Measuring System Latency Differences (Help needed)

[John Aps]

Hello!

I have tried researching how to measure difference in latencies in my setup, but i can't find anything, nor any ai model is helpful and knows about this.
Can you help me as to what i would need to measure my denon avr 2400h difference in latency in its different modes (audyssey on or off, direct vs other dsp presets...)
I have a pc connected through hdmi to my denon which in turn is connected to my speakers.
I have access to a focusrite 2i2 if needed and i also have a ummik-1 mic.

For the life of me i cant figure this one out...
What else do i need to measure these differences?

I thank you in advance!

 

[John Mulcahy]

You will not be able to measure latency with an HDMI connection since you have no access to the unprocessed input signal. You could use an analog input to the Denon fed from the 2i2 and use a loopback connection on the 2i2 to act as the timing reference, measure a line output from the Denon using the 2i2 rather than using a mic.

 

[sm52]

You will not be able to measure latency with an HDMI connection

Why? This makes Serko70 (OCA) in its program working through the API.

 

[John Mulcahy]

Why? This makes Serko70 (OCA) in its program working through the API.

With an HDMI connection there is no access to the unprocessed input signal, which would be required to measure latency. Latency = signal processing delay, the delay for signals cannot be measured without access to the signal. Latency is NOT time alignment.

 

[sm52]

Latency is NOT time alignment.

Yes. I thought about time alignment. You are right.

 

[John Aps]

You will not be able to measure latency with an HDMI connection since you have no access to the unprocessed input signal. You could use an analog input to the Denon fed from the 2i2 and use a loopback connection on the 2i2 to act as the timing reference, measure a line output from the Denon using the 2i2 rather than using a mic.

Hi John,
I want to follow your suggestion using 'a line output from the Denon,' as you said but looking at my receiver's rear panel, I only see: (i attached an image with my specific model backpanel)

• ZONE2 pre-outs
• Subwoofer pre-out
• Speaker terminals

Would the subwoofer pre-out work for this measurement, or is there another way to measure the relative latency differences between modes with my specific model?
I think zone2 preouts isnt usable cause zone2 wont have the same dsp preset modes as the main receiver i believe...

Can i measure latency differences as i intended with my receiver, or as i feared there isn't really a way...?
I appreciate your time and help!

 

[John Mulcahy]

To use a speaker output you would have to set up a resistor divider to protect the 2i2 input. You could try measuring the sub output, but the effects of crossover and LFE low pass filters may make that challenging.

 

[John Aps]

i just want to measure the latency difference between direct and other modes, does it really matter if i use the sub for this? Wont the relative difference always come up the same?
P.S. i tried also searching resistor divider and i am not finding anything...

 

[John Mulcahy]

Nothing to lose by having a go with the sub output.

 

[John Aps]

Thank you i will, i was asking actually, cause i need to buy the rca to trs cables for focusrite and get my hands on the focusrite which isnt mine...
But if you think it may work is enough for me to buy the cables and loan the focusrite to try it out!
Thank you for your time John. Is there anything i should be careful to verify that my i get usable results?
I simply use use a normal measurement with 'use loopback as timing reference'?
(If not is there a guide i can reference?)

I am sorry for the questions but i really cant find anything despite searching hard for a similar case... Measuring latency in such cases is pretty obscure...

 

[John Mulcahy]

Yes, loopback as timing reference. Set the bass management crossover as high as it will go to maximise the bandwidth sent to the sub. Save all the measurements and post the mdat.

 

[JStewart]

Won’t the bass management system and therefore the signal to the subwoofer be passed through the the Denon signal processing and possibly ADC converter if feeding analog to it?
If the OP is wanting to compare latency of “Direct Mode” (i.e. analog pass thru) to processed signals would he be better off using a pre-out? I don’t see a pre-out on the photo of the back panel. I would guess the zone 2 outputs will work. If I’m right about the objective than the OP could consult the manual to see what processing can be applied to the zone 2 outputs if any.
Then use the subwoofer output to measure the latency effects of processing if the unit won’t send processed signals to the zone 2 outs?

 

[John Aps]

@John Mulcahy I will measure and report back John, thank you!
@JStewart I scanned through the manual and as i feared zone2 doesnt have audyssey nor any dsp mode (aka direct or others...), if i have missed something or if you suggest for me to try something else, i would be more than happy to!

 

[JStewart]

@John Mulcahy I will measure and report back John, thank you!
@JStewart I scanned through the manual and as i feared zone2 doesnt have audyssey nor any dsp mode (aka direct or others...), if i have missed something or if you suggest for me to try something else, i would be more than happy to!

Not being sure of what you are trying to learn exactly about the device, my only thought was if you are looking to find the “Direct Mode” latency then you want to use an output with no DSP processing. From your check the zone 2 outputs will accomplish that.

In comparison, with an analog input, the sub output signal will always have passed through ADC & DAC conversions as well as some dsp processing on your AVR model.

With a digital input then the ADC conversion would be skipped, therefore less latency.

What are you trying to get to and how accurate do you need to be? A quick and dirty way to get a change in latency between processing modes only and using hdmi input might be a YouTube available lip sync video.

 

[John Aps]

Actually i am trying to measure for music live playing purposes, and more specifically i am interested to know direct vs multi ch in audyssey mode which must be the least processed when you arent on direct mode.
It would be cool to see the actual latency, although i am not sure if that is possible. What i am after as i was trying to state above is the relative latency difference between direct and multi channel in (audyssey on).
I tend to think that measuring with a tone of 0-250hz or something even with the sub i will probably be able to see the latency difference correctly despite any other processing the subwoofer gets, cause the difference in latency between modes should be (in theory) the same? Maybe it will be a bit longer in the sub? I am not sure...
I will report back with the results this weekend, hopefully with some more insights!

 

[JStewart]

Maybe it will be a bit longer in the sub?

It should be the same. Remember to set all distances for all channels to 0 if you don’t want those settings to contribute to the latency.