advertisement


Pro-Ject Pre Box S2 digital DAC

USB as the DAC operates as the Audio "Master clock" and eliminates the Jitter problems inherent with SPDIF (Coax).
 
One more issue, and I bet some tester is going to be a little embarrassed over this: left/right channel on DSD64 (and I would guess all DSD) is swapped.

Or maybe its my DSD files that are wrong, someone else better test this as well.

hi @MagnusH,
you are sadly right. I tested with regulard (ripped SACD) DSD64 files as well as upsampled PCM->DSD via Roon plus some DSD downloads. in all of the scenarios when playback is DSD Left and Right channels are swapped.
 
Hi Guys,

I've not confirmed the DSD channel swap - but the root of the problem is with Pro-Jects tradition of swapping the orientation if the Left / Right RCA output connectors.

When we designed the S2 DAC series we designed the Audio outputs as you would expect - Left Channel on the Left and Right Channel on the Right - but unknown to us, this was modified by a third party.

To avoid having to redesign the PCB, we swapped the L/R channels in the ESS configuration registers - but its quite possible that the ESS has a design bug and only swaps the PCM channel data and not DSD!

There are two reasons we did not notice this bug:-

1. Early engineering units where hand modified by Pro-Ject's manufacture (unknown to us at the time) - if you look closely at some early pictures you can see Wirelink "modifications" to the rear panel RCA - these units where used for Beta testing.

2. We run our own custom software build with the channels "correct" on the rear panel (Left RCA to Left channel, Right hand RCA to Right Channel) so that we dont keep connecting the power amps incorrectly during A/B listening tests.

If its a confirmed bug in the ESS then we will try to swap the channel data in the XMOS (but there might be issue when transitioning to headphones) - and DSD channels will always be swapped via SPDIF inputs (not sure if this is a real issue).

The best thing we can do is to release a custom software that ignores the rear panel labelling and has the correct connector orientation then we dont need to channel swap in the ESS DAC all will be good (in both PCM and DSD mode).

hi @JohnW,
i think large majority of us would prefer to swap cables assuming all sources/PCM/DSD etc come out in a same way (option 2) ... guess it also might be possible to detect DSD on XMOS and swap channels before it reaches ESS DAC. Not sure if people really use Coax input for DSD ... and you tackled that as well....

So for me to have a 'perfect' unit (dont care about balance, can live witohut USB volume control) is to
1. fix click/pop on master samplerate clock change
2. dsd channel swap to become correct

thank you and would really appreciate if you could tell us if any chance for firwmare upgrade any time soon?

Thank you !
 
So for me to have a 'perfect' unit (dont care about balance, can live witohut USB volume control) is to
1. fix click/pop on master samplerate clock change
2. dsd channel swap to become correct
Agreed! I would not mind USB volume as well, but its no where as important as above.
 
So for me to have a 'perfect' unit (dont care about balance, can live witohut USB volume control) is to
1. fix click/pop on master samplerate clock change
2. dsd channel swap to become correct

Thank you and would really appreciate if you could tell us if any chance for firmware upgrade any time soon?

I'll have Jarek work on it towards the end of the week, maybe during the weekend we will have some code for you to test.

The complication with the L/R XMOS channel swap is that the channel swap needs to be "on the fly" to insure the output to the headphones is correct, and we are not sure how the XMOS will handle this - hence it being better just to have the main outputs as they where designed to be (following the convention of Left RCA to left output, Right RCA to Right output when looking at the unit from the front) - this is how the hardware has been designed.

We will release Beta software following this convention.
 
Agreed! I would not mind USB volume as well, but its no where as important as above.

We also need USB volume control for the MDAC2 project - trouble is Jarek's says its not easy and will require 3-4 days work, trouble with XMOS you never know how long things will take.

Poor Jarek's really not a software guy, and hes really unhappy working with XMOS and feels the pressure when bugs are found as I've pushed him to work outside of his comfort zone.

We really need an embedded software guy to work with us - we are searching but its not easy - forget finding anyone with XMOS experience here in CZ!

