The true experimental tweak thread

[austinpop]

18 minutes ago, moussaobeid said:

OK I ordered this one as it shows both, voltage and Amps

https://www.amazon.ca/dp/B00BYKRETK/ref=pe_386430_126088100_TE_item_image

I will figure out the connections later on, this one seems to have more wires than the one @Cornan shared in the image, maybe because it shows amps as well

 

Coincidentally, I ordered this today too: https://www.amazon.com/gp/product/B00J3JSEG6

 

[austinpop]

Yes indeed, the combo provides some interesting experiments:

1. current draw of Brooklyn from VR Mini
2. current draw of RUR
3. current draw from RUR to Brooklyn on USB
4. Therefore, 3 minus 2 = current draw of RUR internals
5. current draw of tX-USBultra
6. Then 5 minus 3 = current draw of tX internals

Can't wait for your numbers!

[austinpop]

Just now, lmitche said:

You can probably get this one today:

 

http://www.homedepot.com/p/Klein-Tools-Manual-Ranging-Digital-Multimeter-MM300/206517192

 

Larry - the question isn't the multimeter (already have one), but the cables/adapters needed to get it in series to measure current.

[austinpop]

3 hours ago, Cornan said:

 

It is very tempting to buy several grounding boxes when you have heard what it can bring to the table. However, my advice is to run slowly. Suddenly when you get to a certain point of "complete isolation" with both AC/DC and ICs a grounding box will not make a difference and can even make it a tiny bit worse. At that point you will have a spare box to use! ? You can actually use the box as a kind of "nose sniffer" as I have. If it makes a great impact on SQ you will know where the isolation needs to be improved. I do think that despite how much you isolate you will always find a place for it in the safety ground or chassi ground.

 

A tip! If you are picky with your car sound you can actually ground your car as well with great results! ?

 

Can you attach them to excitable people too, to make them "more grounded?" 

[austinpop]

1 hour ago, moussaobeid said:

Initial impressions of SOTM tX-USBultra:

 

Since this is the tweak thread, I will share my initial impressions of SOTM tX-USBultra here. I got delivery of my unit yesterday, then I plugged it in my system in the evening. What I did is to critically listen to 3 or tracks on my system with  configuration (1):

 

(1) Nimitra (JS-2)>>Curious 20cm link>>Intona>>Curious 20cm link>>RUR (LPS-1)>>Curious 20cm link>>Brooklyn DAC.

 

The system simply sounded fabulous, music just flows with a sense of realism and 3D especially after installing my two dedicated circuits (one for digital and another for analogue)

Here I would like to mention that I upgraded my AO from version 2.0 to 2.10 and it made a significant improvement, then I upgraded to the beta AO 2.20 and the sound went to another level (more transparent, wide soundstage and better separation).

 

Then I removed the RUR+LPS-1 and replaced them with the SOTM tX-USBultra powered by the second rail of the JS-2 at 12V (my unit has the 12v option) so configuration (2) is as follows:

 

(2) Nimitra (JS-2)>>Curious 20cm link>>Intona>>Curious 20cm link>>SOTM tX-USBultra(JS-2)>>Curious 20cm link>>Brooklyn DAC.

 

just out of the box without any burn-in, configuration 2 clearly elevated the sound of my system in all directions, most importantly to me, a sense of feeling that the music is so real and so clear you feel that the notes are sharper and more defined in the soundstage, a veil taken off the music! (a cliche but it accurately describes my experience)

 

I didn't try to power it with the VR MINI because I use it on the Brooklyn and if I change this configuration, then it won't be A/B comparison anymore. My logic now is that I can improve it by freeing the JS-2 from feeding two units and keep it only to Nimitra and get a dedicated PSU to power the SOTM tX-USBultra, but too early now to make any decisions before I get my DC ammeter working.

So (RUR+LPS-1) will be staying with me for a short time to do some A/B comparisons after 1 or 2 weeks when the SOTM tX-USBultra has settled in my system and most probably will be listed for sale.

 

My learning is that isolation by ultra-capacitors is important but super clock with a good LPS can be more important.

 

I hope this helps others on a similar path

 

Awesome - thanks for your first impressions. This is very exciting to hear of the SQ bump you experienced.

 

May I suggest removing the Intona from the chain to see what the net impact will be? Also, I do think it would be interesting to swap the VR Mini and the JS-2 - i.e. use the VR for the tX, and the JS-2 for the Brooklyn, to see what combo sounds superior.

 

I'm sure over time, you'll be doing these experiments - it's the tweaker's creed after all! - and I'm sure the tX will improve further with some burn-in too.

