Datasat RS20i BETA test and comparison with ADA Rhapsody/Trinnov TEQ-12!!!! - Page 20

True, but in room corrections there is ample evidence that not only do they "skin the cat" differently, but the sonic outcomes are different. Although Dirac and Trinnov were not part of the Harman study on room correction tools, it did provide some insight into the differences and potential pitfalls of some tools (e.g. Audyssey performed poorly in that study, that does not mean it is always a poor tool, but rather it highlights that the venerable "cat skin" is not always the same when cut differently).

Pete:

Dirac Live uses both IIR and FIR filters. To my knowledge, all versions of Audyssey use FIR filters, only.

Jeff

The significant difference between the correction systems lies in (4) areas:

1) The computation engine- how and what is examined in the collected data to derive the correction filter. Trinnov utilizes our knowledge gained from our work with 3d acoustic fields in a multistep computation process. Ours are quite thorough. The computations for a 7.1 system in typical DSP (TI DA-710), take about 20 minutes to complete. In our PC based systems, a few minutes. Higher channel counts, a bit longer.

2) The resolution of the filter IIR, FIR. The resolutions vary dramatically. Trinnov uses FIR and IIR filters- with the resolution of the filters controlled by the user. As well, one can change the parameters at any time, as both the run time (listening) and computation systems reside in the unit. There is no need to connect external PCs to make any adjustments. This makes it easy for one to rapidly adjust during listening, changing EQs (takes seconds) or the basis for the actual calibration.

3) Spatial Correction, which is equally important to the playback experience to EQ compensation. Trinnov is the only system offering Spatial Correction, based on patented use of Fourier Bessel Transforms. This advancement in audio cannot be understated. Ask anyone who has had the pleasure of hearing it.

(4) Scalability. Most EQ systems are limited in channels. A few have the capability of active crossovers. Trinnov Systems are fully scalable from Stereo to 32 plus channels. Typical systems are usually 8-16 channels. Trinnov features active crossovers (up to 4 way, with automatic matching of drivers), dynamically adapts to any loudspeaker layout and subwoofer count (including subwoofer xovers). An example of this is a custom home theater completed in 2008 that had 17 speaker locations, including 7 fronts and 4 subwoofers with a crossover to an ultra-low frequency "fan" driver. Another example is a home theater with 11 bi-amplified speakers (22 Trinnov channels) with 4 subs. Combine with this the flexibility to interface digitally with some of processors on the market offering digital outs, like Theta and others, or combined in one unit- the ADA Reference, and you have top of the class performance. And it's all transparent to the installer and user. You do see what you get, to the extent you want to look under the hood.

Cheers

Curt, are the advantages described in
Points 1) 2) 3) 4) performed by
Ada teq 8/12 (matched with Ada Cinema Rhapsody Mach 4B)?

Reality is ...

The differences between Trinnov and Dirac Live are fewer than the similarities. Both are cutting edge technologies. Buyers need to listen to them both and decide how well the products incorporating the IP meet their needs and the size of their wallet.
____________
Best Regards,
Carl Huff

It would be interesting to hear Tom Holman or Chris Kyriakakis give their viewpoints on the subject. If anyone knows how to reach them, perhaps they could be pointed to this discussion.

Yes, absolutely, as the ADA TEQs are based on both Trinnov hardware and software. The difference is channel count. The TEQ has a maximum of 12 channels, the ADA Cinema Reference (Processor with built in TEQ w digital interface) has a current maximum of 16 channels, with future expansion capabilities.

ADA provides both RCA unbalanced and XLR balanced I/O. Because of space limitations in the Trinnov two rack unit hi multichannel chassis, Trinnov uses DB25 for all I/O conforming to the Tascam pro audio standard. BTW, DB25 works well - there are lots of off the shelf choices for cabling- google "tascam db25." DB25 for Trinnov can be configured for both balanced/unbalanced connections. (This format not to be confused with the difficult to source Dolby/Cinema DB25 discussed earlier in the thred.)

Only when considering Compensation may the various EQ units be compared.

Trinnov is unique as the sole provider of Spatial Correction, which is the real game changer here. Like EQ correction, Trinnov's Spatial Correction is based on your room specific layout and it has a dramatic impact on your listening experience. You'll find other AVS threads where users of Trinnov Spatial Correction describe their experiences.

