Trinnov Waveforming finally released

[hollis]

Looks amazing, such a shame the studios highpass most theatrical releases. I’m racking my brain trying to figure out how to add BEQ to WF or ART. The minidsp FlexHTx can take 8ch balanced in and let you add eq….

I’d be scared to combine BEQ with ART, boosting an antiphase signal could really damage some subs.

Safer to BEQ in Jriver, but, then one would lose Atmos.

 

[sigbergaudio]

The fact that you can push a bit in infrasonics keeping a relatively linear output speaks volumes on the quality of the subwoofer.

The reason they don't go even deeper is actually because of a high pass filter. The fs of the driver is 17hz. We're experimenting with the limiter to possibly remove the high pass now, or at least have it even lower in frequency. If we're able to pull that off, it will likely have some interesting results, as in significantly more output between 10 to 20hz.

 

[Fidji]

Looks amazing, such a shame the studios highpass most theatrical releases. I’m racking my brain trying to figure out how to add BEQ to WF or ART. The minidsp FlexHTx can take 8ch balanced in and let you add eq….

I’d be scared to combine BEQ with ART, boosting an antiphase signal could really damage some subs.

Safer to BEQ in Jriver, but, then one would lose Atmos.

with WF is actually not a problem - a) you can see the filters applied b) with Pressurization you get lot of headroom - in my room I can get into single digits with 8x18 in [my walls are from concrete] . Actually Pressurization Mode is unsung hero of the whole Waveforming concept - I was one of the beta adopters of WF, so had my WF installed before Pressurization was released and I can tell, it was a gamechanger - you get full woofage out of the system below first resonant frequency.

Re BEQ - I just have 2 presets with Low Shelf boosting sub 20Hz frequencies, and it covers like 90% of BEQ recommendations. Just making sure it is applied in controlled manner [that you need to make conscious effort to turn them on, important, as my wife is often using HT by herself]

 

[Keith_W]

I was reading MSO's "tech topics" series of articles and it had a bit to say about Trinnov Waveforming:

They vehemently claim that their Waveforming system is "not a DBA". To back up this dubious claim, they show a block diagram of their processing, illustrating that it's capable of controlling each subwoofer driver of the front and rear arrays independently.

For such systems, best performance is obtained by achieving the best possible plane wave, and ensuring its propagation is "straight front-to-back" without being redirected in any way. When identical drivers are used for the front and rear arrays, the best approximation of a plane wave is achieved by supplying the same signal to all drivers of a given array. With identical drivers, it's not possible to produce a "better plane wave" by providing different signals to each driver. One cannot compensate for the presence of unavoidable obstructions of the wave such as the seating and audience by providing unequal signals to the drivers either. Under certain conditions, such as non-identical drivers, unequal drive signals could be justified, but figuring out the different drive signals would require near-field measurements, not the far-field ones they specify. Even if they were able to do this properly, one might rightly wonder who the target users are that are willing to rebuild their rooms to accommodate the sub arrays, but not willing to use identical drivers within a given array.

Their approach seems to be using a DBA to minimize software complexity and cost, claiming that their system is not a DBA, then claiming that the performance improvement that's due to the DBA should be attributed it to their proprietary processing instead. It's a remarkably deceptive approach.

My feeling too. I don't see how it is any different to a DBA.

 

[SIY]

I was reading MSO's "tech topics" series of articles and it had a bit to say about Trinnov Waveforming:



My feeling too. I don't see how it is any different to a DBA.

It is generally similar, but not the same. The piece you linked caught one difference (independent control of woofers) but there's more. For example, the front woofers will absorb reflections and sources from the back wall, which is key to getting non-ideal rooms to work better.

I was impressed with DBA (thanks to @Blumlein 88 for some hints on getting it to work), but Waveforming was better. I just wish I could afford it. More to come in the article I'm writing.

When identical drivers are used for the front and rear arrays, the best approximation of a plane wave is achieved by supplying the same signal to all drivers of a given array.

That's not my understanding. Perhaps this could be expanded on. I thought that for an optimal plane wave, the rear would have the polarity inverted and a delay to simulate an infinitely long room.

 

[andyc56]

Perhaps this could be expanded on.

"Double Bass Array" = 2 arrays
Front wall = one array
Back wall = the other array
"A given array" = a given wall

 

[SIY]

"Double Bass Array" = 2 arrays
Front wall = one array
Back wall = the other array
"A given array" = a given wall

I get those definitions but not the assertion that the same signal should be applied to all of them. The reading and experimentation I've done indicate something different, as I mentioned.

edit: OK, I see what you were saying- you only meant the drivers within each array. Apologies for my misunderstanding.

