Topic: DSP buffer size and effect order

A few questions:

1. What is the latency buffer size of the eq and dynamics on say the UFX3? Is it different on other units with DSP?

2. Is adding an eq or compressor to a channel going to add latency to that channel, or is a buffer built in to compensate for phase alignment?

3. Is the effect order is eq into compressor?

Thank you.

Re: DSP buffer size and effect order

I don't think anyone liked my previous questions, but I have another!

UFX3 lists AD conversion at 5 samples, and something about totalmix needing 3 samples. Assuming I'm just using say a mic input to say an ADAT output, am I really only looking at 8 samples latency there? That would be only .16ms latency at 48k which is kind of mind blowing.

Can anyone please confirm or otherwise set me straight?

3 (edited by ramses 2023-04-18 19:45:56)

Re: DSP buffer size and effect order

I think you got it right. Time for AD 5 samples .. processing in the routing matrix .. 3 samples.
Well and then it depends which signal flow you refer to .. for direct monitoring you also need the latency for D/A.
If you calculate routing through the DAW then the RTL also depends on ASIO buffersize (Windows).

You can also have a look at my calculations for a comparison:
https://forum.rme-audio.de/viewtopic.ph … 62#p200962

There you can also see how latency translates into distance between performers on stage or in the studio.

You can also have a look at some examples for RTL using different RME products, all at 44.1 with different ASIO buffer sizes.
https://www.tonstudio-forum.de/attachme … es-v2-jpg/

All in all, all RME recording interfaces are fine products and if you require lower converter latencies, then check the manual and get the products with more up-to-date converters.

BR Ramses - UFX III, 12Mic, XTC, ADI-2 Pro FS R BE, RayDAT, X10SRi-F, E5-1680v4, Win10Pro22H2, Cub13

Re: DSP buffer size and effect order

Hi
But yes, I did like your initial questions but I simply don't know the answers...  but anyway I would like to share the following related to the topic:
Because there are many audio gear users are talking about converter quality and such things, and because I often don't really understand what differences they can hear (can they really?) I did a little and simple test some days ago. I was curious about what AD/DA conversion does with the original sound. On my trusty Fireface 400 I connected several analog outputs using trs cables directly back to inputs and did the routing in TotalMix that way to hear the signal (playback of an audio file on my laptop) after four times going thru DA > AD conversion. Two things I discovered then: 1. there was almost no added latency (I used 48kHz) and 2. beside a tiny little bit of hiss / noise coming from the analog circuits it sounded not really different to the original audio signal.

UCX - FF 400 - Babyface pro - Digiface USB - ADI-2 (original)
Mac mini M1 - Macbook pro - iPad Air2

Re: DSP buffer size and effect order

ramses wrote:

I think you got it right. Time for AD 5 samples .. processing in the routing matrix .. 3 samples.
Well and then it depends which signal flow you refer to .. for direct monitoring you also need the latency for D/A.
If you calculate routing through the DAW then the RTL also depends on ASIO buffersize (Windows).
...
All in all, all RME recording interfaces are fine products and if you require lower converter latencies, then check the manual and get the products with more up-to-date converters.

Thanks. Main questioning here is around using an RME interface for a jam/recording room where the ADAT outs are feeding a Behringer headphone system for a band, so no daw involved in that part.

I shelved my FF800 for years and used some other stuff, but recently went back to the 800 and am really quite blown away with total mix after not using it for some time. Now I'm just trying to talk myself into the ufx3 and making sure I have all my ducks in a row.

Thanks again.

Re: DSP buffer size and effect order

oli77sch wrote:

Hi
But yes, I did like your initial questions but I simply don't know the answers...  but anyway I would like to share the following related to the topic:
Because there are many audio gear users are talking about converter quality and such things, and because I often don't really understand what differences they can hear (can they really?) I did a little and simple test some days ago. I was curious about what AD/DA conversion does with the original sound. On my trusty Fireface 400 I connected several analog outputs using trs cables directly back to inputs and did the routing in TotalMix that way to hear the signal (playback of an audio file on my laptop) after four times going thru DA > AD conversion. Two things I discovered then: 1. there was almost no added latency (I used 48kHz) and 2. beside a tiny little bit of hiss / noise coming from the analog circuits it sounded not really different to the original audio signal.

All good. I think sometimes you can barely hear the differences, and sometimes it just pokes you in the face. I really went kind of crackers with this and why I ended up getting an ADI-pro (something). That got me to hooking it up to my old ff800, then more lights clicking on with totalmix. Now I'm basically ready to move on from the v1 Apollo quad and v2 Apollo 16 and the 48 channel Allen heath mixer, and get a ufx3.

FWIW, the 3 versions of apollos as sound pretty radically different to me. I like the 3rd gen, I like the RME ADI. I really like the AD on my Audient preamps. It's just a rabbit hole, but I think I'm finally coming to my senses.