Smyth Research Realiser A16
Oct 23, 2019 at 11:52 AM Post #7,006 of 15,951
I decided to try and test the manLoud routine last night to test out the youtube video and how user friendly it is/is not... and the bottom line is the video is easy to follow and the results are very good when you complete the manLoud routine....

I decided to use a headphone that I had never used/measured on the A16.. so I chose the Audio-technica ATH-m50s/LE , wanted the A16 to have a little challenge.. :)

I set the A16 on the kitchen counter, plugged it in to the AC wall plug and turned it on..nothing else hooked up to the A16..
To start the routine the A16 needs to know two items.. 1) user name and 2) headphone name... so I went into the menu and verified the user name was Dixter and created a new headphone called M50...that out of the way I pulled out the necessary hardware 1) in-ear mics 2) in-ear lanyard strap 2) grounding wrist strap 3) scotch tape 4) M50 headphones

I un-raveled the cords on the in-ear mics and chose one to be the right side ( placed a small piece of stoctch tape at each end to mark it as right mic) and then I aligned up the cords length wise and placed scotch tape about every 6 inches up the cord and once completed I had a single mic cord with one end both connectors ( one marked as right ) and on the mic end (also one marked as right)

I un-raveled the cord on the grounding wrist strap and adjusted the strap band and placed it on my wrist with the metal part in contact with the bottom of my wrist...and plugged it into the ground plug on the back of the A16

I plugged the M50 headphones into the A16 ... plugged the in-ear mics into the front of the A16, making sure the right mic is in the right input.. I then placed the in-ear mics into my ears ( right side to right ear ) and made sure the foam tips were going to hold the mics correctly I then put on the head phones.. making sure the right cup was on the right ear...

I placed a chair in front of the A16 so when sitting in it the A16 display is at eye level... this is kinda important for this routine as you will be interfacing with that display for the final results...

OK... now that the setup is complete.. on with the routines...

As this was a new headphone to the A16 the first routine that had to be completed is the HPEQ... so I pulled up the A16 manual and went to the HPEQ routine starting on page 73, paragraph 9.2 ... thru paragraph 9.4
I think the only issue you might run into is the actual -in-ear mic level reading... and its not a big issue once you play with it a little... the routine starts out by performing short sweeps to test the -in-ear mics and to verify that the level is correct for them to record... all you have to do is set the gain switch (just right of the volume nob) to the lowest level... run the routine (hit the HP button) and watch the display and if you see the two bars going green then you are good at that gain setting... if the bars are yellow you are too low and you just click the gain switch up a notch until the sweeps go green... pretty easy as you just change the switch and hit the HP button and you can run that all you want until you get them to go green during the initial sweep... once I got the bar to go green, mine ended up at the center gain, I took a large breath and hit the HP button and held my breath until the routine had completed and then hit the Enter button to save it all... a note on the hold my breath thing... its probably silly but I did notice quite a bit of noise from my surroundings and in the microphonics of the cables and breathing... so I decided to stay real still and to hold my breath ... didn't take any measurements with the phone ringing in the back ground and stuff like that... :) sounds silly but at least if the routine didn't go well I couldn't blame it on external noises... with the HPEQ out of the way I go straight for the manLoud routine...

A couple of notes on both of the routines... the HPEQ is ran to characterize the headphones and it just happens that your head is a great place to hold the in-ear mics and the headphones at the same time.. the HPEQ routine creates two filters.. Auto and Flat.... so after the HPEQ routine is ran, one of the variables is taken care of... the other variable ? your ears, and how they actually hear... so that needs to be measured and taken into account.... and thus.. the manLoud routine... this routine is used to measure and take into account how good or how bad your ears can hear.... and you get to compensate for that... the manLoud routine will allow you to adjust for any level variations your ears have over a set frequency range... and if your ears aren't perfectly matched and they hear a little differently then you get to fix the balance between the two ears also...

For this routine I loaded the youtube video and watched it first to get a guest of what I needed to accomplish... its very straight forward and worked like a charm... you should make time for this routine as it does take a while to complete....
The setup is pretty much the same except you won't be using the in-ear mics this time... I just followed the routine as they played it out and that works fine.... when I first started listening to the noise bands, I started at the bottom end and working up the frequencies... it seemed a bit harder once you get to the upper bands... so I found out if I started listening to the lowest band (the reference band) and used the left button on the remote it would jump to the highest band.... so I rocked back and forth from the lowest band to the highest band and adjusted the levels between the two... easy as you just hit the left button then the right button on the remote... once I had that complete I now had a low reference and a high reference... so I started at the bottom end and started to work my way up the frequency bands... I also purposefully tried to not look at the display while adjusting the bands... not sure why that made a difference but it seemed to be easier that way... just listen with the ears and make the adjustments.... in my case while working my way up into the higher frequencies the balance would shift over from center to the left side... the reason is because the right ear does not hear the same as the left ear.... I ignored that for the moment and worked my way through all of the frequencies...once complete I now looked at the display and noted the higher frequencies volume were pushed up quite a bit... the video warns you of this happening and for you to try and not allow this to happen so I went back through the higher frequencies and made fine adjustments and this time I paid attention to the left/right balance and adjusted those out also... so what happens is the right ear gets a little more volume to balance out the ears... and with the fine tuning this time the overall volume came down for the higher frequencies and the balance issue was taken care of.... now I went back at least two times and fine tuned the overall volume between all frequencies... again, not watching the display.... once complete and pretty much happy with the results I saved the manLoad filter...