Cheers,

I'm sure there are differences in hardware...Audyssey, after all, has all but exited the hardware side of things when they stopped offering their outboard boxes (balanced and unbalanced)...but I'm not so sure they would agree with some of the above assertions. Here are some quotes I found which suggest the IIR filters are not without issues...

"parametric equalization methods use a particular type of digital filter called Infinite Impulse Response (IIR) that only attempts to correct the magnitude response in the frequency domain. These filters can cause unwanted effects, such as ringing or smearing, in the time domain particularly as the bands get narrower. MultEQ uses Finite Impulse Response (FIR) filters for equalization that use several hundred coefficients to achieve much higher resolution in the frequency domain than parametric bands. Furthermore, by their nature, FIR filters simultaneously provide correction in the frequency and time domains. FIR filters had been considered to require too many computational resources. But Audyssey solved this problem by using a special frequency scale that allocates more power to the lower frequencies where it is needed the most."

And here is an explanation which suggests that Audyssey does, in fact, also entail spatial correction...

"there is no way to control the sound after it leaves the speaker. However, we can measure the effects of the reflections that arrive after the direct sound by looking at the time domain response. It has a certain pattern to it that will depend on the time of arrival of those reflections. Based on that pattern and the similarity of patterns across multiple measured locations we can identify the problems caused by reflections. Then, a filter is created to invert those problems as best as possible.

The key is to not think of the filtering in the time domain. It's not like a graphic equalizer that can only raise or lower the amplitude at certain frequencies. Our filters are in the form of impulse responses that operate on the audio signal through an operation called convolution. As such, they are affecting the signal in each channel in the time domain--hence the name. An additional benefit of such filters (also called FIR) is that they operate in the time-frequency domain as well. That means they can be used effectively to lower the ring down time of room modes in the low frequency range."

Curt, I'm throwing around the idea of a TEQ.

I'm not convinced a pure digital path, like is seen in the uber$$$ ADA reference piece, would make any real difference in the typical HT sound track. E.g. The extra A/D and D/A that comes with addding a TEQ to your SSP is significant.

Do you agree or what am I [sonically] giving up here adding a TEQ to my CR Mach IV vs the ADA reference?

Thanks!

Trinnov (i.e. Curt) reference to spatial correction is quite different than Audyssey's.

Trinnov does both: Movement of the perceived soundfield by manipulating the decorrelated information between speaker pairs and "correction" of both early and late reflections. I don't know how adjustable Audyssey is on the latter, but Trinnov permits user adjustment of both early and late reflection correction. For early, based on # of cycles. For late, based upon smoothing by octave.

Are you sure about that?

Curt has said that Trinnov remapping creates phantom images where they're supposed to be, and you don't create them with decorrelated information.

Datasat have updated their website and are now listing the RS20i as their consumer product.

My error, Noah. I had meant to say correlated, but on second thought, they're probably panning all of the information. Should've just asked Curt when I saw him yesterday......

The Datasat RS20i begins shipping to customers this week.
_______________
Best Regards,
Carl Huff

WHat model TEQ do you have?

How long did it take to calibrate?

Also, did you have a similar device prior?

Thanks!

Jeff, I'm strictly a 2.x ch guy with an ST-2 PRO. Curt's just down the road and did the cal in person in probably an hour, with several more hours of tweaking, explanation and just plain pleasant conversation. I imagine multi-channel would take longer, esp. if remapping is engaged. Yes, I used a TacT 2.2XP prior to the Trinnov.

Thanks for the info.

I PM'd Curt on his thoughts on the significance of an extra A/D & D/A step using the TEQ (vs the ADA reference). I imagine the real world differences are negligible but am awaiting his thoughts. He's a busy guy!

Thanks!

Jeff, that is great offer- have a listen for yourself. Regarding the extra conversion steps- with the compressed formats, we were giving up so much to begin with that an extra step of a/d/a really mattered. Of course, the extra a/d/a does matter, but less so with the lossless formats. We have several hi resolution installs with the a/d/a insert loss, and getting things right is critical. With d/d you don't have to concern yourself with gain structure (level matching), converters, and the like. If I have the choice of no correction or correction with a/d/a, I'll take correction every time.

