Holo Audio May KTE Dac

[213Cobra]

I have a long history of preferring R2R DACs over delta-sigma and in that context I have had a Holo May KTE in for extended demo with extended comparison to the Bricasti M1LE-Gold and Bricasti M21. The M1LE-Gold is sigma-delta only, whereas the M21 gives you four processing choices: sigma-delta, R2R, DSD as DOP, DSD in analog.

These are somewhat unfair comparisons as the Holo May KTE is ~$5500 depending on how your buy it. The Bricasti M21 is $16k-$19k depending how you buy it. The M1LE-Gold was $15k new when it was in the line. But the Bricasti M3 is in the same Holo May price league with much of the sound of the various M1s. You can occasionally find an M1LE-Gold used for around the price of an M3 or Holo May KTE, new.

I will post a longer summary of comments on the Holo May KTE soon, but in the context of this thread, I will just say that yes, the HM-KTE is warmer. I have not heard a Tambuqui in a clear comparative setting, but while it sounded organic and good, I did not hear it as transformative ala its reviewed reputation. The Holo May sounds highly competitive with other ~$6000 DACs, in terms of low noise, musicality, tone and dynamic elasticity. But I was surprised to prefer the Bricasti delta-sigma 80% of my listening time to the two over a period of more than six months and counting. More to report with more testing.

Phil

 

[dsyzling]

Presumably the comparisons with the Holo May KTE are being done without any form of external upsampling (HQPlayer or pggb) - both to upsampled PCM or DSD (e.g. 256)? Just wanted to clarify how all of these models are being compared. There's quite a lot of praise regarding the May - but quite a few users seem to prefer upsampling using HQPlayer. However most reviews either use the internal over sampling - which most people dislike or use the native R2R implementation which will have its own sound signature without prior upsampling.

 

[Rob181]

Presumably the comparisons with the Holo May KTE are being done without any form of external upsampling (HQPlayer or pggb) - both to upsampled PCM or DSD (e.g. 256)? Just wanted to clarify how all of these models are being compared. There's quite a lot of praise regarding the May - but quite a few users seem to prefer upsampling using HQPlayer. However most reviews either use the internal over sampling - which most people dislike or use the native R2R implementation which will have its own sound signature without prior upsampling.

Oh ye of little faith - you obviously don't know Phil that well - I don't either - never met the guy in fact BUT he is the one of the very few people whose occasional thoughts & opinions I trust - he is a purist & tells it like it REALLY is - without fear or favour & almost everything I have read from Phil I have agreed with...

 

[dsyzling]

Oh ye of little faith - you obviously don't know Phil that well - I don't either - never met the guy in fact BUT he is the one of the very few people whose occasional thoughts & opinions I trust - he is a purist & tells it like it REALLY is - without fear or favour & almost everything I have read from Phil I have agreed with...

Have I disagreed with Phil anywhere within my question? I said 'presumably' because when people don't mention that they are using an external upsampling approach I'm guessing they are using the dac natively - which is fine - I'm just seeking clarification.

I do think one of the reasons people buy the May is to separate the upsampling algo from the dac - if the dac is R2R based with the May's specs, then you can choose the algos and filters externally and feed the dac with the rate you've chosen (as well as PCM/DSD). Now I understand that is an extra level of complication and you may not like the HQPlayer or pggb algos. Equally, you can't always separate the algos with certain implementations of dacs because they are always applied or cannot take higher sampling rates.

 

[213Cobra]

My comparison notes result from evaluating 16/44 - 24/192 via Qobuz. Also Tidal 16/44 and MQA. The streamer is Auralic Aries G2.1; note that MQA is decoded via proprietary, non-MQA code emulating MQA. While I have Roon, the control software used is Auralic Lightning DS. Upsampling in the Aries is turned off. I also have listened to both fed directly by a CDP. No upsampling there, either. The Bricasti M1LE Gold has 15 filer choices. All my listening was done with the "Minimum (Phase) 0" filter. Comparison between May and Bricasti M21 have been with the M21 in R2R mode. I have not gotten around to DSD comparisons yet with any of these DACs.