Good embedded software guys are like Gold dust, if they are any good then the first thing they will do is get the hell out of the Czech Rep! so we are looking for a part time university student or somesuch.
 
hence it being better just to have the main outputs as they where designed to be (following the convention of Left RCA to left output, Right RCA to Right output when looking at the unit form the rear) - this is how the hardware has been designed.

May I ask why for left/right channel choice is from looking at the unit from the rear? If active speakers are used, cables goes to oposite direction.
 
Gints,

Sorry Yes - I meant "from the front" - not how Pro-ject have traditionally arranged there connectors.

We designed the product (before the change by the nameless third party) to be logically "correct" as you describe :)
 
Someone mentioned earlier that there is no galvanic isolation on the usb input is that true?
I think UpTone ISO Regen has galvanic isolation, or the new iFi nano-iGalvanic3.0, igf you want to add it to the S2 DAC (or any other DAC that lacks it, which is most of them).
 
Btw, if you use an vbus isolator, and supply the DAC with powerless USB and connect USB power on the micro, will that work? And if so, will this in effect provide galvanic isolation (or maybe partly)?

Like this:
Computer USB -> vbus isolator -> USB input on the DAC
iFi iPower -> micro USB on the DAC
 
Yes, the DAc is designed to be powered via the Micro USB - if power is present on the Micro USB it uses this supply instead of the Main USB power.

The DAC "Autoswitches" to the Micro USB if power is present on this connector.
 
hi,
i saw articles on Pro-ject S2 PRE DAC which talked (with measurements) about quality of isoltion/filtering (2nd page on http://www.box-designs.com/inhalt/en/pdf/preboxs2digital.pdf).
i do have Intona galvanice isolator and i would need to do lot of listening to see if i can hear any difference - but i did try and answer is not much; so my guess is - powering S2 DAC with iFi iPower and with or without Intona Galvanic USB isolator and looking at those graphs - i doubt anyone gets any real 'noise' into the S2 DAC.
 
I currently use mine with a iFi nano iUSB3, and it seems to make a difference with it but I would guess its mostly because of the better power. The iFi nano helped the DragonFly Red much more.

Btw, the pop/crack at sample switch and the issue with left/right channels in DSD has made me up-sample everything to DSD128, and then switch the left/right RCA connectors. No pops/cracks, and channels are correct (and it sounds awesome) :)
 
I'm just curious why dacs don't have galvanic isolation as std these days, surely it would not add that much to the price.
 
Theres no cheap / simple way to isolate the USB2.0 interface - while its easy to isolate USB1.0

The Pro-Ject DAC is also USB Powered....
 
I got another DSD problem: when I play DSD, or up-sample to DSD, I sometimes get faint click sounds. To reproduce, play a low-volume calm part of a tune, then press pause - play - pause - play (repeat if needed) until it starts clicking in speakers (sometimes in the left, sometimes in the right). It only clicks during music.

I would blame interference, noisy power or something similar if it weren't for the fact that when doing the same with PCM (tried 384khz) there are no clicks.

Note: this could also be an issue with Roon, but lots of people play DSD with Roon so if its Roon related its not something that happens on other DACs.
 
We spent 2 days looking into the Issues:-

DSD channel swap - no problem to resolve.

Sample Rate Clicking is a ESS bug, I need to talk with the designer today to see if there is a work around.

We have managed to reduce the instances to "about" 1 in 15 changes... this might be as good as we can get if there is no better work around.

We will try to recreate the DSD popping...
 
Sounds good, here is how I easily recreate the clicking myself:
Play Brother in arms by Dire Straits, either DSD version or up-sampled to DSD
Go to 1:30 in the tune
Press play - pause - play (repeat one or two more times if needed), and I hear audible clicking noises when playing.

The problem appear on all tunes, not just this one, but its a good test tune in this case. But as I said, this might be a Roon issue, I don't have anything else to play DSD with so can't compare.
 


advertisement


Back
Top