New Holo Audio Spring2 DSD1024 DAC

[matthias]

18 hours ago, Samuel T Cogley said:

Have there been any comprehensive tests done on Kitsune vs. non-Kitsune models to validate the value added by Kitsune? (tests by Kitsune would obviously not qualify)

 

I have so far not seen any comparisons about KTE vs. non-KTE versions.

For me to go for the silver transformer of the KTE does  not  make sense.

 

Matt

[Miska]

2 hours ago, matthias said:

Maybe someone can shed some light on why the Spring is rather expensive in Europe in comparison to the US?

 

Maybe differences in tax system and how prices are presented due to such differences?

 

If you import it from US you'll end up paying customs and VAT on top... In US, there's a possible state tax added on top?

 

I have L2 versions of both Spring 1 and 2.

[gdpr]

23 minutes ago, Miska said:

 

I have L2 versions of both Spring 1 and 2.

 

Hi Miska

On the Spring 1 the Output level with DSD is -6dB compared to PCM.

still the case with the Spring 2?

Dirk

 

[Miska]

5 hours ago, ddetaey said:

On the Spring 1 the Output level with DSD is -6dB compared to PCM.

still the case with the Spring 2?

 

From my tests where I think I used -1 dB 1 kHz input... My figure for PCM, from balanced output is 3.8V RMS. And for DSD, from balanced output is 1.7V RMS. Noise floor level is a bit lower for DSD, but the difference is pretty small.

 

So yes, quite close to 6 dB difference.

[gdpr]

2 hours ago, Miska said:

 

From my tests where I think I used -1 dB 1 kHz input... My figure for PCM, from balanced output is 3.8V RMS. And for DSD, from balanced output is 1.7V RMS. Noise floor level is a bit lower for DSD, but the difference is pretty small.

 

So yes, quite close to 6 dB difference.

Thanks for the feedback. Still scratching my head why they are doing this.

Dirk

[Miska]

12 minutes ago, ddetaey said:

Thanks for the feedback. Still scratching my head why they are doing this.

 

I think it is just side effect for having same reference voltages and analog stages for both PCM and DSD...

 

Only ESS Sabre drives both PCM and DSD at the same level, but then you need to be careful if you encounter overmodulated DSD content (some SACDs have that). In fact -3 dB for DSD would be correct and that is pretty much what TI/BB has and AKM has -3.5 dB. This is because SACD spec allows short term +3 dB overmodulation although it is no recommended.

 

DSD naturally has kind of "-6 dB" level, that's why also HQPlayer has that +6 dB gain option when converting to PCM - when you combine that with the recommended -3 dB volume setting you come up with -3 dB there too.

 

So essentially to compensate, they would either need to have extra 3 dB headroom in analog stages and switch in 6 dB of gain to get in-level. Or alternatively kick in just 3 dB gain to have 3 dB headroom and accept the level difference (which is many other DACs do). But that requires extra electronics to change gains of the analog stage, or switch the analog stage altogether.

 

I don't remember right now how it was for T+A DAC8 DSD...

[gdpr]

2 minutes ago, Miska said:

 

DSD naturally has kind of "-6 dB" level, that's why also HQPlayer has that +6 dB gain option when converting to PCM - when you combine that with the recommended -3 dB volume setting you come up with -3 dB there too.

 

Converting every thing to DSD, here is where I am lost. DSD itself has kind of -6db level and in the Dac another -6db is applied.

If I than, on top of that, apply -3 dB volume setting, I am getting close to -15dB.

Hardly any music left.

Dirk

[Miska]

41 minutes ago, ddetaey said:

Converting every thing to DSD, here is where I am lost. DSD itself has kind of -6db level and in the Dac another -6db is applied.

If I than, on top of that, apply -3 dB volume setting, I am getting close to -15dB.

 

No no, that's not what happens. There's no extra -6 dB, the figures I gave you already include that -6 dB which is not there twice, only once. I tried to explain why, without going too much into details, maybe I didn't manage so well...

 

If you play PCM or DSD, both with HQPlayer volume set to -3 dBFS, the relative level difference between the two at DAC output is the 6 dB. Compared to playing PCM with volume turned to 0 dBFS and DSD playing with -3 dB setting the difference would be 9 dB.

