Topic: RME Raydat very thin live sound with buffers 64 and lower

I have the RME raydat and connected it with my antelope orion studio 2017 through optical connection. I used both ADAT 1 and ADAT 2 for 16 inputs/output.
Everything works as expected but I have a weird issue, basically when I monitor my voice through my daw (studio one 7) whenever I set the buffers of RME lower than 128, the voice sounds incredibly thin. It records fine and all but the live sound is very thin unless I put my buffers to 128 and more, which I would like to avoid cause of latency.

Why is this? what can I do to keep the live sound as warm and nice as it is with lower buffers?

Re: RME Raydat very thin live sound with buffers 64 and lower

Monitoring does not change the sound quality. You might be describing the effect of some comb filtering due to doubled signal, i.e. the signal coming back both through the DAW and Totalmix..

Regards
Daniel Fuchs
RME

Re: RME Raydat very thin live sound with buffers 64 and lower

so what I do if this is what is happening?

Re: RME Raydat very thin live sound with buffers 64 and lower

I think I know what you meant now. You mean I have the live sound from the hardware inputs and the software playback from my daw together? If thats that you mean, this is not happening. My hardware inputs are all the way down. I only have the software playback so I can monitor from the daw

Re: RME Raydat very thin live sound with buffers 64 and lower

Then it would be good if you could somehow record the phenomenon.

Regards
Daniel Fuchs
RME

Re: RME Raydat very thin live sound with buffers 64 and lower

demetrisag wrote:

My hardware inputs are all the way down.

This alone says nothing, the faders are down for which output ?
you have to click onto the headphone output and check again if the input faders are really down for the headphone and same for the monitor loudspeakers.

M1-Sequoia, Madiface Pro, Digiface USB, Babyface silver and blue

Re: RME Raydat very thin live sound with buffers 64 and lower

Good point, forgot to mention that... :-)

Regards
Daniel Fuchs
RME

Re: RME Raydat very thin live sound with buffers 64 and lower

I checked for all hardware outputs individually so I know for a fact I have no hardware inputs open. They are all, all the way down for every output so I know for a fact I have no double signal. Basically all direct signal is 100% whitch off.

The Signal I get from monitoring is anyway different from the direct signal. Direct signal sounds the most natural of all.

But signal from daw sounds the best at 128 samples. higher than there is perceived latency and lower than that it sounds very thin, no body.

This is happening only when I monitor myself speaking from the microphone. The recorded result is of course clean. The sound coming from daw on playback is still not affected from buffer size and also when I play my midi keyboard (vsti) the sound is still not affected.

The only thing affected is the live sound I hear from the microphone on my headphones when of course I go through the daw. The direct sound is still not affected.

9

Re: RME Raydat very thin live sound with buffers 64 and lower

The sound in the headphone is unchanged at lower latency. What you hear is a phased out mix of the headphone signal and directly through your body (mouth to ear). Should be easy to understand.

Regards
Matthias Carstens
RME

Re: RME Raydat very thin live sound with buffers 64 and lower

omg! All I needed to do is flipped the polarity switch!!!! you are the savior of mankind!

Pardon my ignorance, I have never try to monitor through DAW as I always went direct monitoring cause of latency! Recently I bought the raydat for one reason only to start monitoring through daw cause of the massively added flexibility so all this is new to me!

Thank you and thank you again so much!

11 (edited by ramses 2025-01-10 13:14:56)

Re: RME Raydat very thin live sound with buffers 64 and lower

Fully agree. Since RME and TotalMix FX, I do not use the DAWs control room anymore.
Much peace in mind by this and lesser latency in general.

BR Ramses - UFX III, 12Mic, XTC, ADI-2 Pro FS R BE, RayDAT, X10SRi-F, E5-1680v4, Win10Pro22H2, Cub14

Re: RME Raydat very thin live sound with buffers 64 and lower

Surprising that this would just so coincide to work with a simple flip, but good to hear (pun intended).

Regards
Daniel Fuchs
RME

Re: RME Raydat very thin live sound with buffers 64 and lower

Pun very well placed indeed! Thanks guys!

Re: RME Raydat very thin live sound with buffers 64 and lower

I once had a similar experience while testing a figure eight microphone. From the one side it sounded good, from the other it sounded thin.
Recording it and listening to it afterwards showed clearly that there was no difference.
Georg

15 (edited by demetrisag 2025-01-11 06:08:33)

Re: RME Raydat very thin live sound with buffers 64 and lower

Just out of scientific curiosity anyone knows why would this be happening? I mean direct monitoring through headphones doesn't seem to have phase issues. Is like the 2.9ms delay I have with 32 buffers enough to cause such a massive phase issue?

