I tried loops once but I use three rows already, and the additional latency was just far too much.
One thing I didn’t try was having say row 1 output to send 1, and your next row input from return 1. Not sure if that makes any difference?
I guess it would remove the option to switch it though.
We’ll be trying something like this soon.
We’re building 3 separate loops; the third involves sending an empty Lane to Output 3 which will connect to a pedal then to Input 2. Very curious to see how it affects latency
Has anyone measured if the plug-in blocks add additional latency on the QC? Pretty much every single one of them when used in a daw add additional latency. I don’t recall off the top of my head, but some of them are in the neighbourhood of 32 and others up to 70 range. I think for Soldano was even higher
What do you mean by 32-70 range? Assuming you might mean samples? Normally latency is measured in milliseconds, so 32-70 doesn’t seem correct.
The plug in blocks definitely add latency of some sort, since every block does add to the overall latency. It isn’t a fixed amount
Yes I mean samples. Latency can only be measured in mS if the sample rate is known. I’m just wondering if when they ported plugins to the QC whether they have reduced the sample latency. for example I pull up the default preset for a plugin on the QC it will have the same amount as it does when running on a computer or less.
Latency can be measured in ms, the same way it’s measured by samples haha
I fully understand latency. This has nothing to do with what I asked.
32 samples of latency at 48k is double the ms of 32 samples at 96k so when the samples are the fixed entity it makes more sense to use them as the reference. So again, my question is…has anyone measured the sample latency of a preset in the QC vs the plugin. For example the Soldano is 122 samples in a daw. Is the default preset the same in the QC? (if you want to use mS this is 2.5417 ms at 48khz)
I did not measure the latencies of the new amp blocks, will maybe check next weekend.
I don’t understand your explanation that the NDSP plugins add latency in a DAW. Latency in all DAWs I know depends on the buffer size. And you can set it as low as your CPU power allows you to.
That’s just the buffer sample rate. Many plugins add additional latency on top of that, NDSP plugins being one of them. For example NAM, Valhalla, Fabfilter, Helix Native are all 0 additional latency. Tonex essentially is (only 7 samples).
Helix Native isn’t zero latency. It depends on the blocks in the preset….. Especially if you use certain pitch blocks in the preset that latency absolutely changes
Im not much of a Helix user but as far as I understood it outside of snow things - pitch shifting, IR without minimum phase, etc basic blocks like amps, drives, reverbs, delays don’t add additional latency.
Pretty sure that isn’t true….. I’ve measured the Helix stuff for latency and it isn’t just a fixed number. There’s misinformation about that online from people who just think it’s always the same.
Also for the hardware units they have either one or two DSP chips. There’s absolutely similar latency issues with that as the QC. It’s unavoidable
Now I’m confused and have to read more about DAW latency, buffer rate, etc.
I’ll be back 
My plugins are -4 ms of latency
They plays before me
Is the latency lower if I use two different inputs on different lanes, e.g. row 1&2 a guitar rig and 3&4 a bass rig?
Most likely yes, because you aren’t having each signal go between the two dip chips. Lanes 1+2 are on the first chip, and 3+4 are a separate chip. Combining those together for a signal path adds latency
Thank you for your reply. Let me sum up to see if I understood you correctly:
A buffer latency of 2,67ms is only added when I combine the signal path of lane 1 & 2 with lane 3 & 4.
This means if I use a guitar signal path through input 1 over lane 1 & 2 to output one and a bass signal path through input 2 over lane 3 & 4 to output two, no additional buffer latency is introduced. The Quad Cortex is likely to keep the base latency of 1,75ms (or perhaps slightly more because of the latency introduced by the blocks).
Is this correct?