[Rob Watts]

I have worked with a number of semiconductor companies with half a dozen different devices.

There was zero interest in a high end DAC chip as the numbers are too small.

As to a WTA filter for an M scaler, to do that it would cost way too much as an ASIC. It would have to be 40nm or better, and the production mask set alone is 1M USD. Together with design and verification, and prototype costs and initial production run it would be a 2M USD project. You would never sell enough to justify the NRE cost. Moreover, it would eat away at my valuable time...

 

[alphaman]

There was zero interest in a high end DAC chip as the numbers are too small.

Is this really an accurate estimate? Many, many heads and CEOs (= decision makers) of major electronic manufs are audiophiles. And certainly, given the apparent use and endorsement of Chord DACs in major studios ... that reputation --at the very least -- would allow Chord to tap into (no pun intended), venture capital and seed money.
Mr Watts, reach out the folks on Sand Hill Road, in Palo Alto. Or you may even reach out to a military contractor. Make 'em jealous of what dCs has accomplished with "defense" budgets. More accurate audio D/A in solider helmets and drones might be a good pitch.

And to have other aerospace engineers concomitantly refine your product might not be a bad thing.
Then, we could simply put the WattsUp chips into same devices as Topping and SMSL and Cambridge Audio.
This would, unfortunately, separate the wheat from the chaff. I.e., the existing Chord PCB layout and PSU design and output stages would be out-factored from the sonic equation.
Hmmm ... some may use vacuum tubes, while Topping might use 5532's,
Nah ... forget that whole paragraph

 

[musickid]

My d10s usb to optical convertor from my imac to my mscaler into tt2 results in a quieter sound than using usb into my mscaler (through speakers and headphones??). Does anyone know why this is. ie an optical source sounds quieter (much quieter) than a usb source into mscaler with a d10s ddc usb to optical convertor? Strange but its one of these chinese ddc//dac's from amazon.

my douk ddc however does everything perfectly into mscaler apart from 192khz. LOL

 

[580847]

I enjoy your Chord Arc over the years.

Last month, I started noticing major Soundstage depth and width improvements after powering my USB optical cable (strongly recommended by Chord legend romaz) with the best passive unregulated almost zero impendance 5V power supply available (No Mains). Just like Rob Watts's theory if you lower noise floor modulation, other SQs will come to light. For I, it was Soundstage depth and width. But this unexpectedly happened only after six weeks. I guess it was Brain Burn-in. Another RW theory that ended up true. This was only on my dedicated Gaming machine, so it was in a game not Music.

But this gives me motivation to try this for Music in 2024.

The reason I bring this up is because the iPod is not industrial strentgh, so it may be good to plan obsolence. Plus, I don't care for DDCs as that gets messy. I don't have time to explain why Toslink may not be best to source an HMS non-decoupling-wise.

I got sick again last night so cannot comment anymore. I took all of December off work and had December downtime holiday energy, but the bug got be again. I think for 2024, there are USB decoupling alternatives to Toslink.

I'll try to give you a sample of what I'm proposing.

First, I'm very, very pro-toslink so USB is not something I would waste time on in the past. But my Gaming experience significantly improved with USB optical. Initially, it sounds like I'm Gaming in toslink. Then the Brain Burn-in after six weeks with the Soundstage improvements.

I have high standards, so my requirements are CPU Power, RAM and PCI-E (direct path to the CPU so no internal PC noise) along with being powered Off Mains.

I think this can be accomplished in 2024:

1. Off Mains Power CPU <--> Off Mains USB optical
2. Off Mains Power CPU <--> Normal USB cable since source already Off Mains
3. Mains Power CPU <-Power isolated-> Off Mains PCI-E

With the release of the new Raspberry Pi 5 (Made in UK), I think this is worth exploring at a decent value. The only problem is Raspberry Pi 5 is now 5V/5A but I think this can be overcome because my battery power sources scale to at least 100A. Crossing fingers for more 5V/5A battery packs in 2024.