 

One of the open questions is whether the tX can be powered by the LPS-1, but in your case, it's a moot point, since you got the 12V version of the tX, so your PSU choices are different.

[austinpop]

1 hour ago, moussaobeid said:

I will be removing the intona from the chain but not before a week till the tX burns in and I got a sense of the sound. It will make it easy for me to spot changes for the better or worse.

The VR mini always sounded better than the JS-2 in my previous tests which I repeated 2 or 3 times. Maybe it has to do with leakage loops, the question is what about a second VR with tX? that would be something interesting to try but for a lot more $$$. I will try a cheapo 12V battery tomorrow with the tX

 

Another suggested experiment - use the VR MINI to power BOTH the tX and the Brooklyn, using a Y-cable.

[austinpop]

Thanks for trying these experiments, Moussa. Other than @romaz you may be the first to have gotten the tX-USBultra in hand. Maybe this deserves a new thread?

 

As you know, I should soon (in a couple weeks or so) have my own tX-USBultra and modded sMS-200 back from SOtM, so your experiments are helping me prepare and get a preview!

 

3 hours ago, moussaobeid said:

I disconnected my intona so my chain now looks like this:

Nimitra>>tX USBultra>>Brooklyn

 

I have to say that I hardly noticed any difference in sound, maybe a bit more dynamics but a notch brighter. I put it back but again hard to detect differences, if any it has to be subtle.  I feel that it is redundant in my chain now, it doesn't give the big boost that I used to get from it like before.

 

I removed it and soon will be going to the resale market. It seems that the tX is excellent on its own.

 

Great to hear this. I need to sell some stuff to justify the cost of the tX and other mods!

 

Quote

I also tried the battery on the tX, but I didn't like it made it softer and lost microdynamics. Then I used a Y adapter to power both the Brooklyn and tX from the VR MINI, but I didn't like it at all, sounded harsh and musicality was lost. I also tried the 3rd option which is to swap the pwoer supplies to I put the JS-2 on the Brooklyn and the VR on the tX, I prefered the original setup as I felt some magic was lost.

So now the best configuration to my ears is:

NImitra (JS-2)>>tX (JS-2)>>Brooklyn (VR MINI)

 

Very interesting, especially the experiment with the Y adapter with the VR Mini. One possible explanation is that many DACs actually have galvanic isolation between their input and DAC stages, so by using the Y, you may be defeating that isolation. I have no info on what the Brooklyn's design is in this regard. But it's a theory.

 

A similar factor may be coming into play with the swapped PSes. I had forgotten the JS-2 was also powering the Nimitra. Even though the 2 rails of the JS-2 are floating and "isolated" from each other, I wonder if there is a potential for defeating isolation and creating leakage loops by spanning an isolation boundary in the DAC.

 

Quote

My learning is that it may not be a good idea to share the same power supply through Y adapters, but this is only in my system and keep in mind that the Y adapter is a cheapo from ebay not my OCC copper wire with Oyaide connectors.

 

I think it's not necessarily the Y-connector or sharing power, but the specific use case, and whether the components sharing the power supply have the potential for leakage loops.

 

More depressingly, the safest answer seems to be the most expensive - use separate power supplies!

[austinpop]

18 minutes ago, moussaobeid said:

I measured the current of the tX-USBultra and Brooklyn with my new AC/DC multimeter, but I am not sure that the results are correct. I checked the voltage and on both devices I got 12V, so I think my connections are ok.

 

Here are the numbers:

Brooklyn (music playing, Dac only no headphone, screen lights off): 0.05mA

tX-USBultra (music playing): 0.04mA

 

These numbers are way too small, they don't make sense. I attached the pictures for your reference and I used this multimeter https://www.amazon.ca/gp/product/B01NAVAT9S/ref=oh_aui_detailpage_o00_s00?ie=UTF8&psc=1

 

Yeah, those numbers make no sense. Even if you assume the units are amps, not mA, that's still only 40 and 50mA respectively, which is kind of low.

 

Not sure what to tell you. It's a mystery.

[austinpop]

Yes of course!

 

BTW - did you just change your handle name?

[austinpop]

2 minutes ago, mozes said:

Ok here are the current measurements for the tX and Brooklyn after figuring out how to measure current:

-Brooklyn (music playing, Dac only no headphone, screen lights off): 0.843 Amps so at 12V, it consumes 10.116 watts

-tX-USBultra (music playing): 0.477Amps so at 12V, it consumes 5.724Watts. This includes powering the VBUS for the Brooklyn

@austinpop, I think you should be fine powering the tX from an LPS-1

 

 

Ah, very promising!

 

When I get mine, I will confirm it works with my existing LPS-1 and then order another one.

