AVS Forum banner

If splitting signal between different iNuke amps - will time delays be identical?

2K views 25 replies 8 participants last post by  3ll3d00d 
#1 · (Edited)
EDIT: TESTED --- POST 17!
Result: The Inuke DSP 3000 and iNuke DSP 6000 are pretty much identical 'delay' for subwoofer purposes.
---------------------------------------------------




Original Question:


Out of curiosity...

If I split the signal going to an iNuke DSP 6000 and an iNuke DSP 3000 amp what is the likelyhood that time/processing delays are identical on two different iNuke amps? Has anyone tested this? How would you test it?

Use Case Scenario: I have eight sealed 18" ultimax subwoofers. When setting up the front - I'm considering placing five subs under my screen and then placing three nearfield instead of six up front and two nearfield like I currently have.
If I use two DSP 6000 units for four subs, and power the fifth front sub by an iNuke DSP 3000 amp bridged - and split the incoming signal three ways to these 3 iNuke amps - should I expect my time domain to be off without adjusting it in the amp's DSP? Meaning would the output delay be different between the DSP 3000 and the DSP 6000?
 
See less See more
#4 · (Edited)
Place the mic next to the cone, distance is critical for such a test (measure it).

Mute the other amps and see what the timing is of a 90db impulse.

Make sure you repeat it a few times and average the number of milliseconds.
It may vary slightly from test to test, moving mass, don't freak out.

Then repeat it for the other muted amps.

Given enough tests, say 10-20 for each amp channel, the average number should be about the same.


If you plot them in an spreadsheet graph, a pattern should start to emerge.


Just do one cone per amp channel per test; not an array of cones, it will just cause timing headaches.
 
#7 ·
I adjust my inuke with delay set to feet/inches instead of ms. Moving the setting one inch makes almost no difference. Not sure of the ms equivalent. I have to change the delay by at least 3 inches to see daylight between the REW graphs. What I'm trying to say is that if the two amps process the signal in some microscopic difference in time the output from the subs won't be anything to worry about. Imagine moving one of the subs an inch closer to you. That actually would make a bigger difference but still too small to fuss over. On the other hand I'd probably spend a few hours testing anyways because we're speaker nerds. :)
 
#11 ·
An alternative approach, which doesn't involve sending the output from an amp into your soundcard, would be to use the time lock feature from HolmImpulse. There is a good description of how to set it up in http://www.diysubwoofers.org/misc/holmimpulse/ta.html

The same basic approach would be used but you'd measure 2 separate subs instead. For example;

connect 1 inuke to one sub
connect the other inuke to another sub
place the mic equidistant from each sub
connect the inukes so that 1 sub sees channel 1 from your soundcard and the other sees channel 2
take a measurement in holm to channel 1
set time lock
take a measurement in holm to channel 2
review the offset of the IRs, any meaningful delta will mean that one has lower latency than the other

If you're using a USB mic then bear in mind that the time lock drifts over time, if you take the measurements within say 10s then no problem.
 
#13 ·
There probably won't be any delay. If there was, then Seaton customers would probably be complaining since all the Catalyst speakers and Submersives have an A/D, DSP, and D/A stage like the iNukeDSP amps.

You can test with your Omnimic using the instructions:

 

Attachments

  • Like
Reactions: Archaea
#17 · (Edited)
Just tested tonight. The Inuke DSP 3000 and iNuke DSP 6000 are identical for all practical purposes.

Here was my testing method.





Unhook all subs except front two center (of six front) splitting difference between center seat. Play those two subs along with L/R mains in stereo mode on Onkyo AVR. Play track 2 on omnimic. Put mic in center of main listening position. Ensure Omnimic impulse graph (per Desertdome's post) was matched up with a single sine wav looking impulse. This was easy to tell - move the mic an inch to the left or the right and it would start to show two peaks instead of one peak.









The subs didn't have their own peak meaning they were timed appropriately through the Onkyo AVR based on the distance settings setup in the AutoEQ process. I swapped the amp from the 6000 to the 3000 and checked to make sure the impulse peak didn't noticeably change. It didn't. Since, it was easy to throw off the impulse peak by moving my mic just a tad bit from the exact center position I'm pretty comfortable this is not something to worry about for subwoofer use. I didn't test the impulse graph on main speakers alone - but that isn't my use of my iNuke amps and unless the testing I did was invalid because I didn't use the mains -- then I'm comfortable saying the iNuke DSP 3000 and iNuke DSP 6000 can be used interchangeably in the time domain for subwoofer use of the same input signal.


