But Sonny from Phison has made the gap much narrower with the PD2 USB implementation. But let's not give the impression that simply having any old Ethernet interface is superior. What MSB has done is not any old Ethernet interface. But I can assure you it's not a trivial matter to simply implement Ravenna or else Larry in the boys would be using it. It wasn't until Merging came out with their NADAC that it was even introduced to the high end DAC world. However I honed in on it early in the game.
In my experience I've found there's no reason to lie if you're not doing anything wrong
But you may be surprised on how close it comes. When DSD is fed to the MSB, that fancy R2R DAC section isn't even being utilized. There's other diamond owners using HQplayer to upsample PCM to DSD and like it better than the PCM through the R2R. How MSB handles the DSD is very close to how the PD2 does. Only the PD2 has a more direct path as it doesn't have to pass through the SHARC DSP chips. So there's more than 1 way to skin a cat. But I am a believer that clocks make a difference. I found out where MSB sources their uber quality clocks from and I've talked to Sonny about offering clock upgrades like MSB does. But they really are expensive. Many people say they are ripping you off, but clocks that precise are huge money. However if people are willing to pay, why not have the option? I've also got a lead on some new clocks that will be close to the Femto 77 for much cheaper. So we will see how things go.
Well I'm not going to claim that the PD2 will beat the MSB diamond DAC with Ethernet renderer and Femto 33 clock. The clock upgrade alone is triple the price of the PD2. And you get a preamp thrown in as well.
Yes I think your setup would be an incredible benchmark. But the only problem is, if it ends up being better, I won't be able to sleep at night unless I beat it
LOL, we will keep that a secret.
There's actually 2 different paths for the DSD in the PD2. I can use the same block diagram as Srajan used in the review to explain. 1 path is through the AKM chip. If sent through this path it works much like the Merging NADAC as it goes through the SRC/SDM section of the DAC chips so DSP can be applied for things like volume control and additional filtering. The second way is the superior way. It bypasses the processing of the AKM chip, and uses high end discrete components on the output board to filter the DSD. This makes it very similar to the DSD section of the Lampi GG, although the DSD traffic is being directed through the chip. Since this is such a minimalist approach with the purest signal path, using external PCM to DSD upsampling like Hqplayer, and like I will have in my server, transforms the PD2 into super DAC territory. The combination of both makes it more like the Mola Mola, Emm labs, dCS etc DACs, only rather than using low powered FPGA's or DSP chips for the SRC/SDM, (sample rate conversion, and sigma delta modulation) a much more powerful quad core I7 Intel processor is used. This allows far more sophisticated algorithms to be used than are possible on FPGA's or DSP chips. For example the filter I will use in my server for upsampling has 4 million taps!
You can read a bit about taps here and why they matter:
"In order to do this the filters need to have infinite long tap lengths. Currently all reconstruction filters have relatively short tap lengths - the largest commercial device is only about 256 taps. It is due to this short tap length and the filter algorithm employed that generates the transient timing errors. These errors turned out to be very audible. Going from 256 taps to 2048 taps gave a massive improvement in sound quality - much smoother, more focused sound quality, with an incredibly deep and precise sound stage.
The initial experiments used variations on existing filter algorithms. Going from 1024 taps to 2048 taps gave a very big improvement in sound quality, and it was implying that almost infinite tap length filters were needed for the ultimate sound quality. At this stage, a new type of algorithm was developed - the WTA filter. This was designed to minimise transient timing errors from the outset, thereby reducing the need for extremely long tap lengths. The WTA algorithm was a success - a 256 tap WTA filter sounded better than all other conventional filters, even with 2048 taps. WTA filters still benefit from long tap lengths; there is a large difference going from 256 taps to 1024 taps.
The new Chord products using WTA filters all start with 1024 taps. The filters are implemented in FPGAs (Field Programmable Gate Arrays) using a specially designed 64-bit DSP (Digital Signal Processing) core".
http://www.chordelectronics.co.uk/chord-dac-technology.asp
The yellow path is the DSD direct mode. And the other path passes through the chips SRC/SDM section like all DAC's with the Sabre chip.
How MSB does it, they don't talk about much openly. But it goes through the Sharc DSP chips, and either uses discrete components for filtering, or low pass filters implemented in the DSP chip.
http://www.6moons.com/audioreviews/EML45/45s.html
Ok so let me put this in more technical terms. The PD2 can do DSD the old fashion way through regular old DAC chips or it can go a new snazy way which is largely chipless (at least no DAC chip like the Lampi). So this way does not use a traditional dac chip but does use a wiz bang chip for filtering with taps approaching infinity which is well beyond what Andreas, Ed, Paul and the other dsd direct boys have done with field arrays. Furthermore infinite taps beat the snot out of those other less powerful topologies.
If this is generally correct, how does it differ from what Lucasz is doing (setting aside his fine output stage which IMO is the biggest reason people are so enamored with his DAC, NOT the dac itself. DHT can of course be mesmerizing but this has nothing to do with the digital process. It provides an exceptionally musical output stage with all the benefits and negatives of DHTs generally although his execution is quite good.)? How does Lacasz do his filtering and how many of them there taps has he rounded up in his design compared to your pertin-near 4 million?
Lastly, your right about MSB. Its a secret.
