advertisement


MDAC First Listen (part 00111000)

Status
Not open for further replies.
I think i will settle for only passive filters for my next speaker system, digital XO is not really worth it.

I agree with you about using a passive XO for your speakers if it can be designed to fit your requirements - I've been listening to miniDSP processing in "Flat mode" where the miniDSP box just "up samples to 96KHz or downsamples to 96KHz" and there is a significant loss in quality.
 
John,

Will the FPGA allow for digital inputs like S/PDIF and such on the Discrete DAC design?

The Discrete DAC requires a "perfect clock" - when the DAC is clock Master then this is easy - but if we need to cleanup an external clock then the "Jitter attenuation circuit" will endup being more complex then the DAC itself.

Even when "cleaned" an external clock will never be as spectrally pure as a simple Crystal oscillator especially within the audio B/W.
 
The sabre version of the mdac2 offers good performance for the price but the streamer and front panes upgrade does not offer good value for money imo. I would be willing to pay money to get vishey transistors though, i might aswell aim go for full HIFI insanity if i get this dac.

What are your main audio sources? If you only use USB or can live with the "Streamer inputs" then sonically the Discrete DAC offers the best performance.

When the MDAC2 is ready I'd be a happy to send you both DAC versions and then you chose which you prefer and return the unwanted DAC PCB.

WRT to the cost of the front panel update there are tooling costs that need to be spread across only a few units - the capacitive touch panel tooling is US$2300 alone then there are 2x lightquide tooling's - and this is before we factor our engineering time and CNC programming etc.

WRT to the streamer, IMO GBP220 is a bargain for a Bit accurate 384KHz / DSD256 capable device - I cannot think of anything streamer near the price point that offers these features and allows the DAC to be clock master (Jitter free interface). Te Streamer also supports 1080P video, something I'd like to build on.

By the end of September we should be seeing first boards, then the deliveries should start to rapidly expand.
 
I agree with you about using a passive XO for your speakers if it can be designed to fit your requirements - I've been listening to miniDSP processing in "Flat mode" where the miniDSP box just "up samples to 96KHz or downsamples to 96KHz" and there is a significant loss in quality.
Even the expensive groundsound dcn28 is limited to 96KS/s and doesn't even offer FIR filters, i worked with it for a short period and it didn't sound good either. Active XO to subwoofers does make sense though but you do not need expensive digital filterd for that.

What are your main audio sources?

When the MDAC2 is ready I'd be a happy to send you both DAC versions and then you chose which you prefer and return the unwanted DAC PCB.
That's a generous offer but i think i will decline it and just get one finished dac.

I currently mostly listen to music stored on my computer, i never stream music, i stream video via youtube and other sites sometimes but it's very far from HIFI audio then. Now i would only use the USB but in the future being able to use spdif might be valuable.
 
That's a generous offer but i think i will decline it and just get one finished dac.

I currently mostly listen to music stored on my computer, i never stream music, i stream video via youtube and other sites sometimes but it's very far from HIFI audio then. Now i would only use the USB but in the future being able to use spdif might be valuable.

There is some confusion about the term "streamer" as our "streamer" board can be considered a replacement for a PC - you can play your files from USB flash drive, external SSD drive, or stream from other devices & even play YouTube Audio etc. I feel that many will pickup on the streamers advantages once they start to use it.

Here, 90% of my listening is via YouTube as I work on PCB deigns etc.
 
The Discrete DAC requires a "perfect clock" - when the DAC is clock Master then this is easy - but if we need to cleanup an external clock then the "Jitter attenuation circuit" will endup being more complex then the DAC itself.

Even when "cleaned" an external clock will never be as spectrally pure as a simple Crystal oscillator especially within the audio B/W.

So the answer is no?

Would it be possible to clock lock a device with 10MHz input?
 
So the answer is no?

Would it be possible to clock lock a device with 10MHz input?

Yes - we can arrange that, I'm also looking at ways to bring "decent" SPDIF performance for the FDAC (with the non ESS DAC option).
 
So, John, if I understand correctly, the ESS-based FDAC hasn't been cancelled after all?
 