Impulse response of iNuke DSP 6000





Impulse response of iNuke DSP 3000


If you download the two files and swap them back in forth in a picture viewer - they vary ever so slightly. I can get much bigger variations by simply moving the measuring mic an inch to the left or right where two impulse peaks will start to develop and move apart as I move the mic left or right from the exact center position.


Frequency response in stereo mode from measured mic position for both iNuke DSP 6000 and iNuke DSP 3000
 

Attachments

#19 · (Edited)
I am confused about your setup. Did you just run the amps on the subs or also on the mains? If you just ran speakers off of avr and switched amps on subs, I don't think it would show much difference. I think it measures the speakers more on the impulse response. Basically that audyssey got the arrival times of your speakers correct.

"Align two speakers in "time of arrival" at a listening position

Arrange your system so that the Test Track sound plays out of both speakers you are working with, but with any other speakers silent. Choose the Frequency Response page and play the indicated track with the microphone set at the desired position. Set the control that is near the top to "blended" and observe the Impulse Response display. If there is a time of arrival difference between the two speakers, you should be able to see two impulse peaks. The one to the left (near the 0 msec point) is from the speaker with the earlier delay time, the other is for the later one. If you have an electronic delay unit, adjust the delay applied for the earlier speaker so the peaks become superimposed. If you have more than two speakers you wish to align together, choose the latest (the one making the right-most peak) to be the reference and delay the others to align to its peak"

If you didn't use amps on speakers I think that is the way to go. Measure and place you mic with the avr and ensure a single peak. Then put 3000 on left main and 6000 on right main level matched and see if they still match. Don't even need the subs on for the test.

Also amps aren't labeled on frequency response graph, is that just different sub placement or level difference or do they have different roll-off?
 
#20 · (Edited)
I am confused about your setup. Did you just run the amps on the subs or also on the mains? If you just ran speakers off of avr and switched amps on subs, I don't think it would show much difference. I think it measures the speakers more on the impulse response. Basically that audyssey got the arrival times of your speakers correct.

"Align two speakers in "time of arrival" at a listening position

Arrange your system so that the Test Track sound plays out of both speakers you are working with, but with any other speakers silent. Choose the Frequency Response page and play the indicated track with the microphone set at the desired position. Set the control that is near the top to "blended" and observe the Impulse Response display. If there is a time of arrival difference between the two speakers, you should be able to see two impulse peaks. The one to the left (near the 0 msec point) is from the speaker with the earlier delay time, the other is for the later one. If you have an electronic delay unit, adjust the delay applied for the earlier speaker so the peaks become superimposed. If you have more than two speakers you wish to align together, choose the latest (the one making the right-most peak) to be the reference and delay the others to align to its peak"

If you didn't use amps on speakers I think that is the way to go. Measure and place you mic with the avr and ensure a single peak. Then put 3000 on left main and 6000 on right main level matched and see if they still match.

Also amps aren't labeled on frequency response graph, is that just different sub placement or do they have different roll-off?
For the test

Left front main driven by Crown XLS-402
Right front main driven by Crown XLS-402
Two center subs driven iNuke DSP6000 and then driven by iNuke DSP 3000.

Before I swapped the iNuke DSP amps used for the pair of subwoofers I moved the the mic to the centermost position until I had a single impulse peak. That was easy to find because of the realtime sweeps of the omnimic. I simply moved the mic stand left or right until I had one impulse peak instead of two.

As expected based on distance measurements with the tape measure between left and right speaker - the final mic measurement where there was 1 impulse peak was in the center of the center chair (MLP) where I initiate my Audyssey measurements. See pictures in my previous post. I have four chairs - one chair is centered on the projector screen - two chairs on either side and my fourth chair is way off angle and basically a throwaway seat if hoping for optimal audio.

I then took frequency sweeps and impulse measurements with both iNukes assigned to sub duty seperately. Simply swapping the amps out and changing nothing else. The difference between the two sweeps bass SPL levels is because I didn't bother to level match the iNuke amps. I think it's immaterial because with either iNuke amp - the single impulse peak didn't change to two peaks. If one of the iNuke amps had introduced an additional delay or less delay then the impulse would have split back to two peaks again right? The subs were run hotter than the mains as you can see in the frequency response capture --- I think this should be good enough for my purposes - unless I'm missing the mark in a major way?