Re: RME Raydat very thin live sound with buffers 64 and lower

demetrisag wrote:

Just out of scientific curiosity anyone knows why would this be happening? I mean direct monitoring through headphones doesn't seem to have phase issues. Is like the 2.9ms delay I have with 32 buffers enough to cause such a massive phase issue?

2,9ms is a long time.

If you sing a tone of 1000Hz, one phase is 1ms long, shifting the audio 0,5ms can cancel out almost all the sound.
Shifting of 1ms will double the sound (1kHz tone).
Your 2,9ms will have a lot phasing issues.

M1-Sequoia, Madiface Pro, Digiface USB, Babyface silver and blue

17 (edited by demetrisag 2025-01-11 07:00:25)

Re: RME Raydat very thin live sound with buffers 64 and lower

And this is happening because my ears can hear both my voice as it's coming out of my mouth and my voice as it's coming out of the headphones with latency? so I suppose the phase cancelation happens at my ears?

So let's say if I had the worst best noise reduction headphones and couldn't here my voice outside of headphones I wouldn't experience this phase thing?

18 (edited by Randyman... 2025-01-11 07:31:44)

Re: RME Raydat very thin live sound with buffers 64 and lower

demetrisag wrote:

And this is happening because my ears can hear both my voice as it's coming out of my mouth and my voice as it's coming out of the headphones with latency? so I suppose the phase cancelation happens at my ears?

Yep - you hear your own voice through the air, you hear your chest resonances through the air, you hear your vocal chords conducting through your bones and head/nasal cavity, AND you hear the latent sound in the headphones.  You basically get comb filtering - and depending on delay - this can end up with opposite polarity of the fundamental frequency wavelengths causing internal cancellation and the "thin" sound.  Flipping polarity simply allows the latent HP signal to combine more constructively with yourself - but it is still latent at those (and all) frequencies - just combines more favorably in your ear at the fundamentals.

demetrisag wrote:

So let's say if I had the worst best noise reduction headphones and couldn't here my voice outside of headphones I wouldn't experience this phase thing?

You'd still have bone conduction and internal head/nasal resonances exciting the ear drum and nerves in your inner ear which you can't isolate externally or with ANC.

MADIface-XT+ARC / 3x HDSP MADI / ADI648
2x SSL Alphalink MADI AX
2x Multiface / 2x Digiface /2x ADI8

19 (edited by waedi 2025-01-11 07:43:51)

Re: RME Raydat very thin live sound with buffers 64 and lower

demetrisag wrote:

And this is happening because my ears can hear both my voice as it's coming out of my mouth and my voice as it's coming out of the headphones with latency? so I suppose the phase cancelation happens at my ears?

So let's say if I had the worst best noise reduction headphones and couldn't here my voice outside of headphones I wouldn't experience this phase thing?

The noise cancelling is only working good at low frequencies, voices are coming thru.
Use a closed system headphone.
The phase shifting happens also when monitoring thru a DAW and by Totalmix. Be sure having only one signal to the speakers.

M1-Sequoia, Madiface Pro, Digiface USB, Babyface silver and blue

Re: RME Raydat very thin live sound with buffers 64 and lower

No, I am saying theoretically to better understand the nature of this phenomenon. Of course there is only one signal on my headphones so that's out.

My question is if I somehow had the best soundproofed headphones that literally soundproof my voice entirely from outside headphones then I shouldn't experience phase issues right? I am trying to understand if the phase issues are happening because basically the "2nd" signal of my voice is my real voice as it's coming out of my mouth

Re: RME Raydat very thin live sound with buffers 64 and lower

Yes theoretically phasing issue is not possible with only one signal.
But the best headphone can not protect you 100%, it is a level of damping.
This cheap drummers headphone has -22dBA damping :
https://www.thomann.de/de/the_tbone_hd990d.htm

https://thumbs.static-thomann.de/thumb/padthumb600x600/pics/bdb/_19/198642/10569983_800.jpg

M1-Sequoia, Madiface Pro, Digiface USB, Babyface silver and blue

Re: RME Raydat very thin live sound with buffers 64 and lower

Oh I get it now! This is so interesting. When doing direct monitoring cause the latency is super low even with no soundproofed headphones at all the experience is still great but when going though daw then latency comes to play and phase issues become apparent. This actually solved the age long question I had maybe even 10 years ago that I didn't understand why going through daw the live sound is unnatural and basically the reason I bought an antelope Orion cause I could monitor direct with fx and reverbs and all and it's working great on that. But u all know the story with antelope drivers so I decided to buy a raydat to serve as my studio heart and connect antelope hardware through adats and suddenly I found myself back to situation I was in 10 years ago which I entirely forgot all this time!

