Which one to buy: UpTone USB REGEN or Intona USB Isolator?

[Superdad]

Who knows, by then, someone else may soon come out with another device with better data isolation that works in all applications.

 

:) Would not surprise me...

[Superdad]

Mutec has NO usb input and output (like the Intona or Regen). It is just a bit more advanced USB to SPDIF converter.

 

I'm sorry Adam, but this is just not quite accurate. First of all, there is absolutely a need to isolate the USB interface to prevent HF interference from the computer affecting the signal processing and causing jitter and noise problems. Also, the MUTEC MC-3+ Smart Clock USB is a USB interface, audiophile re-clocker, 10 MHz ready master clock, DDC (digital-to-digital converter), and digital switching matrix. So calling it "just a bit more advanced USB to SPDIF converter" is a bit of an understatement to say the least.

 

Best regards,

Julian

 

Hi Julian:

 

The MC-3+ looks to be a VERY fine product, but I think you misunderstand what Adam meant. Since the Mutec does not offer a USB output, it is not useful to people with USB input DACs.

 

(And this thread is not the place to rehash arguments with regards to whether USB or S/PDIF is the more desirable input to a DAC; though just because the Mutec or other converter can produce a excellent integrated-clock S/PDIF signal, does not mean the DAC's own receiver and handling of it is of equal quality.)

 

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

 

For various reasons I have not posted at all in this thread, even though REGEN is in the title. Yet by the way, I too bought an Intona--s/n 53--(though it needs firmware upgrade as it is very finicky with my XMOS-input DACs and some cables).

 

Just to be clear though, the Intona's sole and very effective (audio-interest) function of galvanic isolation does not duplicate or obviate the REGEN's function of improving USB signal integrity and impedance match (or the REGEN's secondary function of providing very clean VBUS power to DACs that need it).

 

And given that the Intona has a single clock oscillator--located on the dirty input side--for the entire device (for both PHY chips, both FPGAs, and fed over the isolator), and that the SilLabs TX/RX RF-type isolator and the FPGAs each add jitter (350ps for the SiLabs alone), the output of the Intona really does benefit from the clean reclocking/generation of the USB signal the REGEN offers.

 

Thus the ideal set-up (short of having a galvanically isolated version of the REGEN) is to place the Intona before a REGEN. Then you get the unique benefits of both. That's what I think quite a few folks are doing--with stellar results.

 

Cheers,

--Alex C.

[Superdad]

For those of you with problems, perhaps the cure is to find a cable that has the shell continuous. How to find.....

 

That is all quite interesting, but it does not explain or justify why the Intona isolator is so sensitive and particular about USB cables.

 

I too have a drawer full of stock Belkin Gold and other quite standard cables, and for the life of me can not get my Intona unit to work with my XMOS-input DAC without a REGEN betWeen the Intona and the DAC. And even with the REGEN in, if I try to use the Curious cable I can not get the DAC recognized as able to do anything beyond 96KHz (USB 1.1 Full Speed). Rather frustrating.

[Superdad]

Maybe your XMOS based DAC needs more power Intona is able to provide ?

 

Actually, my XMOS-input DACs do not require ANY 5VBUS (can tape pin-1 if desired).

 

 

So the LPS was replaced with a 5volt mobile phone auxiliary battery. Wowser! The bass was now restored along with the benefit of the natural sound provided by the REGEN. The surprise was an increase in depth. The result was the best sound ever achieved in my system!

 

 

Swapping the Green and Amber REGEN had little, if any impact on SQ.

 

So Larry, are you saying that you prefer the combo of the Intona feeding the REGEN over just the Intona alone? Given that the title of this thread sets up what I have contended is a false dichotomy between two devices which serve different functions, it would be good to read a bit more about what you hear with and without them paired.

 

BTW, it is zero surprise that when used with the Intona there is not a perceptible difference between the very early original "green" (LED) REGEN and the "amber" production. As always disclosed, the "amber" just added resistance to the REGEN's USB input ground and shield (diverting some ground noise currents elsewhere). But with the full galvanic isolation from the Intona, that diversion is 100% and the extra resistance on the REGEN's ground is irrelevant.

 

Funny thing is, much of what I hear with the Intona is a greater amount of the same sort of improvement we heard between the REGEN green and amber--for both most especially in the bass (which I still don't understand why it is so profound there).

At the time we were working on the original REGEN (late 2014), John and I had an in-depth discussion of what it would take to do a version with full galvanic isolation. Dual PHY chips, an FPGA or two, two power supplies, etc.--similar to what Intona has done. We even looked into availability and licensing of a high-speed USB2.0 hub core (code to embed in an FPGA). It would all have been costly the way we would have wanted to do it, and we dropped the whole thing once we added the resistors to go from the "green" to "amber" (and given the success of 2015 it was the right thing to do at the time).

