Mytek Stereo 192

[labjr]

I'm eagerly waiting for that Tascam to appear in Europe. Korg MR-2000S has been really hard to obtain...

 

$999 street price? Prices are falling!

[Esprit]

I'm trying the S/PDIF input with an economic interface.

IMHO it is very interesting also versus the FW.

The problem is surely the clock.

Listening to PCM you can set manually the internal reclocking and it works, it works very well. With DSD I tried to set up 176.4. The Mytek recognizes DSD files, plays them, sounds good (IMHO better than FW) but, at least with Audirvana, the signal disappears for half a second every 13" of music.

I've changed from "DoP" to "Initial dCS" there was no more interruptions but appeared occasional "static noise".

 

Pity. I'll try to ask Damien if there is a solution.

In this way you would get the best of both worlds without having to spend a lot of money on the interface.

[john925]

I'm trying the S/PDIF input with an economic interface.

IMHO it is very interesting also versus the FW.

The problem is surely the clock.

Listening to PCM you can set manually the internal reclocking and it works, it works very well. With DSD I tried to set up 176.4. The Mytek recognizes DSD files, plays them, sounds good (IMHO better than FW) but, at least with Audirvana, the signal disappears for half a second every 13" of music.

I've changed from "DoP" to "Initial dCS" there was no more interruptions but appeared occasional "static noise".

 

Pity. I'll try to ask Damien if there is a solution.

In this way you would get the best of both worlds without having to spend a lot of money on the interface.

 

Hi Esprit,

 

Waht interface are you using?

[Esprit]

Hi john,

hiFace Two connected directly to the Mytek using one of these:

[Miska]

I'm trying the S/PDIF input with an economic interface.

IMHO it is very interesting also versus the FW.

The problem is surely the clock.

Listening to PCM you can set manually the internal reclocking and it works, it works very well.

 

If you use S/PDIF input and internal clocking and don't synchronize the external S/PDIF source with word clock, you are going to run into problems with both PCM and DSD. Because in this case you have two unsynchronized clocks that will drift versus each other and every now and then cause a buffer overflow or underflow in the DAC...

 

In best case with PCM it means one PCM sample dropped or missing every N minutes while with DSD and DoP it is at least 16 samples (plus loss of state).

[Esprit]

It works there is not audible problems with PCM.

 

From Mytek manual:

"Internal Sync - Incoming data is synced to the converter's ultra-low jitter internal crystal oscillator. This choice assures the best DAC performance and is superior to any (even Atomic) clock supplied from outside because of its close proximity to DAC chipset."

 

"Sync options are only available for S/PDIF, AES/EBU,Toslink, ADAT and SDIF."

[Miska]

From Mytek manual:

"Internal Sync - Incoming data is synced to the converter's ultra-low jitter internal crystal oscillator. This choice assures the best DAC performance and is superior to any (even Atomic) clock supplied from outside because of its close proximity to DAC chipset."

 

"Sync options are only available for S/PDIF, AES/EBU,Toslink, ADAT and SDIF."

 

When using internal clock with S/PDIF you should be using the DAC's word clock output feeding the DAC's clock back to the S/PDIF transmitter so that it will send data at correct speed. Practically all pro-audio interfaces like RME support this.

 

Without this clock feedback DAC should be clock slaved when using S/PDIF. You should always have just one clock in the system, now you are having two... And you will never have two clock that run exactly at same speed.

 

It works there is not audible problems with PCM.

 

Or you have just not noticed with PCM. With DSD it's much more obvious.

[Esprit]

Miska, I'm not a technician but if Mytek added the opportunity on internal sync I think I can use it.

When I use my Echo AF2 I set internal clock as master for all incoming signals from Mac or others device.

The Mac, the sat decoder, the CD player sends out their data and all work perfectly.

So if I connect the Mac via toslink, the CD player via S/PDIF, etc... to the Mytek I don't know why it's useless using the internal sync.

I'd like to know, from Mytek, what they intend for "internal sync"[*] and when I can to use it.

I can only write that I have the best sound ever heard, in my house, from Stereo 192 DSD (S/PDIF + Internal Sync).

If it could also work with the DSD I would be god on earth

 

[*] In the menu int sync can be set for each frequency from 44.1 to 192 and the external WC is a separate option.

[Miska]

Miska, I'm not a technician but if Mytek added the opportunity on internal sync I think I can use it.

 

Of course, most pro converters have the feature, one converter is clock master (internal sync) and others including computer are clock slaves (external sync). That's why there's both word clock input and output. Intention is to use this setting together with the word clock I/O. It is completely normal to have pile of ADCs and DACs in a studio, it would be a complete cacophony if all these devices would be using their internal clocks. They would never be able to agree about correct clock rate and phase. It is usually ADC clock that drives the show. For a mastering studio however it would be the DAC clock and computer feeding it would be the slave (DAC WC out -> computer WC in).

