Smyth Research Realiser A16
Dec 21, 2019 at 3:47 AM Post #7,606 of 15,986
Saturday's project: I'm driving down to San Diego to meet with Darin Fong (master of high-end 2-channel stereo home audio locations for PRIR measurement in the San Diego area), to capture a 2-channel stereo PRIR inside my car!
......

An interesting project indeed: Moving your car into your living room or bedroom...acoustically :) However, I am not sure if you will be able to really achieve that by just measuring 8 PRIRs which means you are essentially simulating only 4 speakers while your HK system likely has (much?) more (e.g. HK system in BMW 3 series has 16 speakers with 8 channels). You may have to measure PRIRs of all individual speakers rather than channels although it may not be feasible without touching the wiring....
 
Last edited:
Dec 21, 2019 at 6:05 AM Post #7,607 of 15,986
Well i have a problem atm..: done a PRIR, run the test which was ok with the different localization of the virtual spkrs etc..then started the Manspk EQ on my initial AutoEQ with my ears to try to get the best from my capture..
Has everyone experienced it before..? my hp are the HD800 and i noticed these details regarding the different HPEq done on my PRIR :
the general sound signature start very brightly with a FlatEQ ..bit less with AutoEQ and still persist with a properly accurate ManSpk.For instance, the high frequencies made from the 'sss' of the dialogue appears to the point of making sibilant noise in my Hp even with the Manspk EQ.. i am surprised bc the PRIR has been done with a pair of speakers which are not bright at all but more soft on the high Freq in a decent acoustic damped room..and i was very happy of the accuracy of my manspkEQ too...

Normally:
Personal made PRIR + normal headphones (no in-ears) --> use AutoHPEQ

If that doesn't sound like the speakers+room that you measured I think that means that the measurement was not 100% successful. Maybe improper mic placement or mics moved at some moment?
Of course it is possible that your manSPKR corrects the "error" somewhat, but ideally I would try to measure again. Do you have another person available that can check the position of the microphones during the process?
 
Dec 21, 2019 at 7:38 AM Post #7,608 of 15,986
As I already said about manLOUD, and I think it's the same for manSPK: There will be no new HPEQ file generated. The manLOUD or manSPK filter will be added to the original HPEQ file you selected as a starting point. You just have to select the correct filter (flat, auto, manLOUD, manSPK) when loading the HPEQ file into a preset.
And there's only one manLOUD or manSPK filter that can be saved in one HPEQ file.
If you take this HPEQ file again as a starting point for manLOUD or manSPK, the old filter gets overwritten.
So if you want to create a 2nd manLOUD or manSPK HPEQ, you first have to create a new flat or auto HPEQ file in which the new manLOUD or manSPK filter wil be saved.

Files in the recycle buffer won't be saved automatically. If the recycle buffer is full, the first file that has been stored there (now on the last position) will get deleted.
After creating a PRIR or a flat or auto HPEQ you should always immediately copy this to the internal memory, and also SD card to have a backup.

If you create several HPEQ files with the same subject and headphone names they will all get the same name consisting of subject and headphone name, so they are only discernible by the time and date they were created. So you have to write it down somewhere which one contains which manLOUD or manSPK filter. Or you can change subject and/or headphone names before creating the HPEQ that will act as a starting point for manLOUD or manSPK.
Same goes for PRIRs with subject and sound room naming.
What I also noted is that file ordering in the internal memory seems to be somewhat random. Not ordered by name or time and date.
 
Last edited:
Dec 21, 2019 at 7:44 AM Post #7,609 of 15,986
However, I am not sure if you will be able to really achieve that by just measuring 8 PRIRs which means you are essentially simulating only 4 speakers while your HK system likely has (much?) more (e.g. HK system in BMW 3 series has 16 speakers with 8 channels).
Quite sure this system only has a stereo aux in, so they could only do a stereo PRIR. So if it is so I don't see a way to create a surround listening room with that.
Have to wait for async signals, if the player of this car stereo plays multichannel PCM files, then could be recorded with A16 probably.

