Phison Audio PD2 Preamp/DAC

Back to the operating system interrupt response! That could be the cause, not the DOP itself.


Sent from my iPhone using Tapatalk

Yeah whatever causes it doesn't matter. Still not as good as native. Even comparing native DSD tracks, I found native DSD sounds better than the DoP. I know DoP is lossless, but it still adds processing on both the computer and the XMOS chip. I think it's this extra processing in the XMOS chip adds a tiny bit of haze to the sound. I've found the same with all DAC's that support both native and DoP DSD.
 
Ouch. Okay.

What is your go to source? Vinyl? Why are you sniffing around the Phison tree? MSB falling short in some way? I know your speakers are laser beam accurate. Are you wanting more body and warmth from your digital? What server/software do you use?

How do you like the Torus? There is a thread on Whats Best Forum extolling the virtues of active power filtration, etc....I know that thing must have been a bitch to install. I was looking at the equivalent Equitech product at one point and installation is a pain....

First understand I have been working on building this system for a couple of years and the room is finally under construction as we speak. It will be replacing an old B&W/ Rotel system I have lived with for many years. I have been slowly evaluating gear and buying pieces I like for a while. I now have everything but an arm.

With that, let me give that question list a shot.

Source- I expect vinyl to continue as my source of choice as it has since the 70's but we will see. The new MSB is awfully good.

Sniffing around Phison - No I think I am going to love the MSB. I am just intrigued by the brain on Blizzard. I have never joined WBF but I have lurked there often and find him a very interesting fellow.

Server - On Paper I really like what Larry and the boys have done with the renderer at MSB. I went to great lengths to avoid all of the half-baked server solutions cuz I don't see anything but a theoretical negative impact to SQ in exchange for a "pretty" user interface. That ain't my thang. Converting straight to i2s from ethernet in a direct topology avoids a lot of problems with ground noise, packet junk, power supply perturbations, etc. I am going mimserver on nas.

Power - My thoughts a pretty simple here. Everything outside the listening room must be isolated by transformer. Were many fail here is they use way, way to small of transformers and kill instantaneous current which is worse many times than leaving the power a little dirty. I am huge believer in the merits of vast amounts of clean current. (Like Caelin over at Shunyata). Once you got the power to the room all thats left is component isolation and proper grounding. I want to isolate the noise between gear with a passive unit. Lots of good ones. I picked the inline Furutech units. The last requirement for component isolation is to make sure the gear doesn't cause line modulation. This is one of the biggest reasons for good power cords. Vince Galbo is pretty smart on this. All power supplies modulate but when the modulation reflects off a high impedance point it buggers up the PS performance. Isolation is done by running either real long, low gauge PC's or better running very good outlets and wires in the walls to a massive transformer (which has zero) impedance. All the PS modulation is eaten up by the transformer and never reflects. Last I got a separate star ground system for all five 20 amp outlets in a cavity right behind my listening wall so each outlet's electrical potential at the ground is identical.

Lastly, yes indeed the specific gravity of ferrous metals is quite high (in other words don't drop a Torus on your toe.)
 
But i have found out that not all machines can handle DOP 256. The hardware handling the USB part is not fast enough


Sent from my iPhone using Tapatalk
 
No Mike , remember it is reclocked.


Sent from my iPhone using Tapatalk

Well it's very very little difference. But I could hear it. Anyways it will be a nice option. But for my clients in North America, at minimum I will send them a free Linux streamer with their PD2 if they don't want my server. This way they can enjoy native quad DSD upsampling from any computer Mac or Windows. Having heard the difference PCM-native quad DSD myself, and feedback from my first client, it's a no brainer.
 
I think the reason your streamer works well, is that it has not a graphic gui to handle as well


Sent from my iPhone using Tapatalk
 
I think the reason your streamer works well, is that it has not a graphic gui to handle as well


Sent from my iPhone using Tapatalk

That's part of the reason. Also very low noise footprint. But that matters less with your awesome reclocker.


But it can do both DoP and native DSD. And I compared them both upsampling to DSD 128 and can hear a distinctive difference between them. However it's a moot point anyways because native is required for quad DSD now and quad DSD with Jussi's new DSD7 256+Fs modulator is in another universe from DSD 128. This modulator only works with DSD 256 or higher.

I have a new streamer now that's better than the Superstream from WBF. And my first PD2 client is buying the Sonore Microrendu to compare to my streamer. It's another option that has HQplayer NAA input that they say is the ultimate. We will see if it's better.

http://www.microrendu.sonore.us
 
Sonny and Blizz, What are you two gunna do when you are both selling similar wiz-bang dacs??????

We won't be. The purestream is a very simple pure DAC with no preamp, or phono stage option. And it will use Sonny's DAC. So it will be a joint venture.
 
