[Rob Watts]

It depends upon the DAC - Dave requires sync, so length is important, but all the other DACs use the AES3 dual data flag so don't require exact sync.

 

[474194]

Hi Rob,

My new Mojo was plateauing so I switched to my goto battery-powered Desktop, but portable footprint HQ optical source I've dedicated to the H2. The pacing, rhythm and timing was immediately noticeable. I cannot believe a change in source can change the Mojo from something that's just outputting sound to something that brings out emotions and gives me tingles on some tracks due to different pacing. It's like the music just slows down just a bit and takes a deep breath and breathes. With the other source, it's more akin to something like a telly optical out rhythm which gets fatiguing over time.

Anyways, onto my question. Since I changed to a more optimal source on the Mojo, my gut was telling me I should be able to play DXD since I'm using a HQ source and a Lifatec glass optical cable (rated max @192Hz, but unofficially?). No worries, it plays @ 352Hz with no clicks or hitches. The issue is the light on the Mojo is light blue. It looks more like 176Hz than the 352Hz on the color scheme. I know optical is max at 192Hz on the Mojo and the Mojo doesn't dither down to 176Hz. Is there a chance I'm playing @352Hz?

http://lifatec.com/toslink2.html

Cheers

 

[Rob Watts]

I think the chances of getting 352k from optical is close to zero - the optical tx and rx are not up to that - even if the FPGA is capable of decoding 352. So Occam's razor suggests that your source is downsampling to 176.4 kHz.

 

[474194]

Thanks RW. That's likely what is happening as the the color indicator light on the Mojo is pointing to 176.4 kHz. DXD is new frontier so needed sometime to process. I intend to stick with Redbook going forward, but it was an interesting experiment.

I double-checked the specs (Yay!, zero USB involvement/interface to tamper with SQ; just pure optical throughput) on the optical decoder and source code to confirm nothing over 192kHz:

https://statics.cirrus.com/pubs/proBulletin/WM8804_WM8805_ProductFlyer.pdf

https://github.com/raspberrypi/linux/blob/rpi-4.1.y/sound/soc/bcm/hifiberry_digi.c

I just enjoy the synergy with this battery-powered optical source and your DAC tech. It's the difference between listening once a month to listening everyday now. I used a $300 USB cable since H2 launch and the Chord experience really didn't resonate until I switched to this optical source last summer.

I wasn't sure if I would be able to replicate maxing the Chord mobile experience with the new Mojo. So I tried a different optical source (maybe same chip, but not as well-implemented) for a few weeks with the new Mojo until I gave in and went with my goto optical source. It makes all the difference in the world.

Anyways, thanks for answering my noob questions. I would of quit this hobby if your DACs had not arrived on the scene.

Time to go on a extended hiatus from this site soon. Looking forward to the next iterations of Chord mobile products.

 

[musickid]

hello Rob,
I'm using a high quality "normal type" power strip from olson. I'd like to know if there is any advantage in investing in a dedicated mains power conditioner or power regenerator which a lot of head'fiers use. If so can you recommend the exact type and maybe a make or brand. Many thanks mk.

On a completely unrelated point i did read that dave also perfectly recreates the original analogue signal albeit from a 2016 review so how does this figure when comparing the mscaler? Unless i am mistaken. It states the WTA filter solves all timing and rhythm problems (wrt to dave).

 

[Rob Watts]

Probably not - and of course mains sensitivity depends upon your gear and how you power your system up. The problem with mains conditioners is that nobody understands how crucial RF noise is, and do not properly filter from 100kHz to 10GHz. It's horribly complicated to do that on a mains filter. Most mains conditioners are only worried about 50Hz or 60Hz power delivery, and that's useless - the approach is likely to create more RF noise, making it sound brighter and worse...

But by all means try them out - strictly before you purchase or on a sale or return basis - and if it sounds brighter, or no difference, then return it. Mains treatment, if it represents an improvement, will always sound warmer, smoother or darker.

The issue of RF noise, and noise floor modulation and how sensitive we are to this problem came to the fore a couple of days ago. I am working on several new projects, and one of them I was doing listening tests on a new HF filter - like the one in Hugo 2 (the white to green filter). Now this filter with 44.1 is technically minute, as HF noise is extremely small when using CD. But I could hear a difference, albeit small. So I ran a single blind test with my son, and asked him to state which one sounded warmer, or if they sounded the same. He could spot the HF filter, identifying it as warmer sounding - although he confirmed it was a small change.

