Smyth Research Realiser A16
Dec 10, 2022 at 2:00 AM Post #14,341 of 16,050
PS: Although the German distributor offered a new measurement for my obviously faulty PRIRs, they never cared to respond to my list of requests to make sure everything is handled more professional next time. I still have not received any answer to my second mail from October 6th. Please be aware that my message to him was in German and that this just a quick translation using deepl.
I think you nailed down very clearly and comprehensively what wasn’t good during your trip at MSM Munich and especially how a PRIR measurement session should be addressed to be really satisfying for a prospective customer.

I do hope that those from for-tune.de and MSM will reconsider their way of thinking and action and will invite you to a new PRIR measurement session.
 
Dec 10, 2022 at 8:56 PM Post #14,342 of 16,050
Solution for 1: Perform the whole calibration as a synchronous measurement with the use of the headtracker. I am very sure that this makes a big difference qualitatively. In my own calibration, in my Pilates studio optimized for audio on the day using diffusers and absorbers, it often took a very long time for my perceived line of sight to match the actual line of sight measured by the headtracker. During the measurement in Paris with Smyth, a head tracker was also used. This also allows a direct comparison between headphones and speakers after the calibration.

I don't think that there's any significant difference between synchronous and asynchronous measurements, at least from a technical perspective. Someone had requested longer delays between view angles in the async signals, but when I asked about specific recommendations for a delay between the prompt and the start of the signal, nobody considered that necessary anymore.

Solution for 2: 12 seconds sweeps instead of 4 seconds sweeps with thus clearly higher signal to noise ratio. Especially with the sensitivity of the inEar microphones of the Realiser very important. In the telephone conversation with Mike Smyth it became clear that he also considers this to be a possible cause of the problems with my calibration. He also confirmed that he can also deliver correspondingly long asynchronous tones, if a synchronous measurement would be excluded.

12s sweeps are already available, but as far as I know, no sweeps with more that 3 look angles, and none at all for elevation != 0°. I think you vastly overestimate the importance of SNR in the measurement sweeps (any noise in there does not manifest as noise when using the PRIR), but given the price, and the overhead to even set up for a measurement, I don't see any good reason not to use 12s sweeps.
 
Dec 10, 2022 at 9:36 PM Post #14,343 of 16,050
but when I asked about specific recommendations for a delay between the prompt and the start of the signal, nobody considered that necessary anymore.
I'd like it! :D
 
Dec 10, 2022 at 11:06 PM Post #14,344 of 16,050
I'd like it! :D
I found the delay between the prompt and start of the signal to be critical to the production of the PRIR that John measured for me in the studio. In fact, it was a significant part of the “secret sauce” that I later realized(no pun intended) to have been missing when I tried making my own PRIR measurements at home. That and the proper placement of the in ear microphones were extraordinarily helpful at arriving at a convincing PRIR. These were, of course, just some of the many elements that worked together to produce a measurement that exceeded my expectations. While the Omega PRIRs do work for me and are an outstanding achievement, my own 65” PRIR is, not surprisingly, even more realistic sounding than the Omegas.
 
Dec 11, 2022 at 12:06 PM Post #14,345 of 16,050
I'd like it! :D
And how much delay would you need? I still don't have my Realiser back, so I can't really test, I just assume that one can insert arbitrary delays, but I could be wrong. Anyone of you who has PCM async files could test that with e.g. Audacity. If you have the Atmos files, but no PCM, just convert them to stereo or 5.1 PCM to test.

Also, Richter Di mentioned that the head tracker would help. Is that thing completely inactive during async measurements? If so, someone with better contacts to the Smyths should file a feature request that it should be active and just provide passive feedback for faster, accurate positioning. Simply showing the current Azimuth and Elevation angles in large, friendly digits on the main display would be great.
 
Dec 11, 2022 at 2:12 PM Post #14,346 of 16,050
And how much delay would you need? I still don't have my Realiser back, so I can't really test, I just assume that one can insert arbitrary delays, but I could be wrong. Anyone of you who has PCM async files could test that with e.g. Audacity. If you have the Atmos files, but no PCM, just convert them to stereo or 5.1 PCM to test.
I think enough of a delay that the user could safely pause the measurement process.

Is that thing completely inactive during async measurements?
It is active... and an extremely valuable tool to keep the user on target during measurements... unfortunately, like most things, the angles are very difficult to see on the display.
 
Dec 11, 2022 at 2:51 PM Post #14,348 of 16,050
I think enough of a delay that the user could safely pause the measurement process.

