Review and Measurements of Okto DAC8 8Ch DAC & Amp

[mdsimon2]

Regarding the Okto's AES/EBU inputs: I have no devices that natively output that format. But from what I have read I think I can convert BNC coax to AES/EBU using an impedance transformer like this:

Neutrik NADITBNC-M - AES/EBU Impedance Transformer

Buy Neutrik NADITBNC-M - AES/EBU Impedance Transformer featuring Inputs 75 Ohm BNC, Outputs 110 Ohm XLR, Matches Balanced to Coaxial Lines, Allows Long Cable Runs for Digital Audio, Simple to Use, Passive Unit, Cost-Effective Impedance Matching. Review Neutrik null

www.bhphotovideo.com

I'm looking to go:

Allo digione signature player > BNC > impedance transformer to AES/EBU > Okto 2 ch input > out USB (to another Rpi running Camilladsp) > back into Okto via USB > out to amps

I've also got at least one Toslink source. I have a cheap Toslink to coax adapter that outputs RCA terminated coax: is there any reason this wouldn't work with the impedance transformer as well?

You shouldn't have any issues with TOSLINK / SPDIF sources and using TOSLINK / coax adapters + transformers when only using 2 input channels on the Okto.

The biggest questions I see in your proposed setup are 1) why use the digione when you can also use the RPi running CamillaDSP directly as a streamer? and 2) how are you going to integrate your TOSLINK source with the Okto?

One thing to keep in mind is the Okto does not have an ASRC and if using it in AES / USB mode you will need to run CamillaDSP at the same sample rate as your source, this can get complicated if your source has variable sample rates.

Michael

 

[dualazmak]

My situation is different in that I use a Mac Mini Win-based PC running JRiver 29, connected to Okto DAC via pure USB connection. With multichannel DSD files (stored as .iso or .dsf from SACD rips), I am bitstreaming as native DSF DOP64, which is the only DSD format that I need seems possible to use with Mac OS and this DAC. No noise problems.

Using DoP also doubles the required transmission banwidth.

FWIW.

Even though I too keep all the DSD tracks (included the extracted/ripped DSF tracks from some SACDs) in intact stereo DSF form in my SSD audio library, I usually play them in 96 kHz or 88.2 kHz PCM into DSP(XO/EQ/delay) EKIO by JRiver MC's on-the-fly format conversion since EKIO cannot directly handle DSD(DSF). I believe the same with the extraordinary expensive TRINNOV Altitude 16/32; all the DSP can be done in up to 96 kHz or 192 kHz PCM.

I have other reasons and rationales for "playing all the tracks in 96 kHz/88.2 kHz PCM"; please refer to my post here for the details.
- Summary of rationales for "on-the-fly (real-time)" conversion of all music tracks (including 1 bit DSD tracks) into 88.2 kHz or 96 kHz PCM format for DSP (XO/EQ) processing: #532

 

[Kal Rubinson]

Even though I too keep all the DSD tracks (included the extracted/ripped DSF tracks from some SACDs) in intact stereo DSF form in my SSD audio library, I usually play them in 96 kHz or 88.2 kHz PCM into DSP(XO/EQ/delay) EKIO by JRiver MC's on-the-fly format conversion since EKIO cannot directly handle DSD(DSF). I believe the same with the extraordinary expensive TRINNOV Altitude 16/32; all the DSP can be done in up to 96 kHz or 192 kHz PCM.

I have other reasons and rationales for "playing all the tracks in 96 kHz/88.2 kHz PCM"; please refer to my post here for the details.
- Summary of rationales for "on-the-fly (real-time)" conversion of all music tracks (including 1 bit DSD tracks) into 88.2 kHz or 96 kHz PCM format for DSP (XO/EQ) processing: #532

Understandable. I, too, generally, convert all the multichannel DSD into PCM in order to apply Dirac to it at up to 24/192kHz.

 

[Labjr]

Understandable. I, too, generally, convert all the multichannel DSD into PCM in order to apply Dirac to it at up to 24/192kHz.

Do you notice a difference between DSD and the PCM conversion?

