Why SACD Disc playback sounds better (less jitter?) than its equivalent ripped DSF file / streaming playback

[Kiko1974]

I'd like to add more information to this thread:
First, Sabre D/A IC's despite their their tested lab specs, it's well known that Sabre chipsets, even its top of the like IC, convert internally DSD to PCM. This will sure has an impact on how DSD sounds converted by Sabre ICs.
Burr Brown or AKM can convert DSD to analogue natively. The SMSL D400 PRO that I own uses the top of the line AKM two IC's solution: AK4191 that does oversampling and Delta Sigma modulation, and an AK4499EX that is the actual D/A IC. This combination as implemented on the SMSL D400 PRO can convert DSD natively (It can bypass digital volumen control or any other kind of processing), and It shows. The SMSL D 400 PRO is an inexpensive D/A converter but It sounds great for the price and It really shines when converting DSD.

Regarding SSD's, I don't buy off the shelve

 

[Blumlein 88]

Regarding SSD's, I don't buy off the shelve

Me neither. I usually buy off the internet.

 

[antcollinet]

I'd like to add more information to this thread:
First, Sabre D/A IC's despite their their tested lab specs, it's well known that Sabre chipsets, even its top of the like IC, convert internally DSD to PCM. This will sure has an impact on how DSD sounds converted by Sabre ICs.
Burr Brown or AKM can convert DSD to analogue natively. The SMSL D400 PRO that I own uses the top of the line AKM two IC's solution: AK4191 that does oversampling and Delta Sigma modulation, and an AK4499EX that is the actual D/A IC. This combination as implemented on the SMSL D400 PRO can convert DSD natively (It can bypass digital volumen control or any other kind of processing), and It shows. The SMSL D 400 PRO is an inexpensive D/A converter but It sounds great for the price and It really shines when converting DSD.

Regarding SSD's, I don't buy off the shelve

There seems to be an echo in here

 

[Mart68]

There seems to be an echo in here

The RT60 is just too high.

 

[threni]

What I mean with that, and that is just my opinión, IS that just like good optical disc drives performs better than cheap ones, the quality of an SSD, the internal flash memory It has, its SATA controller (the Silicon Power SSD I got is supposed to have an Intel SATA controller and Toshiba flash memory) and the quality of the SATA to USB controller, MAY have an impact on sound quality.

Does this extend to photos? Would my photos look brighter, sharper etc if I use this vs that RAM, hard drive, USB controller? Perhaps my accounts will be more accurate too? Of course not - they're just bits. Same for audio. It doesn't matter how they're stored. If you can read data from the optical drive, hard drive, network etc etc in time and without error it'll sound the same.

 

[Scytales]

I'd like to add more information to this thread:
First, Sabre D/A IC's despite their their tested lab specs, it's well known that Sabre chipsets, even its top of the like IC, convert internally DSD to PCM. This will sure has an impact on how DSD sounds converted by Sabre ICs.
Burr Brown or AKM can convert DSD to analogue natively. The SMSL D400 PRO that I own uses the top of the line AKM two IC's solution: AK4191 that does oversampling and Delta Sigma modulation, and an AK4499EX that is the actual D/A IC. This combination as implemented on the SMSL D400 PRO can convert DSD natively (It can bypass digital volumen control or any other kind of processing), and It shows. The SMSL D 400 PRO is an inexpensive D/A converter but It sounds great for the price and It really shines when converting DSD.

I am not sure what you mean by "converting DSD natively", but I just want to point out that it is almost a certainty that ESS Sabre DACs don't decimate DSD input signal into a low sample rate PCM (less than the DSD sample rate, such as 192 or even 384 or 764 ksps).

Proof is the specifications of the IIR (recursive) digital filters implemented by ESS on-chip to filter out the DSD HF quantization noise prior re-modulation by their own delta-sigma modulator [1] :

As can be seen, the frequency response of the 3 available low pass filters goes up to about one and a half MHz. That is roughly the Nyquist frequency of so-called DSD64, which is of course exactly half the DSD sample rate, ie 1.4112 MHz. That means that the digital low pass filter implemented by ESS in the ES9008 doesn't decimate DSD input signal into low sample rate PCM that have much lower Nyquist frequency and would consequently not allow the digital filters to have frequency responses such as those shown above. More recent ESS Sabre DACs accept DSD at much higher sample rates, but the specifications of the filters are scaled depending on the DSD input clock [2]. That means that for each filter, the -3 dB corner frequency and the top end of the frequency response is pushed two times higher in frequency every time the DSD input sample rate is doubled. Thus, the original Nyquist frequency of the DSD input signal to the DAC chip is preserved whatever the DSD input sample rate is.

Obviously, the IIR digital low pass filter do produce a multibit code that is re-modulated back to a lower bit depth signal by ESS's delta-sigma modulator, a typical signal path and conversion technique this day. But fact is that the frequency content of DSD signals is not changed, subject to the effect of the necesary low-pass filter to attenuate DSD shaped quantization noise.

[1] This graph is taken out of the ESS ES9008 DAC chip datasheet (Rev. 1.3 of February 15, 2015) and is identical in the datasheet of the ES9012 and ES9018 (February 11, 2010).
[2] ES9010K2M datasheet, April 27, 2021, and ES9038PRO datasheet, February 8, 2021.