"noise" !?!? OK. The earlier discussed audio signal modulation on the power rail can also be called "noise" - fair enough. 
###
One more comment to the PS situation. Once again.
I've been testing several PS and cables.
The main issue are the characteristics of the (USB-) cables
under higher load conditions!
You guys never ran into issues with your earlier products due
to low power consumption.
That game has changed on two fronts. RPI 3B+ and Katana.
With increased load, following simply Ohm's law, the cable losses rise.
I bought a rather high quality USB-C -> USB 3.1 cable
At 1A and 50cm I'll loose 0.2V.
I do have several cables doing worse!
And the situation gets worse with higher load, greater length and/or quality of the cable.
Now you add to that bill earlier discussed 0.25V diode losses and you'll
be entering dangerous waters.
Your 5.1V PS (fixed voltage) with 3A becomes absolute irrelevant in this context!
What would help is a supply with adjustable output voltage! Most people won't have these at hand though!
All I am saying is, people need to be very careful of how to power the stack!
###
One more comment to the PS situation. Once again.
I've been testing several PS and cables.
The main issue are the characteristics of the (USB-) cables
under higher load conditions!
You guys never ran into issues with your earlier products due
to low power consumption.
That game has changed on two fronts. RPI 3B+ and Katana.
With increased load, following simply Ohm's law, the cable losses rise.
I bought a rather high quality USB-C -> USB 3.1 cable
At 1A and 50cm I'll loose 0.2V.
I do have several cables doing worse!
And the situation gets worse with higher load, greater length and/or quality of the cable.
Now you add to that bill earlier discussed 0.25V diode losses and you'll
be entering dangerous waters.
Your 5.1V PS (fixed voltage) with 3A becomes absolute irrelevant in this context!
What would help is a supply with adjustable output voltage! Most people won't have these at hand though!
All I am saying is, people need to be very careful of how to power the stack!
Last edited:
Mark,
The effect of the filter is to roll-off the highs, with the analog feed into the I/V stage down just a little over 1db at 20kHz. I could hear a very tiny drop in apparent highs.
Some would consider that tiny drop a negative effect, though in my listening going back and forth between the 2 O/P stage versions, I heard more lowering of apparent HF noise and being able to hear subtle cues more effectively.
One thing that puzzles me... why does the addition of that filter cause the noted change in THD+N? I would think filtering out-of-band would have the opposite effect. I don't know the characteristics of what is being measured in the THD+N tests and how different factors impact that. Can anyone provide a good explanation?
I gave quite a bit of feedback from my listening tests, some good, some bad. After listening to the initial physical version with the corrected firmware, I was prepared to say that the Katana only really sounded good with an external +-15V supply.
Allo took all (including mine) feedback seriously, continued adjusting the unit for better measured and sonic performance, and improved the Katana a great deal. AND it now sounds quite respectable with the +-15V being supplied via the onboard DC-DC converters.
Glad to know I still can't make a post without a derogatory response including name-calling from you, Soundcheck. What is up?
Do we need to take this offline? Or can you just do better?
Greg in Mississippi
LOL... seriously, I REALLY don't think I directly hear it. I hear a slight lowering of apparent highs which I am GUESSING is a drop in hash/noise in the highs. But who knows?
In a similar vein, my biggest tip-off on the dynamic ringing was a headache and a bit of ringing and aching in my ears when listening at my normal (actually not very loud) levels. I can't hear the ringing, but could only infer what it might be based on my experience listening to gear over many years. It was only after Allo looked based on my suggestion that it was found.
I recognize the value and limits of listening tests. None of the improvements Allo made would have been possible without detailed and accurate measurements, something I feel they are pretty darned good at.
Greg in Mississippi
In a similar vein, my biggest tip-off on the dynamic ringing was a headache and a bit of ringing and aching in my ears when listening at my normal (actually not very loud) levels. I can't hear the ringing, but could only infer what it might be based on my experience listening to gear over many years. It was only after Allo looked based on my suggestion that it was found.
I recognize the value and limits of listening tests. None of the improvements Allo made would have been possible without detailed and accurate measurements, something I feel they are pretty darned good at.