 

[Kal Rubinson]

Do you notice a difference between DSDand the PCM conversion?

Y'know, I did not bother to listen to the PCM straight. The difference w/ and w/o Dirac was/is significant.
Over the years and with different DSP and different hardware, the improvement provided by room correction has always been significant and welcome and inordinately greater than DSD vs. PCM.

 

[dualazmak]

Y'know, I did not bother to listen to the PCM straight. The difference w/ and w/o Dirac was/is significant.
Over the years and with different DSP and different hardware, the improvement provided by room correction has always been significant and welcome and inordinately greater than DSD vs. PCM.

I fully agree; I, too, have been experiencing the same in my multichannel multi-driver multi-amplifier fully active stereo project. In my setup, not only the DSP XO plus EQ but also the precision (0.1 msec level) time alignment for all the SP drivers greatly contributes the amazing improvement as I summarized here.
- Perfect (0.1 msec precision) time alignment of all the SP drivers greatly contributes to amazing disappearance of SPs, tightness and cleanliness of the sound, and superior 3D sound stage: #520
My latest system setup as of May 30 2022 can be found here.

 

[Deribas]

My situation is different in that I use a Mac Mini Win-based PC running JRiver 29, connected to Okto DAC via pure USB connection. With multichannel DSD files (stored as .iso or .dsf from SACD rips), I am bitstreaming as native DSF DOP64, which is the only DSD format that I need seems possible to use with Mac OS and this DAC. No noise problems.

Using DoP also doubles the required transmission banwidth.

FWIW.

Thank you, Kal. The implication that I am sensing is that if I used something that does NOT run MacOS, the noise problem might not arise. Since switching to a Windows computer seems inelegant in other ways, I will attempt to avoid bitstreaming DSD to Okto DAC and just use the 24/176 PCM conversion within JRiver for SACD-derived material. If that defeats the noise issue, I will add Dirac Live in the audio chain to reap the benefits of room correction which you so eloquently promoted in your columns and online.

 

[gandalfandula]

You shouldn't have any issues with TOSLINK / SPDIF sources and using TOSLINK / coax adapters + transformers when only using 2 input channels on the Okto.

The biggest questions I see in your proposed setup are 1) why use the digione when you can also use the RPi running CamillaDSP directly as a streamer? and 2) how are you going to integrate your TOSLINK source with the Okto?

One thing to keep in mind is the Okto does not have an ASRC and if using it in AES / USB mode you will need to run CamillaDSP at the same sample rate as your source, this can get complicated if your source has variable sample rates.

Michael

Ah, I had overlooked the lack of ASRC on the AES/EBU inputs. I'm still going through your CamillaDSP thread, but: if the sample rates are mismatched, does the dsp output catastrophic noise or static?

Some out loud thinking here:

I mostly use Roon which can resample in software, but for TOSLINK or other external sources this will be a bit of a problem on the AES/EBU side. I can see the benefit of something like the OpenDRC-DI that you have, or maybe going with the Flex which can send out 4 channels of DSP'd audio via USB. It would also provide input switching and the ability to try out DIRAC if desired, all good features. I like the idea going with one box for "everything" - input switching, ADC for phono, and DSP. One could even bypass the Okto's volume control and use the Flex for volume adjustment, which would represent maybe the most integrated off the shelf solution. Obviously it even provides DA conversion so the Okto would not be necessary strictly speaking!

My big issue with the Flex is that it's a 2 in/4 out interface while the Okto is 8 channel. Not the biggest problem now, but ideally I'd build around an 8 channel DSP so that I can expand the system to 4-ways eventually, or perhaps repurpose it to a 7.1 home theater setup in the future. To that effect, I could probably do TOSLINK/coax > USB to get everything into CamillaDSP in an Rpi, but then how do we do input switching at the DSP level? Maybe something could be hacked together there with an IR system? But I saw from your post here:

RPi4 + CamillaDSP Tutorial

Have you checked clocking options in alsamixer? Michael IMO there is none, I have the same soundcard. The adaptive output configuration does not allow syncing to SPDIF input rate.

audiosciencereview.com

That physical capture devices on CamillaDSP have high latency, so maybe this isn't very doable after all? re: a TV TOSLINK output. Maybe option 1 from your post would work then.

