Matrix X-SPDIF 2


That port sounds pretty great to me with the UpTone LPS-1.2!


In the end it is all about what you hear… if it sounds great, sit back and love it… As Darren said in the interview… measurements are good to have, understanding the correct measurements, better… and in the end it is about listening… the reason I pay less attention to opinions and more about sounds… God gave me a good brain and a great pair of ears…,. I never loan their function out to others… trust, but verify…

Forgot - brain and ears are deteriorating every year… getting old sucks… Also, what makes this forum great (1116) is so much sharing of ideas. I get so many great ideas I get to track down and think a lot about…


I think the important take away with this discussion is not a proclamation that the Matrix cannot sound excellent (or even better depending on the case) when compared to the infinite USB power source possibilities.

We are obviously all interested in leaching out the very best from all of our components, the Matrix included. This is illustrated by the fact that this thread is not 1117 posts long. I am a tweaker to the Nth degree and if there is even a remote possibility that something within my control is causing an issue or limiting my system’s potential, I will address it. But I know this level of OCD is not within all of us and that is perfectly fine. Again, the Matrix is a fine device even if I question how they implemented their external power option.


The AQVOX LPS route you are taking is of great interest to me as well. Certainly the more affordable option if only the power needs addressing. In my case, I’m assuming that the USB into the Matrix from a Mac may benefit more from ifi’s full bag of tricks. But I’m also skeptical of daisy chaining reclockers from my earlier attempts at ‘fixing’ USB audio.

For what I hope are good reasons, I’ll try ifi first.


Before I bought the LPS1.2 I used a six volt battery. The LPS1.2 sounds better, which wrecks my brain.

1 Like

I can certainly believe it. Uptone makes excellent kit. And for all of the power discussion, I agree with @JeffofArabica that this is in service of wringing out the last bit of SQ from an already stellar device. What fun would it be if I already had a perfect system with nothing left to try?

1 Like

Nice your present streaming setup via BridgeII delivers that much quality and joy to you, now just beware to be ready for the Matrix treatment :sunglasses:

1 Like

@wijber - perfectly put… then why am I doing this if my streaming is better than my DMP. The answer is like @JeffofArabica talks about… I am a shinny box whore and did not like the fact that I2S is double and BII is single. I asked about the BIII future and got the feeling that it will be tied into the Octave approach. Not hanging my approach on Octave till that stuff gets really wrung out… Now, It seems to me that Ted does not like to go above double DSD and I’m OK with his rational. I’m good to hit the sweet spot for DSD. Getting the BII out of the DSD has been touted as a good thing for SQ… I don’t buy that cause my SQ streaming is better than DMP without even any filters in place… This is a 2-3 year stop gap before I go PF 2.16 streamer of perhaps the PSA Octave “vaporware”… I may trade up to BIII if they just modify the BII to include double DSD instead of $2K for streaming I don’t want cause I absolutely love my Nucleus+ for SQ and the SW is just AMAZING! I have owned my Nucleus+ for (6) months and never (not one time) had a single issue for one second… while streaming and I use it 15hrs a day and have it running 24hrs a day… the Sbooster that powers it is hooked to my P20 with an AC12 power cord… The Roon ROCK OS is the bomb! I hated the Mconnect dropping hourly and doing all kinds of funny stuff with so I looked for a better approach back in fall 2018. Just my experiences…

Forgot - I’m waiting for Ted to push back the TSS new FPGA with more resources and hopefully triple DSD with HW mods… A DSD Grandpa in between the DSD Sr and TSS. I hope at a minimum he plows back some of his insight from the TSS project into a nice “everestmass” build in spring 2020… and is not FPGA resource bound to do it… who knows… just the babbling of a madman… :slight_smile:


My OCD is controlled via meds so I’m having a lot of trouble getting excited about any of this. I have enough trouble trying to resolve issues that actually exist.


Be careful and make sure you read the warning label on the box.


Try before you [don’t] buy!


My, or any other’s, journey doesn’t require your excitement, so you are more than welcome to continue one with what you want to do. The sharing of experiences here are for those that have the capacity and desire to learn. If you don’t want to participate, then your participation is not required.

You don’t see a problem. I see a problem. You chose to not devote time vetting any of this out. I chose to objectively analyze and work toward a better solution. You are you. I am me. All so very simple.


Relax it’s all good - have fun…


As I read the later half of this thread it was clear that people were jumping to unwarranted conclusions. Then I thought “good, people are getting back on track”. But then again there was more weirdness:

The Matrix is obviously reasonable quality, they know when to use an LDO (which is always linear) and why not to use a switching supply without a lot of thought. The picture of the board shows no signs of a switching supply (e.g. where’s the inductor?) The reason that the external input has a lower voltage limit greater than 5V is that a LDO (low dropout) linear regulator isn’t a no-dropout linear regulator. It needs a small (or not so small) voltage difference between it’s input and it’s output to do a good job of regulation (the dropout voltage is where the regulator “drops out” if the input voltage is too low.) The reason it has a maximum input voltage limit is that all of the extra voltage has to be dissipated as heat right at the regulator. USB 2.0’s maximum allowed current draw for a device is 500mA, so if the Matrix was using 500mA it would be dissipating 0.5A x 5V = 2.5W at max, but if the input voltage was 9V then the regulator itself would need to dissipate (0.5A x (9V - 5V)) = 0.5A x 4V = 2W. Given that the maximum heat dissipation of a typical LDO in that range is 40 to 60 degrees C per watt when mounted on a board in still air, 2W will get HOT.

So some input voltage between 6, 6.5 or 7 V is probably best if you use the external voltage input.

The LDO will reject a reasonable amount of noise so 6V has a good chance of being quieter than being USB powered, as long as the outboard power supply delivering the 6V is clean, e.g. a linear power supply itself.




Great…I already ordered a 9V LPS :^ /


Yes? That clearly means that you can use the external power inlet to avoid the crappy USB power that’s almost universally generated by switching supplies. It also clearly states that they are using LDO’s internally (not switching supplies) even on the USB power which is also great.


@tedsmith - so the answer is there is a DC/DC circuit… switched… now if you use a “data only” with a good LPS through the USB-B, it is better than allowing the external PS to pass through the DC/DC switched… now, I don’t have schematics and neither do you… this is all your/my speculation…


Hey Ted,

Not sure if you saw this post measuring noise on the external port vs. USB. Someone posted a link to it way up in this thread. That was the catalyst to the discussion of noise with regard to the Matrix external power port.

I’d be interested in reading your thoughts if you have a moment to give it a look.


That DC/DC circuit is in source of the USB in your computer or wherever, if you use an well designed USB HUB that uses an LDO for it’s power supply that HUB needs to use better regulators than the Matrix (which is certainly possible, but not guaranteed.) But it’s clear that the Matrix doesn’t have any internal switched supplies.

I only briefly scanned the link provided, but it’s not a surprise. Unless I missed something the Matrix is doing it’s job on all accounts. In general USB power sucks, but any device can do a good job of cleaning it up. An external power input allows giving a unit a better power source to start with…

But I was just chiming in when people seemed to me to be suggesting that the Matrix had DC/DC converters (switching regulators/supplies) internally when it was clear that they know when to use them and when not to. Requiring a voltage between 6V and 9V for the external power input almost guarantees that you are using an LDO or else you are using a very bad switching supply - switchers can easily regulate, say, 40V down to 5V, LDO’s can’t come close.