• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Review Questions for Flagship Processors: Trinnov, Storm, JBL, Lyngdorf, Datasat

Dimifoot

Addicted to Fun and Learning
Forum Donor
Joined
Mar 27, 2019
Messages
506
Likes
746
Location
Greece
This is an excellent point about layering on processing to elevate the movie watching experience
It’s also very important in the imaging of multichannel music reproduction.
We have to stop thinking: 2 channels=music, multichannel=movies.

I have tens of thousands of multichannel music tracks.
 

Dimifoot

Addicted to Fun and Learning
Forum Donor
Joined
Mar 27, 2019
Messages
506
Likes
746
Location
Greece

Dimifoot

Addicted to Fun and Learning
Forum Donor
Joined
Mar 27, 2019
Messages
506
Likes
746
Location
Greece
The question is what objective measurements embody the virtues unique to AVR/processors that are completely missing from a simple DAC device? More importantly, what tools in Amir's arsenal can be purposed to uncover these data points?
I can only think of one way to test this:
We need a 5.x.4 setup at least, put the center channel speaker 15degrees lower than it should be, and calibrate the setup properly with the 3D microphone.

Then blind test the dialogue/singer position with the 3D remapping on and off. It’s just a switch on the menu. I think the dialogue is going to be the easier and the most obvious to test.
 

StuartC

Active Member
Forum Donor
Joined
Mar 8, 2021
Messages
154
Likes
398
Location
Rugby, United Kingdom
Correct, that's the target placement - just thinking it through, if I was to design such a measuring device, it would definitely require microphone triangulation much like what Trinnov is doing with its 3D microphone. Then you would play specific test tones via Atmos processing to have that tone (sound "object") move in a perfect hemisphere around the MLP, maybe 60 points around and equidistant such that any variation from this target distance is identified as an error - assuming all speakers were properly placed to begin with.

This is exactly the method I was thinking of. I fully expect that it exists, but is most likely an in-house tool used by the likes of Dolby, DTX, Sony and Auro. Creating it would definitely require intimate knowledge of the theory/software code that places 3D sound objects.

Dolby gets their coins by encoding the audio and selling people the decoder for home devices and the encoder to the content creators. Not so much by regulating room acoustics in everyone’s home.

Unfortunately, this is the nail on the head. Most of us cannot implement perfect speaker placement due to room size, WAF, room use etc etc. Its annoying that only the high-end processors allow you to program the speaker position accurately in 3D space rather than just with basic distances. Unless someone knows of a more moderately prices processor that does so?
 

wavesharp

Member
Joined
Mar 5, 2021
Messages
16
Likes
41
[QUOTE =“ Dimifoot,文章:711816,成员:7039”]尝试一些向下倾斜的方法[/ QUOTE]
this is target curve.
1616060779555.png

this is optimazed curve.
1616060895489.png


Because of my room reason, two standing waves could be showed in 30hz and 90hz.
My speakers are KEF Blade Two.
 

Andysu

Major Contributor
Joined
Dec 7, 2019
Messages
2,887
Likes
1,478
Send my that trinnov I do it though fairly to see if it is worthy or gets the cat litter box. But trinnov for front panel gets cat litter box, It has no direct push button for video mode or audio buttons to press so you can engage the sound format instantly not fumbling around with the remote control cos most of the other features are on the remote control. (sigh)

How fast does the video mode switch from one mode to the other? m/s or few seconds cos if it is few seconds it gets the cat litter box. If it was fast m/s then that would be okay.

Projection in the day was fun and audience has no idea on how much is done between end of trailers and start of the movie even when the film is on a cake platter. If it was A B projector (not had experience with that) cake platter makes it bit easier if all goes well. Fading the light down and muting cinema processor whi
[QUOTE =“ Dimifoot,文章:711816,成员:7039”]尝试一些向下倾斜的方法[/ QUOTE]
this is target curve.
View attachment 118883
this is optimazed curve.
View attachment 118884

Because of my room reason, two standing waves could be showed in 30hz and 90hz.
My speakers are KEF Blade Two.
You're speakers are they matched LCR or mismatched. Pictures of speakers.
 
OP
Spocko

Spocko

Major Contributor
Forum Donor
Joined
Sep 27, 2019
Messages
1,621
Likes
2,999
Location
Southern California
Why? I doubt that very strongly.
It's been mentioned by manufacturers previously, and I believe @amirm has acknowledged this problem; for every layer of processing, whether decoding, DSP, bass management, room calibration, that processing layer lowers overall signal levels by a fixed amount that cannot be changed such that when all processing is done by the various implementations required, you're way below the original signal before processing. It's unavoidable and the blame is not on the AV manufacturer but the individual software processing layer. Where the AV manufacturer has control is their physical circuit board design in terms of introducing noise or decreasing signal levels even further because bad design or amplifier implementation or noisy volume knob, etc.
 
