MDAC First Listen (part 00011110)

[Mikas]

Good point! But with 8 bits being chopped off should that not corrupt to influence the output? There should be missing info I guess. The files plays fine and the sound quality is great.

(Sorry for my questions but I have not so in-depth technical knowledge of the matter but are very interested )

There's no missing info.
All audio info is stored by the bits frequency.

The bits depth affects only the dynamic range.
With 'only' 16 bits, the dynamic range is limited to 96dbs.
Take a look:
http://en.wikipedia.org/wiki/Audio_bit_depth

Michael

 

[Dodeth]

Hi Dodeth,

no need for power.
I haven't tried MDAC using anything else than USB, for me Asynch USB is just "right"

Thanks for the help with this. Started using USB at the weekend and, to me, it appears to make a significant improvement. I'm well aware of the 'placebo effect' though!

 

[JohnW]

Dodeth,

Glad its working for you - out of interest, do you have any problems when powering off and re-powering the SBT / MDAC system - I'm trying to understand the issue encountered by Simon (Wintoid) - feel that I've not done enough to understand / resolve the problem.

http://www.pinkfishmedia.net/forum/showpost.php?p=2274341&postcount=570

 

[JohnW]

I just had a realisation that with the DSP version we can offer a Filtered / Sub Woofer Digital output - we have been asked before for some form of Bass management, with the DSP option this would know be possible...

Its not something that I'm set up to try with my system - so once we start to develop the DSP code we would ask for PFM'ers input on this feature.

 

[JohnW]

Today a small box of 100 tubes have arrived, so once I've completed Tony Rogers MDAC upgrade - I'll test a few on the little test PCB I built earlier - there's only so many hours a day one can keep productive designing the MDAC2 PCB layout... I short break during the day working on something else not so challenging refreshes the mind

 

[Reenberg]

I assume you have seen the work done on the es9018 with Broskies ( tubelab.com ) unbalancer.

There is a thread on diyaudio.com where apparently very good results are obtained by taking the bal outpit directly from the chip and feed it into Broskies tube unbalancer ( of course the result is only a SE signal ).

http://www.diyaudio.com/forums/digital-line-level/216917-biii-broskie-unbalancer-vs-ee-minimax-plus-my-experience.html

 

[Mikas]

...
I short break during the day working on something else not so challenging refreshes the mind

Although not really less challenging, don't forget to walk your team of assistants too.

Michael

 

[JemHayward]

Dodeth,

Glad its working for you - out of interest, do you have any problems when powering off and re-powering the SBT / MDAC system - I'm trying to understand the issue encountered by Simon (Wintoid) - feel that I've not done enough to understand / resolve the problem.

http://www.pinkfishmedia.net/forum/showpost.php?p=2274341&postcount=570

I don't actually think its a MDAC problem, I think it may be a "problem" with the SBT/EDO - if the DAC isn't "there" when it wakes up, then it decides to use another output, if that output is there, then it sticks with that output, which is probably logical behaviour. Its just how the ALSA system seems to work on Linux. Luckily, for me, the MDAC boots faster than the SBT, so if you power both up by switching on the mains, it all works fine.

Maybe we could specify a clean 5V output on MDAC3 for powering linked streamers? [note MDAC3 not MDAC2 ]

 

[JohnW]

I assume you have seen the work done on the es9018 with Broskies ( tubelab.com ) unbalancer.

There is a thread on diyaudio.com where apparently very good results are obtained by taking the bal outpit directly from the chip and feed it into Broskies tube unbalancer ( of course the result is only a SE signal ).

http://www.diyaudio.com/forums/digital-line-level/216917-biii-broskie-unbalancer-vs-ee-minimax-plus-my-experience.html

To be quite honest I've have not done much searching on tube designs on the internet as there are so few real experts - tubes seem to attract a disproportionate amount of "Garage Amateurs" (He says as the "garage building work" proceeds outside for the new lab ).

I have what I believe is a novel design that simulate and works very well on the lab bench - looking forward to hearing it once I've completed the design of the MDAC2 analogue board.

 

[JohnW]

Although not really less challenging, don't forget to walk your team of assistants too.

Michael

Oh they drag me out into the sunshine, and then they as a "Micro Gang" proceed to cause general havoc - but its still a little too chilly ATM for me.