Greg in Mississippi
Could be due to the series resistors. People have experimented with that before and found it increased distortion.
Mark,
Thanks.
I've actually personally tried the same filtering 5 times in the past:
- With a Sony HAP Z1-ES I was repairing for a friend, where I bypassed the much more complex Sony output stage with a simpler one using the OPA1632 for balanced output (he didn't need single-ended).
- 2x with two different modified I/V stages on a Buffalo-IIIPro 9028 build, one based on the IVY and the other the Legato.
- 2x with two different transformer output stages on a prototype ES9028Q2M DAC from Ian Canada. I used it on 2 different transformer output stages, one using Lundahl LL1570XL and the other using Onetics 600R:600R transformers.
All of these (except the HAP Z1-ES) exhibited similar changes in SQ before/after adding the filtering as I heard on comparing the 2 Katana output stages... more musicality and dimensionality with the filtering. I left the filtering installed in all cases. I still need to try it on my Buffalo-IIIPro 9038 build.
On the HAPZ, I didn't try the unit without the filtering, so no comparison.
Of course this is in my audio setups with my listening preferences, I don't expect everyone to prefer the same sonics I do. Sorta nice Allo is giving us a choice!
Greg in Mississippi
Thanks.
I've actually personally tried the same filtering 5 times in the past:
- With a Sony HAP Z1-ES I was repairing for a friend, where I bypassed the much more complex Sony output stage with a simpler one using the OPA1632 for balanced output (he didn't need single-ended).
- 2x with two different modified I/V stages on a Buffalo-IIIPro 9028 build, one based on the IVY and the other the Legato.
- 2x with two different transformer output stages on a prototype ES9028Q2M DAC from Ian Canada. I used it on 2 different transformer output stages, one using Lundahl LL1570XL and the other using Onetics 600R:600R transformers.
All of these (except the HAP Z1-ES) exhibited similar changes in SQ before/after adding the filtering as I heard on comparing the 2 Katana output stages... more musicality and dimensionality with the filtering. I left the filtering installed in all cases. I still need to try it on my Buffalo-IIIPro 9038 build.
On the HAPZ, I didn't try the unit without the filtering, so no comparison.
Of course this is in my audio setups with my listening preferences, I don't expect everyone to prefer the same sonics I do. Sorta nice Allo is giving us a choice!
Greg in Mississippi
Last edited:
Still having some issues at 2A and more...we are debugging. Its been one of the hardest projects every..
My first impressions about the sound of the new Katana 1.2:
The SQ of the new Katana has changed dramatically - for the better.
In a direct comparison of the sound quality between the OPPO Sonica DAC and Katana 1.2, the OPPO was broken to pieces, losing in all respects.
I compared, after some warm-up, both OPAMP cards "Sound Quality" and "Pure THD" version. Both sound great.
The board which is made for "Pure THD" version, I liked much more. Very natural sound. The whole range sounds great and balanced.
Nothing bulges, nothing shouts, everything is in its place, everything sounds soft, but at the same time detailed and dynamic.
The bass is very deep and very prominent. As conceived by the sound engineer, as he saw his sonic canvas while working on mixing the record.
I think that finally now Allo got a great device with a great sound.
The version for "Sound Quality" is also very convincing and should be enjoyed by young music fans. Bass there is VERY deep, beautiful, very good to do a massage for the ears and body
Now we have to work hard to create the necessary power supplies, with very high parameters, to fully cover the potential of the new device from Allo.
Congratulations, you have gone through a difficult path, we all waited and hoped very much and at the end of the path you achieved very good results. Bravo!
The SQ of the new Katana has changed dramatically - for the better.
In a direct comparison of the sound quality between the OPPO Sonica DAC and Katana 1.2, the OPPO was broken to pieces, losing in all respects.
I compared, after some warm-up, both OPAMP cards "Sound Quality" and "Pure THD" version. Both sound great.
The board which is made for "Pure THD" version, I liked much more. Very natural sound. The whole range sounds great and balanced.
Nothing bulges, nothing shouts, everything is in its place, everything sounds soft, but at the same time detailed and dynamic.