Phil

 

[Al M.]

Today I compared my May KTE with the Tambaqui.

I really wanted to prefer the Tambaqui, as I'm looking at upgrading to get to that clear next level. But I just didn't.

Unfortunately I only had an afternoon with the Tambaqui, so my thoughts aren't definitive. However, during the time I compared the two, the Tambaqui seemed to be, in comparison, crisper with more pronounced treble. It was also dry and didn't soundstage as well. Images were less solid, less dense, more diffuse, and sounded less real.

The May was warmer. The Tambaqui brighter. Overall, it just didn't do it for me. No matter how much I wanted it to.

The search continues. Hopefully I get to hear a Discrete, Rockna Wavedream, or a Lina soon.

You said you only had an afternoon. Was the Tambaqui properly warmed up, for a day or so at least?

Not writing to defend the Tambaqui which I haven't heard, but just to get an idea about the comparison.

 

[Tuckers]

Presumably the comparisons with the Holo May KTE are being done without any form of external upsampling (HQPlayer or pggb) - both to upsampled PCM or DSD (e.g. 256)? Just wanted to clarify how all of these models are being compared. There's quite a lot of praise regarding the May - but quite a few users seem to prefer upsampling using HQPlayer. However most reviews either use the internal over sampling - which most people dislike or use the native R2R implementation which will have its own sound signature without prior upsampling.

So, I played with HQPlayer and upsampling with the May. I couldn't upsample past 768K PCM in my tests as my computer had wrong type of CPU (I had a top end AMD processor, the HQPlayer only works on higher upsamples with Intel). With this limited testing, using the NOS setting on the May, the results sounded much like other upsampling DACs etc I have heard in the past. Sort of an artificial sharpening at the expense of the organicness and flow. The higher end upsampling is supposed to be much better, but my own lesson is that I much prefer that organicness, life and flow that the NOS provides, and that it wasn't worth investing a sizable amount of money and time to build a PC that would support those higher rates (and be more audiophile compatible).

Also, I definitely thought the May sounded better connected to a streamer than a PC. I even preferred streamers that cost $300. Let alone to my current Lumin U2 Mini with an LPS, which was significantly better again.

So my redux is that those that prefer and pursue high processing power upsampling had different audio priorities than I do.

 

[dsyzling]

So, I played with HQPlayer and upsampling with the May. I couldn't upsample past 768K PCM in my tests as my computer had wrong type of CPU (I had a top end AMD processor, the HQPlayer only works on higher upsamples with Intel). With this limited testing, using the NOS setting on the May, the results sounded much like other upsampling DACs etc I have heard in the past. Sort of an artificial sharpening at the expense of the organicness and flow. The higher end upsampling is supposed to be much better, but my own lesson is that I much prefer that organicness, life and flow that the NOS provides, and that it wasn't worth investing a sizable amount of money and time to build a PC that would support those higher rates (and be more audiophile compatible).

Also, I definitely thought the May sounded better connected to a streamer than a PC. I even preferred streamers that cost $300. Let alone to my current Lumin U2 Mini with an LPS, which was significantly better again.

So my redux is that those that prefer and pursue high processing power upsampling had different audio priorities than I do.

It does seem to be a common criticism or I should say experience - increasing transparency and detail can sometimes lead to reduced body - sounding 'thin'. Clearly there are different approaches to digital to analogue conversion with no free lunches, each may have their benefits and compromises within the perspectives of the listener and dac designer. Of course listeners favour certain characteristics and I'm sure sometimes people just want a change - if you listen to a component for a number of years, it may be that suddenly a new approach is more interesting and engaging for you.

Completely understand what you mean about artificial sharpening - reminds me of TV processing - sometimes the motion looks so unnatural even though it's incredibly sharp - often too sharp and doesn't mimic real life.

