Trinnov AltitudeCI Review and 32 Channel Build Thread

[etc6849]

When you define a source (e.g. "HTPC") in the CI, you have to pick a single audio protocol/audio connection type for the inputs and a single type for the outputs.

I don't think you can have Dante inputs and AES inputs active at the same time. You can have AES and HDMI active at the same time, where HDMI will just pass video, while you use the AES inputs, but even then you can't use HDMI for audio in and AES for audio in at the same time.

My guess is the AES loopback is doing nothing if you defined your selected source to use Dante. I don't use Dante, but would consider looking at their virtual driver to see if it can do loopback function?

If that doesn't work, you can try splitting it up the test if you have an AES in/out card in your PC:
CI AES in and CI Dante out
CI Dante out and AES in

PS: regarding the 16-channel Tascam cable comment... For others installing the CI in your theater and planning to use AES, you want a cable like this for the AES DB25 output:
https://www.bhphotovideo.com/c/prod...6_00_DB25_XLRM_DigiSnake_Analog.html/overview

And one like this for the AES DB25 input:
https://www.bhphotovideo.com/c/product/769733-REG/Avid_9940_30427_00_DB25_XLRF_DigiSnake_Analog.html

These aren't the TASCAM standard as they don't have 8 channels in/8 channels out. Technically this is considered the "Doremi/DCI Digital Pinout" as Trinnov points out here: https://kb.trinnov.com/en/knowledge-base/db25-pinouts-for-trinnov-products

Just bringing this up in case anyone ever wants to use the AES inputs or outputs... I know for sure the DB25 AES cables I linked to work, as that's what I bought and use in my theater. I know many have called these TASCAM cables before, including myself

I'm attempting to test a newly-received Altitude CI (16 licensed channels) which is connected via Dante Virtual Soundcard (DVS) to a Macbook Air M5 running REW. I've mapped M5 DVS outputs 1-16 to CI inputs 1-16 and mapped M5 DVS inputs 1-16 to the CI outputs 1-16 using Dante Controller. A 16-channel Tascam digital cable loops the CI AES output 1-16 DB25 into the CI AES input 1-16 DB25. The CI source is set to Dante.

When REW generates a 1 kHz tone on channels 1&2, corresponding signals are visible on the CI input and output meters. But, the REW spectrum analyzer detects no input signal from the CI. Web searches confirm that the 16 AES inputs on the CI are supported.

Any guidance or suggestions would be appreciated.

 

[darth]

Doesn't tascam/AES59 also allow 16 channels in 1 direction? According to this it does https://assets.ctfassets.net/4zjnzn...69cfcf4471/Genelec_9301B_Operating_Manual.pdf

"The AES59 standard offers a configuration to deliver 16 channels of digital audio in one direction."

Asked chatgpt to crate a pinout comparison, seems same I guess:

DB25 Pin | Genelec 9301B AES OUT | Trinnov Doremi/DCI AES3
---------|------------------------|-----------------------
1        | Ch 15/16 +             | Ch 15/16 +
2        | Ch 15/16 Shield/GND    | Ch 15/16 GND
3        | Ch 13/14 -             | Ch 13/14 -
4        | Ch 11/12 +             | Ch 11/12 +
5        | Ch 11/12 Shield/GND    | Ch 11/12 GND
6        | Ch 9/10 -              | Ch 9/10 -
7        | Ch 7/8 +               | Ch 7/8 +
8        | Ch 7/8 Shield/GND      | Ch 7/8 GND
9        | Ch 5/6 -               | Ch 5/6 -
10       | Ch 3/4 +               | Ch 3/4 +
11       | Ch 3/4 Shield/GND      | Ch 3/4 GND
12       | Ch 1/2 -               | Ch 1/2 -
13       | No Connection           | No Connection
14       | Ch 15/16 -             | Ch 15/16 -
15       | Ch 13/14 +             | Ch 13/14 +
16       | Ch 13/14 Shield/GND    | Ch 13/14 GND
17       | Ch 11/12 -             | Ch 11/12 -
18       | Ch 9/10 +              | Ch 9/10 +
19       | Ch 9/10 Shield/GND     | Ch 9/10 GND
20       | Ch 7/8 -               | Ch 7/8 -
21       | Ch 5/6 +               | Ch 5/6 +
22       | Ch 5/6 Shield/GND      | Ch 5/6 GND
23       | Ch 3/4 -               | Ch 3/4 -
24       | Ch 1/2 +               | Ch 1/2 +
25       | Ch 1/2 Shield/GND      | Ch 1/2 GND

The significant difference is mostly terminology: Genelec describes the conductors as AES/EBU pair +/− and shield, while Trinnov calls the third conductor ground. Electrically, the arrangement is the same. The Genelec manual's AES OUT pinout matches the Doremi/DCI mapping published by Trinnov.