 

[Reenberg]

( the 2 300 ohms should prob. be 50, haven't build it yet ( time time time ) )
This is the schematic... And newer mind, I think the tube part sounds greats.


John, since the price oof the unit is increasing a bit I would suggest one or two more ' down payments ' before it's done. So you have some cash to work with, and we don't get a big surprise 4 months from now ( take my money request )


Theis

 

[JohnW]

John, since the price oof the unit is increasing a bit I would suggest one or two more ' down payments ' before it's done. So you have some cash to work with, and we don't get a big surprise 4 months from now ( take my money request )

Theis

Theis,
Thank you for the offer

The amount of development has increased with the addition of the Galvanic USB and Analogue isolation, DSP + BT & the Analogue attenuator, however the extra GBP100 installment will cover the extra development time - I'll request the next installment once I've issued the prototype PCB files for the the "heart" of the Analogue stage, which will be in about 30 days.

If the design works as expected then it can be added to a few MDAC owners units for "first evaluation" of the MDAC2 design - although I stress its only a portion of the planned upgrades.

 

[fusion5]

If the design works as expected then it can be added to a few MDAC owners units for "first evaluation" of the MDAC2 design - although I stress its only a portion of the planned upgrades.

I've got 3-4 weeks holidays starting end of may

 

[JohnW]

I've got 3-4 weeks holidays starting end of may

Now that sounds like a trip to Czechs in order.... with one of your units to sacrifice itself for the greater good

 

[Stuart Frazer]

Jim,

The Analogue input (Analogue bypass) is fixed level, so you will need a method to adjust the level of your analogue source (a simple passive pre-amplifier will do)

We will make an external Analogue attenuator available at a later date if your need a variable (adjustable level) input based on the extremely HQ Khozmo 48 step stepped attenuator - which will be controlled via the MDAC2.

Oh Yes - I can promise the MDAC2 will be out THIS year

John, can this not developed in conjunction with the existing work so that this is available when the MDAC2 is ready? I think it's important to some of us as it may mean we can loose a preamp box, which will help fund the MDAC2 purchase.

 

[JohnW]

John, can this not developed in conjunction with the existing work so that this is available when the MDAC2 is ready? I think it's important to some of us as it may mean we can loose a preamp box, which will help fund the MDAC2 purchase.

Stuart,

We will have a prototype tested when we release the MDAC2 so that everything in place for the external analogue attenuator - but my priority is to get the MDAC2 out

 

[jirij]

if the DAC isn't "there" when it wakes up, then it decides to use another output, if that output is there, then it sticks with that output, which is probably logical behaviour. Its just how the ALSA system seems to work on Linux.

Well, kernel ALSA doesn't select the sound output IIRC, that's done by the application (perhaps via libasound2, honoring aroundrc conf) and even then, there's no "sticks with" - when no default is explicitly specified in the config, it defaults to (AFAIK) lowest sound card "index".
What might "stick to" some output is the application (which can try to open devices in some order and stick with first one that successfully opens (snd_pcm_open)).

Getting back to the referenced post, it *may* actually be a MDAC firmware issue, I've seen some behavior rather unique to the MDAC in terms of usb packet scheduling. However SBT apparently has its own USB issues, so it's entirely possible that it's not an MDAC issue.

If I get the chance (JohnW/Dominik), I'll ensure that MDAC2 is properly tested on a variety of linux kernel versions before the final firmware versions is released. This might positively impact SBT as well (along with various NAS/player boxes), although I don't have the hardware to test with SBT directly.

 

[Dodeth]

Dodeth,

Glad its working for you - out of interest, do you have any problems when powering off and re-powering the SBT / MDAC system - I'm trying to understand the issue encountered by Simon (Wintoid) - feel that I've not done enough to understand / resolve the problem.

http://www.pinkfishmedia.net/forum/showpost.php?p=2274341&postcount=570

I've tried a number of different conditions and it appeared to be important that the MDAC is switched on before the SBT is activated (via the orange button on its remote), and that the SBT is de-activated before the MDAC is switched off. Otherwise it looks like the MDAC is working but there is no audio. However, there does not appear to be any consistency. Sometimes the sequence mentioned above works, sometimes it doesn't. Weird. Not sure I've been any help....!