The reason I was doing this listening test was the Hugo 2 HF filter gently rolls off at 20kHz - it's about -1dB down. One reviewer commented that this was the reason for it sounding warmer, and I thought that was unlikely - although it's possible. The new filter actually has 20kHz compensation built in, and nett boosts 20k by +0.4dB (to compensate for other filters). But even with this boost, it still sounds warmer, proving that noise floor modulation is the cause of the HF filters change in sound quality.

Moving onto your point about a review stating that Dave perfectly recreates the analogue signal, then "perfectly" certainly didn't come from me or Chord. I am very careful to, when possible, define how good a given process is; that's why I have said that Dave is about 13 bit accurate worst case compared to the ideal sinc function, and the M scaler is better than 16 bits. I always find it very amusing looking at some competitors DACs with perfect this and perfect that nonsense. Nothing is ever perfect. And from a business POV, once you have created the "perfect" DAC, how can you promote the "more perfect" DAC in a couple of years time when the replacement comes along?

 

[musickid]

Great response cheers Rob. wishing everyone on headfi all the best through these very sad times.

 

[HeeBroG]

The reason I was doing this listening test was the Hugo 2 HF filter gently rolls off at 20kHz - it's about -1dB down. One reviewer commented that this was the reason for it sounding warmer, and I thought that was unlikely - although it's possible. The new filter actually has 20kHz compensation built in, and nett boosts 20k by +0.4dB (to compensate for other filters). But even with this boost, it still sounds warmer, proving that noise floor modulation is the cause of the HF filters change in sound quality.

Hi Rob.

How did you reduce RF noise in the second scenario to make the filter with the nett +0.4dB boost sound warmer?

Geoff

 

[Rob Watts]

Its a digital filter using two bi-quads - I simply adjusted the Q on one of the sections to give a small boost at 20k. It is overall 4th order set to 50kHz, so it gives good attenuation above 100k but without the drop at 20k. The warmth is down to reduced out of band noise, giving less noise floor modulation - although in both instances (with or without) it wasn't a measurable change in modulation.

 

[musickid]

Hello again,

As the days are long now i've been reading around the source i use which is a battery powered ipod touch 7th gen as a roon endpoint usb into hms. In the hms manual it states 5v power rail is needed on the usb input for hms. Is it possible that my class 2 certified usb cable attached to the apple adaptor is providing far less than 5v power into hms? If this was the case with usb would you just get no signal with usb into hms?

Also out of interest with dave resolving to 13 bits at worst what deficiencies are there in SQ when compared to hms which resolves at greater than 16 bits. This point is very important for me to understand. Thanks again.

 

[Rob Watts]

Don't worry about the VBUS voltage - the USB decoder and galvanic isolation will work as low as 3.3v. VBUS (the USB power voltage) will be better than 4.75v at the very low current being drawn even with poor cabling.

The difference in SQ with reconstruction of timing at better than 16 bits against 13 bits depends upon the listener - but I personally can't do without the M scaler.

 

[Uncle Monty]

On a similar point (I think), the Antipodes EX has the option of connecting via USB 5v ON or 5v OFF sockets - which should I use to connect to my M-Scaler?

 

[Rob Watts]

The USB on all Chord desktop DACs and M scaler need 5v on the USB VBUS, as this powers the galvanic isolation. The other benefit to this is the USB decoding process is on the source domain, so the FPGA just sees data, and isn't affected by the USB decoding noise. So you must turn on USB 5v.

On the portable DACs VBUS must also be on, as this acts as a flag to tell the FPGA that a USB is connected.

 

[474194]

^^

Yes, I made this error on my first USB cable. I want things as cut down as possible so opted to have the 5V cable removed (gold / bronze thin cable). Without it, the Hugo2 could not do a USB handshake. PITA to ship to OZ.

Some USB devices much more VBUS-friendly:

 

[Siz.]

...

The issue of RF noise, and noise floor modulation and how sensitive we are to this problem came to the fore a couple of days ago. I am working on several new projects, and one of them I was doing listening tests on a new HF filter - like the one in Hugo 2 (the white to green filter). Now this filter with 44.1 is technically minute, as HF noise is extremely small when using CD. But I could hear a difference, albeit small. So I ran a single blind test with my son, and asked him to state which one sounded warmer, or if they sounded the same. He could spot the HF filter, identifying it as warmer sounding - although he confirmed it was a small change.
...

Rob - does this apply even if the source to DAC signal is through optical?