TOPIC:

Interesting hypothesis re: normalization issue but I followed the steps outlined in @dreite's thread which did not change my results.

@Entripy and @john.reekie - you guys are going to have to come up with something better than user error or a REW bug here. Frankly, it's kind of bizarre and is starting to feel like trolling. Please review my first post and see the impulse and step responses for Roon>Chromecast>SHD via Toslink. They are textbook for far field measurements. The stimulus signal, conditions, and measurement system are identical for all that were evaluated. These measurements are repeatable over and over and over again. Clearly Volumio managed signals are not truthful to the source.

I have an open ticket with minidsp and will continue to update the forum on any progress that can be made. That said, if this is affects all SHD Studios (and possibly all SHDs) this is a MAJOR issue as it is clearly audibly impacting.

Clearly it doesn't affect my SHDs or I think I would have noticed. Your impulse response was just so bizzare for the volumio case we're just brainstorming. Volumio should be almost unlistenable with that response, not just a little wrong.

Understood. Appreciate the clarification.

Wow. You've uncovered something interesting, but accusing people that look into it of trolling? That's a new low for this forum I think.

entripy wrote:

Clearly it doesn't affect my SHDs or I think I would have noticed.

While it's possible that it varies with units (or version of Volumio), the impulse responses I displayed earlier were from a unit that also displayed the weird impulse when measured with the REW offline sweep. Two different measurement methods, two different answers, ....

dreite wrote:

Did you guys see my post about the volume normalization bug??
I suspect the effects of that are what you're seeing here with the impulse measurements.

Dave.

Interesting thought but the recorded sweeps I have that produce the weird impulse response are constant in amplitude. (Except for the beginning and end per the stimulus.)

john.reekie wrote:

dreite wrote:

Did you guys see my post about the volume normalization bug??
I suspect the effects of that are what you're seeing here with the impulse measurements.

Dave.

Interesting thought but the recorded sweeps I have that produce the weird impulse response are constant in amplitude. (Except for the beginning and end per the stimulus.)

The normalization issue could be isolated in a much more straightforward way than the type of testing employed in this thread. However, when I first noticed this I happened to be doing some impulse testing via ARTA. 'Some very weird effects that had me confused for awhile until I found the Volumio problem.
Regardless, I was just throwing it out there as a potential cause to investigate.....since we're spit-balling here.

FYI for everyone, this normalization issue has still not been fixed in the latest Volumio iteration (Version 1.038), so all SHD users who haven't performed the little workaround procedure I posted have Normalization active in their system with anything played via the Volumio portion.

Dave.

joelkarp wrote:

@Entripy and @john.reekie - you guys are going to have to come up with something better than user error or a REW bug here. Frankly, it's kind of bizarre and is starting to feel like trolling. Please review my first post and see the impulse and step responses for Roon>Chromecast>SHD via Toslink. They are textbook for far field measurements. The stimulus signal, conditions, and measurement system are identical for all that were evaluated. These measurements are repeatable over and over and over again. Clearly Volumio managed signals are not truthful to the source.

You've got a couple guys trying to help you here since devteam is out for the weekend. Yet, you act like a smart-alec and accuse them of trolling.

Yeah, that's a good way to go.

Dave.

I spent several hours looking at the issue, I wouldn't call that spit-balling... aaanyway thanks again, I'm still curious as to the root cause.

john.reekie wrote:

I spent several hours looking at the issue, I wouldn't call that spit-balling... aaanyway thanks again, I'm still curious as to the root cause.

Spit-balling in the larger context.......as in including everybody who's trying to help and throwing ideas out.
I thought that was obvious. C'mon John.

Dave.

Sorry, slightly ticked off...

Anyway, here's another plot. The "impulse response" at the left is from a 256k sweep, at the right is from a 2M sweep. If these were real, the length of the impulse response wouldn't vary with the sweep length. (Ergo, in case it's not obvious to everybody, they aren't real.)



I suppose something JohnM might want to look into. Why it happens with Volumio on the SHD only, I have no idea. It doesn't happen with Volumio on an RPi. I'm guessing maybe clock drift of some kind?

john.reekie wrote:

I'm guessing maybe clock drift of some kind?

That does seem to be it. If I compress the recorded sweep slightly in time then do the import in REW I get impulse responses more like what you'd expect. Green is with no timing reference, orange is with the timing reference. REW doesn't seem to be getting the right clock adjustment.

Wow, excellent work John. I still don't really understand why REW/Volumio is so susceptible to this though.

Don't know. I've discovered that you can induce the weird impulse response just by stretching and padding the stimulus and using that in lieu of a recorded sweep. I've given up on it now tho due to null pointer errors in the import.

I didn't mean to ruffle any feathers. The thought of this issue being a measurement anomaly seemed and still seems incredibly implausible to me. With little explanation as to how or why this could happen - especially considering how two inputs are measuring so different with all other conditions being equal - I incorrectly assumed the efforts were meant to discredit this thread. I now realize this was not the case.

As an update, I do have one response from miniDSP tech support but not much information yet.

I did decided to post to the REW forum as well to see if JohnM or someone else can chime in. Here is the link in case anyone is interested: www.avnirvana.com/threads/rew-measuremen...io-volumio-bug.5937/