[austinpop]

BTW - I got my Drok meter today as well.

 

This is the USB inline version, so I used it to measure the current flowing from my RUR to the Codex DAC on the USB bus.

 

The highest I'm seeing is 250 mA.

 

[austinpop]

22 minutes ago, mozes said:

Ok here are the current measurements for the tX and Brooklyn after figuring out how to measure current:

-Brooklyn (music playing, Dac only no headphone, screen lights off): 0.843 Amps so at 12V, it consumes 10.116 watts

-tX-USBultra (music playing): 0.477Amps so at 12V, it consumes 5.724Watts. This includes powering the VBUS for the Brooklyn

@austinpop, I think you should be fine powering the tX from an LPS-1

 

 

Can you measure the current on the USB drawn by the Brooklyn?

[austinpop]

1 hour ago, mozes said:

Very low power consumption on the Vbus from the tX to the Brooklyn, 0.06A does it make sense?

 

 

Yes absolutely, it helps.

 

Based on your earlier reading, that tells me the intervals of the tX draw approx 477 - 60 = 417 mA.

 

Add to that the 250 mA my Codex draws, we're up to a current demand of 667 mA.

 

In terms of PSU sizing, the other unknown is that my tX will be driving all 4 clock taps, 2 to the sMS-200, and 1 to the Zyxel switch. This could add up to more current.  But it would have to be around 400mA to overload an LPS-1.

 

So for now, I am very hopeful an LPS-1 will do the job.

[austinpop]

34 minutes ago, austinpop said:

 

Yes absolutely, it helps.

 

Based on your earlier reading, that tells me the intervals of the tX draw approx 477 - 60 = 417 mA.

 

Add to that the 250 mA my Codex draws, we're up to a current demand of 667 mA.

 

In terms of PSU sizing, the other unknown is that my tX will be driving all 4 clock taps, 2 to the sMS-200, and 1 to the Zyxel switch. This could add up to more current.  But it would have to be around 400mA to overload an LPS-1.

 

So for now, I am very hopeful an LPS-1 will do the job.

 

Of course, my calculations make an important assumption that the current demands stay fairly constant over the supported voltage range. You have the "12V version" of the tX-USBultra, while I will have the standard version, which is said to operated in the 6-9V range. I'll be running the LPS-1 at 7V.

[austinpop]

2 hours ago, austinpop said:

 

Yes absolutely, it helps.

 

Based on your earlier reading, that tells me the internals of the tX draw approx 477 - 60 = 417 mA.

 

Add to that the 250 mA my Codex draws, we're up to a current demand of 667 mA.

 

In terms of PSU sizing, the other unknown is that my tX will be driving all 4 clock taps, 2 to the sMS-200, and 1 to the Zyxel switch. This could add up to more current.  But it would have to be around 400mA to overload an LPS-1.

 

So for now, I am very hopeful an LPS-1 will do the job.

 

I got another data point from @romaz where he's seeing ~200mA current drawn at 9V by his modded dX-USB HD Ultra.

 

All this data suggests the LPS-1 should be great with the tX-USBultra.

 

Will let you know once mine gets in.

[austinpop]

Hey @Cornan

 

Have you checked out the new Aries Mini 5.0 beta1 firmware, along with the beta Lightning 4.2 beta app?

 

I am very impressed with Auralic - this is a huge upgrade - for free for existing customers. See:

 

 

[austinpop]

5 hours ago, Cornan said:

 

Now I have tried firmware 5.0 beta 1 (5.0 beta 2 still not available for me) with LightningDS beta 4.2. Listening to Bat For Lashes "The Bride" in MQA is a major jump in sq. Still a bit "edgy" compared to BubbleDS Next but I'll guess Auralic will improve it further until it runs out of beta. Overall a very nice upgrade indeed. Love it! ?

 

Great! If you want to find more Tidal MQA content, this crowd sourced Google Sheet - https://docs.google.com/spreadsheets/d/10VtON9VjMAt3uyHC2-Oo2MjIa3orv9DKZfwiRQKmTAA - while messy, is a good place to look.

 

BTW - Wang says that they're improving their MQA upsampling, and the version currently in the beta is just a test implementation. So he clearly thinks it will improve.

[austinpop]

6 hours ago, Cornan said:

 

Thanks Rajiv! 

If LightningDS will improves further I might even start using it instead of my beloved BubbleDS Next. It is not far off I tell you and MQA sound marvelous compared to normal FLAC! 

Thanks for the pointer to the Google sheet. I will check it out! ?

 

BTW. Is Aries Mini really Roon-ready now? Looks like it to these eyes! ? If that's true I might consider to buy myself a NAS and get Roon as soon as my wallet let me! ?