The bass is very deep and very prominent. As conceived by the sound engineer, as he saw his sonic canvas while working on mixing the record.
I think that finally now Allo got a great device with a great sound.
The version for "Sound Quality" is also very convincing and should be enjoyed by young music fans. Bass there is VERY deep, beautiful, very good to do a massage for the ears and body
Now we have to work hard to create the necessary power supplies, with very high parameters, to fully cover the potential of the new device from Allo.
Congratulations, you have gone through a difficult path, we all waited and hoped very much and at the end of the path you achieved very good results. Bravo!
There is a little low level RF coming out of the dac outputs. Depending on components in the output stage, that RF may cause some unpleasant distortion. Filtering out as much RF as possible before going into the I/V can help if there is a problem. Instead of resistors in an RC filter, inductors can be used to make an LC filter. That might help with distortion caused by added resistance in some cases. Up to the dac designer to figure out the best trade offs.
My first impressions about the sound of the new Katana 1.2:
The SQ of the new Katana has changed dramatically - for the better.
In a direct comparison of the sound quality between the OPPO Sonica DAC and Katana 1.2, the OPPO was broken to pieces, losing in all respects.
I compared, after some warm-up, both OPAMP cards "Sound Quality" and "Pure THD" version. Both sound great.
The board which is made for "Pure THD" version, I liked much more. Very natural sound. The whole range sounds great and balanced.
Nothing bulges, nothing shouts, everything is in its place, everything sounds soft, but at the same time detailed and dynamic.
The bass is very deep and very prominent. As conceived by the sound engineer, as he saw his sonic canvas while working on mixing the record.
I think that finally now Allo got a great device with a great sound.
The version for "Sound Quality" is also very convincing and should be enjoyed by young music fans. Bass there is VERY deep, beautiful, very good to do a massage for the ears and body
Now we have to work hard to create the necessary power supplies, with very high parameters, to fully cover the potential of the new device from Allo.
Congratulations, you have gone through a difficult path, we all waited and hoped very much and at the end of the path you achieved very good results. Bravo!
Frankly , I am very happy to hear that.
Very nice and extremely communicative short review!
Curious... what power supplies and configuration did you use?
TIA!
Greg in Mississippi
Can you give us some insight as to the goals and issues that you have encountered with the development of the Nirvana ? The fact that you are now working on an LPS in addition to the SMPS suggests that your initial goal of 5V/3A low noise power may not be attainable. How good is Nirvana at ≤ 1.5A which would be sufficient to power the DAC stack?
I agree bvlabs..My first impressions about the sound of the new Katana 1.2:
The SQ of the new Katana has changed dramatically - for the better.
In a direct comparison of the sound quality between the OPPO Sonica DAC and Katana 1.2, the OPPO was broken to pieces, losing in all respects.
I compared, after some warm-up, both OPAMP cards "Sound Quality" and "Pure THD" version. Both sound great.
The board which is made for "Pure THD" version, I liked much more. Very natural sound. The whole range sounds great and balanced.
Nothing bulges, nothing shouts, everything is in its place, everything sounds soft, but at the same time detailed and dynamic.
The bass is very deep and very prominent. As conceived by the sound engineer, as he saw his sonic canvas while working on mixing the record.
I think that finally now Allo got a great device with a great sound.
The version for "Sound Quality" is also very convincing and should be enjoyed by young music fans. Bass there is VERY deep, beautiful, very good to do a massage for the ears and body
Now we have to work hard to create the necessary power supplies, with very high parameters, to fully cover the potential of the new device from Allo.
Congratulations, you have gone through a difficult path, we all waited and hoped very much and at the end of the path you achieved very good results. Bravo!
after 4 days of burn in for sound quality optimized opamp.. the bass deparment dramatically improved.
I am confirming also the other findings of Katana v1. 2 users that you will get a less fatiguing sound.. i can say that sound presentation is very gentle in my ears.
ill try the THD optimized opamp today.. Im gonna benchmark this with my marantz HD DAC.. with ifi purifier2 and jitterbug.. usbridge as my transport..
Now my chord Hugo has a serious match at 70% lesser than its price tag..
its the controller.. the THD optimized is underneath that.. either of the two will have the glued cap