So now that the routines were completed it was time to hear some music and see if it worked or not.... so a super easy hookup, I took my iphone XS with a headphone adapter and plugged it into the A16 at the stereo jacks on the top rear and loaded the Onkyo HF Player app and played back some nice DSD files for the testing.... I can report the routines worked perfectly and those head phones have never sounded better... I also went into the apps EQ settings and fine tuned the A16 to the iphone for a great pairing.....

The only issue... it does take a bit of time to get this all done.... is it worth it.... I don't see how you can get away with not doing it.... :)

Sorry for the long... drawn out post... but I do hope it helps some...

I did the manLOUD HPEQ for both of the preloaded PRIRs and it made a world of difference. I don’t have access to any Atmos rooms to capture, but after doing the calibration the built in ones sound really convincing. Convincing enough that I have no problem using them for Atmos content until I can measure a real room myself.
 
Oct 23, 2019 at 12:04 PM Post #7,007 of 15,951
I created 9.1.6 room by building rig for the ceiling channels with camera stand and steel beam and mounted loudspeakers at both ends.

First I measured 5.0 room with my current setup, then proceeded with additional channels until I had all loudspeaker positions for 9.0.6 measured, then I compiled one room from resulted 4 PRIRs containing different measurements. Since I don't own actual subwoofer I just picked one from supplied BBC -room. It was quite a job to do, but result is very good. Too bad my living room acoustics are not too good so its not very good for music, but its very convincing for movies, illusion is very good.

For music I measured 2.0 pair on my computer desk in a room with better acoustic. I hope to get possibility get measurements in a real dolby atmos setup with suitable rooms in the future.

But building this rig for ceiling channels was well worth the effort, I'll link picture here: http://sid.fi/~grue/IMG_2115.jpg
 
Oct 23, 2019 at 1:24 PM Post #7,008 of 15,951
I did the manLOUD HPEQ for both of the preloaded PRIRs and it made a world of difference. I don’t have access to any Atmos rooms to capture, but after doing the calibration the built in ones sound really convincing. Convincing enough that I have no problem using them for Atmos content until I can measure a real room myself.
Told you so...

The manLOUD procedure is really good and well implemented.
@Dixter: As I already wrote for manLOUD I would start with a flat HPEQ filter which can be generated via one button press. No need for the auto HPEQ. For manLOUD this could even be a worse starting point than a flat filter.


@Sander: This little Thomann DSP is also one of several I consider buying to make own measurements. It is already known in german home cinema circles, look here, including measurements (although in german): https://www.heimkinoverein.de/forum/6-hifi-a-surroundanlagen/17905-the-t-racks-dsp-4x4-mini

I had some doubts though because it's only 48 kHz sampling, but then again as an engineer I believe in Nyquist. But there are all these audiophiles who think even 96 kHz isn't enough and you start to doubt...
What I'm also concerned about a little are the multiple AD DA conversions, don't know if and how this deteriorates the signal.

For the sub you could just use one of the factory PRIRs as Grue pointed out here, no problem.
Or you could just record one of the fronts again as sub. And you ecen could eq it to have a lower frequency limit, as long as it makes no problems when the sweep is played. You just need the frequency response.
But using a sub from a factory PRIR would be better I think.
Or, I always forget: Use direct bass, that's theoretically the best bass you can get. It is totally free of coloration or reverb. But one has to get used to this.
 
Last edited:
Oct 23, 2019 at 1:38 PM Post #7,010 of 15,951
Another tip: When you separately record top(ceiling) loudspeakers as Grue did: Sit on the floor!
The top loudspeakers would then be higher above you in the resulting listening room.

@Grue: Normally is recommended that the tops should be angled to the listener.
 
Oct 23, 2019 at 2:44 PM Post #7,011 of 15,951
[QUOTE="audiohobbit,

The manLOUD procedure is really good and well implemented.
@Dixter: As I already wrote for manLOUD I would start with a flat HPEQ filter which can be generated via one button press. No need for the auto HPEQ. For manLOUD this could even be a worse starting point than a flat filter.[/QUOTE]

I went with the auto HPEQ because of the way I read the manual... the auto creates both auto filter and flat filter with the in-ear mics.... the non auto flat filter was written for iems as you can't use the in-ear mics with iems so they opted to call it flat filter...

from the manual:

6.4.2 Content
This describes the EQ information available in the HPEQ file. Currently there are four sets of data, one set (autoEQ) taken with the automatic HPEQ measurement procedure, one set (flatEQ) generated at the same time as the autoEQ, and two optional sets that are typically manual adjustments to the autoEQ or flatEQ data. The flatEQ filter was designed to be used as the base filter for IEM-type headphones.

autoEQ: All HPEQ files contain autoEQ data, measured using the automated EQ procedure.
flatEQ: This EQ filter is flat and is typically used as the base filter for IEM-type headphones.
manLOUD: Manual EQ adjustments determined using the Equal Loudness EQ measurement procedure.
manSPKR: Manual EQ adjustments determined using the External Speaker EQ measurement procedure.
 