 

[SIY]

"Double Bass Array" = 2 arrays
Front wall = one array
Back wall = the other array
"A given array" = a given wall

Thinking about this a bit (and recalling some measurements on the Waveforming function), is that still true for non-ideal rooms rather than symmetrical rectangles?

 

[Keith_W]

It is generally similar, but not the same. The piece you linked caught one difference (independent control of woofers) but there's more. For example, the front woofers will absorb reflections and sources from the back wall, which is key to getting non-ideal rooms to work better.

Why would the front array need to absorb reflections from the back wall, if the back array is there to absorb bass from the front array and prevent reflections?

If there are residual reflections that the back wall has not cancelled out, I don't think it would be too difficult to manually design a front array to cancel them. Essentially, it would be similar to setting up a VBA. If I was asked to create one, I would turn off the rear array and tune the front array like a VBA. The critical factors in setting up a VBA are delays and attenuation. The delays can be measured or calculated using a tape measure and the speed of sound, the attenuation has to be guessed and refined through iteration. Then turn off the VBA. Then I would turn on the rear array, set the delays and attenuation. Then turn both front and rear arrays on and do final tuning of front array attenuation to completely cancel out any residual reflection from the rear array.

I have set up a VBA before but I do not have enough subwoofers to form a plane wave, so my experiment has not been successful.

I was impressed with DBA (thanks to @Blumlein 88 for some hints on getting it to work), but Waveforming was better. I just wish I could afford it. More to come in the article I'm writing.

I look forward to your article!

 

[SIY]

Why would the front array need to absorb reflections from the back wall, if the back array is there to absorb bass from the front array and prevent reflections?

It can and does, but since it isn't covering the entire wall, there's inevitable reflection, albeit at a reduced level.

If there are residual reflections that the back wall has not cancelled out, I don't think it would be too difficult to manually design a front array to cancel them.

You could do that, but doing it in software is certainly less onerous.

I look forward to your article!

This has been fun to write. It will be less fun packing the Altitude back up. It will be sorely missed, but I'll do the best I can with a DBA and a lesser processor.

 

[Blumlein 88]

It can and does, but since it isn't covering the entire wall, there's inevitable reflection, albeit at a reduced level.

You could do that, but doing it in software is certainly less onerous.

This has been fun to write. It will be less fun packing the Altitude back up. It will be sorely missed, but I'll do the best I can with a DBA and a lesser processor.

Are you also going to do an article with info on doing DBA? There are other articles, but perhaps you example with lesser processor(s) would be helpful to people.

 

[sigbergaudio]

Why would the front array need to absorb reflections from the back wall, if the back array is there to absorb bass from the front array and prevent reflections?

If there are residual reflections that the back wall has not cancelled out, I don't think it would be too difficult to manually design a front array to cancel them. Essentially, it would be similar to setting up a VBA. If I was asked to create one, I would turn off the rear array and tune the front array like a VBA. The critical factors in setting up a VBA are delays and attenuation. The delays can be measured or calculated using a tape measure and the speed of sound, the attenuation has to be guessed and refined through iteration. Then turn off the VBA. Then I would turn on the rear array, set the delays and attenuation. Then turn both front and rear arrays on and do final tuning of front array attenuation to completely cancel out any residual reflection from the rear array.

I have set up a VBA before but I do not have enough subwoofers to form a plane wave, so my experiment has not been successful.



I look forward to your article!

I am not an expert on this, but my understanding is that with the help of more advanced processing, they get it to work even in less than optimal rooms, and with less than optimal placement of the subwoofers. And each individual subwoofer actively work to cancel out reflections.

 

[SIY]

Are you also going to do an article with info on doing DBA? There are other articles, but perhaps you example with lesser processor(s) would be helpful to people.

There was already a really good article in AX about that a year or two ago. If time permits, I'll try to compare them, but I suspect that time will not permit.

 

[hollis]

Re BEQ - I just have 2 presets with Low Shelf boosting sub 20Hz frequencies, and it covers like 90% of BEQ recommendations. Just making sure it is applied in controlled manner [that you need to make conscious effort to turn them on, important, as my wife is often using HT by herself]

Thats great info, thank you.

And I agree, there could be some kind of "universal BEQ."
On my old Audicontrol AVR I found I would often leave BEQ on the 2x4hd set to the wrong movie.
I'd flip between presets: off, beq action movie, beq disney, and special beq just for movie x

 

[Fidji]

I was reading MSO's "tech topics" series of articles and it had a bit to say about Trinnov Waveforming:



My feeling too. I don't see how it is any different to a DBA.