But at least this time finally I understand what is happening! it might seem like obvious knowledge to some of you but unfortunately it wasn't for me.

Anyway so glad again u guys could help with your experience and knowledge!

Re: RME Raydat very thin live sound with buffers 64 and lower

Glad you can work.
Happy recordings

M1-Sequoia, Madiface Pro, Digiface USB, Babyface silver and blue

Re: RME Raydat very thin live sound with buffers 64 and lower

On a side note how do you guys track live vocals/instruments?

Are u going though daw and phase flip whenever needed or doing direct monitoring?

25 (edited by Randyman... 2025-01-11 08:27:57)

Re: RME Raydat very thin live sound with buffers 64 and lower

I always track/monitor wet unless I'm in a situation where the project won't run cleanly at 32 or 64 samples.  This includes Drums (mainly me being the drummer).  I will also check phase when tracking drums for headphone issues (with me as the drummer) since the same phenomenon happens where you hear both the direct drum sounds and latent HP sounds.  Of course - this is ultimately "put back" to where the FULL MIX sounds most coherent after the drummer's headphones are no longer needed.

I'm not a singer (be glad lol) - but I will often ask the singer(s) if normal or inverted polarity sounds best to them in the cans...

One caution on monitoring through the DAW - Be CAREFUL with high-latency plug-ins.  You have 2 choices with them - use them with PDC on and maintain aligned sound but with a butt-load of latency, or disable PDC and have a temporary alignment mess on your hands.  Or better - substitute a low-latency/zero-latency version of a Plugin for tracking, and revert to the higher latency Plug-in at mix stage.  This caught me off-guard once when I inserted a plug-in I was not aware had so much latency - and the Drummer casually mentioned he was having a delay in his headphones (this was AFTER a few takes lol).  I sat down on the kit, put on the cans, and had egg on my face!  Removed the problematic plug-in, and back to low-RTL wet monitoring heaven!

MADIface-XT+ARC / 3x HDSP MADI / ADI648
2x SSL Alphalink MADI AX
2x Multiface / 2x Digiface /2x ADI8

Re: RME Raydat very thin live sound with buffers 64 and lower

I do direct monitoring thru Totalmix and afterwards moving the recording manually to fit.
It can happen in the DAW the recording is not at the perfect place due to latency.

M1-Sequoia, Madiface Pro, Digiface USB, Babyface silver and blue

27 (edited by ramses 2025-01-11 09:02:53)

Re: RME Raydat very thin live sound with buffers 64 and lower

demetrisag wrote:

Are u going though daw and phase flip whenever needed or doing direct monitoring?

I am using TM FX for monitoring.

I have some use cases like recording guitar through VSTi where I have a little unavoidable delay
which needs an ASIO buffer size which delivers less than 14ms latency (32..256 at single-speed).

Latencies <= 10 ms feel more comfortably for me to play, but up to 13,5ms is still possible.
So I could also use buffer sizes of 256 samples and still play.
With higher latencies, you immediately slow down, impossible to play and hold tempo.

Here are some latency values and the equivalent of distance between 2 musicians on stage.
Recording interface: UFX III at 44.1 kHz sample rate.
Sound travels through air at approximately 343 m/s at 20°C.

Buffersize     RTL     Distance
[samples]     [ms]       [m]

32                 2,993       1,03
64                 4,444       1,52
128               7,347       2,52
256             13,152       4,51
512             24,762       8,49
1024           47,982     16,46
2048           94,422     32,39

You can see here .. for pure recording you use normally higher ASIO buffersizes for more stability
to compensate any potential issues with your computer during recording.

ASIO buffersizes >= 512 up to 2048 samles would cause too much delay for monitoring.

So monitoring over TM FX makes in all cases most sense.

BR Ramses - UFX III, 12Mic, XTC, ADI-2 Pro FS R BE, RayDAT, X10SRi-F, E5-1680v4, Win10Pro22H2, Cub14

Re: RME Raydat very thin live sound with buffers 64 and lower

And don't forget to include plugin delays as they will easily double and quadruple RTL's if not choosing wisely (some can add thousands of samples of delay!).

MADIface-XT+ARC / 3x HDSP MADI / ADI648
2x SSL Alphalink MADI AX
2x Multiface / 2x Digiface /2x ADI8

Re: RME Raydat very thin live sound with buffers 64 and lower

I don't know if this has been said, but you hear your own voice mainly through your head and bone structures and partly the air, that is why you recorded voice sound so different then the way you hear it yourselves and not when recording others. The phase cancelation is mainly with the audio through your head not air. How much of your air voice you hear is very dependent on acoustics too.