Cheers,

There is truth in this, it's why the best compensation systems have FIR filters, but it's more complicated than it may appear on the surface. FIR filters are relatively easy to design, with IIR being more difficult to do very well. The best systems we see use a combination of both. Question- at what filter resolution??


Pete, thank you for bringing up the subject of spatial correction, because if is often misunderstood and this is an area where anyone who wants better sound could benefit from understanding a bit more about it. Turns out this quote of yours doesn't really address spatial resolution directly- it addresses time domain issues, which are different, but related, because cleaning up time domain issues does ultimately have an effect on spatial issues. There is something very important here, and RUR was trying to take a swat at it, by discussing correction of both early and late reflections. Thanks to you both. Given the importance of spatiality, I'll give it a post that stands on its own, rather then having the reader get lost in the quotes.

The following is a discussion regarding the benefits of loudspeaker correction with respect to spatiality- where it helps and where a different kind of correction is needed.

Spatial Detail

The highest spatial detail comes from a point source- like one loudspeaker. The amount of reproduced spatial resolution will be how big the image appears. Often we see speakers placed away from walls- that would cause early reflections and reduce spatial detail. Most spatial detail comes between pairs of loudspeakers, and again, many are placed away from walls to reduce reflections. Here we have the added problem of needing the speakers to be matched- to each other and the room in order to have utmost spatial detail. If the speakers don't sound the same at the listening position(s) (in both amplitude and time), then there is no hope to have robust spatial detail.

Enter loudspeaker correction. The best improves spatial detail between the speakers. To my way of thinking, this benefit is bigger than the target listening curve. I'd easily give up my favorite curve to have better spatial resolution, because it has every thing to do with seeing into the performance, revealing subtle details. Some recordings have layers of spatial detail that can easily be lost without the right balance of speaker, room, and correction. For stereo reproduction, this may be all you need, but for surround, there is more to the puzzle then speakers, room, and EQ correction.

Angular Resolution

So we have our spatial detail between two speakers. Now make it surround-with any number of speakers. You have a completely different spatial problem, even if we assume you have the spatial detail problem (outlined above) solved. That problem is what I'll call Spatial Angular Resolution. Will images appear and move where they should? Will you hear only the images, with the speakers seemingly disappearing? You could achieve this, for one dedicated playback format, only if you have exact placement of your speakers to the format and spatial detail correction outlined above. Then, as soon as you change formats (there are several - movie, ITU misic, etc), you'll need to change everything: move the speakers to positions required for the new format, re-calibrate for the new positions otherwise you will have lost your spatial angular resolution. Maybe you're thinking, Ah that won't be so bad, I can live with that!. You can, but here's what you are missing, and it's easy to tell: you'll be listening to loudspeakers, not images. Why?

Angular Resolution is very important to Spatial Resolution because you are listening in what we call an Acoustic Field. It's 3 dimensional. In the studio, just as there are layers of sounds created by the harmonic textures of instruments and voices, there are also often spatial layers of sounds created by more then one loudspeaker that form together in the acoustic field and appear to come from an intended direction. The loudspeakers are spatially coherent, supporting the 3d acoustic field. If everything is right, then you hear the images originate in the field from the intended directions. If it's not right (the loudspeaker/room placement is off), the waves from the loudspeakers will not be coherent and won't properly combine in the acoustic field. The result is that you just hear sounds from speakers and the imaging is lost (or incorrect).

Listening

What to listen for? Those who have heard Trinnov Spatial Correction (separate from room correction) understand the difference, but what if you haven't? Maybe here's one way to describe it, and those of you who have heard Trinnov Spatial Correction can respond. I'm going to use a rough example to get the point across. Here goes.

First consider a stereo recording you've heard that has one singer singing a duet with themselves, their voice recorded separately to each channel. The harmony is excellent, so listening on average stereo, it appears as one voice, but clouded somewhat and lacking detail. As you improve Spatial Detail through a better playback system, you now hear the two voices distinctly.