The CPU Power is 2.5X-3X it's predecessor so getting close to NUC territory. Memory latency (RAM is vital for SQ) and PCI-E has improved:





So imagine a PCI-E USB card instead of a NVME drive in the much improved PCI-E connection.

I'll try this out in 2024 when I'm feeling better. I was hoping I would be bug-free in December but it just didn't work out. If I get Soundstage improvements and it sounds like Toslink, I will consider it a success. But I have all the passive tools to attempt this feat.

It can work as your Main or an endpoint.

PCI-E is used for AI, Graphics Cards, etc. so it's a solid Audio interface:





Previously, I was considering a Intel Xeon Off Mains but the value is terrible and you have to scale large passive power supplies. Note the PCI-E slot and green 2-pin 12V Power input:

But now with the Raspberry 5, I think the value performance is there to attempt a solid Off Mains USB source without breaking the bank. The Intel Xeon get's close to $3000 so zero value just for Music.

Ciao for now.

 

[580847]

I stumbled upon this Car Lead Acid Battery BMS:

I can recharge via Solar power. Output 12V and 5V. Not a recommendation as there are better power solutions now. My Passive Off Mains Power Supply destroys Car Battery Power. But seems like a fun project if you have a car battery collecting dust.

Yes a car battery is the best possible PSU - huge dynamic current, ultra low impedance and no noise. But the real benefit is batteries, and isolation from the mains, so a car battery is not needed; just use a USB power bank.

https://www.head-fi.org/threads/watts-up.800264/post-15600310

In 2020, car battery was the best possible passive PSU. In 2024, I don't think that is the case anymore as more passive power supplies arrive on the scene (huge dynamic current; check dynamic enough to power the dynamic low and highs of public buses/trains/vehicles, ultra low impedance; check almost zero impedance and no noise; check zero noise).

I'll check back in March 2024 after recharging my health again. The Raspberry Pi PCI-E connection boards not released yet but hoping early 2024. Thanks again romaz for the USB optical suggestion. It ended up working out, but only after powering it with super clean passive 5V power.

EDIT: Oh, I believe technically I have bulletproof galvanic USB isolation now since the USB PWR is a completely off mains and USB optical decouples. Bulletproof USB Galvanic Isolation >>> USB Galvanic Isolation

 

[musickid]

very interesting reading. however my technical limitations are present lol. im still looking for a usb to optical ddc that does 192khz flawlessly. after comparing optical to usb in in on mscaler and putting aside all the impressions by other members optical has the edge over usb for me in that it sounds more natural and lifelike without any colouration. i hope you fight off this bug very soon. mk

 

[580847]

Technically, USB optical is Toslink non-glass optical. So zero colouration. No RF pollutants that supposedly increases transparency because no traditional cooper wiring is used, it's optical. Same material used in Toslink cables.

I was not a USB believer either. It was suppose to be just for Gaming. But if I can replicate the increase it Soundstage SQ on the Music side, then I'll be a proponent of USB optical for Music.

USB DDCs are for convenience not performance. That's why I want to experiment if you can duplicate Toslink benefits via USB optical. Pre-2024, I understand it's all about Toslink but that may change if this USB 2024 Project works out.

USB DDC is easy. Just "Add to Cart" and checkout.

Thanks, I had all of December planned to be healthy and full of energy. But at least I got out my proposed plan and now it's time to grind and burn-in and hopefully can reproduce the USB optical magic I achieved for Gaming (PC on Mains).

EDIT: Reminder, I believe RW mentioned he uses a 2M Lifatec Toslink for 192 kHz. Anything under 2 Meters, I don't think he attempts 192 kHz with a cheaper, generic Toslink cable. I could be mistaken since I don't have time to research.

A problem I see with USB optical is you need a USPCB full size micro 'B'. Uptone suspended production.

https://uptoneaudio.com/products/uspcb-a-b-adapter

Mistaken. I thought .5m for generic not 5m:

I have the 5m KabelDirect and 2m Lifatec - both work perfectly with 192k. I was pretty astounded that a 5m plastic fibre that is inexpensive could do 192k.