If it appears that Sonny and I are not seeing eye to eye on a few points it's because Sonny still has no idea how good his own DAC can even sound. He hasn't even heard PCM upsampled to quad DSD with it yet. I have been testing his prototypes since October and telling him all about it, but he still hasn't tried my streamer, or bought a server with an I7 quad core processor that can handle upsampling PCM to quad DSD with Hqplayer. Once I send him my server, he will be in for a shock. :)
 
I have. [emoji851][emoji41] remember I have a Linux


Sent from my iPhone using Tapatalk

Okay explain the setup you are using for quad DSD upsampling? What do you think of the DSD7 256+ modulator compared to the others, and what is your favorite filter? Also how do you like the Roon interface? Do you find the quad DSD upsampling is a game changer with Tidal like both my clients and I do? How about Jussi's new MQA filter. Have you tried it with Tidal MQA upsampled to quad DSD with the DSD7 256+ modulator?
 
Okay it's official. Starting today, I will be including a free Linux based streamer with all PD2's sold in North America. Once ready, for an additional $2500 upgrade cost, the streamer can be upgraded to my SOTA server. The $2500 will include 1 year subscription to both Roon and Tidal. This streamer is an upgraded version of the Superstream streamer that was featured here:

http://www.whatsbestforum.com/showthread.php?19221-Streamer-to-kill-the-big-boys-for-under-250

It's a very compact unit that can be tucked away behind the PD2 so you won't even see it. Just like the Sonore microrendu. I will make a feedback section on the streamer on my website where users can share their experiences with and without the streamer.
 
Hi guys. I seen that the 6 moons review was updated with HQplayer experimenting. For 1 he is using the worst filters the software has. The -2 filters are for low powered computers and sound is like nails on a chalkboard to my ears. The DSD5 modulator is very dull sounding and not to my tastes either. Also I can see that the volume knob. is not in the max position. This indicates DirectSDM isn't checked, and the sound is being processed through an additional layer of sound degrading filtering. And finally he's only upsampling to DSD 128 over DoP. Please anyone trying HQplayer with any DAC don't use these settings. If your computer can't handle processing the right filters/modulators, use a different media software.

For the PD2 my recommended filter is Poly-Sinc-Short-MP and modulator DSD7 256+Fs. And very important to have "DirectSDM" checked in the settings.

c4134667294e7ebfc521c289d0e85db8.jpg
 
Blizz, As a a purely theoretical matter, talking strictly about sound quality (NOT INTERFACE), is it not true that the the best this server could sound would be equal to but never better than the MSB solution? If not can you explain why?
 
Blizz, As a a purely theoretical matter, talking strictly about sound quality (NOT INTERFACE), is it not true that the the best this server could sound would be equal to but never better than the MSB solution? If not can you explain why?

Are you talking about the filters? If so its all about the algorithms used. Put it this way. The algorithms that MSB uses in their DSP chips could run on a quad core I7 Intel processor at probably 1-2% CPU load. But there's no way on earth that HQplayer's best algorithms could run on the DSP chips. Not only that, you have all the extra circuitry to run through when you use DSP chips. DSP chips are not 100% noise and jitter free. If you are using a computer as a server anyways, you are running the data through it regardless. A well implemented USB or Ethernet interface will block all noise from a connected computer. If you look at the PD2's DAC board for example, you'll notice that there's no FPGA's or DSP chips. This allows the data path to be very short with low noise and jitter. I think this is 1 reason it sounds so good.

Now if you read the 6 moons review you will see that Srajan says he prefers "PCM as PCM" well saying that proves that he has no idea how a modern SDM DAC chip works. The only DAC's that process "PCM as PCM" are R2R DAC's like in the MSB. With his experiment using the lousiest filters/modulators in HQplayer, all he is saying is he prefers the SRC/SDM built into the AKM chip, over the particular SRC/SDM combination chosen in HQplayer. That's like giving someone a set of speakers to audition with a blown midrange and expecting a positive result. But the worst part is 99% of people reading that review who don't know any better, are going to walk away thinking that the PD2 is lousy at DSD.
 
What do you mean by saying DSP is not 100% noise free?


Sent from my iPhone using Tapatalk

There is some noise added by a DSP chip. You can even see in the datasheets. It may be very low, but it's still some. If there's no DSP chip at all, there will be less noise would there not be? What's the shorter path, a 2cm copper PCB trace, or a SHARC DSP chip?
 
They do not add noise, unless the DAC or ADC is integrated in the DSP. The rest is in the digital domain. If as I am working on are choosing a DSP you just have to be sure that the MAC run with enough bits.


Sent from my iPhone using Tapatalk
 
Back
Top