[audiohobbit]

3 thoughts about the fixed vertical offset:
1. I have no user experience except for the short demo I had. So question to the long time A8 users: did you ever feel the need for this, did you ever lay your head backward a bit on the couch and experienced the virtual speakers "lifting up" and did this downgrade the whole listening experience for you?
2. It is also possible to simply measure an extra PRIR while holding your head in this position (for the people who still have to make their PRIRs or have the opportunity to re-do their PRIRs without too much hassle, besides: also in Stephens proposed scenario you need new measurements with vertical lookangles anyway).
3. If this is not very time consuming for the Smyths to implement then please, go ahead. But if this would considerably slow down all the other things on their to-do list (including the complete normal vertical head tracking) then skip it.
So actually I fully agree with this:

1. yes. It was my specific question to Stephen. I recognise this when watching movies and lean back in my chair that the voices go up. But I have a huge screen in front of me so voices are still inside the picture, only at the top edge. But I think for people with "small" TVs voices will come from far above the TV then.
So for me it's not that important, I think more important is that you at least have the vertical measurements. But this seems to get easily very complicated so I tend to drop this and stick to just horizontal measurements.

2. Would be possible but difficult. And one reason for me asking Stephen was for measurements that you can't easily re-do (in a studio or high end home cinema etc.)

3. I don't know how time consuming this would be for them, but every new piece of code introduces the chance to get new bugs etc. I already told Stephen that I think most important is to have the possibility to at least measure vertical look angles.
But with Stephens answer I'd have a million new questions..

 

[audiohobbit]

ust for reference, I will connect the RCA analog headphone outputs to my Stax amp (which supports both RCA and XLR inputs via INPUT switch) and test this out

Do you only have Stax headphones?
No normal ones that you can connect directly to the amp in the A16 just for testing?

 

[dsperber]

Do you only have Stax headphones?
No normal ones that you can connect directly to the amp in the A16 just for testing?

Before I had my external DAC I was using the RCA headphone outputs on my A8 to feed my Stax amp, and that was working fine. Better quality sound with the external DAC, but the original RCA approach also was fine. Re-trying this was my plan, to eliminate optical output and my external DAC as being at the root of the symptom.

Although I've never done it before I also have Shure SRH940 closed-back headphones I can connect directly to the A16. I suppose I would need to create an HPEQ for it first.

But simply hearing from others regarding the SIX separate situations I described above, as to whether you do or do not have those six same crack/pop anomalies using your own setups, this would be very helpful to confirm or deny that either my setup is somehow totally unique or that others do in fact hear one or more of these sounds same as I do, but with their own equipment and configuration.

 

[audiohobbit]

I suppose I would need to create an HPEQ for it first.

I think not for the cracking/popping test. Don't think that this can come from HPEQ (well but you never know).
So just start without HPEQ. Just connect it to the A16 amp. If there's no cracking etc. then make a HPEQ.

By the way just to be sure: We're talking about LOUD cracks/pops, yes?
Some minor clicks/pops when switching something wouldn't alarm me.

With the A8 (connected via RCA analog to external amp) I hear a relatively loud noise when switching off the A8, but I don't bother since it's not that loud that it could damage headphones.

 

[dsperber]

By the way just to be sure: We're talking about LOUD cracks/pops, yes?
Some minor clicks/pops when switching something wouldn't alarm me.

The power-on sounds Crack1-Crack3, are relatively minor. Power-off "pop" Crack4 is even less. These are all 1-syllable and short, but definitely present and noticeable, and immune in loudness to whatever is the variable Volume setting of the A16 at that moment.

But preset-selection "WAIT" Crack5 and Crack6 (both identical) are 2-syllable explosions and very loud. If you have your headphones on and all of your external DAC/amp powered on (as you probably will) and you've got your A16 and headphone amp volumes set for listening, and you're now ready to listen to something after you've selected the correct preset/input, this 2-syllable explosion whenever you select a preset (either randomly or sequentially) will truly startle you it is so loud.

Assuming the sounds are related to the use of optical data feeding external DAC, the 2-syllable Crack5/Crack6 soiund might be (a) turning off current optical data output as part of the preset selection process, and then (b) turning on optical data output for use with the now selected and loaded preset. That could conceivably induce a "crack" response from the DAC for each of these actions. I'm just speculating, of course.

 

[frank2908]

Im not sure we are having the same "crack", in my experience there is lot of distorsion in the following case, listenping to p m 2 channel
Fiio m5>optical out>coax to mojo, line out 3V mode > 3.5 cable to rca> stereo input of a16
There is lots of distorsion, but then I find out that maybe the 3v line out of the mojo was too high, so i change the set up chain as following
Fiio m5,line out mode> 3.5 to rca cable> stero in a16: all distorsion is gone, sound very clear, but my headphones lacks a bit bass and body, so I add my mojo to the chain using spdif out, and there is more body and bass, but the localisation and imaging suffer.
My guess is the best system for 2 channel music listening is to either remain simple, dap source to optical into a16, and headphones out from a16. Or dap to optical in, then a16, then a very clean amp from headphone amp at the back of a16.

