[humanflyz]

I started thinking about upgrading my source about a month ago, and I had certain criteria for the DAC that I will buy:

1) No tubes and/or op-amps in the output stage; must be fully discrete
2) I wanted an upsampling DAC
3) fully-balanced outputs

My budget was around the $1000 mark, and after doing some research, I have my choices limited to the Lavry DA10, Stello DA220, and Lite Audio DAC-38.

I just wanted to know what people thought about the three choices that I'm contemplating. But first, let me say something about what I want out of whatever source I'll be getting. One, the DAC must have a black background with as little noise as possible. Two, I want my DAC to be neutral, because I believe a source's first priority is to pass the signal as purely as possible. Three, I want a source that is tonally even, not favoring any one region of the frequency spectrum. Four, if there is one thing I can't stand, it's sibilance and harshness in the highs. And finally, I want the DAC to be revealing, but not to the point of being "in-your-face" and bright.

So based on my sonic criteria, what do you guys think the best DAC out of the three that I have listed is?

 

[slwiser]

My choice is shown in my sig. for what it is worth. I got the Lavry for it's flexibility, headphone amp is consider fairly good, balanced outputs, and Dan Lavry's reputation as designer after checking out his history.

Good luck, you can't miss with any of these selections.

 

[Iron_Dreamer]

Quote:

Originally Posted by humanflyz I started thinking about upgrading my source about a month ago, and I had certain criteria for the DAC that I will buy: 1) No tubes and/or op-amps in the output stage; must be fully discrete 2) I wanted an upsampling DAC 3) fully-balanced outputs My budget was around the $1000 mark, and after doing some research, I have my choices limited to the Lavry DA10, Stello DA220, and Lite Audio DAC-38. So based on my sonic criteria, what do you guys think the best DAC out of the three that I have listed is?

The Lavry fails on criteria 1 and 2. It is not upsampling (though it does use an ASRC in the "wide mode"), and it does use opamps in the output path. That said, I wouldn't say it sounded better or worse than the Stello, just different.

The Stello does satisfy all three, but of course it is much larger and heavier, whereas the Lavry could be considered transportable. I've no knowledge of the Lite DAC.

I found the Stello to have a bit more weight to the upper bass and lower mids, as compared to the Lavry, and a slighter smoother (but not rolled off) high end, with upsampling engaged. The differences were quite subtle, and the two perform similarly otherwise. Neither has any background noise problems, nor do they accentuate sibilance.

 

[Jon L]

Quote:

Originally Posted by humanflyz 1) No tubes and/or op-amps in the output stage; must be fully discrete 2) I wanted an upsampling DAC 3) fully-balanced outputs My budget was around the $1000 mark, and after doing some research, I have my choices limited to the Lavry DA10, Stello DA220, and Lite Audio DAC-38.

Well, Lite DAC38 is NOT an upsampling DAC. If you want an upsampling Lite DAC, it would be DAC39, DAC68, or DAC192.

However, I would personally pick DAC38 or DAC60 over the upsampling ones. If you want non-tube, DAC38 with some proper mods should do the trick.

BTW, I know somebody who has both DAC38 and Lavry, and he thinks both sound very similar in stock form...

Stello DA220 MkII is a brand new animal, and I do have a good feeling about it, with defeatable upsampling and USB input (although USB-spdif).

 

[humanflyz]

The DA220 MK2 is a bit above my budget, so I'm more interested in the original DA220.

As for Lavry, I don't know where I heard that it has discrete output stage, but thanks to Iron_Dreamer for clearing that up.

 

[Ori]

Quote:

Originally Posted by humanflyz and I had certain criteria for the DAC that I will buy: 1) No tubes and/or op-amps in the output stage; must be fully discrete 2) I wanted an upsampling DAC 3) fully-balanced outputs

Reminds me the committee for an ideal horse for the cavalry, which came up with the three criteria:

1) has to be tall (so the rider can spot the enemy faster).
2) comfortable to sit on.
3) durable, can go all day long without drinking.

They ended up buying a camel...

Criterias aside, go listen to the options you picked. I'm sure the answer will be quite clear to you, and it may also be very different from another person's preference!

 

[slwiser]

For those that do not want to take the time to review the information available in Dan Lavry's forum concerning the DA10. Here are some notes that I have collected. I have attempted to capture all of Dan Lavry's comments in italics.

The headphone output amp is discrete per Dan Lavry. The extensive speculation about the Lavry design has resulted in some misunderstanding of how the opamps are used in the Lavry DA10 I think. Per Lavry:"2. Yes, the headphone can drive 300 Ohm. The driver is made out of discrete transistors. In fact, it can drive 20dBu level (that is really huge for a headphone) into 60 Ohms!"

Another Dan Lavry comment about the speculation made here in on this web site of a review of the Lavry DA10: 4."The analog circuitry of the LaVry consists of discrete transistors with a few OP275 opamps, presumably outside the signal path." Lavry's response: The comments regarding OPamp is based on common ignorance. Transistor circuits are better? An opamp is a repackaging of a transistor circuit! "

The XLR balanced output I do not know about. Dan has been quite about the design on this side. People keep on asking Dan Lavry about his circuits but since he developed his own for his own company he purposely leaves some information hidden for what is a normal practive in keeping "trade secrets." For those who actually take the time to develop their own products and maket them this should be completely understandable. He says; "As a rule, I do not talk about the internal circuitry of my designs."