 

 

Yes, that is a welcome and quite unexpected surprise. The Aries Mini is Roon Ready. I'm playing with it now.

 

Since I run Roon Core on a W10 box, optimized with AO, to my ears at least, Tidal files (regular, not Master) sound better via Roon rather than the native Tidal in Lightning running on the Mini.

 

The fake-MQA-decoding on the Mini changes that again, since comparing fake-MQA decoded Tidal Masters files vs. the untouched Masters files through Roon, the balance shifts back to the Mini.

 

However, Roon has stated that they are licensing the MQA SW decoding from MQA, which performs the first unfold, just like the TIdal app on WIn/Mac. So hopefully soon, Roon will be able to do the real MQA SW decoding on Masters tracks. I'm convinced this will tilt the balance back to Roon.

[austinpop]

Just now, mozes said:

Can Roon right now pass through MQA files to MQA davs like the Brooklyn?

 

Yes, absolutely.

[austinpop]

2 minutes ago, mozes said:

That's great news, have you done any comparison between Jriver and Roon purely from a sonic perspective?

 

I've never used JRiver. I was previously using MinimServer, running on both my NAS, and more recently on my W10/AO-optimized bridged box.

 

I've done comparisons of streaming via RAAT - i.e. Roon Core to Roon Ready endpoint - vs. DLNA using MinimServer, and cannot hear any difference. Others have insisted they think RAAT sounds better. Not to my ears.

 

But there are enough functional reasons to like Roon.

 

And of course, this does not factor in upsampling or DSP, which is a whole other can of worms. I wrote about this in my recent Aries Mini vs. mR vs. sMS-200 thread here:

 

 

[austinpop]

I should write up my experiences with Roon and post it somewhere.

 

In a nutshell, Roon has a steep learning curve. Yeah, the basics are simple - point to the library, point to the endpoint, boom, done - right? Wrong! It takes time to learn the nuances. I went from being a deep skeptic, convinced it wasn't for me, to a lifetime subscriber, over the course of a 60-day trial.

 

But it is certainly not for everyone. For example, if you're primarily using Tidal as your music source, I won't try to convince you the cost of Roon is worth it.

 

All I can say is - if you see potential, be patient and keep digging into it. There are a lot of features and neat ideas below the surface, and over time you will have some aha moments. Or not - in which case, you shouldn't waste your money.

[austinpop]

I can't answer all your questions, but will try to help point you to where to look/ask.

 

51 minutes ago, mozes said:

-Does Roon support Native DSD256 playback?

 

Yes, the capability exists, AFAIK.

https://kb.roonlabs.com/FAQ:_Does_Roon_support_DSD%3F

 

51 minutes ago, mozes said:

-If yes, do I have to use the Brooklyn windows ASIO driver? I can only see the Wasapi driver to connect to

-If someone is using Roon with the Brooklyn with Roon, pls share your audio settings

 

All this Wasapi and ASIO driver stuff is one of the reasons I prefer using endpoints! A couple of places to look for expertise:

1. Roon community forum: https://community.roonlabs.com
2. Brooklyn thread on CA: https://www.computeraudiophile.com/forums/topic/25353-mytek-new-dac-brooklyn/?do=findComment&comment=466563

51 minutes ago, mozes said:

-Now I can play DSD either via DOP or transcode to PCM

 

Does it work natively for DSD64 and DSD128?

 

51 minutes ago, mozes said:

-I can't run Roon control on my laptop, I updated the driver for the display adapter (Intel HD graphics 4000) still no luck!

 

More Windows weirdness. I honestly just futzed with my display drivers to get it to work. Mine is working with Intel HD 4600. Here is what mine looks like. Make sure you reboot - sounds lame!

 

 

51 minutes ago, mozes said:

-MQA unfolding as of now is only showing 44.1/24 with MQA blue light on on Masters when I set exclusive mode.

 

That's how it should be. Roon MQA unfolding support is not yet out. So it should be passing through MQA content at 24/48 or 24/44.1 to the Brooklyn. The success is if the Brooklyn indicates it is receiving an MQA stream. I don't know the Brooklyn, but it hopefully shows you an authenticated MQA "blue" light, as well as the unfolded sample rate?

 

51 minutes ago, mozes said:

On the positive side of things, It is clear that Roon is above all the other audio software I have experienced in terms of features, interface, graphics, integration with Tidal, etc.

 

Indeed!