OP
Spocko

Spocko

Major Contributor
Forum Donor
Joined
Sep 27, 2019
Messages
1,621
Likes
2,999
Location
Southern California
[QUOTE =“ Dimifoot,文章:711816,成员:7039”]尝试一些向下倾斜的方法[/ QUOTE]
this is target curve.
View attachment 118883
this is optimazed curve.
View attachment 118884

Because of my room reason, two standing waves could be showed in 30hz and 90hz.
My speakers are KEF Blade Two.
Wow, that wide-Q peak of 10dB at 40-50Hz where much of the movie's soundtrack occurs would be quite audible, boomy and annoying I'd imagine!
 
OP
Spocko

Spocko

Major Contributor
Forum Donor
Joined
Sep 27, 2019
Messages
1,621
Likes
2,999
Location
Southern California
I can only think of one way to test this:
We need a 5.x.4 setup at least, put the center channel speaker 15degrees lower than it should be, and calibrate the setup properly with the 3D microphone.

Then blind test the dialogue/singer position with the 3D remapping on and off. It’s just a switch on the menu. I think the dialogue is going to be the easier and the most obvious to test.
The possible issue here is choice of center speaker. A well designed center from Revel, Genelec or even Monoprice will have both excellent frequency response and controlled directivity which will greatly assist the DSP, but then maybe it sounds just as good with the DSP off! So to truly test the effectiveness of the correction software, maybe you have to get a speaker with known bad FR curve but excellent directivity like the Klipsch RP-600M. This allows the DSP to do its magic with the FR while the excellent directivity allows you to hear the improvements to object placement off-axis.
 

Dimifoot

Addicted to Fun and Learning
Forum Donor
Joined
Mar 27, 2019
Messages
506
Likes
746
Location
Greece
for every layer of processing, whether decoding, DSP, bass management, room calibration, that processing layer lowers overall signal levels by a fixed amount that cannot be changed such that when all processing is done by the various implementations required, you're way below the original signal before processing
I don’t know if this is the case with 3D remapping.
With DSP in general, you might lose some barely audible dynamic range dbs, but you gain massively audible improvements, like better bass response.
The possible issue here is choice of center speaker. A well designed center from Revel...
I think you are missing the point here.
3D remapping is not a matter of frequency response, but (virtual) location of the speaker.
Another example: Let’s suppose that where your Right Surround speaker should be located, there is a door in your room, and you have to move it further to the back, outside the recommended range.
With 3D remapping you will hear it where it should be. And this is independent from its Frequency Response.

I hope I have explained it better this time.
 
OP
Spocko

Spocko

Major Contributor
Forum Donor
Joined
Sep 27, 2019
Messages
1,621
Likes
2,999
Location
Southern California
I don’t know if this is the case with 3D remapping.
With DSP in general, you might lose some barely audible dynamic range dbs, but you gain massively audible improvements, like better bass response.

I think you are missing the point here.
3D remapping is not a matter of frequency response, but (virtual) location of the speaker.
Another example: Let’s suppose that where your Right Surround speaker should be located, there is a door in your room, and you have to move it further to the back, outside the recommended range.
With 3D remapping you will hear it where it should be. And this is independent from its Frequency Response.

I hope I have explained it better this time.
Got it, this does address the irrelevance of FR but wouldn't "bad speaker placement" be partly alleviated by speakers with good directivity as they still sound better off axis (aka bad placement) than speakers with terrible directivity. This will allow the virtual location of the speaker to be more easily corrected I would assume.
 

Dimifoot

Addicted to Fun and Learning
Forum Donor
Joined
Mar 27, 2019
Messages
506
Likes
746
Location
Greece
3D remapping doesn’t address how well the speaker sounds, but where it sounds from.
Of course you will always get better results with well designed speakers.
 

RichB

Major Contributor
Forum Donor
Joined
May 24, 2019
Messages
1,946
Likes
2,611
Location
Massachusetts
It's been mentioned by manufacturers previously, and I believe @amirm has acknowledged this problem; for every layer of processing, whether decoding, DSP, bass management, room calibration, that processing layer lowers overall signal levels by a fixed amount that cannot be changed such that when all processing is done by the various implementations required, you're way below the original signal before processing. It's unavoidable and the blame is not on the AV manufacturer but the individual software processing layer. Where the AV manufacturer has control is their physical circuit board design in terms of introducing noise or decreasing signal levels even further because bad design or amplifier implementation or noisy volume knob, etc.

There is also the potential of bass management for 12+ channels. I think the original DD for 5.1/7.1 systems has 10 dB attenuation for bass management.

I have seen as much as 18 dB attenuation applied to account for bass manage (extra channels) and Dirac (10 dB). I don't think there is a way around this except rolling the dice or attenuating to avoid clipping. It is another reason that AVPs benefit/need SINAD headroom.

- Rich
 

lemnoc

Member
Joined
Jun 9, 2020
Messages
70
Likes
97
I would repeat some measurements (graphs of: module & phase, intermodulation, harmonic distortion, bit resolution, crosstalk) without DSP, L+R channels, for different levels of output: 2 and 4 Vrms for XLR and 1 and 2 Vrms for RCA, as already stated by Kernelpanic.