 

When I use my Echo AF2 I set internal clock as master for all incoming signals from Mac or others device.

The Mac, the sat decoder, the CD player sends out their data and all work perfectly.

 

I don't remember about Echo AF2, but I think it can use S/PDIF input as "word" clock input (that's at least how E-MU and M-Audio support it). But for example you can use Echo AF12 together with Mytek in two following ways:

 

1) Set Mytek to internal clock and Echo AF12 to external clock, connect word clock cable from Mytek clock output to AF2 clock input. This doesn't necessarily make so much sense...

2) What makes more sense. Set Echo AF12 for internal clock (master) and connect clock output of AF12 to clock input of Mytek.

 

In both cases there's just one clock in the system and all others are slaves and you can run all simultaneously in nice harmony.

 

I can only write that I have the best sound ever heard, in my house, from Stereo 192 DSD (S/PDIF + Internal Sync).

 

You should be just using an S/PDIF or AES/EBU interface that supports word clock input to slave the transmitter clock to fix the problem...

[pacwin]

Hey. I am strongly considering to get the Mytek and was lurking here for a long time.

Yesterday a thread of another forum got to my attention and I wonder if the tested gear was defect or not since their conclusion seems to be quite the opposite what I was reading elsewhere.

Here is the link h**p://www.changstar.com/index.php/topic,883.0.html

 

According to their "methodology" this is how they tested it.

 

Evaluations were performed using the Mjolnir as an amp / pre to the following:

> Fostex fe166 rear loaded horn high efficiency wide-banders with Hsu sub and Fostex T90 supertweets. Speakers are EQ'd for listening position using the EasyQ plug-in in JRiver.

> Crest CA2 power amp > HiFiMan HE-500s with fuzz pads.

 

The Mjolnir is a bit of an odd device to use as a preamp: There aren't even any clear specs: But based on these two measures:

 

THD: <0.006%, 20Hz-20KHz, at 1V RMS

SNR: >104db, unweighted, referenced to 1V RMS

MJOLNIR - $749.00 : Schiit Audio, Headphone amps and DACs made in USA.

 

I suppose its a reasonable assumption that the Mjolnir is not going to be expected to accept more than 1 or maybe 2 volt RMS. Certainly the headroom is not specified That's considerably less than what the Mytek puts out over balanced connections. One suspects overloading of the front end of the headphone amp. If they drove the Crest CA2 power amp directly, and no suggestion they did, again another unusual "audiophile" choice to feed headphones then this amp has a standard 0.775V input sensitivity with factory options of 0.866V and 1.73V. Even with the latter tweak its potentially going to overwhelm the amp.

 

One presumes then that the most plausible explanation is mismatched equipment and way too much input voltage as Ted has suggested as well:

 

"The volume control was set to Bypass. The balanced outputs from the DAC were used to feed the balanced input of a Schiit Mjolnir headphone/HE speaker amplifier. The internal gain jumpers of DAC were not touched. All the other DACs in the comparison were set up similarly (JRiver WASAPI, USB, balanced outputs, etc.)"

http://www.crestaudio.com/media/pdf/ca2_11-25-97.pdf

 

The Headphone amp also has extremely wide band frequency response 2Hz-400KHz. There may well be ultrasonics hitting their brains especially with the supertweeters they used (35kHz). Who knows.

 

Changstar is a private forum for DAC geeks. It doesn't pretend to be a professional testing centre. The secret of computer audiophile is to be able to sort out the wheat from the chaff. The Internet has way to much chaff.

[Esprit]

Miska, WC has nothing to do with the internal sync in the Mytek.

Please read the manual or use your Mytek DAC.

[Esprit]

You should be just using an S/PDIF or AES/EBU interface that supports word clock input to slave the transmitter clock to fix the problem...

What problem? (Dirty quote?)

[blaine78]

Ted_b - your Concert Fidelity is one beautiful pre amp, luck you and congrats.

[valenroy]

iFi iUSBPower with Mytek and Tweak

 

Just did some assessments on the main Mytek setup with the iUSBPower that is hooked up to my 2010 MacBook Pro 13. This is a interesting case as most followers of this thread are aware I am running two separate runs of USB and Firewire respectively. The USB cable has the VBus and GND connectors blanked out with small, little stripes of sticky tape to act as the differential signals +/- only conduit whereas the Firewire is tasked to act as the dedicated power conduit with no connectors blanked out. I was interested to find out whether the power from the supplied iUSBPower is comparable to my current setup, where the influence of the power line on the signal ones is null in this instance. Right off the bat... my current setup is still the best sounding to me. The addition of the iUSBPower provides no audible benefits and on the contrary, deteriorates the sound reproduction ever so slightly with it in the chain. I suspect one of the possible reasons is the additional component (iUSBPower) that is added to the signal chain. The deterioration of the signal could have manifested over the additional length of USB cable that runs from the MBP to the IUSBPower and/or the internal components of the iFi unit itself that degrades the signal. The other possibility is more subjective - perhaps the iFi needs running in too?! To add further, the IsoEarth function makes no difference in this setup. I am not surprised though, as with all apple mbp, the power adapter is never grounded so I won't think grounding issues will be a concern. So in a way, this is good news for Mytek users as you don't need such a device to enjoy the benefits of signal/power line separation, just an extra el cheapo USB or FireWire cable will get you better benefits than with the iFi!

