Interesting, that would explain why I had to pad the test tone in the DAW first, and it lines up with Avantronica’s experience using the CV output from the A4, too. I can switch the inputs between -10 and +4 reference but not the outputs so it wasn’t really an option for me.
This might explain why some people find it really easy to clip the OT’s inputs and others don’t have trouble with it. I generally assume that any gear aimed at professional users is going to be +4 but it actually makes sense for the OT to run at -10 because it will be used for a wide variety of sources and it’s always better to attenuate a source that’s too hot than to boost a source that’s not hot enough.
According to the manual, my Casio CZ-101 has a maximum output of 1.2v RMS, which according to that calculator would be 1.8dBV - well under the 5.8 dBV (1.95v) that the OT’s inputs are rated for, but I can easily drive the OT’s into audible digital clipping with it in the project I’ve been using lately. I’ll have to double check and see if I’m clipping the inputs or if my gainstaging in the OT can be tweaked to keep it from happening.
I was surprised about not using +4 and rapped with Don from Apogee who was very knowledgeable at explaining why the -10 setting was preferable…
One of the key points I picked up is that the +4 and -10 settings are a reference setting.
Every audio interface will have a max output at each setting, and it’s this max output that is produced when your meters in the DAW hit 0…
So by consulting manuals and using the converter I posted above, you can find out how many dBu, dBV, or volts an audio interface produces from its outputs when the meters hit 0 in the DAW for each reference setting, and then determine the appropriate setting to use for whatever the device your feedings input specs are…
Just wanted to get to the heart of why this was all of interest to me, it’s just being able to rely on the immediacy of using the OT to capture short ideas/moments without resorting to hooking up an interface or field recorder etc
Buoyed on by the epiphany that previous assumptions are no obstacle I’ve arranged a quick’n’ dirty A’n’B test for willing participants, it’s hidden just now, just to keep it to focussed replies … essentially it’s a chance for all those folk banging on about how bad the OT inputs/mixer are to show in a blind test that they’re hearing this degradation or not depending on what they listen to (as oppose to expect)
To be honest, it was quickly assembled and I mostly just hear my own personal Mika Vainio gig in my ears all day anyway besides the passing traffic, so I’ve not had a chance to listen properly, but I do know what’s what and whether I’ve been mischievous or not - what do your ears tell you ?
I’m thinking it could all be fixed point maths inside… which do those DSP chips preferred anyway? I actually wouldn’t mind a fixed point design, it has a ”sound” lol
If anything, this thread has already clarified some key points for me:
I will not be using mixer input gain settings from now on (no point) better to adjust pre OT
track level faders are attenuators only
Okay to redline the inputs a bit
I really appreciate all this guys! I wish this sort of curiosity was more commonplace tbh (s there something wrong with ne? )
Back when I was trying to decide between getting a JV1080 and JV2080 I downloaded a bunch of level-matched, high quality, raw waveform comparisons that were posted on a site and purported to show that both machines sounded identical, and was able to tell the difference between them with about 90% accuracy (which was great because I consistently preferred the sound of the 1080, so I saved myself $100). Which isn’t really relevant, except to say that I’ve used it and it works really well for self-administering blind ABX tests.
EDIT: topic got split while I was typing this, oops!
Yeah, I would be surprised if it wasn’t, it wouldn’t make sense to convert between the two and it wouldn’t clip internally if it was all floating point. There are a whole lot of beloved hardware effects that use 16, 18 or 24 bit fixed point DSP, there’s certainly no reason it wouldn’t sound good but it does mean that gainstaging is different than the gainstaging in a modern DAW, and since the OT has a more complicated signal path that a lot of digital hardware, there are a lot more opportunities to accidentally clip stuff, especially if you’re used to floating point gainstaging where you really only have to worry about clipping at the inputs and outputs.
I was also thinking about it having a “sound” too, and also that knowing it’s possible to clip the actual DSP signal path at various points, there’s potential for creative abuse. Of the top of my head, if your input signal or sample is peaking near 0BFS already, the AMP volume suddenly becomes a digital clipper for example - use an LFO modulating AMP volume to rhythmically push it in and out of clipping, and put a compressor in effect 1 set to smash it a bit to keep the output level even and round off the clipping a bit. Haven’t tried it, but it might be a cool trick.
So. Did you guys ever reach a conclusion on how to get a signal going through the octatrack without loosing volume or quality. A newb guide would have been awesome. Was it posted elsewhere?
If so, where? and if not… What are the settings that work? Where do you boost the signal to keep it clean??
From the new project default settings all you have to do is turn the master volume up to +12 to get unity gain through the OT. Or you can leave it at the default and turn individual tracks up to 127 to get unity gain on a per-track basis.
I just turn the master to +12 and go by ear from there.
Can someone explain what in the gain structure is in the analog domain, and what’s on digital domain? It is highly unclear to me…
It would be nice to avoid driving the analog amps if it is not necessary to avoid tone loss or noise, and try to maximize the digital signal without clipping instead.
Hi,
Got an answer from the support.
On the mixer page, “gain” “cue” and “main” are digital volume controls.
The headphone knob on the front panel is an analog attenuator.
So I picked up an Akai S-5000 recently, and I thought in would interest people here to know that it also defaults to -12dB to preserve headroom, just like the OT (although in its case it’s applied as a global setting for the outputs, rather than with the individual samples’ parameters).
)