[austinpop]

1 hour ago, mozes said:

Here are my impressions about Roon:

It took me sometime to make it work with the ASIO driver of the Brooklyn so I can play native DSD, but I managed after one day of trial and error. My system is as follows for those who don't know:

Nimitra>>SOTM tX-USBultra>>Brooklyn

 

What I liked:

1-Excellent user interface

2- Makes you discover your music in new ways

3-Seamless integration with Tidal

4-fluid navigational experience

 

What I didn't like:

-Sonically, it is not the best I heard, in direct A/B with same tracks from my SSD, Jriver 22 is clearly superior. Jriver 22 is more transparent such that Roon sounded veiled in comparison. As for Tidal streaming, Bubble DS Next with Jplaystreamer is more musical and less digital or metallic. The only exception is when playing Masters files to the Brooklyn (MQA lit blue), then it becomes so close to Bubble DS Next with Jplaystreamer and then we are talking about differences in presentation. Roon has deeper bass, but Bubble DS Next has more natural vocals, this is way too close that I can live with either. If Bubble DS Next with Jplaystreamer can support MQA then I believe that this will be well ahead of Roon.

-A bit slow to load on my control app on iphone and Android tab (maybe my hardware, don't know)

-Frequent disconnects and I have been asked to enter my username and password many times. almost 1 out of 3 or 4 attempts to launch the control app

-Overwhelming with information that I felt I am not in control of my music library, it is like Roon wants to be the BOSS and not me, I don't like that, I have to be in control so less suggestions, pushed recommendations is what I like. When I want to explore, I leave it to streaming solutions like Tidal to do that in the cloud not when I want to mange my library and play music

-Very fussy with hardware (couldn't launch it on my W10 laptop de to some video card driver compatibility issue).

-Took some time for it to recognize Brooklyn ASIO driver, not sure why.

 

Overall, these are my subjective impressions on my system using my subjective ears, so take it with a pinch of salt.

 

For now, I uninstalled Roon from my Nimitra, iphone and Android tab and back to my old trusted Jriver and BubbleDS Next/Jplaystreamer.

 

I will still keep an open mentality towards Roon as it is a great software for music management and playback, but now it doesn't suit my needs.

 

 

Very interesting report.

 

Since I do not live in the world of PC-based music players, I cannot comment on the sonic differences you're hearing. Clearly there is more optimization for Roon to do. The only other suggestion I could make is to run the Roon processes (roon, raatserver, etc) as High priority. I use Process Lasso in Bitsum High Performance mode, but the priority settings can be done manually in Task Manager.

 

Now, as to the functional issues, I do think the feeling of Roon being boss is just being early in the learning curve. Once you find your way, I do not feel that way any more.

 

As to the speed, I've had no issues. The main thing is to ensure Roon's DB is located on an SSD. And maybe it's because I use a Roon Ready endpoint, I've had no functional errors or hiccups at all.

 

Glad you tried it - even if it isn't your cup of tea. At least you won't be wondering.

[austinpop]

1 hour ago, Cornan said:

I took a big step in my little audio world today. I decided while listening to Death Cab For Cuties MQA album "Kintsugi". I have changed my Android tablet with BubbleDS Next for an iPad with LightningDS for the first time since I bought my Aries Mini. The MQA unfolding with "Smooth" setting sounds really great in my setup and the normal FLAC have actually improved enough for me to feel comfortable with the switch. Finally I am sure LightningDS will catch up BubbleDS Next when it leaves beta! ?

 

An unexpected thing was that my low budget setup sounds better as well now. I connected the D-Link DAP-1620 to a separate channel and connected my Android tablet to it. Now both tablets are on separate channels. Voila! ?

 

That's great!

 

I always love when a HW vendor adds functionality and SQ to an existing product through firmware or software without additional cost. That's the kind of thing that builds brand loyalty.

[austinpop]

54 minutes ago, mozes said:

I am now thinking of which way to go

1- get 2 isolators and place in front of the modem and router. This is easy but expensive

2- change the cables to BJC Cat6a. More affordable but hard to do

 

any recommemdations for affordable ethernet isolators. I assume these don't have a task as critical as the ones before my Nimitra

 

2 is a trade off. When we built my house 17 years ago, I had it wired for Ethernet, but it was all Cat 5e at the time. 

 

I'm not able or willing to rip all that out. I just go BJC 6a within a room, i.e. 5e for long runs in-wall, but 6a for everything else.

 

i also worry about the effect of cheap isolaters on signal integrity.

 

Try it and see. I have honestly not heard anything to matter upstream of my bridged W10 box. 

 

I've tried grounding boxes, LPSes, Baaske isolator. Maybe I'm lucky.

 

This is always the question isn't it. If you take care with isolation next to your DAC, then how far upstream of your DAC does it cease to matter?