Yes - we can arrange that, I'm also looking at ways to bring "decent" SPDIF performance for the FDAC (with the non ESS DAC option).

It is a pre-requisite for me as I intend to continue to use a CD transport as well as the streamer.
 
So, John, if I understand correctly, the ESS-based FDAC hasn't been cancelled after all?

I don't want to define FDAC to much until we have the feedback from MDAC2 - so please don't get to concerned about the exact features - but FDAC will have to support SPDIF somehow.
 
Apologies, I have no doubt this has been covered, but things change so quickly over here ;)

I am really interested in 'getting into' some form of room correction. I stream music from LMS/Squeezebox setup. Via USB to my MDAC (for now, awaiting MDAC2). I also have analogue. Currently, both MDAC and phono are amplified via an ATC SIA2-150 into Impulse H2's. My listening room is not ideal. the room is laid out like this: (BTW I have speakers in position A but toed in)




I think the bass and the forward treble (somewhat controlled by PAC's magic work on the crossovers) in these speakers could be better tamed with some DRC. The question is with the planned MDAC2 with Streamer will there be any way to do this in the digital domain.

I have been reading up on this last few days, so am a complete newbie to this. But the most interesting option to me is to try using something like REW to measure and then produce filters for BruteFIR which I think I can 'plugin' to LMS. But I may be way wrong. the advantage of this approach is it would happen upstream of MDAC2. is there another way?
 
There is some confusion about the term "streamer" as our "streamer" board can be considered a replacement for a PC - you can play your files from USB flash drive, external SSD drive, or stream from other devices & even play YouTube Audio etc. I feel that many will pickup on the streamers advantages once they start to use it.

Here, 90% of my listening is via YouTube as I work on PCB deigns etc.
Funny how we value features differently, my listening is 0% YouTube.

I will never use the streamer as a PC and will therefore not connect a SSD, CD/DVD or use the HDMI to connect it to my TV (hate have the TV switched on while listening to music).
The only way I could possibly use the HDMI port is to stream multi channel audio.
I sit 4 meters from the DAC so will remote control it from my armchair.

I still see the streamer+front display as giving great value for the following reasons (My favorites):

  • DAC to be clock master (Jitter free interface) this alone makes it worth every penny.
  • It can function as a RoonBridge
  • Volumio distribution is using node.js which the Roon API happens to support which means it is fairly easy to add cover art an track info to the UI (like SB Touch)
  • It will be possible to use different kind of remotes which includes the standard ir remote to tablet and laptop using a app or browser. Very flexible!
  • I believe that it will be fairly easy to add functionality in the future such as widgets.
What I am trying to say is that I do not understand people complaining about features they do not want. Just ignore the specic feature and be happy with the ones you use :)
It is OK to point out functionality that you do need/want like spdif, AV bypass etc.

I don't want to define FDAC to much until we have the feedback from MDAC2 - so please don't get to concerned about the exact features - but FDAC will have to support SPDIF somehow.
Agree it is better to leave that for when some of us have first hand experience with the SS9028Pros MDAC2+streamer+Touch display and spdif that some need.
It is difficult to have a objective discussion until then about FDAC / discrete DAC.
 
I am checking out the competition, it is getting better.

http://archimago.blogspot.se/2017/06/measurements-oppo-sonica-dac-ess-sabre.html it supports DSD512 and 768/32 but the dynamic range is only 110dBA.

On http://mdac2.lakewestaudio.com/ it is stated that 124dBA dynamic range is the goal with the mdac2, for the fdac 135DBA dynamic range was the goal earlier.
Its been a feeling for awhile now that the limiting factor is the ESS DAC's - it not to say they are bad DAC's but just that the MDAC2 / FDAC are ultimately being held back by the ESS's.

