This is perhaps true for me also. I purchase FLAC content 'studio masters', mastered in 24/96 and they sound amazingly good!
The only reason I was enquiring about 192KHz, is because now, increaslingly you can select 24/192 when purchasing content,, so my concern was about 'future proofing' my music collection..
But anyway, I am happy now with knowledge that everything is converted to 96K for processing by the DSP..
Ta
CM
Hi all,
1. The internal data rate of the DCX is always 96 kHz independent of digital data input. So if you use 192 kHz sample rate the sample rate converter (SRC) of my SRC/clock mod would down sample your data stream to 96 kHz.
Typical CDs or MP3 files use a 44.1 kHz sample rate. It does not make sense to up sample this date to 192 kHz. Sound quality is not getting better but worse because of the additional and unnecessary up sampling which adds distortion.
So it’s recommended to use the sample rate of the data source (e.g. 44.1 kHz for CD) and not to up sample data to avoid unnecessary sample rate conversion.
2. The SRC/clock mod is not only an improved sample rate converter (SRC) but an ultra low jitter clock too, which replaces the internal 24.576 MHz high jitter clock of the DCX built with a simple crystal.
3. The SRC/clock mod replaces the erroneous (dull sound) CS8420 chip. It needs one additional wire (+8-9V) for the two on board ultra low noise and high PSRR power supplies. These supplies are used for the improved PLL and sample rate converter and for the ultra low jitter clock oscillator.
The improvement of the SCR/clock mod was measured by Ergo: http://www.pilghamaudio.com/uploads/PDFs/Jitter.pdf
The diagram shows much better performance compared to the original DCX.
Please also have a look at the attached table which shows the different available mods for the DCX.
Good luck
OK, you have no experience with the mod.
The measurement http://www.pilghamaudio.com/uploads/PDFs/Jitter.pdf not only shows a better noise floor.
With the original DCX clock the 12 kHz test input spreads up to some extend to a frequency spectrum between 11 and 13 kHz (yellow line). With the low jitter clock mod you get a clear 12 kHz output only (green line). That makes the more precise sound.
The measurement http://www.pilghamaudio.com/uploads/PDFs/Jitter.pdf not only shows a better noise floor.
With the original DCX clock the 12 kHz test input spreads up to some extend to a frequency spectrum between 11 and 13 kHz (yellow line). With the low jitter clock mod you get a clear 12 kHz output only (green line). That makes the more precise sound.
Irrelevant.
It is still under audibility thresholds for every jitter test I have seen, eg Dunn, so simply saying it is better because of a small difference >100dB down is no proof of anything. Might as well be homeopathy.
Last edited:
Frank, I have no experience with the mod, but the improved measurements seem nice. However, to be fair, the differences are all in the -120dB range, which seems a bit of a stretch to claim audible benefits. If it weren't so expensive, I'd probably do it just for the theoretical advantage.
In the spectra you showed, is the ONLY difference the clock? I certainly saw some measurable improvement (cheaply) when I replaced the output stage with Jan's passive RC solution.
In the spectra you showed, is the ONLY difference the clock? I certainly saw some measurable improvement (cheaply) when I replaced the output stage with Jan's passive RC solution.
Actually the distortion level is about -70dB (difference to fundamental). Because of the zoomed diagram (cut off at the top) you can't see this. I could hear the difference clearly. There is a lot of positive feedback like:
As I mentioned numerous times, your board brings DCX's performance to a completely different level that is clearly audible and in my mind presents dramatical improvement equal to eliminating opamps in output. The very same excitement about the board I had was confirmed by very advanced users; to mention few - Jan Diden and Ergo. We spoke about this in this thread on numerous occasions, and we measured and posted results. If none of that was not helpful for someone to have a sure path to a success than nothing will.
So I seem not be the only one who can hear a clear improvement. But all the experiences might be based on Beraneks Law, which means all mods you are doing yourself are homeopathy.
The attached list shows all (or at least most) of the available mods for the DCX. It's up to you to find out which are the best for you.
As I mentioned numerous times, your board brings DCX's performance to a completely different level that is clearly audible and in my mind presents dramatical improvement equal to eliminating opamps in output. The very same excitement about the board I had was confirmed by very advanced users; to mention few - Jan Diden and Ergo. We spoke about this in this thread on numerous occasions, and we measured and posted results. If none of that was not helpful for someone to have a sure path to a success than nothing will.
So I seem not be the only one who can hear a clear improvement. But all the experiences might be based on Beraneks Law, which means all mods you are doing yourself are homeopathy.
The attached list shows all (or at least most) of the available mods for the DCX. It's up to you to find out which are the best for you.
The yellow line in the diagram http://www.pilghamaudio.com/uploads/PDFs/Jitter.pdf shows the unmodified DCX. The answer depends on whether a -70dB sideband is impeccable or not.
But interesting that the 44k1 test shows notjhing like the same peak level.
And I may be mistaken, but by far the majority of digital sources that will ever be played into a DCX will be 44k1, simply because the number of sources at that sampling rate will outnumber all others by several orders of magnitude. How much 48k source material is actually out there?
Talk about selective advertising.
And I may be mistaken, but by far the majority of digital sources that will ever be played into a DCX will be 44k1, simply because the number of sources at that sampling rate will outnumber all others by several orders of magnitude. How much 48k source material is actually out there?
Talk about selective advertising.
I’m sure you have also seen this diagram http://esken.net/ergo/blog/wp-content/uploads/2008/01/19k-20k-imd-yelloworig-greenergos.png which is based on 44.1 kHz. The difference between signal and first sideband is about -70 dB.
So let’s talk about selective advertising. BTW: When you buy a SRC/clock I personally loose money because I donate the PCB and the small earnings are for a school in Ghana AfricaAction - Gumyoko - Montessori School
So let’s talk about selective advertising. BTW: When you buy a SRC/clock I personally loose money because I donate the PCB and the small earnings are for a school in Ghana AfricaAction - Gumyoko - Montessori School
Attachments
And the difference there is still not going to be audible. It was still very selective advertising because out of the two tests you could have chosen, 44k1 or 48k, the one which looks worse, and most people are never even going to use except rarely was the one selected to show the "worth" of this mod, with no link to the other to show fairness. It was entirely presented to show the mod in it's best light even if that was not how most people were ever going to use it.
That's nice, but entirely irrelevant to the discussion on the worth of the mod