On the M scaler after doing careful AB tests, both cables sounded absolutely identical.

https://www.head-fi.org/threads/hug...nics-the-official-thread.885042/post-17288140

https://www.head-fi.org/search/15814951/?q=lifatec&c[users]=Rob+Watts&o=relevance

I just think instead of sinking more costs into Toslink that opporunity cost may be better serve going USB optical or Off Mains Source + USB cable. But since you are a zero colouration fan, USB optical. The USPCB provides zero colouration also since zero traditional cooper wires.

 

[580847]

I moved on from discussing details about my production chain and focusing on the USB 2024 Project. I just don't have the energy if something is already in production, but I have motivation for present/future projects. So I'll try to give some comments on what to look for on SPDIF devices.

For SPDIF devices, to get to 192KHz, you need specific OCXO clocks (24MHz, 49MHz or 98MHz). Double-check the specs on your DDC if it has one of these clocks.



If it's a embedded clock (no OXCO), you have to hunt down the datasheet to check the frequency of the oscillators it contains (PLL). An example of a popular embedded SPDIF I guess is WM8804 chip. It's been years since I thought about this stuff so due diligence recommended.

https://www.cirrus.com/products/wm8804/



Your SQ mileage varies heavily with the quality of clocks. That's why I'm trying to get USB decoupled and working like Toslink. IMO, it will be much cleaner into an HMS because Toslink up-sampling gets messy. Not enough to label toslink with cheap, generic clocks a bottleneck into an HMS yet. But DDC is possibly a convenience, isolation solution at the cost of pure SQ. I believe it's potentially possible to have the best of both worlds with USB optical. No worries in the back of your mind about the signal integrity of toslink feeding an HMS to up-sample. USB optical shouldn't taint the signal integrity. Your basically building a transport carrier to feed an HMS. That's the job of this source. I haven't thought about it too much but toslink feeding HMS could be like feeding an analog signal into a 100% digital device (HMS). There can be variance. USB would keep at 100% digital feeding into a 100% digital device (HMS). If your feeding into a DAC direct, toslink is perfect but I'm starting have second thoughts about toslink feeding to up-sample.

Again, I only started thinking about USB optical as an alternative to Toslink last month when I experienced wowness on the Gaming side. Before, it was exclusively Toslink since the romaz, ray-dude days. I had the USB optical cable for years, it's just this Summer I gave it pure native 5V passive power. Previously, the USB optical ran on Mains for convenience reasons.

Since I only focus on Redbook, my Toslink runs at 5MHz to get to 44.1KHz to max out on Toslink Redbook performance. Works super well on well-mastered Redbook and the Chord DAC accepts the toslink input like a champ. Also, running exclusively @16-bit Redbook. Again, the Chord DAC processes the toslink input like a champ. No need for unnecessary 32-bit conversion before feeding the Chord DAC.

That's why for Redbook, I focus on the max clock to get max Toslink Redbook performance. But for SW/HW up-sampling, I think that use case is better served going USB optical. All the benefits of Toslink (isolation, perfect neutrality with zero colouration, decoupling) without the liabilities of Toslink. I only started thinking about this once I read the new HMS might be around the corner and I already conquered Toslink sources along with now Gaming. So the next major challenge is an HMS source. Again, I could be completely wrong as I only started considering how I would feed a HMS this month.

Okay, I'm spent, check back in 2-3 months.

Ciao

 

[GHOSTUK]

I have a question for Rob if you don't mind.

Will your Ultima DAC have the ability to have its volume control (assuming it has one) adjusted by the streamer?

I use Roon and it has the ability to control a USB DACs volume control and I would love to see this. For reference I use a Lumin streamer though its Roon sending the volume change instruction.

Thank you, huge fan of your work.

 

[580847]

Another issue I just considered is the reliability of feeding HMS an optical input @192KHz. Toslink was not officially designed for 192KHz, so it may work on a few tracks but for "Production Use" feeding an HMS? Having a HQ clock will improve reliability, but still possible variance feeding an HMS use case?