For me personnally, although I like car stereo, and I heard very good systems, this will never come close to a good stereo speaker setup in a good room with a wide sound stage.
But that's personal preference of course
 
Last edited:
Dec 21, 2019 at 11:39 AM Post #7,610 of 15,986
Quite sure this system only has a stereo aux in, so they could only do a stereo PRIR.
Correct. Multiple speakers, but operating in 2-channel stereo mode. And that's precisely what I'm hoping to duplicate, namely how it sounds in my car when I play stereo FLAC music from my Cowon J3 feeding AUX input to the car..

For me personally, although I like car stereo, and I heard very good systems, this will never come close to a good stereo speaker setup in a good room with a wide sound stage.
But that's personal preference of course
In the end, you may be right. And although the PRIR may be a faithful capture, hearing that sound in a different way may end up being disappointing. A novelty item (like the Egyptian Theater PRIR turned out to be).

Nevertheless, he persisted.
 
Dec 21, 2019 at 6:41 PM Post #7,611 of 15,986
As I already said about manLOUD, and I think it's the same for manSPK: There will be no new HPEQ file generated. The manLOUD or manSPK filter will be added to the original HPEQ file you selected as a starting point. You just have to select the correct filter (flat, auto, manLOUD, manSPK) when loading the HPEQ file into a preset.
And there's only one manLOUD or manSPK filter that can be saved in one HPEQ file.
If you take this HPEQ file again as a starting point for manLOUD or manSPK, the old filter gets overwritten.
So if you want to create a 2nd manLOUD or manSPK HPEQ, you first have to create a new flat or auto HPEQ file in which the new manLOUD or manSPK filter wil be saved.

Files in the recycle buffer won't be saved automatically. If the recycle buffer is full, the first file that has been stored there (now on the last position) will get deleted.
After creating a PRIR or a flat or auto HPEQ you should always immediately copy this to the internal memory, and also SD card to have a backup.

If you create several HPEQ files with the same subject and headphone names they will all get the same name consisting of subject and headphone name, so they are only discernible by the time and date they were created. So you have to write it down somewhere which one contains which manLOUD or manSPK filter. Or you can change subject and/or headphone names before creating the HPEQ that will act as a starting point for manLOUD or manSPK.
Same goes for PRIRs with subject and sound room naming.
What I also noted is that file ordering in the internal memory seems to be somewhat random. Not ordered by name or time and date.
Thank you again for clarifying the nuances between these HPEQ filters Christian, now it s clear for me.
i am actually wondering if the autoEQ is really effective at neutralizing the different HP sound signature...I mean that i noticed so far that me and Rene who use HD800 (neutral to bright) are always experiencing bright sound in the beginning with our virtual room from proper PRIR and HPEQ.. on the other side, from what i heard and read ,you had too much bass in your virtual room as a LCD (bassy HP) user?
 
Dec 21, 2019 at 9:55 PM Post #7,613 of 15,986
i am actually wondering if the autoEQ is really effective at neutralizing the different HP sound signature...I mean that i noticed so far that me and Rene who use HD800 (neutral to bright) are always experiencing bright sound in the beginning with our virtual room from proper PRIR and HPEQ.. on the other side, from what i heard and read ,you had too much bass in your virtual room as a LCD (bassy HP) user?
Maybe my conclusion was a bit too quick, I now remember a few things: Maybe the problem is that AutoHPEQ isn't working very precise in the low end (what was it again, 0-500 Hz is one band or something like that?). And/or that the treble peak of the HD 800 isn't precisely compensated.
 
Dec 21, 2019 at 11:26 PM Post #7,614 of 15,986
On a personal note I think if you use the HD800 with the A16 the HD800 really requires the ManLoud routine and Bass Management needs to be turned on and adjusted properly for the HD800 to deliver the entire frequency range... it responds well to that setup... and the A16 just really handles the lower end so well it makes the HD800 sound like a completely different headphone in a good way...
 