EDIT: that being said, if I ever go trinnov route, will have to get a custom cable built for the input side with DB25 to 2x DB25 split, gah

                 ┌──────────────────────────┐
                 │      TRINNOV AES OUT     │
                 │   Doremi/DCI, 16 ch      │
                 └──────────┬───────────────┘
                            │
                 ┌──────────┴───────────┐
                 │                      │
              Ch 1–8                 Ch 9–16
                 │                      │
                 ▼                      ▼
       ┌─────────────────┐    ┌─────────────────┐
       │ GENELEC AES IN 1│    │ GENELEC AES IN 2│
       │ TASCAM digital  │    │ TASCAM digital  │
       └─────────────────┘    └─────────────────┘

 

[FreMo]

I'm attempting to test a newly-received Altitude CI (16 licensed channels) which is connected via Dante Virtual Soundcard (DVS) to a Macbook Air M5 running REW. I've mapped M5 DVS outputs 1-16 to CI inputs 1-16 and mapped M5 DVS inputs 1-16 to the CI outputs 1-16 using Dante Controller. A 16-channel Tascam digital cable loops the CI AES output 1-16 DB25 into the CI AES input 1-16 DB25. The CI source is set to Dante.

When REW generates a 1 kHz tone on channels 1&2, corresponding signals are visible on the CI input and output meters. But, the REW spectrum analyzer detects no input signal from the CI. Web searches confirm that the 16 AES inputs on the CI are supported.

Any guidance or suggestions would be appreciated.

What is the purpose of such a loop between REW and CI?

 

[etc6849]

Kinda, but you're missing part of what I said, and part of what I didn't talk about.

There is no digital TASCAM DB25 pinout for 16 channels in or for 16 channels out. There is a digital TASAM DB25 pinout for 8 channels in and 8 channels out.

There is an analog TASCAM pinout that allows for 16 channels digital in/out or 8 channels analog in/out, but if you buy a random analog TASCAM breakout cable, it may not have the right impedance (110 ohm) unless it is clearly stated it is for both analog and digital. We can argue about whether that matters, but you absolutely can get reflections, especially at higher bitrates. Not sure if Benchmark's application notes talk about this, but I vaguely remember one of John Siau's posts or podcasts where discussed this.

I haven't tested this personally, I assume people like John Siau have, but I do have an EE degree and it's not BS. Cables can absolutely matter. EE's take electromagnetic theory courses discussing the topic of transmission lines, and reflections absolutely occur at higher frequencies. Yes, the bitrate is at most 192kHz. However, the data rate is 12.288 Mb/s, and the cable must preserve timing and edges based on a 24.576 MHz half-bit interval.

But back to the main topic you asked about:
There is however a Doremi/DCI digital pinout standard that has a similar pinout to the analog TASCAM pinout. So if a cable follows this standard, it's impedance would be appropriate for the digital use case (unless the manufacturer doesn't design for 110 ohm impedance, which would be strange if they didn't). Pin compatibility does not establish impedance compatibility.

https://www.tascam.eu/en/docs/DB-25_Pinout.pdf

Doesn't tascam/AES59 also allow 16 channels in 1 direction? According to this it does https://assets.ctfassets.net/4zjnzn...69cfcf4471/Genelec_9301B_Operating_Manual.pdf

"The AES59 standard offers a configuration to deliver 16 channels of digital audio in one direction."

Asked chatgpt to crate a pinout comparison, seems same I guess:

DB25 Pin | Genelec 9301B AES OUT | Trinnov Doremi/DCI AES3
---------|------------------------|-----------------------
1        | Ch 15/16 +             | Ch 15/16 +
2        | Ch 15/16 Shield/GND    | Ch 15/16 GND
3        | Ch 13/14 -             | Ch 13/14 -
4        | Ch 11/12 +             | Ch 11/12 +
5        | Ch 11/12 Shield/GND    | Ch 11/12 GND
6        | Ch 9/10 -              | Ch 9/10 -
7        | Ch 7/8 +               | Ch 7/8 +
8        | Ch 7/8 Shield/GND      | Ch 7/8 GND
9        | Ch 5/6 -               | Ch 5/6 -
10       | Ch 3/4 +               | Ch 3/4 +
11       | Ch 3/4 Shield/GND      | Ch 3/4 GND
12       | Ch 1/2 -               | Ch 1/2 -
13       | No Connection           | No Connection
14       | Ch 15/16 -             | Ch 15/16 -
15       | Ch 13/14 +             | Ch 13/14 +
16       | Ch 13/14 Shield/GND    | Ch 13/14 GND
17       | Ch 11/12 -             | Ch 11/12 -
18       | Ch 9/10 +              | Ch 9/10 +
19       | Ch 9/10 Shield/GND     | Ch 9/10 GND
20       | Ch 7/8 -               | Ch 7/8 -
21       | Ch 5/6 +               | Ch 5/6 +
22       | Ch 5/6 Shield/GND      | Ch 5/6 GND
23       | Ch 3/4 -               | Ch 3/4 -
24       | Ch 1/2 +               | Ch 1/2 +
25       | Ch 1/2 Shield/GND      | Ch 1/2 GND