Anyways, I'm too curious if I can replicate the "magic" I had with Gaming on USB optical with Music on USB optical with no HMS in the mix. It's amazing how adding more Soundstage depth and width adds to the Gaming experience. I definitely prefer USB optical Gaming > Toslink optical Gaming as it streamlines the chain and let's the Chord PLL do it's thang with no variance.

Here's an recent example of the headaches of 192KHz with a HMS/TT2 setup:

https://www.head-fi.org/threads/chord-m-scaler-optical-input-refusing-192-khz.969447/

Cheap embedded clocks with the Bluesound + various Toslink cables. It's a challenge. So even if it's working, I'm skeptical on the reliability for "Production Use".

I just wanted to give you another parallel universe to consider.

---------------

Not Rob, but from what I remember you can set software volume control to vary if that's what one really desires, you just won't have bit-perfect output.

I fix my output at 100%. I can set it to vary, but under <100% it's not bit-perfect feeding a Chord DAC.

For example, I created my own low latency realtime Audiophile OS DAP with an iPhone dongle (Modern 2024 iPod since it uses an Apple DAC), it's volume control can vary.

But if I remove the iPhone Dongle and plug in a Mojo, I have to set volume fixed to 100% otherwise I will not get bit-perfect. I can set volume to vary, but it's not feeding the Chord DAC correctly.

---

BRB March 2024

 

[GHOSTUK]

Here is the device volume option I'm referring to. If supported it will control the devices built in volume control and not in roons DSP.

 

[580847]

Here is the device volume option I'm referring to. If supported it will control the devices built in volume control and not in roons DSP.

Sorry, I won't be of much help:

1. I'm not in the HMS/TT2 ecosystem
2. I don't stream, no Roon
3. Not enough information to be able to process my own statement. The fixed volume looks good selected in the drop-down box, but the other variables "Device volume" and "Fixed volume" is an unknown.
4. If you are using software DSP, that is another variable that affects volume.

So too many unknown variables for I to make a statement. I don't know why one would even want to fiddle with software volume when you have a physical remote in the HMS/TT2 ecosystem. If the ergonomics of the Chord default remote are not friendly, you can program your own remote to your liking or use a third-party remote control. Just set software volume to 100% fixed and get bit-perfect goodness. Software volume ergonomics are usually not too friendly.

I believe Logitech Harmony Remote Control is discontinued, but just using it as an example. Learning remotes are readily available.

But there are plenty of others in the HMS/TT2 ecosystem that can answer. I recommend searching bit-perfect volume control to understand why software volume may not be ideal:

I did a quick search and ran into some interesting:

It is always better to give Mojo bit perfect files and let Mojo do the work, as the processing within Mojo is much more complex and sophisticated than a mobile or PC.

So when you have an app that has a volume control, and no bit perfect setting, then set it to full volume on the app on the assumption that this will keep the data closer to the original file.

The volume control function on Mojo is much more sophisticated than the PC as I employ noise shaping and I do the function at a very high internal sample rate. Hopefully using the volume set to max on the app will mean the volume coefficient is 1.0000000... so it will return the original data.

Rob

https://www.head-fi.org/threads/chord-mojo-1-dac-amp-☆★►faq-in-3rd-post-◄★☆.784602/post-12274209

That sounds very strange - all sample rates (44.1, 48, 88.2, 96, 176.4, 192) would be supported if it was bit perfect. I don't use Roon, but would suggest you try JRiver on a 30 day trial, and that will tell you if it's bit perfect and will also enable DoP for DSD tracks so you can test it that way.

https://www.head-fi.org/threads/hug...nics-the-official-thread.879425/post-17339655

Sir Rob doesn't seem to use Roon either...

Good luck, I'm offline due to injury but I will check back in March 2024. I was hoping to do some Holiday postings, but it didn't work out.

 

[580847]

The reason I don't stream is I don't want to go down that rabbit hole. It will be like creating another hobby to jump into by procuring HQ Networking equipment. Plus, the recent layoffs in the Music Streaming industry doesn't give me confidence. I like to have horse-blinders on and focus on one task at a time when focusing on a project.

