT in the hardwere : Input S/PDIF clock internal. The MPX input sorce S/PDIF and clock source External. Create two Aux track one for the "Rev Out " whith input Buss 1/2 and ouput interface S/PDIF , the second one "MPX500" input 1/2 S/PDIF and ouput to Analog 1/2 that in the M-Box is the only ouput available. I get all the system working but the aux track will easily feed back and is not as loud as the other one.
How do you use the MPX 500 as insert just by using a Y inset cable and not via S/PDIF ?
Thanks.

[accession]

Originally posted by Chompers:
MAny claim that the spdif fx method works great. However there is latency, that will sound like crap. Most dont notice this out of lack of experience.

<font size="2" face="Verdana, Arial">oooh! How controversial! [img]images/icons/grin.gif[/img]

You've got me thinking...

Every digital FX unit will have latency. So how much latency is desirable? Then how much latency does the S/PDIF transfer induce?

I wonder about the need for phase accuracy. Reverb isn't phase accurate to begin with. It's a simulation of sound waves traveling at the speed of sound bouncing within a space.

Take the speed of sound to be 343metres/sec, or 34.3cm/ms, or about one foot per millisecond.

What's the difference between a reflection bouncing off a wall half a foot further away (total additional distance traveled = 1 foot) and a 1ms latency in the recorded reflection?

Anyway, just thinking out loud, Chompers. I'll do some tests, nudging reverbs back in time and compare for myself.

Phil

[Dirren]

i totally agree with the post above...

[Chompers]

Yous guys are missing the point.
EVERY digital fx does not have latency.
THere may be latency somewhere in the signal, but not in the direct feed.

The latency that mucks up the sdif method is not compareable to the distance of a wall. That is not the same issue. We are talking about the time difference bwtween the dry signal and the earliest fx signal. The spdf method introduces a large gap that creates a very noticeable sound.
What do you mean all reverbs are out of phase? THat is not true. The tails are not out of phase just because they are playing later in time.
Just set up your spdf reverbs and try to get a really good reverb sound. IF you have ever used a decent reverb before, you will hear the issues.
I dont think it is a useless way of getting reverb, Its just not a very good way to get quality verb. I just wish someone had pointed this out to me before I spent hours discovering it.
IF it doesnt bother you, go for it.