Even if Andy deserves a statue for his tireless work on MSO and I am grateful to him for some years with good bass, I think to judge the tech based on presentation and use descriptions like "dubious" "deceptive approach" is bit unnecessary. I try to give benefit of the doubt, if company has record of not bulls...ing, where I think Trinnov is one of the better guys out there in that respect.

I will try to describe factual situation some months after full WF release:
1) It works [as also seen on lot of before/after measurements]
2) it indeed is based on steering each channel individually, as per Trinnov claim
3) it works also on non-DBA setups, with certain limitations
4) Trinnov is pretty transparent about performance levels with each of subs placement options - and yes you get best results with DBA placement

What would be really interesting - to get somebody to test e.g. ART vs WF vs MSO on same configuration of subwoofers. I know, there was at least one instance of ART vs WF comparison, but results are not shared - intentionally.

All I can say - in the same physical configuration DBA vs WF is no match - for both measurements and sound quality. At least in my not perfect room and placement.

 

[Keith_W]

All I can say - in the same physical configuration DBA vs WF is no match - for both measurements and sound quality. At least in my not perfect room and placement.

Please post more details. I would like to see how you physically configured your subwoofers (i.e. how much of a DBA is it?), and the measurements comparing DBA to WF. A diagram or photo showing physical placement of subwoofers would be very helpful.

I am not doubting you, I am genuinely curious.

 

[Fidji]

Please post more details. I would like to see how you physically configured your subwoofers (i.e. how much of a DBA is it?), and the measurements comparing DBA to WF. A diagram or photo showing physical placement of subwoofers would be very helpful.

I am not doubting you, I am genuinely curious.

Measurements posted on previous page. Horizontally placed in 1/4 & 3/4 of the wall, only real outlier are front lower subs, that need to accommodate the screen. There are 18in subs, just for the perception of the scale.

Photos:

Front:

Rear [console is REALLY flat]:

+ one of later iterations, mid bass slightly boosted by design for one specific preset.

 

[Keith_W]

Measurements posted on previous page. Horizontally placed in 1/4 & 3/4 of the wall, only real outlier are front lower subs, that need to accommodate the screen. There are 18in subs, just for the perception of the scale.

Thank you. I might have missed it, but what DSP did you use to create the DBA?

 

[Fidji]

Thank you. I might have missed it, but what DSP did you use to create the DBA?

Trinnov - it has active crossover capability - you can group up to 4 channels into one, and then adjust volume, delay and polarity (and PEQ, if necessary) for each channel.
I think there was some sub grouping used to get all the 8 subs on 1 channel

It was installed by one German installers company, that specializes on DBA installations, they also made custom subs for me - you can see the position of the drivers is assymetrical in the subs in order to fit the best 1/4 3/4 placement. And they tuned it in on site.

Liked the outcome - only "issue" was bit lack of energy below 18Hz, first world problems ;-).

I would have never believed it - but there is SIGNIFICANT difference between bass reaching to 20Hz or going down to single digits. Makes the stuff real.

 

[andyc56]

I try to give benefit of the doubt, if company has record of not bulls...ing, where I think Trinnov is one of the better guys out there in that respect.

This may seem hard to believe, but I was of the same view for a time. That changed after watching the video linked in this post. I think that video was well summed up in this post by NTK. Much of the information presented in the video qualifies as BSing in my book. That video is what prompted my write-up, which would never have happened without it.

What would be really interesting - to get somebody to test e.g. ART vs WF vs MSO on same configuration of subwoofers. I know, there was at least one instance of ART vs WF comparison, but results are not shared - intentionally.

Because of the physics involved, I have no doubt that a DBA or DBA-like configuration would outperform an MSO-based setup for almost any parameters except SPL capability for the same number of subs. They are oriented toward different goals: one for dedicated, purpose-built rooms and the other for applications where one has little or no choice of sub locations.

All I can say - in the same physical configuration DBA vs WF is no match - for both measurements and sound quality. At least in my not perfect room and placement.

A "DBA" is not a DBA if it doesn't meet the requirements for sub location that give partial cancellation of vertical and width modes. Much of the DBA discussion I've read is like the old joke that many old people like me will recognize: "Put two grooves in it and call it an ashtray."

That said, I think the use of DSP to compensate for non-ideal sub locations for "close-to-DBA" configurations (when possible) is definitely worthwhile.