The ESS version measures better and supports PCM to 768KHz (verses 384KHz) and also volume control with DSD.
I trust your measurements and facts more than your feelings, you never explains how the PMD would be better, seams like money and effort will be spent to make the mdac2 worse with later iterations.
based on listening tests without a moments hesitation Jarek and I choose the discrete DAC version over the ESS.
I assume the tests whern't blind, i suggest you do some proper (blind) AB testing, i do not have proper testing environment for that(my computer isn't totally quiet, no good speakers, bad amp).
 
http://archimago.blogspot.se/2017/06/measurements-oppo-sonica-dac-ess-sabre.html it supports DSD512 and 768/32 but the dynamic range is only 110dBA
Wow, this seems surprisingly similar to the base MDAC2 feature set (ADC, streamer, form factor, etc.). I guess that's the next "evolution" step in ~$1000 DAC boxes, though missing out on the headphone output would be a step backwards, in my opinion.
I'll have to look for something cheaper until MDAC2 comes, as my current MDAC is getting harder and harder to revive after every OS reboot and it's getting really annoying.
 
Actually -more like 120dB with XLR and probably at or beyond the limits of Archimago's measurements
If that is the case i no longer see any valid reason to get the mdac2, sure it lacks a few options(dac filters and headphone jack) but i still want my money back. Remember: this is a dac for 800$.

EDIT: I still need a good ADC so maybe i should not try to sell my pledge.
The "Dacapo Anniversary" MDAC2 DAC PCB builds upon the discrete DAC topology used in my first commercial DAC design (ignoring a handful of “Westlake Audio DAC’s” sold before I was approached by Pink Triangle) – it has certain limitations (such as “only” supporting PCM to 192KHz / 384KHz without level control, and no level control with DSD playback) and also measures worst (~110dB dynamic range / THD 0.001%), but I’ll leave it for the forum to listen and decide which sounds the best – based on listening tests without a moments hesitation Jarek and I choose the discrete DAC version over the ESS.
I am pretty sure it was placebo, i do not really want to be honest now but this is insanity.
 
Apologies, I have no doubt this has been covered, but things change so quickly over here ;)

I am really interested in 'getting into' some form of room correction. I stream music from LMS/Squeezebox setup. Via USB to my MDAC (for now, awaiting MDAC2). I also have analogue. Currently, both MDAC and phono are amplified via an ATC SIA2-150 into Impulse H2's. My listening room is not ideal. the room is laid out like this: (BTW I have speakers in position A but toed in)




I think the bass and the forward treble (somewhat controlled by PAC's magic work on the crossovers) in these speakers could be better tamed with some DRC. The question is with the planned MDAC2 with Streamer will there be any way to do this in the digital domain.

I have been reading up on this last few days, so am a complete newbie to this. But the most interesting option to me is to try using something like REW to measure and then produce filters for BruteFIR which I think I can 'plugin' to LMS. But I may be way wrong. the advantage of this approach is it would happen upstream of MDAC2. is there another way?

Nice speakers those, from memory they use a focal inverted dome which has a reputation for being brightish Martin Clark who posts regularly in the DIY forum has some so I suggest you ask him about them,he also has a web site with a section devoted to Impulse.
REW is a great tool to understand what your system and room are doing, its easy to use, I had a big 50 Hz boom which made some tracks unlisenable but using a DSP plate amp for bass with a notch filter at that frequency has solved that problem.
There is a temptation to use DSP to get an overall flatter response but this from my experience tends to suck the life out of music so only treat the peaks that genuinely degrade your listening and ignore the troughs.
 
Wow, this seems surprisingly similar to the base MDAC2 feature set (ADC, streamer, form factor, etc.). I guess that's the next "evolution" step in ~$1000 DAC boxes, though missing out on the headphone output would be a step backwards, in my opinion.
I'll have to look for something cheaper until MDAC2 comes, as my current MDAC is getting harder and harder to revive after every OS reboot and it's getting really annoying.
The Oppo sonica DAC does not have an ADC but if you just want a dac without any headphone jack it seams to be fine. Another limitation is that it only supports 2 channels but good multichannel processors cost a lot of money.

I did check out the test more and the poor dynamic range measured(110dBA) is the same for almost all dacs to it must be his setup that is limited.

I would still sell my pledges for 1700$, i do not need a better adc now and i am not into buying stuff for the future. I want to keep my possessions to a minimum.

http://www.aes.org/e-lib/browse.cfm?elib=14195
 
Status
Not open for further replies.


advertisement


Back
Top