And even if it is not an "official" measure, I usually also check the step response, obtained from the frequency response measured with a swipe from 10Hz to 96KHz (192KHz sample rate).
Inspecting the transient, you can have an evidence of how neutral the device is: how fast it is and if it "softens" the oscillations (and then the sound) of the band-limited input signal, overlaid for comparison.
Qualitative data useful when they are compared to other devices in the same situation.

I would be checking its 500Hz square wave response at various locations around the room after it's been optimized to see how good their room optimizer really is ? After all they are the ones making the claims ;)

Also Amir should be checking if this Trinnov thing is really audiophile quality or is it just a PC in a fancy case with some audio hardware tacked on to it. I can't imagine it being that good with a noisy PC power supply running it but we won't know until we see the test results.

cheers

img_20200321_012330-jpg.1294707
https://www.avforums.com/attachments/img_20200321_012330-jpg.1294707/
 

peng

Master Contributor
Forum Donor
Joined
May 12, 2019
Messages
5,615
Likes
5,168
I hear you but my analyzer can only test two channels at once. To test a third channel, then I need to re-run all the tests again with that one channel. Is that what you are asking? Things like jitter, IMD vs level, etc., etc.?

I would think just one or two (max.) more channels to be tested for IMD, THD+N vs level would be great. The extra work may not be warranted for budget AVPs and AVRs but for >$3,000 flag shop preamp processors, if I were the potential buyer I would like to know more about the "flagship" device before paying so much premium for one. Actually even for the budget priced AVPs/AVRs it would be nice to at least test one more channel, say the all important center channel.
 

RichB

Major Contributor
Forum Donor
Joined
May 24, 2019
Messages
1,946
Likes
2,611
Location
Massachusetts
I would think just one or two (max.) more channels to be tested for IMD, THD+N vs level would be great. The extra work may not be warranted for budget AVPs and AVRs but for >$3,000 flag shop preamp processors, if I were the potential buyer I would like to know more about the "flagship" device before paying so much premium for one. Actually even for the budget priced AVPs/AVRs it would be nice to at least test one more channel, say the all important center channel.

Earlier in the thread, @amirm mentioned that the AP tests two channels so the suggestion was to run an additional test with the center and on surround channel.

- Rich
 

dualazmak

Major Contributor
Forum Donor
Joined
Feb 29, 2020
Messages
2,823
Likes
2,951
Location
Ichihara City, Chiba Prefecture, Japan
...
Also Amir should be checking if this Trinnov thing is really audiophile quality or is it just a PC in a fancy case with some audio hardware tacked on to it. I can't imagine it being that good with a noisy PC power supply running it but we won't know until we see the test results.
...
I feel the same...

Just before starting my multichannel multi-amplifier project, I have seriously considered and web-reviewed Trinnov Altitude 32, and my understanding at that time was Trinnov would be an entry-level ATX PC (with a few cooling fans in it!) running Unix OS and digital software crossover (up to 192 kHz 24 bit) and multichannel DAC+preamp in one EMF noisy box.

At least for what I would like to achieve (stereo music audio, 5-way 10-channel crossover and multi-amplifier), I have concluded that I can achieve the same functionality of Trinnov with my completely silent (fan-less, spindle-less) Windows 10 PC and digital software crossover EKIO sending the crossovered channels into Okto DAC8PRO, and actually I could successfully completed such multichannel system as shared here.

And the sound quality of my system would be mainly depending on DAC8PRO (already so nicely measured and reviewed by amirm) and also on the amplifiers I am using, as well as on the speaker drivers...

Now, I feel again that Trinnov would be an entry level small computer with multichannel DAC+preamp together in a small EMF+fan noisy box.

Even though many people here are very much interested in the AVR functions and capabilities of Trinnov including the Atmos control in movie listening which greatly depending on the SPs and the room environment, I myself is rather interested in the fundamental SQ (sound quality) evaluations and the reliabilities of its software crossover including the delay and phase features.

I believe amirm can check the delay (and phase) control of Trinnov in digital level and line-out level just just like I did for software crossover EKIO as shared here.

I also would like to know the basic crossover reliability and reproducibility, i.e. simple Fq response curves, of Trinnov in digital level and in analog XLR output level just like I did and shared here.

I do hope if I could have chance to evaluate and compare Trinnov in my setup as the replacement of my PC+DAC8PRO, but the possibility of such occasion would be very small.

I really looking forward to amirm's review and report on Trinnov, therefore, which may give me some indirect comparison in Stereo Sound Quality between my multichannel multi-amplifier system and the same system possibly using Trinnov as replacement for PC+DAC8PRO.
 
Last edited:

peng

Master Contributor
Forum Donor
Joined
May 12, 2019
Messages
5,615
Likes
5,168
There is also the potential of bass management for 12+ channels. I think the original DD for 5.1/7.1 systems has 10 dB attenuation for bass management.

I have seen as much as 18 dB attenuation applied to account for bass manage (extra channels) and Dirac (10 dB). I don't think there is a way around this except rolling the dice or attenuating to avoid clipping. It is another reason that AVPs benefit/need SINAD headroom.

- Rich

Agreed, if it can do two at the same time, do the center and one surround, but not two surrounds.
 
Top Bottom