Vincent, Amsterdam
https://soundcloud.com/thesecretworld
BFpro fs, 2X HDSP9652 ADI-8AE, 2X HDSP9632

30 (edited by demetrisag 2025-01-11 16:23:28)

Re: RME Raydat very thin live sound with buffers 64 and lower

Which I guess kinda explain why all those big star recording studios mostly record through those massive mixing consoles. Cause I guess phase cancellation is real kicker when it comes to artists. I was thinking of starting to monitor through daw and that's why I got the raydat cause of the really good low latencies but now I am learning this low latencies are just enough to cause the most horrible phase issues. Of course at the flip of the polarity switch everything sounds just normal again (but it's not really fully normal) so when I have to record a band that might open a new can of worms whereas before I never even have to think about that.

I might be wrong though, and I just let it frighten me cause I never really recorded professionally before through daw monitoring it was always direct.

What do u guys think about that?

Re: RME Raydat very thin live sound with buffers 64 and lower

The fact that a polarity flip simply fixes the issue here is a coincidence. There's no law that says monitoring through the DAW is somehow polarity reversed for any given signal, unless something is incorrectly configured in the input signal path. You will find it also depends on pitch and the latency you've set. If regardless of any of these factors, the flip will always fix the issue, check the signal path.

Regards
Daniel Fuchs
RME

32 (edited by demetrisag 2025-01-11 21:59:02)

Re: RME Raydat very thin live sound with buffers 64 and lower

RME Support wrote:

The fact that a polarity flip simply fixes the issue here is a coincidence. There's no law that says monitoring through the DAW is somehow polarity reversed for any given signal, unless something is incorrectly configured in the input signal path. You will find it also depends on pitch and the latency you've set. If regardless of any of these factors, the flip will always fix the issue, check the signal path.

Yeah I know what you mean, in my case polarity "fixed" it cause it just so happen to get comb filtering at those frequencies with that certain latency. And of course it never fully fixes it cause comb filtering is a big complication, but it just makes it sound more natural as it affects other frequencies that are not as important to me. The best sound is of course and always will be direct monitoring

Re: RME Raydat very thin live sound with buffers 64 and lower

You can of course use direct monitoring and fx like reverb and delay from the DAW. You can of course not use direct monitoring and compression or eq from the daw. Reverb and delay are just not as latency sensitive.

Vincent, Amsterdam
https://soundcloud.com/thesecretworld
BFpro fs, 2X HDSP9652 ADI-8AE, 2X HDSP9632

34 (edited by demetrisag 2025-01-12 19:28:05)

Re: RME Raydat very thin live sound with buffers 64 and lower

vinark wrote:

You can of course use direct monitoring and fx like reverb and delay from the DAW. You can of course not use direct monitoring and compression or eq from the daw. Reverb and delay are just not as latency sensitive.

Finally I ended up back to direct monitoring but this time with a twist. As I have both Orion studio and raydat I have two options which both utilize the amazing routing of total mix and super flexibility of Orion studio. As totalmix can can route all individual input to all outputs separately this gives me the option to take all 12 hardware outputs of Orion (8 line outputs and 2 stereo HP outputs) and create 6 live headphones mixes and also route hardware reverb from Orion to all of those for live monitoring. Also in this setup as I have 16 adat inputs I could if I wanted if I didn't utilize them all, I could send some of them to the afx section of Orion for real time effects compression/eq and other stuff. So in my daw I would record the clean preamp but the singer will listen another output with fx and reverbs.

Or option two if wanted to route all 16 outputs to afx section of Orion then I could switch my daw to use Orion for recording and record straight from Orion but the band would hear the headphone mixes from raydat. The drawback from this is that I would lose one headphone mix cause I would need a way to send audio from daw to the headphones.

Either way both options are good as for me it's of the utmost importance the artist to get the most natural sound possible to their headphones.

Before raydat, I could still do up to 4 mixes but it was horrible doing them through Orion panel as there was no way of controlling it externally but now with raydat I can do up to 6 and totalmix works beautifully with Mackie protocol so I can control it with my icon controller which makes it a breeze and so fun!

Re: RME Raydat very thin live sound with buffers 64 and lower

Great!
Cheers
Vincent

Vincent, Amsterdam
https://soundcloud.com/thesecretworld
BFpro fs, 2X HDSP9652 ADI-8AE, 2X HDSP9632