 

[BlueIguana]

Just wondering if the Smyth will work with IEM's like the Audeze LCD-i4 as well? I'm not a huge fan of cans.

 

[bcwang]

Yes, I use optical output from the A16 to my external DAC, which then feeds XLR to my Stax amp. And, like you, I am suspicious that the cause of crack is the result of how my DAC reacts to the instantaneous disappearance/reappearance of optical digital data being fed to it. I don't know if there is a more graceful way the A16 can deal with this.

Just for reference, I will connect the RCA analog headphone outputs to my Stax amp (which supports both RCA and XLR inputs via INPUT switch) and test this out. I will also run a similar set of tests on my A8.

NOTE: turns out there are actually SIX different situations in which anomalous cracks (of either 1-syllable or 2-syllable) or pops of different fixed volumes (unrelated to the variable Volume setting of the A16) are occurring, which I will name Crack1, Crack2, etc to avoid ambiguity in future discussions:

(1) power-on, after the Smyth logo screen disappears and the Home Menu screen appears and begins to populate. The first thing that happens is that "wait" appears in the UserA preset line, at which time 1-syllable Crack1 occurs.

(2) power-on, while UserA preset is being loaded and "loading" is displayed, there is silence. But when preset loading completes and "activated" is presented a 1-syllable Crack2 occurs.

(3) power-on, after the UserA preset process completes the UserB preset process is now perfomed. Again, the state goes from "wait" to "loading" and then "activated" at which time a 1-syllable Crack3 occurs.

(4) power-off, at which time a short 1-syllable Crack4 (really a "pop") occurs

(5) numeric keypad DISABLED, so that new 1.75 functionality supporting direct-select of presets using 1-9 keys is facilitated. When Speaker Map is displayed and 1-9 keys are used to directly select a preset, there is a large "WAIT" displayed on the screen, at which time the loud 2-syllable Crack5 occurs. Doesn't matter if the same preset is selected as is already currently loaded or if a different preset is selected, when the "WAIT" appears the Crack5 sound occurs.

(6) numeric keypad DISABLED or ENABLED (doesn't matter), standard sequential preset browsing/selection functionality always is available using the +/- ADJ keys when the Speaker Map is displayed. Whenever the +/- ADJ keys are pressed and "WAIT" is displayed, the loud 2-syllable Crack6 occurs (same as Crack5 from direct-select of preset).


It is certainly possible that these unexpected cracks/pops are due to how my external Audio-GD NFB9 DAC is responding to whatever optical data is being fed to it by the A16. There may or may not be the same results using a different brand of external DAC, or when using the A16 built-in DAC/amp. I am limited in what I can test, which is why I was asking for supporting comments from others, to complete and flesh out the total set of test data points and so that Smyth can address all the situations with relevant firmware fixes.

In my experience, I have not experienced any cracks or pops in any of the scenarios that you would have the headphone on your head. I have not listened during power-on because I use that point to have my headphones still and laying there in the storage position for the head tracker calibration warm up.

But I do believe strongly that the pop comes from your external DAC, the reasoning is this. You said the variable volume of the A16 makes no difference to the pop. That means you are using a fixed volume on your post dac side and adjusting volume with your A16. This would be the exact scenario where a pop due to your dac digital stream being interrupted would translate to a fixed volume pop from the Dac. If your post dac has fixed amplification, the pop would be a fixed volume as coming from your DAC.

One thing you could try is to also see if your DAC pops if you disconnect and re-connect the optical cable while a stream is playing. This would show how your dac handles a broken stream, maybe with a pop. But even if this test passed, it doesn't mean there isn't some other data stream that might cause a pop, like if the A16 sent some invalid data during a mode switch. Other DACs might ignore as junk while your DAC still processes it and plays it as a burst of noise.

 

[dsperber]

But I do believe strongly that the pop comes from your external DAC, the reasoning is this. You said the variable volume of the A16 makes no difference to the pop. That means you are using a fixed volume on your post dac side and adjusting volume with your A16. This would be the exact scenario where a pop due to your dac digital stream being interrupted would translate to a fixed volume pop from the Dac. If your post dac has fixed amplification, the pop would be a fixed volume as coming from your DAC.

I haven't had a chance yet to test things completely bypassing the use of the external DAC, instead routing the RCA headphone output of the A16 directly to the RCA input of my Stax amp. This would of course now be using the built-in internal DAC and HPamp (as a preamp to the Stax amp) of the A16. So that particular piece of experimental result is not yet known.