EDIT: that being said, if I ever go trinnov route, will have to get a custom cable built for the input side with DB25 to 2x DB25 split, gah

                 ┌──────────────────────────┐
                 │      TRINNOV AES OUT     │
                 │   Doremi/DCI, 16 ch      │
                 └──────────┬───────────────┘
                            │
                 ┌──────────┴───────────┐
                 │                      │
              Ch 1–8                 Ch 9–16
                 │                      │
                 ▼                      ▼
       ┌─────────────────┐    ┌─────────────────┐
       │ GENELEC AES IN 1│    │ GENELEC AES IN 2│
       │ TASCAM digital  │    │ TASCAM digital  │
       └─────────────────┘    └─────────────────┘

 

[welwynnick]

IIRC, it's AES 59 that has provision for both analogue and digital DB-25 connections.

I don't have a reference, but I think AES 59 also allows both input/output, all input, and all output, for digital connections.

TASCAM is the most widely adopted DB-25 standard. Yamaha is the other, which is used by quite a lot of manufacturers like Antelope.

An AES59 analogue interface is an 8 channel balanced analogue audio connection that is always either all inputs, or all outputs. I haven't found any exceptions.
This is used by everyone except Yamaha.

An AES59 digital interface is a multi-channel balanced digital audio connection that has essentially the same pin-outs, and can use the same cables.
However, it's almost always an I/O connection, with 4 AES3 inputs (8 audio channels) and 4 AES3 outputs (8 audio channels).
This standard is used by the following audio interfaces: AVID, Antelope, Audient, DAD, Digidesign, Focusrite, Grace, Merging, MOTU, RME, Tascam and Universal Audio

The Yamaha standard for DB25 digital audio connections is used by Apogee, Lynx and Yamaha.
Datasat, JBL, Trinnov and Lyngdorf use different digital interfaces, typically with 8 AES3 input channels on one connector, and 8 AES3 output channels on another connector.
Genelec use something different again, using AES59 / Tascam connections for the inputs, and outputs only for the output. Storm Audio are also completely different.
Useful post by Darth:

Possible solution for DTS/Dolby/Atmos+eARC+HDCP to AES/EBU - via Dante? (for Okto DAC8PRO etc)

8 aes channels carry 16 channels of lpcm. After a closer look I see it’s input and output on the Card-AES-96. 8 channels each. The H16-AES is output only. So not sure why they needed 2 DB25 connectors.

www.audiosciencereview.com

 

[etc6849]

Yep, my bet is AES59 may very well cover all the use cases we are discussing, no way to be sure though as I'm not going to pay money to read it. And fair point, I didn't say it directly, as I didn't think I needed to, but don't buy a Yamaha breakout cable for a Trinnov

But no, there aren't just two types of digital DB25 pinouts... It's not that simple, so hopefully AES 59-2012 fixes that.

IIRC, it's AES 59 that has provision for both analogue and digital DB-25 connections.

I don't have a reference, but I think AES 59 also allows both input/output, all input, and all output, for digital connections.

TASCAM is the most widely adopted DB-25 standard. Yamaha is the other, which is used by quite a lot of manufacturers like Antelope.

 

[Kal Rubinson]

Kinda, but you're missing part of what I said, and part of what I didn't talk about.

There is no digital TASCAM DB25 pinout for 16 channels in or for 16 channels out. There is a digital TASAM DB25 pinout for 8 channels in and 8 channels out.

There is an analog TASCAM pinout that allows for 16 channels out or in, but if you buy a random analog TASCAM breakout cable, it may not have the right impedance (110 ohm) unless it is clearly stated it is for both analog and digital. We can argue about whether that matters, but you absolutely can get reflections, especially at higher bitrates. Not sure if Benchmark's application notes talk about this, but I vaguely remember one of John Siau's posts or podcasts where discussed this.