Have you tested pausing the async measurement process? Sounds like a potential problem if the playback device stops and starts the output clock.
It is active... and an extremely valuable tool to keep the user on target during measurements... unfortunately, like most things, the angles are very difficult to see on the display.

Do you mean the LCD display? With async signals, you can place that as close to your face as you want, can't you?
 
Dec 11, 2022 at 2:57 PM Post #14,349 of 16,050
Solution for 1: Perform the whole calibration as a synchronous measurement with the use of the headtracker. I am very sure that this makes a big difference qualitatively. In my own calibration, in my Pilates studio optimized for audio on the day using diffusers and absorbers, it often took a very long time for my perceived line of sight to match the actual line of sight measured by the headtracker. During the measurement in Paris with Smyth, a head tracker was also used. This also allows a direct comparison between headphones and speakers after the calibration.

I don't think that there's any significant difference between synchronous and asynchronous measurements, at least from a technical perspective. Someone had requested longer delays between view angles in the async signals, but when I asked about specific recommendations for a delay between the prompt and the start of the signal, nobody considered that necessary anymore.
Also, Richter Di mentioned that the head tracker would help. Is that thing completely inactive during async measurements? If so, someone with better contacts to the Smyths should file a feature request that it should be active and just provide passive feedback for faster, accurate positioning. Simply showing the current Azimuth and Elevation angles in large, friendly digits on the main display would be great.
It is active... and an extremely valuable tool to keep the user on target during measurements... unfortunately, like most things, the angles are very difficult to see on the display.
For me is the head tracker a part of a perfect measurement process. Otherwise it is total guess work were you are looking. I can see this daily in my Pilates studio that people have no idea how far their head or body rotates and how much is “simulated“ by the eyes turning with you. There might be no difference between synchronous and asynchronous measurements from a technical standpoint but in my opinion there is a huge difference when it comes to using the headtracker or not. For Tune, the German distributor, in the msm studio setting did not wanted to use a head tracker and that it my critique.
 
Dec 11, 2022 at 3:05 PM Post #14,350 of 16,050
Solution for 2: 12 seconds sweeps instead of 4 seconds sweeps with thus clearly higher signal to noise ratio. Especially with the sensitivity of the inEar microphones of the Realiser very important. In the telephone conversation with Mike Smyth it became clear that he also considers this to be a possible cause of the problems with my calibration. He also confirmed that he can also deliver correspondingly long asynchronous tones, if a synchronous measurement would be excluded.

12s sweeps are already available, but as far as I know, no sweeps with more that 3 look angles, and none at all for elevation != 0°. I think you vastly overestimate the importance of SNR in the measurement sweeps (any noise in there does not manifest as noise when using the PRIR), but given the price, and the overhead to even set up for a measurement, I don't see any good reason not to use 12s sweeps.
Never said that it would lead to “noise” while using the PRIR.
What is the base of your claim that I am “vastly overestimating the importance of SNR in the measurement sweeps”? Did you do any tests comparing 4 to 12 second sweeps?
 
Dec 11, 2022 at 3:08 PM Post #14,351 of 16,050
Have you tested pausing the async measurement process? Sounds like a potential problem if the playback device stops and starts the output clock.
I haven't noticed a problem

And yes... I meant the LCD. It's difficult for most to read at any distance.🤪
 
Last edited:
Dec 11, 2022 at 3:17 PM Post #14,352 of 16,050
For Tune, the German distributor, in the msm studio setting did not wanted to use a head tracker and that it my critique.

I have had some say that they have no intention on using headtracking while listening. I would still insist on using it for measurements. As it is an invaluable tool in lining up the center position of all channels... Like @Richter Di says... Without it. You are guessing.
 
Dec 11, 2022 at 3:22 PM Post #14,353 of 16,050
I have had some say that they have no intention on using headtracking while listening. I would still insist on using it for measurements. As it is an invaluable tool in lining up the center position of all channels... Like @Richter Di says... Without it. You are guessing.
Thank you. Maybe your voice reaches those putting their hard earned money on the table. And maybe they just insist.
 
Dec 11, 2022 at 4:46 PM Post #14,355 of 16,050
The Realiser is still relevant and people today are buying the same thing you preordered. Doesn't excuse their behavior but it's still the #1 product.
Its been 5 years since I dropped $1700 on a preorder. Istn there a legal obligation of "good faith" to get the preorders filled now?
Its not a reasonable expectation to have to wait 5 years.
 

Users who are viewing this thread

Back
Top