 

46 minutes ago, ddetaey said:

Hardly any music left.

 

6 dB difference is not much in the end. In fact, you just have less attenuation by pre-amp volume setting before power amp's about 26 dB of gain...

 

Once again, one feature I'd call "gain matching". If the output level would be higher, you would need to attenuate it more with preamp's volume control. These days pre-amps don't usually actually amplify anything, instead they are attenuator devices... You are probably not anywhere near running out of volume control range?

 

With the current DSD output level, highest listening volume on my preamp is set to -26 dB... At night time it is around -50 dB. This with HQPlayer volume set to -3 dB. I would probably get better SNR by dropping HQPlayer volume to -12 dB or even more.

 

[gdpr]

Hi Miska

I follow your reasoning, and I am aware that a preamplifier is in fact attenuzting the signal.

The scale of my Ayre KX-R goes from 0 to 60 in steps of 1dB ( with 60 equals no attenuation)

With my T+A 8 DSD dac my average volume setting is 38 for PCM upsampled to DSD. For native DSD bolume setting is about 44 as music is recorded at a much lower level (in fact around - 6 dB most of the time).

When  I used the Holo Spring however, I needed to increase my volume setting with an extra 6 dB to have the same music loudness level, 45 for upsampled PCM and (at least) 50 for native DSD.

All with the recommended - 3dB settinh in HQplayer.

So what I do not get, since on loudness level , music distributed in DSD in general has already a much lower loudness level than PCM.

So why does a Dac manufacturer applies another 'safety margin'  in there.

I believe you have both dac's yourself. Maybe you could compare their output level?

Dirk

 

[4est]

21 hours ago, matthias said:

 

I have so far not seen any comparisons about KTE vs. non-KTE versions.

For me to go for the silver transformer of the KTE does  not  make sense.

 

Matt

I'd have to agree about the silver transformer. It just never made sense as to why one would plate a 50/60 Hz power transformer winding with silver. It would seem to me that they would have been better off replacing the toriodal one with an R-core (or ?) that has true separation of the primary and secondary to reduce capacitive coupling of noise. What do I know though...

[Veri]

1 hour ago, 4est said:

I'd have to agree about the silver transformer. It just never made sense as to why one would plate a 50/60 Hz power transformer winding with silver. It would seem to me that they would have been better off replacing the toriodal one with an R-core (or ?) that has true separation of the primary and secondary to reduce capacitive coupling of noise. What do I know though...

 

But it's made of silver!...

 

[Miska]

5 hours ago, ddetaey said:

So what I do not get, since on loudness level , music distributed in DSD in general has already a much lower loudness level than PCM.

So why does a Dac manufacturer applies another 'safety margin'  in there.

I believe you have both dac's yourself. Maybe you could compare their output level?

 

Yes I have both. I think I found measurement screenshots for DAC8 DSD where I can see the signal levels. It was quite a while ago and I'm not 100% sure what exact volume setting I used in HQPlayer, if it was -1 (some DACs get upset by exact 0 dBFS so I try to stay consistent and keep same settings always to make the results comparable). But anyway relative level is what matters here not the absolute value. So for DAC8 DSD: PCM=4.3Vrms, DSD=4.1Vrms (balanced out). So not exactly same but pretty close. So they are likely employing some gain for DSD and maximum peak level with DSD will exceed those of PCM. 

 

But the difference you are talking about is more related to mastering. As you know PCM largely suffers from loudness wars, so record companies tend to use some amount of compression to make it sound louder. And then they use normalization so that peaks hit exactly 0 dBFS - or commonly grossly exceed it causing horrible amount of clipping. While DSD producers usually try to record with some safety margin away from 0 dBFS and don't employ compression or normalization. So even if DSD content actually has peaks exceeding nominal 0 dB level they may not sound as loud due to lack of compression. From the PCM albums I have (most of my music anyway) there are massive differences between albums how loud they sound with same volume setting, although they all reach 0 dBFS periodically.

 

Switching from some DXD material to regular RedBook pop makes me turn down volume by almost 20 dB...

 

Usually; not sounding as loud = less compression. That is a good thing, peaks can be high, but average level is lower. That's one area where difficulty of processing DSD helps us.