How will the latency be through the desk when you add for example FX and compressor to a channel?
Channel-based DEEP processing models (compressors, GEQ and more) won’t add any latency to the channel path. DEEP ‘embedded plugins’ are available on all channels on the fly.
Inserting a RackFX from the FX rack (such as a pitch shifter or transient controller) will add latency dependent on the FX type.
Inputs and groups are all delay compensated and phase coherent to the Mix Outputs.
Is that already the same in iLive and GLD with the inputs being delay compensated and phase coherent to the Mix Outputs?
it depends…
The latency is so low that it is a non-topic at this stage, The new FPGA processors are so fast that the system latency is under
.6ms
6 or 0.6 ?
ddff
The latency is 0,58ms to be accurate. But 0,6ms is used be many including A&H
That is impressive indeed.
ddff
Is that already the same in iLive and GLD with the inputs being delay compensated and phase coherent to the Mix Outputs?
No, iLive and GLD have phase coherent busses, in that, for example, an input mixed to two groups will have the same phase in both. If you mix these two groups to a main, you should not have any comb filtering. However, if you mix an input to a main and to a group, then mix the group into the main as well, you could experience comb filtering.
In dLive, everything except the FX is delay compensated. You should not experience any comb filtering no matter what combination of mixes and inputs you have feeding another mix. The only exception is that if you have a signal mixed post delay, with the delay not set to zero, this manually added delay will not be compensated for.
[Edit] Please also refer to this knowledgebase article: https://support.allen-heath.com/Knowledgebase/Article/View/1202/0/what-are-phase-coherent-mixes
Hope this helps
- Jeff, A&H
The only exception is that if you have a signal mixed post delay, with the delay not set to zero, this manually added delay will not be compensated for.
I would expect exactly these… 
can anyone here tell me approximately what latency I should expect if had a bunch of input connected to DM48 routed via Tielines to MADI Slot card (to a digico SD8), and the MADI returns routed via tielines directly to the outputs of the DM48? Any documentation on these sort of things anywhere?
there is a technical data sheet where some latency scenarios described…
I will try to find it
Latency 0.7 ms (MixRack XLR in to XLR out, Input to Mix)
- 5 samples, Surface to Mixrack (GigaACE hop)
- 8 samples, DX32 to Mixrack (DX hop)
Samplerate is 96kHz
Yeah, I was looking for the Tieline functionality, which is not subject to processing… So should be lower I would assume…?
it is lower
converter latency plus GigaACE hop