Dec 21, 2019 at 11:32 PM Post #7,615 of 15,986
Maybe my conclusion was a bit too quick, I now remember a few things: Maybe the problem is that AutoHPEQ isn't working very precise in the low end (what was it again, 0-500 Hz is one band or something like that?). And/or that the treble peak of the HD 800 isn't precisely compensated.
It is an interesting opinion..will love to hear from the others and also the Smyth regarding it bc they also use HD800 and HD800S as their main HP...
Maybe I should try the Superdupon mod on my HD800 to see if it can help or not?..
Bit disapointed atm bc spent a lot effort and time to get my first full PRIR done included a proper autoEQ with manspk filter etc....ended up having these annoying treble picks and sibilant for my first Atmos movie with the A16..
One thing that i am impressed so far is the precise localisation and soundstage of the virtual speakers including the thunderous LFE on my HD800...
 
Dec 21, 2019 at 11:35 PM Post #7,616 of 15,986
On a personal note I think if you use the HD800 with the A16 the HD800 really requires the ManLoud routine and Bass Management needs to be turned on and adjusted properly for the HD800 to deliver the entire frequency range... it responds well to that setup... and the A16 just really handles the lower end so well it makes the HD800 sound like a completely different headphone in a good way...
Correct me if i am wrong but we don t need to perform Manloud with our own PRIR and autoEQ HP ..just need the Manspk EQ bc we have our physical spk to compare with our HP..
 
Dec 22, 2019 at 1:13 AM Post #7,617 of 15,986
It is an interesting opinion..will love to hear from the others and also the Smyth regarding it bc they also use HD800 and HD800S as their main HP...
Maybe I should try the Superdupon mod on my HD800 to see if it can help or not?..
Bit disapointed atm bc spent a lot effort and time to get my first full PRIR done included a proper autoEQ with manspk filter etc....ended up having these annoying treble picks and sibilant for my first Atmos movie with the A16..
One thing that i am impressed so far is the precise localisation and soundstage of the virtual speakers including the thunderous LFE on my HD800...
I am using the Superdupont mod on my HD 800 and have been since the first week I bought it. The reason I use it is because the 6k peak you and everyone else notices is not just a peak, but also an oil can resonance excited when certain frequencies play. It's very narrow in its broadband, and, personally, I didn't mind it that much because it seemed relatively well damped and short in duration, but it is unmistakable there and impossible to ignore when excited. Eq will ameliorate but not eliminate it. And the SD mod? Yes it does address it and very effectively. After you use it the only brightness you will experience is the actual voicing of the phone which I believe is also not correct, but you will be able to fix that with the eq. What you will have left is one highly transparent transducer, which will be a near perfect instrument to impersonate the variety of speakers you capture with your PRIRs.
 
Dec 22, 2019 at 2:26 AM Post #7,618 of 15,986
Is there a place where people are posting their PRIR for download. I found sander's and Rene's (by the way thank you) and tried manLOUD routines with both of them with varied results. Wondered if I could try with someone else's

Also I'm having a weird issue when performing the manLOUD routine and curious if its the same with everyone else. When adjusting the frequencies from 1 - 32 then reaching the end, and then cycling through which takes you back to 1, the audible volumes are not retained and become very quiet. The thing is the set graph doesn't change, only the volume. Therefore it doesn't allow me to go through all 32 frequencies again to double check, because doing so messes up the previously adjusted numbers while attempting to bring the volume back up. This means that I can only go through the process on one go and have to save it after that. The reason I figure this is an issue is because in the youtube instructional video, they are shown re-cycling through the frequencies to make additional changes.
 
Dec 22, 2019 at 3:28 AM Post #7,619 of 15,986
Is there a place where people are posting their PRIR for download. I found sander's and Rene's (by the way thank you) and tried manLOUD routines with both of them with varied results. Wondered if I could try with someone else's

Also I'm having a weird issue when performing the manLOUD routine and curious if its the same with everyone else. When adjusting the frequencies from 1 - 32 then reaching the end, and then cycling through which takes you back to 1, the audible volumes are not retained and become very quiet. The thing is the set graph doesn't change, only the volume. Therefore it doesn't allow me to go through all 32 frequencies again to double check, because doing so messes up the previously adjusted numbers while attempting to bring the volume back up. This means that I can only go through the process on one go and have to save it after that. The reason I figure this is an issue is because in the youtube instructional video, they are shown re-cycling through the frequencies to make additional changes.
I confirm your observation....totally and deeply know how painful it is to try to complete the ManLoud...real torture bc we still try find a general method to potentially make it work..but atm there is no such thing yet so...we all need to rely on our ears and brain..which are unique to everyone...
 
