To me the APS transformer upgrade is a worthy mod since the sonic improvement was quite nice after I have it installed. After the Massive update the MK2 is sounding incredible. I now think the initial “warmer” bottom was just a normal break-in process because now the unit is barely warm. In fact, the stock unit ran quite warm at beginning too as I remembered.
I have not experienced any problem with my MK2 so far, and with the “normal” temperature now I expect it will serve me for a long time🕺🏻
Hi folks, I finally get a DS MK2 (tomorrow), happy bday week to me 
Looking forward to seeing how it does. Heavily worked DS MK1 holdin’ meanwhile of course, playing right now 
We shall see.
TK
We will be interested to hear your thoughts after your new toy has been given a chance to break in.
Me too, thank you. Delivery is delayed now sigh lol. But I think we should have some idea by the weekend.
TK
Well she’s still breaking in but I’m happy to say I do like what I’m hearing, a lot. The MK2 is more “present”, forceful and engaging than the MK1. There’s seems to be more beefy but natural upper midbass drive to the music, makes it more engaging. Vocals are smooth, nice and real sounding. There’s also more but natural detail extracted. Happy to say I don’t find much anything to complain about the MK2 fw Good job TS!
Still getting a handle on it and letting it break in.
Friend DW has offered to send me the Studio MK1 so I can compare. I’ll take him up on it. Meanwhile he’s getting some new gear too for the new house listening room and will use the Esoteric n01xd se DAC for now. Each time I’ve heard the Esoteric stuff in his system I’ve had him ban it from the room basically lol while I’m there But, it’s a remnant. AND for grins he’s getting a Gryphon Diablo 333 to go along with some new Focal Diablos, along with a pair of huge Rel subs 
Should be fun at least. Lives in the Bay area now so I can’t just drive up anymore 
But he’s also very interested in what I find with the MK2. To be continued…
TK
Well, received the Studio Mk1 back yesterday from DW. And I about fell out of my chair on first hearing. That thing is truly so d@rn good it’s not even funny. And my first time to hear it with a max performance preamp and the Ultra_EXB7 MB amps in my own system here. Incredible, unprecedented clarity, huge imaging, etc.
I’m not talking about a minor increment from changing caps or trafos or something like that. No, we’re talking a whole new class level of performance. The reason is, the Studio analog section is truly ~ best possible. I call it “ultra analog” for a reason. So we’re essentially hearing the DSD stream as it is fed out of the re-clocker with ~ 0 imprint on it from the downstream analog stage.
And thus the only way to substantially improve the Studio Mk1 further from here would be to substantially improve the PCM->DSD conversion itself. And here is the new dilemma; I have heard now the improved fw of the Mk2. And it does in fact seem to be very much improved to my ears, very much. My hunch estimates it’s ~ close to as good as one might be able to get on that process now.
Soooo, it seems what we will have to do ultimately is pretty clear We’ve already been through all the progressions on the Mk1, done that, so no need to duplicate that again here. I mean for example the stock M6 trafos → high nickel trafos → good caps → neither; ADxxxx → ADyyyy → Burr-Brown ZZZZ, etc, etc.
Though I admit to being a bit apprehensive about jumping into it on an expensive new DAC. Meanwhile, think we’ll just let things play a few weeks, enjoy some and ponder life. Is it even possible to do an all new OPS with bi-polar PSU in this case on the Mk2? Not sure yet…
But we’ve heard what we needed to already I guess, and we can’t unhear it now. And my mind is a bit boggled for now by just how good the Studio Mk1 sounds (truly), and what might be to come then with a “Studio” Mk2. If we could get the better DSD processing on the Studio Mk1 we’d be done. But can’t have that.
But if we combine now a very real, present, meaty, lifelike ~ best available DSD signal with ~ 0 imprint analog OPS, well it’s not too hard really to imagine what that result might wind up being 
Should probably put on pads or something for the inevitable fall out the chair first moment I think
Anyway, just how I see it and my own experience. As usual YMMV. Happy Friday!
TK
Turbo, I have performed some minor mods on my MK I, which certainly have helped. Do you have any idea which combination of mods constitutes a “Studio MKI” ?
Like you I just ordered a MKII under September’s deal, and I am awaiting its arrival.
Oops that was supposed to be a private msg.
Thanks, TK
Hi, it wasn’t that much cost actually. Probably about ~ $750 or so.
TK
Well it’s been awhile but after Axpona and such I just want to comment about several things around the DAC(s).
First, congratulations to PS Audio on a great show and hugely interesting product introductions 
Nice work guys.
Next, just going to say what I feel basically, mostly from my own experience and efforts. As always JMO and YMMV.
Yes, as hinted to in the Mk1 mods thread the DS Mk2 is also quite good with the transformers removed and caps substituted. The Mk2 processing is sonically improved IMO and has a bit more meat on the bone in the sound vs the Mk1. I’m serious when I say the music has been unleashed with the trafo replacement.
Some of you may recall I spent a lot of time and effort taking a DS Mk1 as far as it could go basically, including replacing the entire LPF/output stage with what I call “ultra analog” along with DC coupling. That’s a lot of work to engineer all that in there 
BUT it was a huge improvement. My high end buddy (in the Berkely area now) still has that “studio” DAC.