Is my testing invalid for some reason -- ultimately I don't really care if they aren't identical for mains because I'm using them for subs, and as best I can tell - the impulse peak just looked nearly identical (not exactly, but very nearly) identical between both amps --- so to me that means they don't have different enough processing delays to matter if I just split subwoofer out signal to both iNuke DSP 6000 and iNuke DSP 3000 amp. Meaning - I don't have to worry about the 3000 amp processing slower or faster and muddying my bass when sharing a split signal with the two 6000 amps I'm using.
 
#21 ·
It is not clear but if your mains were running while you were looking at the IR then your test is invalid. The magnitude of the IR peak from a main speaker is orders of magnitude bigger than that from a sub so all you will really see is the mains. You should turn off the mains when doing this and just compare the subs. Alternatively use a limited bandwidth sweep (but the same bandwidth for each speaker)
 
#22 · (Edited)
If splitting signal between different iNuke amps - will time delays be identical?

Should change title to "If splitting signal between different iNuke amps - would there be any time delay causing an issue with my subs? :p
Time delay's won't be as important or noticable with wavelengths of 60 ft vs 1 ft. I doubt there is a difference between the amps, but using the speakers would be the more precise way to actually see if there is any difference. I don't see people using impulse response to set delay to there near-field subs, instead using frequency response in the modal region.

IMO you should just go with four 6000's. You know same power delivered to each sub, can group them front or back in any multiple, can set eq's to be the same between all subs, set same delay's to all near subs the same, and swap them back and forth easy to try different things out... Just seems the simplest setup route.

I thought your next meet would be the near field sub meet since that's been big on the forums lately. Try all up front, then two in back, four, six, 8. Could even try near subs at different distance from seats. Different crossover points and how running subs hot or level affects it. May be some people who prefer all subs up front. I don't know it could be fun. :D
 
#24 ·
The mains were playing at the same time as the subs --- all I was trying to figure out was did the amps differ at all in processing delays to screw with the subs timing and potentially cause muddy sound. The purpose of the test was so I could feel comfortable splitting a single subwoofer out connection between two iNuke DSP 6000, and one iNuke DSP 3000 to power five subs on my front row under my screen.

But if the mains playing means the subs weren't even being measured with the impulse reponse - then I failed and will have to try it again.


So to repeat the test --- I guess I could setup the left speaker being powered off one iNuke DSP 6000, and the right speaker being powered off one iNuke DSP 6000.

Then get the mic stand setup so the imulse peak is just a single peak.

Then swap the iNuke DSP 6000 on the left channel for the iNuke DSP 3000.
Then verify the impulse peak is still a single peak with two different amps.


Right?
 
#26 ·
But if the mains playing means the subs weren't even being measured with the impulse reponse - then I failed and will have to try it again.
the IR is of the combined response which will be dominated by the high frequency output. It is feasible to compare a HF IR with a LF IR but it's quite fuzzy due to the shape of the 2 IRs & you need a timing reference when you take multiple measurements. To illustrate, here are 2 separate measurements overlaid, one full range (which appears as a v short, sharp spike) and one a sub on it's own.

Text Green Line Plot Diagram


You can align speakers or subs using the IR alone though, you just need to use the same frequency range for the sweep (example here) and be careful with how you interpret the shape of the IR (example here)

So to repeat the test --- I guess I could setup the left speaker being powered off one iNuke DSP 6000, and the right speaker being powered off one iNuke DSP 6000.

Then get the mic stand setup so the imulse peak is just a single peak.

Then swap the iNuke DSP 6000 on the left channel for the iNuke DSP 3000.
Then verify the impulse peak is still a single peak with two different amps.
I think that sounds reasonable.
 
#25 ·
Do the mains, get a precise measurement. If time arrival is off on mains, which are more precise for arrival time measurements than subwoofers (why audyssey gets speaker distance correct and often doesn't the subs), then use delay to set them the same. You can now use this delay with the amps to make sure amps are sending signal at the same time whether it affects your subwoofers integration or not. Like others have said, the 6000 is two 3000's and they use the same software so I would highly doubt they would be different. This is just how I would do it, maybe I am totally off base. :confused:
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top