Dec 22, 2019 at 5:18 AM Post #7,620 of 15,986
Does A16 allow multiple virtual speakers to be connected to the same channel?
Not specifically. But when the LEFT channel input is fed to the car's audio system however that sounds (via all the speakers of the car) to the microphones in my ears is what gets measured as the LEFT channel in the resulting PRIR. Same with feeding RIGHT channel input, as played through all the speakers of the car, ending up being the RIGHT channel in the resulting PRIR. A 2.0 PRIR only has two channels, so that's it.

The total set of speakers in a listening environment constitute what gets measured. If the room only has front L/R speakers and that's all that gets energized when LEFT channel input is fed to the electronics, then it's only the front L speaker which is reflected in the PRIR, etc. But if multiple speakers are energized together when the LEFT channel input is fed to the electronics (e.g. 5-channel STEREO or 7-channel STEREO in Yamaha AVR's, which feed the single input channel source to all of the "left front/side/rear speakers", etc.), then that's what the PRIR reflects.

So in the car whatever multiple speakers are energized for LEFT channel input, that's the LEFT channel of the PRIR. Same for right.


Now... to report the results of today's goal of producing a PRIR from my car, I sadly report that the project was 100% unsuccessful. Not because the concept or objective was flawed, but because for some reason there was a major buzz induced when connecting the L/R outputs of the A8 to the AUX input of the car's audio system. This wasn't a 60-Hz ground loop hum, but a buzz.

We were in the parking garage of a building, and using a 3-prong power outlet on the wall. I had brought a "lifter" with me (that eliminates the ground prong) so we tried that, although the buzz was not really a ground loop hum sound. As expected it had no effect in eliminating the buzz. We also tried using jumper cables to ground the car, connecting the car to some pipes that were hanging on a nearby wall, but again there was no effect and the buzz persisted.

We then hypothesized that maybe it was caused by connecting the 120v AC Realiser to the DC AUX input. Perhaps using the car's 12V DC system to power the Realiser might help, so off we ran to Fry's Electronics to buy an "inverter" (to generate 120v AC from the car's 12V cigarette lighter). In passing I comment that the once mighty Fry's (whose flagship store in San Jose has 32 cash registers) is now a "carcass" of a store (thanks to online shopping). On this Saturday before Christmas there were six cars parked in the huge parking lot. Inside the store there were five customers, and seemingly two employees, in what seemed like acres of store interior space. 90% of the pegboard pegs (to hold items for sale) were empty, and there was hardly any stock. Couldn't find any adapters, connectors, or extension cords for 3.5mm/RCA plugs which I wanted. Amazingly, they actually DID have a number of "inverters", so I bought one.

Unfortunately the inverter also accomplished nothing as far as eliminating the buzz. Of course just because an inverter is the source of the AC power, the Realiser is still an AC device and is still feeding a DC AUX input, if that is the cause of the buzz.

We did test the "solo" output of the A8 using the headphone outputs (both the phono plug on the front, as well as the RCA outputs on the rear), and there was no buzz. The buzz was definitely caused by connecting the L/R main RCA outputs on the rear to the 3.5mm AUX input of the car. And we tried both Darin's A8 as well as my A8, with both producing the buzz.

In theory, if we had enough cables and adapters we perhaps could have tried going from RCA to balanced, and then balanced back to RCA, as a way of perhaps eliminating the buzz. Unfortunately we didn't have the necessary converters to try this. But really we were at a loss to fully explain what was happening, other than to face the reality that unless the buzz could be eliminated there was no way to run a calibration.

Anybody have an idea what might have induced this buzz, and how to eliminate it?
 

Users who are viewing this thread

Back
Top