Oct 23, 2019 at 3:27 PM Post #7,012 of 15,951
@audiohobbit:
I need the sub signal in the fronts when using the A16 as a pre-pro with the real speakers. I am aware audiophiles and high enders will be thoroughly unimpressed with my set-up, but you should understand my objectives:
  • Have some fun without breaking the bank, because it is possible!
  • Be able to give simple and quick demo's to visitors (for fun). Be able to do a quick 9.[0/1].6 measurement and directly A B compare the complete real with the virtual system, with the psychoacoustic support of the real system being present in the room (also for myself during normal use of the A16 with headphones). Even if the system is "mid-fi" I can still demonstrate how well the A16 can mimic it.
  • Some of the imperfections of the real system won't be captured so the virtual system might already sound a little better than the real system.
  • See how much improvement of the virtual system is possible using direct bass (or generic subwoofer), and reduced reverberation time.
  • Maybe use a PRIR from this system to personalise a high quality BRIR on the exchange website.
Besides all this I may get a top notch measurement done in a studio or something. And maybe after a while I am going to sell all those old speakers and amps again.
Even if there was/is an audible difference from the extra AD DA conversion I don't think it would influence the PRIR measurement.
(Maybe I will do a liitle comparison test: comparing a "high res" stereo signal direct with that signal looped through channels 1 and 2 of all 4 DSP 4x4 Mini's, and once again through channels 3 and 4 of all 4 DSP 4x4 Mini's so that it passes 8 AD and DA conversions in a row. If that doesn't give a clear obvious audible difference then there is no need to worry about 1 extra AD and DA conversion.)
 
Oct 23, 2019 at 3:51 PM Post #7,013 of 15,951
Ok, understood about the bass.

And for the rest you're saying I do agree. I'm also a fan of the pareto principle.

I'm already relatively happy with ghe manLOUDed Surrey PRIR. But if I find the time I will set up 3 speakers with a sub in a 60 degree triangle at a good spot in the room, together with a sub and treat the room with some foam etc. This is easier for a few speakers than for 8 or more. With one of the t.racks DSPs I could drive this and do full bass management so I could measure every channel as full range. And I will record those 3 speakers 4 times in 90 degree increments. So I end up with a speaker every 30 degrees and can select what I want to combine, correct 5.1, 7.1, 9.1.x. And I will do this at the correct distance (about 3 m) so I do not need to play with delays and you hear a correct surround circle with 6 m diameter and are not limited by the room size.
For the top speakers I'll make a separate setup with 3 speakers in one row at the ceiling and record this two times to get the x.x.6. As I said I can sit on the floor for those two PRIRs so they will be higher than normally possible.
And I think I'll do this with the synchronous mode because only there the HT guided procedure will work.

If you do your multi AD-DA experiment I would be interested in the result.
 
Oct 23, 2019 at 9:23 PM Post #7,015 of 15,951
What order number are they up to on the Kickstarter orders?
 
Oct 23, 2019 at 9:47 PM Post #7,017 of 15,951
Does the windows driver work for anyone? I get a Code 52 warning in the device manager, the driver apparently isn't properly signed.
The ASIO control panel for the A16 can be opened, but doesn't detect any device.

I have this same problem. You can restart Windows with driver signature enforcement disabled and see if it works.
 
Oct 24, 2019 at 1:15 AM Post #7,018 of 15,951
I tried USB 2 and 3 with two cables and two PCs, no success. What version of Windows do you use? I tested on Windows 10 Education 64 Bit version 1903 (Build 18362.418) and Windows 10 Pro 64 Bit version 1809 (Build 17763.805).

I have windows 10 professional, latest version (I cannot remember the build, one that was released on may?). I have quite recent and powerful setup with i9900k cpu and z390 chipset.
 
Oct 24, 2019 at 1:19 AM Post #7,019 of 15,951
Another tip: When you separately record top(ceiling) loudspeakers as Grue did: Sit on the floor!
The top loudspeakers would then be higher above you in the resulting listening room.

@Grue: Normally is recommended that the tops should be angled to the listener.

I forgot to mention earlier that I was indeed sitting on the floor while doing measurements and what comes into tilting loudspeakers, I'm quite happy with the fact I was able to measure them at all at the moment :wink:

But now I have quite nice atmos room measured to my own ears while waiting for proper measurements.
 
Oct 24, 2019 at 1:56 AM Post #7,020 of 15,951
5 days ago backer nr. 81 received his A16 (he said on the kickstarter comments page).

Oh dang, still a long ways to go for me then (in the 200's, lol)
 

Users who are viewing this thread

Back
Top