TT will accept 32 bits quite happily. The potential SQ losses is in the conversion from 16 to 32 bits, not TT receiving a 32 bit signal.

https://www.head-fi.org/threads/cho...c-amp-impressions-thread.756029/post-12789279

1. I'm 100% Well-mastered Redbook, that's why I run my clock @ 5MHz/16-bit. These well-mastered media will not be around forever and I want to do my best to collect them all before they disappear. Pure 16-bit, pure clock performance so it's "Mission Accomplished" for my Off Mains Toslink Redbook chain.
2. Gaming. Something I never considered, but by following RW teachings, a wild "Soundstage" Pokemon appeared! after a few months of Brain Burn-In. With the Soundstage improvements by feeding a USB optical cable the best unregulated native (no up/down conversion) 5V power supply in the world, Gaming SQ is now a thing for I.
3. Hires. I believe Hires is just a tool to sell products, but my 2024 Resolution is to build a Off Mains HQ PCI-E source. With or without the USB optical since the source is already Off Mains. If it works out, it should work well for a possible new HMS in the future. I have nothing in the tool chest to feed a hardware up-sampler ATM nor to playback basic Hires. As an alternative, I can max out Hires by procuring a 12MHz/24-bit clock for optimal performance but that's a rabbit hole I cannot justify. Opportunity costs better focused on a new HMS.
4. Future 1M+ taps Chord source. Since it's "Mission Accomplished" for my other immediate needs. I'm starting to turn my attention to the new HMS. My requirements are a Off Mains USB optical source, Solid Core BNC cables w/ PTFE and the best Passive 12V Power Supply in the world. I already bought some resistors for my 12V Power Supply to experiment in 2024. Dealbreaker if I had to use a Off Mains Toslink source and/or cringey BNC w/ ferrites with the new HMS. I could just easily punt / pass if I had to use the latter. I've grown a new appreciation now towards PTFE since RW mentioned they ward off evil spirits and emphasising PTFE's valuable properties. If the solid core w/ PTFE BNC was not an option, the project would come to a grinding halt as I don't like weak links in the chain.

I'm really good with just my Toslink Source as I consider it a alternative to the current HMS so wait and see on new HMS. In the Chord threads, some prefer standalone Toslink over standalone HMS. Super optimising a Toslink Source just adds dimensions to being a current HMS alternative. It remains to be seen how a Super Optimised Toslink feeding a HMS versus USB optical feeding an HMS will work out. Again, I would lean towards USB optical since zero variance. It's use case dependent whether for Redbook, Gaming, Hires or Up-sampling for I. No Toslink/USB bias now as long as Off Mains and decoupled. I did have a Toslink bias before I experienced USB optical for Gaming last month. Once the Soundstage leveled up, it opened news doors on my future Source trajectory. Of course, perfectly neutral while Gaming as it's optical. Nothing false, no colouring, no added preservatives, no trojan horse RF transparency variance. It's grass-fed optical.

 

[Rob Watts]

I have a question for Rob if you don't mind.

Will your Ultima DAC have the ability to have its volume control (assuming it has one) adjusted by the streamer?

I use Roon and it has the ability to control a USB DACs volume control and I would love to see this. For reference I use a Lumin streamer though its Roon sending the volume change instruction.

Thank you, huge fan of your work.

Thank-you for your kind comments.

For volume to be adjusted inside the DAC from streamer commands I would need two things to happen - firstly for Chord to request such a feature, and secondly (and more importantly) the USB decoding device to have a command stream to send to the FPGA so that volume can be set. From a control POV I would have the minor headache of sorting out what is the master volume control - the streamer or the DAC. People would be upset if their DAC remote volume control no longer worked....

This is not something I am planning with Ultima though. I have too many much more important things to worry about - like getting better SQ/musicality!

 

[Reactcore]

I would need two things to happen - firstly for Chord to request such a feature, and secondly (and more importantly) the USB decoding device to have a command stream to send to the FPGA so that volume can be set.

Or go the opposite direction.. let the player software incorporate your already designed volume protocol in its audio stream (but that would mean licensing out some IP)

Let them do the work..