If we want the most flexible system, it looks like we would need to go:

Flex input switching coax/analog/TOSLINK > 2 ch USB out > CamillaDSP > 8 ch USB out > Okto > amps

So we are essentially downgrading the Flex to just an input switcher and ADC in this system, using the CamillaDSP Rpi for all processing.

 

[manisandher]

In my setup, not only the DSP XO plus EQ but also the precision (0.1 msec level) time alignment for all the SP drivers greatly contributes the amazing improvement as I summarized here.

I agree. Accurate time-alignment of all drivers makes a massive difference to the sound. It's like having a camera lens that's slightly out of focus vs. one that's well-focused.

Mani.

 

[mdsimon2]

Ah, I had overlooked the lack of ASRC on the AES/EBU inputs. I'm still going through your CamillaDSP thread, but: if the sample rates are mismatched, does the dsp output catastrophic noise or static?

At least on the Okto I think all it does is say "fs mismatch" and it will not output anything. It's been a while since I tested it out but I think that is what happened, I don't remember any catastrophic noise.

I mostly use Roon which can resample in software, but for TOSLINK or other external sources this will be a bit of a problem on the AES/EBU side. I can see the benefit of something like the OpenDRC-DI that you have, or maybe going with the Flex which can send out 4 channels of DSP'd audio via USB. It would also provide input switching and the ability to try out DIRAC if desired, all good features. I like the idea going with one box for "everything" - input switching, ADC for phono, and DSP. One could even bypass the Okto's volume control and use the Flex for volume adjustment, which would represent maybe the most integrated off the shelf solution. Obviously it even provides DA conversion so the Okto would not be necessary strictly speaking!

The Flex is definitely a nice front end for a CamillaDSP system. Of course you could go with the digital Flex and output directly in to the AES input of the Okto, advantage here would be you would avoid the need to enable rate adjust and resampling in CamillaDSP as the Okto would be clocked by the Flex. Of course using USB would also work (and you could do this with an analog output Flex if you wanted) you just need to enable rate adjust and resampling to avoid buffer over/under runs. The other point to consider is if you are using the Flex as a front end you could potentially get by with a cheaper pro audio DAC (like the MOTU Ultralite Mk5) that doesn't have IR volume control / display, although you said you were using compression drivers so the lower noise of the Okto is likely appreciated.

My big issue with the Flex is that it's a 2 in/4 out interface while the Okto is 8 channel. Not the biggest problem now, but ideally I'd build around an 8 channel DSP so that I can expand the system to 4-ways eventually, or perhaps repurpose it to a 7.1 home theater setup in the future. To that effect, I could probably do TOSLINK/coax > USB to get everything into CamillaDSP in an Rpi, but then how do we do input switching at the DSP level? Maybe something could be hacked together there with an IR system? But I saw from your post here:

In terms of TOSLINK to USB cards I really like these:

Hifime S2 Digi -Optical SPDIF input and output to USB

The Hifime S2 Digi is an optical in/out device. It has an optical SPDIF input and also an output which can both be used at the same time.

hifimediy.com

Hifime SPDIF Optical to USB converter, record DAT/minidisc to computer

The UR23 accepts SPDIF optical sources and transfer this to USB. Suitable to record DAT tapes, minidisc or other digital source to a computer.

hifimediy.com

They are cheap, handle signal interruption well and at least have the potential to detect sample rate changes. They will detect sample rate changes as long as the sample rate of the card is higher than the sample rate of the input. If it is lower CamillaDSP will end up in a stalled state. I've thought about writing a Python script to restart CamillaDSP at 192 kHz if it is in a stalled state and then check detected sample rate and reset the configuration to the correct sample rate. Still something with an ASRC is a lot easier to deal with.

What TOSLINK sources do you have? Are they variable sample rate?

That physical capture devices on CamillaDSP have high latency, so maybe this isn't very doable after all? re: a TV TOSLINK output. Maybe option 1 from your post would work then.