Second, let's consider a stereo recording with one voice. If the system lacks Spatial Detail, then the voice can appear cloudy, possibly swishy, and as though there are two voices- a duet- even though it is just a single voice. You are hearing both loudspeakers, rather then the single voice. This is an example of what to listen for: do you hear loudspeakers or images? In stereo, spatial detail comes from matching speakers and room. In surround, detail comes from this and speaker placement. If placement is not exact, one can use Trinnov spatial correction, called Spatial Remapping. The result is seamless imaging. This is at the heart of what Trinnov Spatial Correction provides as a benefit, and it should never be confused with room correction alone.

Cheers-

This sounds backwards to me. The higher the resolution of the source, the more able it is to be diminished by other aspects of the playback chain. A low rez system helps mask the effects of audio coding. That's one reason MP3s are acceptable to the masses, but not to people who listen on better gear -- the horrors are on full display.

It may be backwards ! Are we talking MP3(what rate?!), DD, DTS, it's all relative. How one deals with "the horrors," is a matter of personal reckoning. Forget digital. Just adding another analog devise to a hi rez analog chain has issues. I'll repeat what I believe is important: in my experience, inserting correction in the analog domain beats not having it. And yes, when dealing with compressed audio, remaining digital matters more to me perhaps because I'm trying to listen for as much resolution as possible. Maybe I should take your approach, as it might be less fatiguing in the end!

Cheers,

We're talking about whatever you meant by compressed formats.


That is not in question.


If by my approach you mean having the EQ in the same all-digital path as the source signals, then I'd agree it is the best way to go. Otherwise I do not compute what you mean by "your approach."

Curt.. Trinnov sounds awesome, and based on what I read from you and others (very cool that you are here by the way and so open and informative) it is the cream of the crop in room correction.. but in my opinion you guys HAVE to get it into more products. Obviously many in this forum section have the means to buy some pretty darn expensive items, but I think its missing the boat to have so few options for Trinnov. I doubt many here are going to purchase the receiver, and at $40k, that processor just doesnt seem like a good value. Its funny when I read what the Bland said about faroudja processing from a few years ago. I immediately thought about their $20k processor when I saw the ADA's price. I sold faroudja at the time, and it was an extremely short time after that that very similar tech was in dvd players. I think audio and video processing advances too fast to invest too heavy a sum. A few members in this section have expressed the same feelings to me. Now, the standalone options actually seem somewhat reasonably priced to me for what you get, especially since they are "first on the block". So I get that pricing, but it just seems that the kludgy setup, not to mention the A/D/A, will push quite a few people away. I actually wouldn't mind one A/D/A, but I will already be doing it with my Seaton Catalyst speakers, so I think a second one is not going to happen.

So... are there plans in the works to get Trinnov into $5k - $20k "complete" processors with an all digital interface? We all know a less tweakable version can be put in a cheap receiver, so what about versions that climb the ladder with features at some new price points?

Just my opinion, but I just think that three options; cheap, kludgy, and uber expensive isn't the way to get this awesome tech out there.

Just to compare, I use the Integra 80.2 for my pre pro now. Inexpensive at about 2k street, but nothing else seems worth the upgrade. I was pretty set to give Lexicon my money on the MP-20, but they blew it of course. So now I am looking again.

How about an Integra 90.1 with 12 channel Trinnov for $10k?

I'm referring to your post...


Cheers-

I have to admit I don't follow what you mean by Rogers's Approach? To put the earlier question a little differently, in a blind experiment are you confident you could consistently pick out the Rhapsody/ TEQ combination compared to the Reference on high resolution audio assuming equal calibrations and level matching?

I bought the Sherwood 972 just to play around with Trinnov and I've been extremely impressed. Now I want that sound in a more stable platform. I assume that the unit that I have doesn't have an internal a/d/a conversion and I wonder how much I'd lose in sound quality (surely I would gain additional flexibility) by going to the Rhapsody/TEQ combo.

Thanks

How easy is it to create height channels?

Hi Curt,

I have a fast track cinema overseas that needs to be able to be operational by September.

I want to implement a pseudo Neo x or PLIIZ frontal configuration 5 Quested Main speakers behind a cinemascope screen: Left wide, Left, Center and mirror image on right... plus above the Left and Right I want Height Channels.

I know doing the width channels is easy, how can you create height channels with the ada or the trinov pro piece?