I'm not sure I'd assume that increasing beyond 786k would lead to some kind of nirvana. Even the author of HQ Player is slightly dubious of claims that DSD 1024 on the May is out of this world. He's said in the past he'd really like to measure the output because it's pushing the limits of data bandwidth and hardware and he wouldn't be surprised if the measurements show a slightly worse performance with distortion - leading to a more euphonic sound that might sound pleasing to some.

 

[lordcloud]

You said you only had an afternoon. Was the Tambaqui properly warmed up, for a day or so at least?

Not writing to defend the Tambaqui which I haven't heard, but just to get an idea about the comparison.

The Tambaqui was plugged in and turned on for about 45 minutes prior to listening. I've read it needs about 30 minutes to warm up.

 

[Al M.]

The Tambaqui was plugged in and turned on for about 45 minutes prior to listening. I've read it needs about 30 minutes to warm up.

Thanks for the info. Yes, I am doubtful that this is sufficient. I would consider an overnight warm-up of a DAC reasonable. Usually DACs are optimized for running at a certain internal temperature, and that equilibrium is unlikely to be reached in short time spans. I keep my own DAC powered on at all times, per recommendation by the manufacturer.

Having said that, a longer warm-up of the Tambaqui might have given the same result. Emphasis on "might"; it adds an uncomfortable level of uncertainty to the observations, in my view.

 

[lordcloud]

Thanks for the info. Yes, I am doubtful that this is sufficient. I would consider an overnight warm-up of a DAC reasonable. Usually DACs are optimized for running at a certain internal temperature, and that equilibrium is unlikely to be reached in short time spans. I keep my own DAC powered on at all times, per recommendation by the manufacturer.

Having said that, a longer warm-up of the Tambaqui might have given the same result. Emphasis on "might"; it adds an uncomfortable level of uncertainty to the observations, in my view.

Luckily we're not trying to cure cancer or anything.

While I'm sure a longer warm up would have been better. I'm not so sure it would have made enough of a difference to change the sound significantly.

The owner of the DAC thought it warmed up long enough and even touched it to see if it felt warm enough. He was satisfied and he felt that the DAC sounded as he would have expected.

The Tambaqui didn't sound bad by any means. I can't even say the May is a better DAC. But I preferred some aspects of the sound of the May. And I don't think the difference in sound is worth over twice the price.

The Tambaqui lacked the warmth and organic sound that I would expect from a really good r2r DAC. I believe, while I want as much transparency and neutrality as possible, that all dacs lean a certain direction. I just needed the Tambaqui to lean a little more warm. And by all accounts, it lacks the last word in soundstage depth. The May is a champ in this department.

Now.....I will say one thing. And this may be big. I changed the output voltage to 6, from 2. My very first thought when I played a song through the Tambaqui, was that it sounded amazing. But then volume was much lower. So I switched to 6v. It sounded good, but nowhere near as good. I chalked it up to it just not sounding as good at higher volume levels. But it could be the voltage. As a result, in going you try and listen to it again in my system, at the lower voltage. I have a feeling things will be different.

 

[acousticsguru]

Luckily we're not trying to cure cancer or anything.

While I'm sure a longer warm up would have been better. I'm not so sure it would have made enough of a difference to change the sound significantly.

The owner of the DAC thought it warmed up long enough and even touched it to see if it felt warm enough. He was satisfied and he felt that the DAC sounded as he would have expected.

The Tambaqui didn't sound bad by any means. I can't even say the May is a better DAC. But I preferred some aspects of the sound of the May. And I don't think the difference in sound is worth over twice the price.

The Tambaqui lacked the warmth and organic sound that I would expect from a really good r2r DAC. I believe, while I want as much transparency and neutrality as possible, that all dacs lean a certain direction. I just needed the Tambaqui to lean a little more warm. And by all accounts, it lacks the last word in soundstage depth. The May is a champ in this department.