As far as the variable volume of the A16 makes no difference to the loudness of the pop/crack, your explanation makes perfect sense. After the A16 (and its own variable output volume, which influences the volume implicit in the optical data stream sent to the external DAC) there is the DAC and its volume setting, as well as the Stax amp and its volume setting. So if the the volume of the crack is a constant even when the A16 volume is way down, that implies the constant volume must be the result of the fixed volume settings on the DAC and Stax amp, which further implies that the source of the crack is at the DAC. And this clearly must be due to how it behaves when the optical data stream is suddenly discontinued or resumed, in some kind of a "non-graceful" way if there is such a thing as "graceful" instantaneous loss/resumption of data.

So, it still seems that the A16 is kind of the real source of the data-interruption issue and consequential behavior of the DAC, instantly disappearing and then instantly resuming optical data going to the DAC as a result of sequential/direct preset selection processing (and also those other four minor power-off/on situations). Perhaps it might be possible to just substitute a continuously output "null" optical data stream of bytes even during the preset selection process moments, with a payload that represents sound at ZERO VOLUME, instead of shutting off and on the spigget. Once the preset gets selected actual data flow must clearly resume, with this ZERO VOLUME implicit in it. Obviously normal "silent" DAC operation resumes after the crack, so clearly something must be starting again after the preset selection process finishes, even when there is still no actual source content being played yet through the A16. Actually I don't know that this is how things really work. Pure speculation.

In other words, whatever is really going on optically in "steady state" and with nothing playing, headphone output via the external DAC is silent. Content can start and there's no new crack just because real data started, so it must simply be that there is now actual real optical digital data reflecting actual sound with a real volume that is flowing from A16 to DAC. So why can't that same continuous zero volume data stream (or nothing at all) as is always going on and which produces perfectly silent DAC behavior even when no content is playing, why can't that same stream or state be kept up and operational even while the preset selection process is going on in the A16? Then there would never be any data cutoff/resumption to trigger the crack from the DAC.

Anyway, I've had several emails back and forth from Smyth on this, although I don't know if they know yet exactly what they will do to fix it. But they say it should be "simple" and they expect it to be in the next firmware release. It does seem to definitely be related to use of my external DAC, but why not also when using the internal DAC? I will know later tonight from my own RCA experiments if it is uniquely from my external DAC or not.

 

[castleofargh]

Just wondering if the Smyth will work with IEM's like the Audeze LCD-i4 as well? I'm not a huge fan of cans.

there is a head tracker to put your head, and the measurements are done by placing little microphones in your ear canal, where an IEM would be. so no IEM solution.

 

[blubliss]

Wrong.

 

[BlueIguana]

there is a head tracker to put your head, and the measurements are done by placing little microphones in your ear canal, where an IEM would be. so no IEM solution.

Thanks for the response. My understanding was that the microphones are only used during the measurement process, so not sure why that would negate the use of IEM's... And you dont absolutely have to wear the tracker, right?

 

[sander99]

Just wondering if the Smyth will work with IEM's like the Audeze LCD-i4 as well? I'm not a huge fan of cans.

IEMs can be used. You can do a manual HPEQ, the manLOUD HPEQ, or use a neutral HPEQ. The head tracker can be worn using a head band. (I even remember having seen a big plastic clip - in Evshrug unboxing video I think - that can be used, that would have been included for head tracker guidance during PRIR measurements.)

 

[castleofargh]

Thanks for the response. My understanding was that the microphones are only used during the measurement process, so not sure why that would negate the use of IEM's... And you dont absolutely have to wear the tracker, right?

as @sander99 said, you can always bypass some parts. at some point nothing stops you from measuring with one headphone and then use that simulation with an IEM, but then you won't have measured the IEM itself in your ears and what you'll get will never be as close to the original room you're trying to copy. and the alternative is having the IEM and the mic in your ear at the same time for the measurement, which is going to be physically difficult. ^_^

as for the head tracker, if you're not going to use it, I'd argue that the A16 has lost half of its appeal. I guess it's a matter of personal opinion, but if the A16 didn't have head tracking, I wouldn't have cared for it. now if you're fine with fixed position, better checks https://www.head-fi.org/threads/recording-impulse-responses-for-speaker-virtualization.890719/ the cost will be to procure mics, and .... that's it.

 

[sander99]

as @sander99 said, you can always bypass some parts. at some point nothing stops you from measuring with one headphone and then use that simulation with an IEM, but then you won't have measured the IEM itself in your ears and what you'll get will never be as close to the original room you're trying to copy.

In Munich I have spoken about this with Mike Smyth personally. He said you can use them. In fact with IEMs it is mostly one resonance of the ear canal that has to be compensated. Normal headphones interact with the outer ears and that's what needs more complicated compensation.
And even IF you would not manage to get exactly the same timbre as the original loudspears+room it would not degrade the out of head experience.
And the manLOUD HPEQ procedure works well if we can believe audiohobbit.
[Edit: see 9.1.3.1 in the manual v091.]