 

 

Does anyone here know how firmware updates of the Intona are accomplished? I am assuming that the unit needs to ship back to Germany for the procedure. Guess I'll wait a little while and let them get everything sorted before asking to send mine. Don't want to do it more than once.

 

Ciao,

--Alex C.

[Superdad]

Now if I can just find a substitute for that damn battery!

 

Thanks for the details Larry.

As for that last request, well you know we are hard at work on that:

http://www.computeraudiophile.com/f27-uptone-audio-sponsored/uptone-audio-regen-power-supply-add-24963/index16.html#post468055, with a more recent update from John here: http://www.computeraudiophile.com/f27-uptone-audio-sponsored/uptone-audio-regen-power-supply-add-24963/index21.html#post497891

[Superdad]

Alex,

You should try using the USB A/B adapter you provide with the REGEN to connect the Intona directly to your DACs that use an XMOS USB input, rather than a USB cable (any USB cable).

Regards,

Glisse

 

Thanks Glisse. I can try, but that will be physically very awkward. The Intona box is a LOT bigger than a little REGEN.

[Superdad]

If you want the benefits of galvanically isolated power, there is really no other way.

 

Well, as you know, there will be in just another couple of months.

[Superdad]

Hi,

it's my first post here and I'm pleased to have the opportunity to share my thoughts and my audio experiences with the CA members.

 

I have been reading this thread from the beginning so i would like to ask if someone has tested this isolator that it seems to be like Intona but in lower price.

It's a high speed usb isolator with interesting features and it provides an input for external power. So it solves the issues that coming from power hungry usb inputs...

 

Hi John:

 

Welcome to CA!

And great find for what looks to be only the second high-speed USB galvanic isolator, USB 2.0 High Speed Isolator Adapter and Screw Lock Cable - CoolGear. Just $149 and includes provision for injecting higher-current capable downstream bus power for devices that need it (though the cleanliness of that power is not known).

Wondering if they are doing it with an FPGA ala Intona, or some other method.

[Superdad]

FYI. The Intona creates an I2S stream from the incoming USB, transverses the barrier by capacitive coupling and re-creates the USB waveform.

 

Actually, that is not correct. The SiliconLabs Si866X isolators used by the Intona use tiny RF transmitters and RF receivers separated by a semiconductor-based isolation barrier (see: http://www.silabs.com/Support%20Documents/TechnicalDocs/Si866x.pdf).

 

[bTW, those isolators add 350ps of jitter which needs to be reclocked out. John Swenson greatly prefers the NVE GMR (giant-magneto-resistive) isolators as they are quieter and add less jitter.]

[Superdad]

So it is,as described on page 22. Where did I get the capacitive info from them? Now you need to quantify "less jitter" akin to "whiter than". Anyway, it may not be important if the data is treated with re-clocking on the output side, doesn't matter (for audio) if the delays are through the barrier.

 

You were probably thinking of the Texas Instruments digital isolators. Those do use a capacitive barrier.

 

[As for the clocking--and I really don't want to get into picking apart someone else's product (nor do I have the expertise to do so)--I just wonder if the reclocking (which Intona says they do inside the output-side FPGA) can be very effective if it is based on same one clock that is used for the entire device (both PHYs, both FPGAs) and located on the "dirty" upstream side. I know (from our considering such project over a year ago) that that is not how John would have done it.]

 

Ciao,

 

--Alex C.

 

P.S. I just ordered one of those new CoolGear high-speed USB2.0 isolators. Just too curious!

[Superdad]

I have no experience with this product but found a cheaper source. For anyone interested in trying it, it's available at cooldrives.com for only $129.95 with free (standard) shipping.

 

That is funny since cooldrives.com is really the same company, run from the same offices. Same for at least a dozen other USB and accessory sites that CoolGear owns. I had a long chat with the owner today. He said they used to operate over 300 e-commerce sites all selling similar stuff.

 

The product is brand new and they only ever had a trial run of a few in. More are on the way from overseas and will be available in about 3-weeks--after the Chinese New Year holiday is over.

[Superdad]

Is there any news from the CoolGear devise?

 

Mine is being delivered on Monday. Ask me again late next week.

[Superdad]

These isolaters are also phy-level receiver/isolator/reclocker/repacketizer whereas from my understanding, the Regen does all of those except isolate.

 

If you are referring to the CoolGear isolator, then that is incorrect. There are no USB PHYs and no clocks in the CoolGear. I own one and have become friendly with Rene, the owner of CoolGear. There is a somewhat long and confidential backstory to that product. And long before it was ever announced, I was working with the maker of the isolator chip it uses.

