Yes it wille be like a dream to get pcb to dam 1021 1121, with good one freq clock like an option who want more in music use. This will be good option for you to get profits,and as who want it more. please consider this option. Take a look of terminator r2r dac, it have twos fixed clocks.
Well no one red my novel it seems 
So i will ask the question again, is it possible to have both a programmable clock to handle spdif and usb, and an alternative i2s input that bypass all of that (isolator, fifo etc) and also allow an external masterclock ?
You keep the advantage to have low latency input and if people want long fifo for no jitter with expensive clock or whatever; (but with latency...) they are happy...
@bmxmen : What you said i also said it, i think a lot people would like this option.
However i don't think you should compare to other R2R. Again i am no electronic, but for the few i understand some choice of these brand seems complete non-sense to sell over-expensive audiophile BS.
For example audiophonics that sell a rebrand holo spring. I find it strange as the original is suppose to be 4k€ and they sell it 1.5k... anyway...
I like the 400va transfo sold as a commercial argument... Also the dac use a ak4137 for the filter. I am no expert but isn't this chip built to feed a Delta-Sigma dac ? I guess custom filter build specifically for a r2r dac will do better ?
And for the terminator dac, well... the bank of caps thing...
So i will ask the question again, is it possible to have both a programmable clock to handle spdif and usb, and an alternative i2s input that bypass all of that (isolator, fifo etc) and also allow an external masterclock ?
You keep the advantage to have low latency input and if people want long fifo for no jitter with expensive clock or whatever; (but with latency...) they are happy...
@bmxmen : What you said i also said it, i think a lot people would like this option.
However i don't think you should compare to other R2R. Again i am no electronic, but for the few i understand some choice of these brand seems complete non-sense to sell over-expensive audiophile BS.
For example audiophonics that sell a rebrand holo spring. I find it strange as the original is suppose to be 4k€ and they sell it 1.5k... anyway...
I like the 400va transfo sold as a commercial argument... Also the dac use a ak4137 for the filter. I am no expert but isn't this chip built to feed a Delta-Sigma dac ? I guess custom filter build specifically for a r2r dac will do better ?
And for the terminator dac, well... the bank of caps thing...
I understand your rationale behind and I appreciate that. There must be compromise in some area to make a product more affordable. Your DAC really has good quality to certain degree. When it comes to ultimate performance, no compromise design, it would be available for a niche market only.
I hope someone (or you) in future could build another DAC for the DIY community which is less compromise.
Seems like you still don't understand the Sign Magnitude R-2R principle....
What transformers ? I'm talking about the vref supplies.... People are already having fun using batteries there, but I don't consider batteries that practical.
I can hear that you have no idea what you can find on youtube nowadays in high resolution. For example, I found some Halestorm videos, recorded live in a studio at high resolution, pretty amazing....
Last edited:
I already have too many SKUs, and as the complexity and price goes up, the customer base go down.... I am just one person and need to figure out the best use of my limited time....
Overall I have to say that, whilst one always wishes for more/better, Soren is to be congratulated for having been brave enough to introduce and develop the Soekris range of DACs. To do so was not a minor undertaking and he has to make a return on his investment in order to earn his daily crust so I don't think it should be a surprise or a disappointment that he has made compromises that some of us disagee with from our audio nerd perspective. Under the circumstances is it reasonable for just a few of us here to expect Soren to make the investment to satisfy our niche requirements?
One way forward might be for a group development/buy for a no compromise R2R DAC though I'm not sure, with a group arrangement, that reaching an agreed design position would be easy even if the necessary skills and experience to develop the solution were available. With a small niche demand it would likely be expensive too. I couldn't bring any useful skill to such a party.
BTW, I have a dam1121 and it really does reproduce extremely good music.
One way forward might be for a group development/buy for a no compromise R2R DAC though I'm not sure, with a group arrangement, that reaching an agreed design position would be easy even if the necessary skills and experience to develop the solution were available. With a small niche demand it would likely be expensive too. I couldn't bring any useful skill to such a party.