I don't have any latency issues with physical inputs on my CamillaDSP setups. My main setup is an AppleTV 4K -> HDMI extractor -> OpenDRC-DI -> Okto AES input -> Okto USB output -> RPi w/ CamillaDSP -> Okto USB input -> Okto analog output.

In my office I've used a variety of different sources / setups but currently use an older AppleTV 3 -> TOSLINK output -> MOTU Ultralite Mk5 -> MOTU USB output -> RPi w/ CamillaDSP -> MOTU USB input -> MOTU analog output.

Both work very well with no latency. I've also used other input cards like the one mentioned above without issue.

Michael

 

[gandalfandula]

I got the Okto today. From order date to delivery was a mere one week! It came with firmware 1.42.

Initial listening impressions over pure USB from my PC (HD650 out of the headphone jack) are very positive. It seems a bit crisper than my now ancient Benchmark DAC1, though I still love that unit. Excited to get it going with my speaker setup this week.

I'm still somewhat amazed that they can deliver this level of build quality, features, and of course sound quality from a device that costs only "marginally" more than a decent 2 channel DAC these days. And it's made in Europe!

 

[ChristianN]

Any okto research pro 8 Owners that have compared the “sound” of the dac with an miniDSP HD2x4 or miniDSP Flex and can put som words on what to expect if shifting to okto ?

 

[Kal Rubinson]

Any okto research pro 8 Owners that have compared the “sound” of the dac with an miniDSP HD2x4 or miniDSP Flex and can put som words on what to expect if shifting to okto ?

More channels.

 

[waynel]

Any okto research pro 8 Owners that have compared the “sound” of the dac with an miniDSP HD2x4 or miniDSP Flex and can put som words on what to expect if shifting to okto ?

Lower noise floor, especially if you have sensitive speakers and a low noise amp. This is not audible at the listening position.

 

[ChristianN]

Lower noise floor, especially if you have sensitive speakers and a low noise amp. This is not audible at the listening position.

Thx. Was looking to Include more subs via the extra channels, but also gain some more intangible sound improvements, seems as if the latter is not realistic as I don’t here any noise in the current setup.

 

[waynel]

Thx. Was looking to Include more subs via the extra channels, but also gain some more intangible sound improvements, seems as if the latter is not realistic as I don’t here any noise in the current setup.

A few considerations:
1) I think that usually you are better off running your subs in mono so you don’t need a channel of dac per sub
2) the xlr outputs are good for preventing ground loop noise especially when running multiple subs, having a separate xlr for each sub is nice (you can sum to mono digitally)
3) the rotary volume knob, display, remote , and trigger out are very nice to have
4) you won’t find a better sounding dac (although you’ll find some identical sounding ones)

 

[ChristianN]

A few considerations:
1) I think that usually you are better off running your subs in mono so you don’t need a channel of dac per sub
2) the xlr outputs are good for preventing ground loop noise especially when running multiple subs, having a separate xlr for each sub is nice (you can sum to mono digitally)
3) the rotary volume knob, display, remote , and trigger out are very nice to have
4) you won’t find a better sounding dac (although you’ll find some identical sounding ones)

Hi thx,
Ad point 1-3 that is also how i run it today with a balanced minidsp flex, however with fever subs.
Its the point 4 thats intereseting, if is sounds so good, is it better then the miniDSP or are all dacs more or less the same. I belive you said it would not be noticable.

 

[juliangst]

Does the DAC8 Pro require software like Voicemeeter for USB/AES mode or how does it loop the AES input through the PC?

 

[waynel]

Hi thx,
Ad point 1-3 that is also how i run it today with a balanced minidsp flex, however with fever subs.
Its the point 4 thats intereseting, if is sounds so good, is it better then the miniDSP or are all dacs more or less the same. I belive you said it would not be noticable.

Yes , it will sound identical to the dac in the MiniDSP flex , I’m assuming you’ll be using a MiniDSP flex digital or SHD studio if you switch to the Okto.

 

[Kal Rubinson]

1) I think that usually you are better off running your subs in mono so you don’t need a channel of dac per sub

Only if you must. Running them individually and using DiracLive Bass Control is preferable, imho.
I am OK with 2-4.