 

The improvements were there with the Regen, but wasn't as dramatic as the isolator for my case. I would describe it as sounding a little more analog. Even if together netted additional improvement, I'd doubt it would be greater than the sum of its parts.

 

I agree but also disagree (despite my obvious commercial interest here I am being very honest about what I hear and what we have found). The isolator chip in the CoolGear adds tons of jitter--probably more than most basic digital isolators and a bit more than a typical FPGA. With the CoolGear alone it can be heard VERY clearly--smearing of time and soundstage--exactly what excessive jitter sounds like.

 

With the CoolGear, one does get the benefit of galvanic isolation--much "blacker" backgrounds and startling bass (wish I understood why blocking ground current noise from the computer always shows up so readily in the bass--same thing happened, to a lesser extent when last May we made the "amber" REGEN by adding resistance to the USB ground/shields). But what you describe as "more analog" is very euphonic and quite typical of excess jitter. At least that I what I hear when using the CoolGeear by itself.

 

But put the REGEN after it, and my oh my, the REGEN takes care of the CoolGear's added jitter and everything comes together beautifully.

 

It is still not perfect. The CoolGear powers both sides of the isolator from the VBUS--using a DC-DC switching regulator. And despite it having a 5V jack to optionally power the downstream side externally, I do not find that it helps. Also, the CoolGear is fussy about what cables it will drive or be driven by. My 1m Curious Cable won't work on either side of it (though the 20cm one does).

 

I have been quiet about this for a couple of reasons:

1) Rene at CoolGear asked me not to talk up the product just yet as this presently is a "soft launch" (his words), which I took to mean that either the product will change (the board does look pretty beta) or that they are just ramping up production;

2) Remember above I mentioned that we have been working with the isolator chip maker (since December)? Does not take much to see where this is going--but we are not quite ready to drop our bombshell about it. (Though perhaps I just did.)

 

I know this is also the Intona thread, but please don't ask me for a comparison between the Intona and the Coolgear by themselves as my Intona unit needs firmware upgrade and thus at this time won't function without the REGEN after it. But I am pretty sure that used alone the Intona will sound better than the CoolGear, because despite its various not-designed-with-audio-in-mind engineering flaws, at least the Intona attempts some reclocking at its output.

 

Shutting up now…

--Alex C.

[Superdad]

+1 for the Intona and Regen combo, it is one part of my chain that will not change for the foreseeable future unless unless Uptone come up with a surprise!!

 

I agree, it is a great combo. And our goal is to surpass it (which based on other enhancements, we already have with the protos).

 

Here is my set-up with the Intona and funny looking REGEN PCB (note the moat around the input jack).

 

This is NOT the announcement, but you guys make it hard to keep cats in bags. Due to significant parts costs increases (there are other surprise enhancements that will be revealed at official time and place), price will be $100 higher.

 

That's all I'm saying for now.

 

[Superdad]

It seems most who own the Intona and the Recovery have the Recovery after the Intona. Has anyone tried the Recovery before the Intona? Or won't it work that way?

 

It should work, but it is really more optimal to have the better clocking after the Intona.

 

 

I now have the updated Intona, updated Recovery and Mutec MC-3+ USB . My source, a Bryston BDP-2 is in for repair and I'm dying to get it back.

So. I'm just wondering how to start the chain from the Bryston BDP-2 to my PS Direct Stream Dac.

 

My suggestion is: Bryston>Intona (right at the back of your BDP-2 if possible)>USB cable>Recovery>Mutec>PSA DAC.

 

Which input to your DS DAC are you using? Coax S/PDIF or AES/EBU S/PDIF?

 

And have you tried your DAC's USB input with just the Recovery into it? How does that compare to using your Mutec USB>S/PDIF converter?

[Superdad]

Just an FYI, I received Daniel's field programmer and was able to update the firmware in my Intona yesterday...

 

Is there an official circulation list for the field programmer? I really need it as mine is one of the early units and I can not get it to work without a REGEN after it. Please PM me if it is possible to borrow it. Or tell us if there is a formal procedure to go though to get on the list.

Thanks!

 

--Alex C.

[Superdad]

When will that Mysterious PSU be available?

 

6-8 weeks if all goes smoothly. 206 parts--including an FPGA and a microSD card slot--mounted on both sides of a 4-layer board, all in a 4-inch x 4-inch case. No wonder it has taken 7+ months. Hopefully the real fun and conversation can begin soon. Trademark filed; patents may come later!

[Superdad]

Alex--Daniel at Intona has a list, I emailed him a few weeks ago and was placed on it. I recall I was #3, and I'm shipping off to the next person today.

 

Thanks Randy. I'll shoot an e-mail to Daniel. I wonder how long the list is...

[Superdad]

Any suggestion where to buy a male/male USB A/B solid adapter plug, like the one that ships with the Regen?