BTW, I have a dam1121 and it really does reproduce extremely good music.
SPDIF Spec:-
Level I: high accuracy mode
The transmitted sampling frequency shall be within a tolerance of ±50 ppm
Level II: normal accuracy mode
The transmitted sampling frequency shall be within a tolerance of ±1 000 ppm
Technically your correct that the SPDIF spec Level II requires a tolerance of ±1 000 ppm, but I'm sure your well aware that any decent product will ultimately be Crystal oscillator based and you have to be having a VERY VERY bad day and very bad Quality to be outside of ±50 ppm let alone ±100 ppm, almost all decent products well be within / near Level I: high accuracy mode of ±50 ppm.
Its possible with my Audiolab MDAC design to indicate the external PPM clock error (Within an absolute accuracy of ±10 ppm) – and I've really never seen worst then ±100 ppm on any decent product – INCLUDING Digital TV / streaming products (which typically have horrid Jitter levels).
Between the many IC projects I've been involved with and “end products” (Audiolab, Cambrdige Audio, ProJect, Quad etc) I've never had a problem with a typical ±200 ppm clock range, and its relativity easy to design an affordable (less then US$10) dual frequency clock circuit that has a ±200 ppm to ±250 ppm pull range and offers atleast a magnitude better PN performance over the SI PLL based parts...
Lets face it, your selling a premium DAC solution, nobody's going to use it with a source that's outside a ±200 ppm clock range – saying otherwise is just never going to be a valid argument!
If you where in the frame of mind, you could offer a High performance clock option with the caveat that it will “Only” function with digital source with clock accuracies better then ±200 ppm.... Trust me, its a non issue!
Last edited:
Depends what you mean by cheaply - US$5 to US$10 then it can be better by atleast a magnitude - if you understand the PN graphs numbers then its apparent how pretty poor it is when when compared to even a basic XO circuit...
You cannot expect such a PLL design to have decent close-in Phase noise...
As I said, the si570 is a high performance solution, already used by high end DAC manufacturs like MSB Tech and Rockna.... For my digital PLL to work, I want the 1000 ppm glitchless adjustment, don't have anything to do with SPDIF....
I just need the clock performance to be good enough so the clock is not the limit for the DAC performance....
So thanks for your offer, but I'm going to stick with the SiLabs si514/si570 programmable oscillators.
Hello mayby you will help. I was testing this system: roon core on intel nuc, roon bridge raspberry pi4 with moode audio 6.2 ,next bu usb to xmos , next via i2s tp soerkis dam1021 dac. Im plaing with upsampling to 352/384khz in roon audio, it plays laouder and more detailed then normal 44khz. Next test intel nuc with roon core ,roon bridge raspberry pi4 with moode audio 6.2 via i2s to dam 1021 dac. And in roon core seting it is limited tp 192khz,of corse i dont se much difrence in sound quality by 44 and ups to 192khz, it just play poor vs prev config. My dam 1021 dont have i2s isolators, with xmos it plays clearer without isolators. Did rasp pi 4 can play via i2s more then 192khz? Plan is to play via i2s with you reclocer. But im little scary that will be not good as xmos with ups to 384khz. Please help me!
I'm not an electronics designer, so maybe I'm missing something, but it seems you are comparing a fixed clock to a variable one. As far as I know it is not possible to use a fixed clock for reclocking without either a long FIFO or ASRC. Is this correct?
No, I'm suggesting a dual frequency VCXO with a guaranteed +/-200ppm pull range (typically nearer +/_ 250ppm).
We already use such circuits on some of our more "fun" DAC designs
...Its all rather academic as Søren does not wish to explore the possibility (as I say, I'm sorry I posted on this thread as originally I had no idea that this was a commercial endeavour).
Last edited:
Typically I see 2dB - 3dB variation in PN across the pull range close to the carrier - further out from the carrier (say below 1KHz) the "Increase" drops below the NF - so nothing thats too worrisome, especially in consideration to the huge leap in performance over the SI PLL based solution....
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.