Lately though he also has a ~ $120K retail MSB DAC, and is at their factory today in CA in fact for some checking and updating of the unit. To move to this DAC he let the complete Ideon digital stack go (before I could hear it, boo).
But the Mk2 with caps, it’s not hard to do really. And the only reason it even needs those in the output is we have to block DC. As we’ve said before, no cap is as good sonically as no actual cap (ie DC coupled). But IME caps are generally more transparent than trafos and we can get very close to “no cap” with some care in what we put in.
I have 2 pair of V-Cap ODAM in it currently ( V-Cap ODAM 400VDC 5.6 uf). About the largest value that made sense to try and fit physically in there.
BTW if anyone does likewise in their DS Mk2 (and they should IMO and IME Please say hello to Chris V and tell them TK in TX sent them
Oh and almost forgot, I also have 2 pr of VCAP 0.01uF CuTf caps as bypass in there too. Call it the max effort; go right or go home. Don’t use a different value bypass, use 10nF.
To clarify, that’s 4 caps total (plus 4 bypass caps) for balanced output.
There is some room because we’re replacing large trafos after all, but not as much real room as one might think. For example, reasonable value high end Mundorf are generally too large dia to fit under the board above it. Assess the fit carefully ahead of any potential candidate.
What I mean by reasonable value is, about as large a uF value as we can fit basically. The distortion, especially down in low freq is reduced as the cap value goes up. In fact that can even be measured these days with the ultra high res AP systems. Doug Self published a nice graph or three in his amplifier book(s) showing the effect clearly. I’ll see if I can find some of that online and insert it here.
In my mind the effect is mostly due to Xc (capacitive reactance) vs freq; as freq goes down, Xc goes up. To “short out” the capacitor sound imprint we need Xc to be as close to 0 as we can across the freq band. Ie, a component can’t affect the sound much if it is paralleled with a short circuit..
In other words, a 1uF might “serve” the task in a simple freq response analysis sense, but we’d prefer to have something like at least 4u7 IMO. Bigger would be better.
Also, with film caps there are no leakage concerns that go along with the larger values; film caps ~ don’t leak but others do and it causes real, practical issues.
But I also have a cap I might ultimately prefer; it’s what i started with before trying these other film caps. It’s a different, unique technology but has some real advantages, ~ very high CV with small size, but also ~ no leakage concern even with quite large values. I really loved how it sounded too; very out of the way with little sonic imprint. I’ll put them back after I’m sure I “know” the ODAM sound well enough and we’ll see which one ultimately stays.
Regarding the new DAC announcement at Axpona. I think this may make Mk2 owners envious, but after some thinking I just wanted to say a few things. Of course I haven’t heard the new DAC yet, and I’m sure it is quite good either way.
BUT- going by what we know so far, the new DAC has a different approach. It converts PCM input to DSD; then hands off to new piece (the SpectraWave 64-Tap Sequencer) that is said to be like a ladder DAC and/or a filter, some shift registers, I dunno. Then the whole thing gets passed to the ouptut stage, and I think they paid a lot of attention to that (as they should). And it’s DC coupled, I think they said (good).
But to make the point, let’s compare to what’s happening in the DS Mk1 and Mk2. In these, incoming PCM is converted to DSD, then that is handed directly to an active output LPF/buffer/gain stage. There’s no intermediate “DAC” stage between the DSD and the output LPF.
And IMO, there’s doesn’t need to be either. In fact, as I’ve said one could literally take the DSD stream and feed it into the preamp and there would be music; no “DAC” per se. To me that is the majority of the huge appeal of DSD, because I know we can make an active output LPF, DC coupled no less that has very, very low sonic imprint on the music. Very low imprint, I don’t call it “ultra analog” for nothin’
-Maybe meanwhile PS Audio design can educate us as to the need for the intermediate stage. I’m all ears
But hopefully this may help keep some Mk2 owners from feeling like they are suddenly obsolete and out of the hunt. IMO you’re not. And until (possibly) educated otherwise I would suggest it is even still the preferred topology, until proven otherwise. A very pure, direct conversion process chain (good IMO) with the best fw processing the designer could muster all things considered. After all, the actual DSD d/a conversion occurs in that PCM-> DSD fw process. The rest of the hdw then simply strips away (and buffers the music portion with some gain) the unwanted DSD high freq portions.
I hope this maybe helps some,
Thanks, TK
The DS Mk1 “Studio” playing a few days ago in DW’s current super system:
A representative graph of rising capacitor distortion with falling freq from Doug Self’s amplifier deign book. Audio Power Amplifier Design Handbook, 4th ed.
It can really massacre it at low freq; thus the need for larger than “usual” values to mitigate it.
So yesterday we put the other caps back in and joila, the music is very much unleashed and unmolested again. It’s about as close to DC coupled output now as one can get IME, without actually being DC coupled. And it sounds really good
In fact, had about the best digital systems results yesterday I’ve ever had in my own ~ modest system. Main thing is, besides the caps, finally got HQPlayer running on my streaming setup, and thus can upsample to 8x streaming my Qobuz playlists for example. 4x is also quite good.