 

P/s: The Weavers - Reunion At a Carnegie Hall is a fantastic release from Analogue Productions with excellent sonics! Used this SACD and rip for the above listening sessions. "Guantanamera" and "Wimoweh" sure bring back old school memories...

[Esprit]

why You use/prefer USB to Firewire?

[valenroy]

why You use/prefer USB to Firewire?

 

I do actually prefer the USB connection because of the wealth of cable options available to this interface. My previous observation using the Audioquest Carbon USB and FireWire cable is that the FireWire is slightly more gentle and softer on the presentation versus the USB. The USB connection tends to provide more bite and attack to the performance. And unfortunately in my case with my MBP, both connections requires the presence of vbus/gnd for the MBP to detect the DAC. I am sure a lot of other folks on FireWire have good experience with the interface with the power lines blanked out only, and not needing another run of the USB cable for the vbus/gnd connections.

[Miska]

Miska, WC has nothing to do with the internal sync in the Mytek.

Please read the manual or use your Mytek DAC.

 

Of course it does when you use internal sync + SPDIF/AES input. It doesn't say anywhere that it wouldn't?

 

Either you

1) Use "Input sync" and the SPDIF/AES input alone

2) Use "Word clock" and SPDIF/AES input + word clock input together

3) Use "Internal sync" and SPDIF/AES input + word clock output together

 

Option (3) lets you do kind equivalent of "async" input with SPDIF/AES.

 

I don't want to argue about this more. But don't be surprised if something doesn't work properly when you don't follow these configurations...

[petecare]

Hi john,

hiFace Two connected directly to the Mytek using one of these:

[ATTACH=CONFIG]5175[/ATTACH]

 

Hi Esprit,

 

With the Hiface 2, did you had "Integer Mode" ?

[Esprit]

But don't be surprised if something doesn't work properly when you don't follow these configurations...

Excluding the DSD (I do not even know what should be its clock frequency and I assumed 176.4) everything works perfectly. Regarding point #3 I do not think that the WC output has something to do with internal sync. Anyway, I asked Adam.

[Esprit]

With the Hiface 2, did you had "Integer Mode" ?

Yes (for all files: from SD to HD, DSD included).

I'm thinking about a better interface like Matrix X-SPDIF.

Again no cable but an adapter like this:

[Miska]

Excluding the DSD (I do not even know what should be its clock frequency and I assumed 176.4) everything works perfectly. Regarding point #3 I do not think that the WC output has something to do with internal sync. Anyway, I asked Adam.

 

DSD works perfectly with "input sync" and probably would also with "internal sync" + WC out. According to diagrams WC out outputs the internal clock when in internal sync mode.

 

For (3) it should, that's how all the other pro converters I've used in past 15 years have worked. From Mytek DAC side the output is kind of unrelated, but it is of course related from the transmitter's perspective. The device sending the data to Mytek should be using Mytek's clock in case the sending device's own clock is not being used at the recipient side.

 

Point is to have single clock, either at Mytek side, or the sender's side. Since SPDIF/AES is unidirectional, WC is needed to send the clock to opposite direction.

[Esprit]

DSD works perfectly with "input sync".

Not here. I have to choose one option (all OR, none AND).

[Miska]

Not here. I have to choose one option (all OR, none AND).

[ATTACH=CONFIG]5184[/ATTACH]

 

Just select the "Input sync" (highlighted red in the picture) and play DSD. Works fine for me, either from MuFi V-Link192 with AES cable or M2Tech hiFace with coax, or even from ASUS Xonar DX with Toslink fiber.

 

If you use internal sync, select 176.4 and use word clock output to the transmitter (for example RME or Lynx card) and configure the transmitter side to use external sync.

 

Of course when using "internal sync" on Mytek it expects so see data appearing exactly at it's own clock pace which means that you have to send Mytek's clock back to the sending device so it can sync with Mytek's clock.

 

Mytek <-- data -- computer

Mytek -- clock --> computer

 

 

With "input sync" it's the standard

 

Mytek <-- data+clock -- computer

[Esprit]

Miska in all my posts I'm only talking about "internal sync" not "input sync".

I know input sync works. I don't like, with hiface2, what I obtain (i don't like its clock...).

I understand your explanation, but in this case there are very few interfaces that accept WC input.

[Esprit]

One is this but it is not XMOS

I just have to hope in Evo Two...