 

No luck at Monoprice or Amazon.

 

Here you go:

https://www.amazon.com/gp/product/B00PVAJTPG

[Superdad]

Agree. My Brooklyn DAC also Galvanic isolation built in. SQ is still improved after Intona connected inbetween.

 

John and I have posted in detail about this--and he about exactly what digital isolators in DACs do and do not do, and how they also add jitter of their own, some worse than others. But the important point to remember is that digital isolators inside DACs (used quite often, but only some manufacturers highlight their use as "galvanic isolation") are ALWAYS placed AFTER the USB PHY and processor/FPGA USB protocol engines. They are placed in the I2S signal lines after the USB input stage.

 

But since it has been proven (at least by those with ears) that USB signal integrity, impedance match, and galvanic isolation into the DAC matters (because it helps the DAC's USB input stage generate less ground-plane and packet-data noise--which can sail on the data lines right through the DAC's isolators to affect the master clock), devices such as the Intona and REGEN still make a difference.

 

SI, impedance, and ground issues are also why USB cables still make a difference with DAC's that claim galvanic isolation of their USB input. So if a crappy USB cable makes any difference with an exaSound or Mytek or whatever DACs (most these days) have digital isolators at the output of their USB inputs, then you have the proof of what I am saying right there.

[Superdad]

1. Intona doesnt feed the dac, it uses the 5vbus from the dac to power the DAC side of the USB galvanization. Basically there are 2 boards inside the Intona unit. With a galvanization chip between them, no current is transferred between the boards. Any 5vbus noise is coming from the dac itself. The computer powers the PC side inside the Intona unit, and the DAC powers the DAC side of the Intona unit.

 

Sorry, that is incorrect. No DAC provides VBUS power to an upstream connected device. Both sides of the Intona's single board are powered via the VBUS from the computer.

 

In the below photo you can see the Coilcraft transformer sitting at the bottom, across the moat. This is how they transfer VBUS power across without loosing the galvanic isolation. It is also why their VBUS power (when connected to a standard 500mA limited computer USB host port) is limited to 300mA.

(Sadly, they also place the one clock--which is used for both FPGAs--on the upstream "dirty" side. And then the Silicon Labs digital isolators use add jitter. Hopefully they do some sort of further rechecking in the downstream FPGA, but it must still be based on the clock coming from the upstream side through the isolators.)

 

Don't get me wrong, the Intona is effective as a galvanic isolator (I've had one for over a year). But it sure sounds a lot better with a REGEN (powered by an LPS-1 so as not to defeat the isolation) after it.

 

[Superdad]

i asked because i wanted to double check but still don't understand how intona's works with microusb.

 

Works fine. You just need either a USB A>microB cable to run from the output of the Intona to your Hugo, or a solid adapter such as this (assuming you can position the Intona and Hugo jacks at the same height):

 

https://www.amazon.com/gp/product/B00TAM0MZW

 

[Superdad]

On 4/8/2017 at 6:23 AM, Em2016 said:

 

Hi mate, sorry to revive an old thread. Just reading through this as I await the Uptone ISO REGEN.

 

What Asus Xonar card are you referring to and can I ask what about it's design makes the SPDIF well implemented?

 

Genuinely interested because I'm looking for a quality SPDIF source for a secondary system.

 

Cheers!

 

Sean:

 

Alex Ketel (sandyk) has ceased posting here at CA, most likely due to seriously failing health (I've been meaning to e-mail him or one of his close friends to find out how he is).  So I do not think a reply to your questions will be forthcoming from him.

[Superdad]

7 hours ago, carlmart said:

How do they compare the Intona with the Hifime in audio quality?

 

Well I can tell you that the raw parts cost alone for our ISO REGEN exceeds the retail price of the HiFiMeDIY isolator.

Ultra-low phase noise clocking, five of the world's lowest noise integrated voltage regulators (LT3042), additional 1A 5VBUS dedicated regulator, specially selected hub chip, and expert attention paid to signal integrity and impedance matching.

We invite you to audition for yourself.

 

[Superdad]

4 minutes ago, CheapSplurge said:

Is the uptone regen allow hqplayer upsampling to 768k?

 

Oh sure.  Even higher. Even DSD512 or DSD1024.  Audio data packet rate for high-speed USB 2.0 is not particularly fast and the packets are never full. ISO REGEN knows/cares nothing about sample rates, channels, data types.  It might as well be printer data or file transfer.

Our device works by greatly improving USB signal integrity and impedance match, and by providing galvanic isolation.  All these things (including low-phase-noise clocking) result in lower ground-plane-noise and less clock-threshold-jitter INSIDE the DAC.  That is why it is audible. 

If you hear any difference between USB cables (which also comes from a measurable modification of signal integrity), then you will hear a much greater different with our device.