I haven't tested this personally, I assume people like John Siau have, but I do have an EE degree and it's not BS. Cables can absolutely matter. EE's take electromagnetic theory courses discussing the topic of transmission lines, and reflections absolutely occur at higher frequencies. Yes, the bitrate is at most 192kHz. However, the data rate is 12.288 Mb/s, and the cable must preserve timing and edges based on a 24.576 MHz half-bit interval.

But back to the main topic you asked about:
There is however a Doremi/DCI digital pinout standard that has a similar pinout to the analog TASCAM pinout. So if a cable follows this standard, it's impedance would be appropriate for the digital use case (unless the manufacturer doesn't design for 110 ohm impedance, which would be strange if they didn't). Pin compatibility does not establish impedance compatibility.

https://www.tascam.eu/en/docs/DB-25_Pinout.pdf

View attachment 547045

Your musings parallel mine. I simply moved an AES-59 Tascam (8 channel, in/out) cable from my Hapi II to the AES/EBU output of the CI for the AES/EBU output from JRiver. With the Hapi, the two channels of output for Ls/Rs were routed (in Aneman) to channels 7/8. With the CI, the same two channels were specified as 15/16.
(My other channels, at present, run through the CI's balanced analog outputs via a standard analog/Tascam DB25 snake.)

 

[jhenderson0107]

What is the purpose of such a loop between REW and CI?

I just intended to augment the test results provided by the OP. In my application, I'll drive Genelecs base layer speakers via AES and use analog outputs to subs and overheads. I thought I might embellish this thread by add CI analog output measurements. I am using Dante rather than HDMI since it's conveniently available on my bench.

 

[jhenderson0107]

...PS: regarding the 16-channel Tascam cable comment... For others installing the CI in your theater and planning to use AES, you want a cable like this for the AES DB25 output:
https://www.bhphotovideo.com/c/prod...6_00_DB25_XLRM_DigiSnake_Analog.html/overview

And one like this for the AES DB25 input:
https://www.bhphotovideo.com/c/product/769733-REG/Avid_9940_30427_00_DB25_XLRF_DigiSnake_Analog.html

Thank you for the comprehensive and thoughtful response. I did considered your cable suggestions, but since I intend to expose the base layer AES and subwoofer signals via XLR from the back of my rack, I'm instead using a Tascam BO-16DX/OUT 16CH D-Sub to XLRM Breakout Box. The CI connects to the Tascam via two, conventional DB25 snakes. The six overhead analog signals are routed back into the rack via XLR to an ATI 528 amp.

 

[etc6849]

Oh nice! Yep, you can mix and match output types (AES/Analog/Dante, etc...). You just can't mix Dante inputs and AES inputs (and probably some others I haven't tried).

Currently, I have all 8 of my subs running off a pro-audio DSP that time aligns and level matches them. The CI just sends a single sub channel to them... Works great and sounds fantastic right now. However, I dream of an all Benchmark and Trinnov theater with no outboard DSPs.

Thank you for the comprehensive and thoughtful response. I did considered your cable suggestions, but since I intend to expose the base layer AES and subwoofer signals via XLR from the back of my rack, I'm instead using a Tascam BO-16DX/OUT 16CH D-Sub to XLRM Breakout Box. The CI connects to the Tascam via two, conventional DB25 snakes. The six overhead analog signals are routed back into the rack via XLR to an ATI 528 amp.

 

[jhenderson0107]

Attached is the measured performance of Altitude CI DAC channels 1&2 using it's internal sine generator at output amplitude +17 dB into a Cosmos ADC via Cosmos scaler (min gain).

 

[jhenderson0107]

Altitude CI digital channels are textbook-perfect when playing signals via Dante - identical to the HDMI test screenshots provided by the OP.

 

[etc6849]

Wow, nice test bench. The CI's analog outputs look impressive indeed.

Alitutde CI digital channels are textbook-perfect when playing signals via Dante - identical to the HDMI test screenshots provided by the OP.
View attachment 547531

 

[welwynnick]

Attached is the measured performance of Altitude CI ADC channels 1&2 using it's internal sine generator at output amplitude +17 dB into a Cosmos ADC via Cosmos scaler (min gain).

I think that's the best FFT ever measured for any processor.

 

[GXAlan]

Impressive! Looks like the Altitude CI is running the ARM based setup from the Nova?

Those Klipsch Palladium’s are impressive! How do they measure in-room before correction?

Have fun with the Trinnov.

 

[etc6849]

@GXAlan They are great speakers. They can measure perfectly flat if you just use Dirac and the passive crossovers. Going full active and crossing to 8 subs improves them even more. About +/-6dB measured from the MLP before any room correction. The room is well treated though. The acoustically measured THD at the MLP surpasses those of the world's best headphones too.