-It also works very well sonically as a standalone files player. And I have a lot of ripped music over the years, as many others probably do as well.
In fact I’d say it was sounding better as a standalone player than even a CDT3, first time I’ve ever said that as well actually.
I have the DAC connected straight to the project super amp, using the DAC vol control and using the best sounding cables I currently have here. No analog preamp. And in my case I generally have a Gaia DDC in between the streamer and the DAC, using the DAC I2S input. But need to try the DAC USB input soon as well.
We’ve found that the DS Mk1 and Mk2 do respond very much sonically to highest quality upsampling input, for whatever reasons. So be it; feed it what makes it sing the best.
Highest quality upsampling processing is rather compute intensive though which is why it generally needs to be done on a computer of decent power, though the embedded solutions seem to be catching up a bit meanwhile.
In my case I’m running it on a ~ 2018 vintage mid-upper range nb, as I was heavily involved in nb developmengt for 15+ years in my career and have a few old engineering samples laying around. But it seems to get the job done for now at least, and I had only a couple of dropouts along the way last night, though they may not have been related to computing hp (dunno yet, tbd).
The HQ Player sw can be rather buggy and non-user friendly but I think there’s no going back on this now, once heard. It truly sounds much higher resolution than the usual, very reminiscent of analog now and that’s about the highest praise I can give it. Richer, fuller, higher resolution and much more like analog. Ahh, it feels good to finally type that! lol.. I’ve never bestowed that praise on my own digital system before. Ever. But now, it does approach that.
So looks like I’ll probably be forking over the ~ $340 for HQ Player finally, as I hit the 30 min trial limit several times last night and that is annoying while enjoying music to say the least I do wish they’d fix some of the major bugs though at least, and will be reporting several lol..
For example, to use Qobuz with it you have to also run the client app. But then the remote control app on my phone won’t Fwd to the next song, it crashes it. Though Pause works. Sigh.
Anyway, also thinking about the DAC digital volume control; it is working quite well in my system as is. But am thinking we could possibly improve that situation a bit by reducing the gain of the output stage, such that we have less digital attenuation (and thus loss of resolution) required for a given system volume. It would put us up higher in the attenuation curve as well, not quite as design intended but I think that might be workable. The lower portion of the curve seems rather slow in my system anyway.
Last night I was up in the 50’s-60’s range on volume; loud but wasn’t blasting it (yet).
Generally most amps clip by about 1800mV input or so; there’s no point really in the DAC putting out much higher max voltage than what clips the amp. If we get that dialed in then the attenuator would be running less attenuation for our listening habits and system, basically. Which might be beneficial. Not sure if its sonically ~ worth exploring, but maybe we’ll see sometime.
TK
TK DS Mk2 being fed 8x via HQ Player:
I am confused. Why upsample the signal before feeding it to the DirectStream? Isn’t the point of the DS DAC to upsample any incoming signal, on any input, to DSD 256, then go to wideband DSD (20x DSD or DSD1280) before going back to DSD 256 through the low pass filter? It’s also kind of a one-shot deal with a single source, right, you can’t feed your transport’s signal or TV, etc. through HQPlayer also. I hope this doesn’t come across as me picking on you, multiple people here have had the same success using HQPlayer or use something like the Grimm MU1 to feed the DirectStream. It just doesn’t make sense to me. If I were to go the HQPlayer route I’d just get a NOS DAC like the Holo May, possibly even converting everything to PCM to take advantage of the R2R ‘network’.
Why not try it and see. It’s free to try.
Thus also the qualifications about high quality upsampling requiring a lot of compute hp that embedded generally can’t match.
Whatever the case, the DS processing seems to benefit from being fed higher freq sampled data. The Mk1 and the Mk2 IME. Try it and see for yourself.
TK
interesting to see you are sending high rate PCM to DAC - wouldn’t DSD256 be better?
Try it.. find out! Report back
TK
BTW I think we have a forum gentleman in CA who is likely about to try updating his Mk2 as I have described. Hopefully he’ll share his findings as well. TK
I already tried the caps instead of the transformers.
unless you modify the op-amp circuit to limit the high frequency you will blow tweeters as I did. Also it didn’t sound as good as the transformers. Too harsh at times. We tried many high-end caps costing upwards of $2000 for all four needed. I scraped this idea after the tweeters blew on both of my B&W 801D. That was a $2000 mistake. I ended up getting the APS transformers installed. The MK2 sounds perfect to me now. Way better than stock.
Something else I forgot to mention, If you are not using the USB input you can turn the power off for it. After setting the USB to auto you need to manually go through all the inputs to activate this power off for the USB. Once done go to your i2s or other input you are using and toggle from auto to on to auto for the USB input while playing a song. You can hear the big improvement in sound quality when set to auto. This tip was from APS.
Give it a try.
Based on the above post, anyone care to comment on the USB input on mkii vs I2S?
Typically on the older models, USB to Matrix, I2S to DAC was of benefit. Is that still true?
Im figuring there may be a few mkiis out there soon.
Thanks..