Now.....I will say one thing. And this may be big. I changed the output voltage to 6, from 2. My very first thought when I played a song through the Tambaqui, was that it sounded amazing. But then volume was much lower. So I switched to 6v. It sounded good, but nowhere near as good. I chalked it up to it just not sounding as good at higher volume levels. But it could be the voltage. As a result, in going you try and listen to it again in my system, at the lower voltage. I have a feeling things will be different.

Be sure to check the input sensitivity of your amp or preamp, in order not to overdrive its input circuitry. 6V outputs are mostly used in studio settings where cable lengths may run several hundred feet (to compensate for series resistance). The problem is that when one "mildly" overdrives the input of an amp or preamp, it may not be enough for the listener to be able to tell the input stage is distorting, but enough that it sounds "off". Amps with an input sensitivity greater than 2V are relatively rare (but they do exist). Basically, that number is supposed to be equal or greater than the output voltage of the source component.

Greetings from Switzerland, David.

 

[lordcloud]

Be sure to check the input sensitivity of your amp or preamp, in order not to overdrive its input circuitry. 6V outputs are mostly used in studio settings where cable lengths may run several hundred feet (to compensate for series resistance). The problem is that when one "mildly" overdrives the input of an amp or preamp, it may not be enough for the listener to be able to tell the input stage is distorting, but enough that it sounds "off". Amps with an input sensitivity greater than 2V are relatively rare (but they do exist). Basically, that number is supposed to be equal or greater than the output voltage of the source component.

Greetings from Switzerland, David.

That wasn't an issue. I believe the input sensitivity of my amps are 9.8.

 

[Gjo]

I will post a longer summary of comments on the Holo May KTE soon, but in the context of this thread, I will just say that yes, the HM-KTE is warmer. I have not heard a Tambuqui in a clear comparative setting, but while it sounded organic and good, I did not hear it as transformative ala its reviewed reputation. The Holo May sounds highly competitive with other ~$6000 DACs, in terms of low noise, musicality, tone and dynamic elasticity. But I was surprised to prefer the Bricasti delta-sigma 80% of my listening time to the two over a period of more than six months and counting. More to report with more testing.

Phil

Those impressions align with mine.

At the time of my Holo May KTE ownership, I also owned a Bricasti M3 with LAN. I volume matched the DACs, and after a few days played some familiar tracks and switched back and forth using my preamp's remote (SMc Audio VRE-1C). Neither my wife nor I could tell which DAC was playing. They sounded virtually identical. Considering the Bricasti M3 had the advantage of an onboard Network card, it seemed the logical choice to keep.

 

[audioresurgence]

Hi,
Has anyone tried a Sonore UltraDigital with the May, to get i2s input?

I just went from using USB to adding the ultradigital and it sounds really nice. Only problem encountered is that it sounds phasey when passing DSD from my Aurender N10.

Clearly I need to change a pin configuration either in the UD or the May KTE. Haven’t figured this out yet and wondered if anyone else has?

I do have a new Iconoclast AES/EBU cable coming soon so perhaps this will exceed the sound quality of the UD and i2s making this a nonissue.
Anyone using the UltraDigital care to share their setup?
Cheers
CAH

 

[olivier]

Hi,
Has anyone tried a Sonore UltraDigital with the May, to get i2s input?

I just went from using USB to adding the ultradigital and it sounds really nice. Only problem encountered is that it sounds phasey when passing DSD from my Aurender N10.

Clearly I need to change a pin configuration either in the UD or the May KTE. Haven’t figured this out yet and wondered if anyone else has?

I do have a new Iconoclast AES/EBU cable coming soon so perhaps this will exceed the sound quality of the UD and i2s making this a nonissue.
Anyone using the UltraDigital care to share their setup?
Cheers
CAH

I use the Holo Red with i2s into the May - a real good match. The Red needs about half an hour to warm up
Best