Some of the clearest lifelike sound I've ever heard, and the headroom is amazing. It's sad that Audiovox bought Klipsch out in 2011, then slowly discontinued them and moved a lot of stuff over seas. Sure Jubilee's are great speakers too, but no way can anyone fit five of them in a room or afford them.

I'm almost ready for phase 3 of the setup. Just need a few more DAC3 B's!

 

[welwynnick]

Impressive! Looks like the Altitude CI is running the ARM based setup from the Nova?

I was also hoping the AL CI might be using a similar architecture to the Nova, but it seems it's more complicated than that, and uses a hybrid architecture.
Yes, the AL CI does have an ARM processor, but it still uses a PC-type CPU for the core audio processing, leaving the ARM and the FPGA for other tasks.
10:07

 

[Frunse]

Cheep (günstig) ????? LOL 16 Channels cost 14000 to 16000 € my Car has cost so much LOOOOL

Waiting for Hyperion DPR-16 or Nuprime HX16-AES 5k are still not cheep for Endconsumer gear !

An Onkyo 5007 has cost 3000€ lomg time ago, but it has 9 Amps and DAC's in so why an pure digital Processor without DAC's or AMp Modules at all still this expensive ????

Could only be the License stuff that makes it expensive i bet!

If the Industrie like to sell more HDMaster or Hires Surround Media they realy need to think how this will happen if the Licenses still high and no digital Processors on Endconsumer Level aviable for an full aktive surround system.

That Surround Onkyo 5007 AVR was sounding so bad in comparence to an DSP full aktive Stereo System that i never used it, looool, and gave it away.

Makes no sense for hires Media if the normal CD sounds already so bad on them LOOOOL

 

[etc6849]

Yep. I don't blame you.

Trinnov's room correction is actually really good. I had Dirac Live and bought the PC license, but was never happy with it. The Trinnov lets you make custom excursion curves if want to lower correction above a certain frequency, etc... Even in my extremely well setup room, side imaging is better than without the CI. Atmos albums are honestly breathtaking because of it. Also no one else has a 3d mic that can tell where your speakers are in 3D space. It is needless to say very cool from a DSP standpoint.

It is expensive for sure, but shop around. No one pays MSRP for the CI (in the US anyway), and it is not unheard of to shave as much as $4-8k off MSRP, depending on number of channels, etc...

Who knows what agreements they have with Dolby? I'm sure there's a reason no one else has a certified software based Atmos decoder that runs under Linux. I have no idea of the expense it takes to develop such a thing, overhead costs for staff, certification and licensing costs, etc... and besides that there is really low production volume.

That said, I bet prices come down the second they have more competition. It's not by coincidence and Storm Audio ISP Elite with AES comes out to about what a CI would cost. Thankfully, this is a lot less than the Trinnov Altitude 32 was, so it is one of the cheaper options if you have DACs already and must have 32 channels.

That said, I've had several brands of processors such as Onkyo, Emotiva, and Marantz. None of them could compete with just using a pro audio DSP/speaker processor such as the Xilica XD4080 and my PC, let alone compete with a Trinnov CI. Maybe things are different now, but the difference was night and day when I just started using my PC as a decoder. The Trinnov AltitudeCI is essentially very similar to that.

Cheep (günstig) ????? LOL 16 Channels cost 14000 to 16000 € my Car has cost so much LOOOOL

Waiting for Hyperion DPR-16 or Nuprime HX16-AES 5k are still not cheep for Endconsumer gear !

An Onkyo 5007 has cost 3000€ lomg time ago, but it has 9 Amps and DAC's in so why an pure digital Processor without DAC's or AMp Modules at all still this expensive ????

Could only be the License stuff that makes it expensive i bet!

If the Industrie like to sell more HDMaster or Hires Surround Media they realy need to think how this will happen if the Licenses still high and no digital Processors on Endconsumer Level aviable for an full aktive surround system.

That Surround Onkyo 5007 AVR was sounding so bad in comparence to an DSP full aktive Stereo System that i never used it, looool, and gave it away.

Makes no sense for hires Media if the normal CD sounds already so bad on them LOOOOL

 

[darth]

Curious how well trinnov handles multi codec playlists. Very common to have dolby atmos music interleaved with regular music (dolby encoded vs pcm) so when track switches from one to another do you hear loud screeches, blips or something? Or is it just muted? And how long?

That's another thing I hate about H16-AES is that I feel it will blow either my genelecs or my ears at times. My dogs also don't like it.

Apple music test so to speak with continuous mode turned off if using Apple tv