Here is what Dan says about the use of opamps: "Having a well designed op amp in a good circuit is fine. An opamp is really not all that different then discrete transistor (or fet) circuit. It is mostly just a "repackaging job" of transistors resistors and small caps. Most opamps (not all) are oriented towards low currents. There are times when it is better to go to discrete components. "

Concerning his headphone amp: "The reason I did a discrete 20 transistor circuit for the headphone is - I needed some serious drive current with very low distortions, beyond what opamps would yield. There is time for discrete, and there is time for opamps, and different cases call for different circuits or opamps. That is why we have thousands of different types of transistors and thousands of opamps… "

Sounds like Dan Lavry is a circuit designer understanding the design complexities needed to bring a product to market. Here are some other ideals from Dan Lavry about the DA10:
___________________________________
So what you have here is a year and a half of careful detail work by a seasoned designer with much experience in DA design, no matter what some audiophiles choose to think or believe.

If my goal were to please the “run of the mill” audiophile critics. I would need to do the following:

The list is well known:
1. Use linear supply, no matter what.
2. Make a big fancy looking box with “retro” look to it
3. If possible, use at least one tube
4. Use only discrete circuits, no matter what
5. No aluminum electrolytic in series, regardless of the circuit
6. No DC servo circuits.
7. Use open loop circuit, no negative feedback

I COULD ADD MANY MORE THINGS TO THIS LIST.

If I did the above, I would have a lesser product and the approval of the “generic audiophile critic”.

Depending on circumstances, the above can be true or false. It is ridiculous to design a unit based on an accumulation of “notions” that many years ago MAY HAVE BEEN TRUE IN SOME PAST DESIGNS. The goal should be the END RESULT, and that is what I work towards.
_____________________________________

Concerning his pedigree:

lavrye wrote:

I did get a kick from the comment about the Lavry designers.

I designed everything including historically the original DAC module sold (repackaged) by Ultra Analog to OEMs Mark Levinson, Wadia, Pacific Microsonics.(1980s) My design put in their gear was sold under the name UltraAnalog DAC D20400. I then designed the much improved DA924 (Lavry Gold) used by the many of the top mastering facilities. I also designed the LavryBlue DA which, as I mentioned, shares the same supply, crystals and oscillator circuit…

These have been some things that I thought should be noted about the Lavry DA10 before more speculation incorrectly gets posted around here.

None of these things suggests that the Lavry DA10 is a better sounding product which is a completely subjective thing anyway. The other units noted in this thread may very well match up to your preferences better than the Lavry but poor speculation concerning a unit does no one any real good in these discussions. If one were so inclined, speculation for any particular unit can takes us anywhere and be much less useful to us all.

Quote:

Originally Posted by humanflyz The DA220 MK2 is a bit above my budget, so I'm more interested in the original DA220. As for Lavry, I don't know where I heard that it has discrete output stage, but thanks to Iron_Dreamer for clearing that up.

 

[slwiser]

The speculation that the Lavry does not upsample may be wrong based on the following note found in a the LavryBlue manual concerning how the CrystalLock and Narrow ranges work for D to A, especially since the DA10 and Blue are supposed to have similar circuit design using the similiar designs for the CrystalLock and narrow modes of operation.

The DACs always operate at double-speed mode in narrow lock or CrystalLock™: normal speed data is first upsampled by the on board DSP to 88.2 or 96KHz. The digital to analog conversion utilizes two converters per channel, operating in a differential mode, thus roviding cancellation of many artifacts and improving the noise performance by 3dB.


[size=medium]So based on speculation, criteria 1 and 2 may not be available in the Lavry DA10. But based on what is found in the manuals and Dan Lavry forum, maybe the Lavry does meet those criteria.[/size]


Quote:

Originally Posted by Iron_Dreamer The Lavry fails on criteria 1 and 2. It is not upsampling (though it does use an ASRC in the "wide mode"), and it does use opamps in the output path. That said, I wouldn't say it sounded better or worse than the Stello, just different.

 

[humanflyz]

For the people whose advice is for me to audition the DACs themselves, I totally agree, but the problem is, I don't know any one that has them, and there aren't any dealers in my area. So I pretty much have to buy blind here; that's why I'm trying to gather as much info as I can.

As for the Lite DAC-38 not being an upsampling DAC, I looked on Pacific Valve's website, and their specs for the DAC-38 indicate that it is a "solid state asynchronous up sampling" DAC.

And also, I do not really care if the Lavry's headamp output is discrete or not, since I won't be using that feature.

 

[Cronstead]

you might want to check this out
http://www6.head-fi.org/forums/showt...&highlight=omg

I believe the standard setting is less than 1K

 

[Ori]

Quote:

Originally Posted by slwiser Another Dan Lavry comment about the speculation made here in on this web site of a review of the Lavry DA10: 4."The analog circuitry of the LaVry consists of discrete transistors with a few OP275 opamps, presumably outside the signal path." Lavry's response: The comments regarding OPamp is based on common ignorance. Transistor circuits are better? An opamp is a repackaging of a transistor circuit! "

First, we all have our differing ideas on opamps vs. discrete circuitry. An opamp is made of low tolerance transistors and non-linear resistors with a common substrate. Do I need to get into the Physics of semiconductors to illustrate that there is a difference? Just look at the common schematics of an opamp and count the number of stages in the signal path. That ought to tell you enough and if you need more detail then I'll direct you to the literature regarding delay-induced distortions in feedback circuits. This is not specific to opamps, just inevitable with most of them.

Now to the 'guts'. That was a comment I made, giving the DA-10 the benefit of the doubt, based on information from a DA-10 owner. I've seen that vague response by Dan and I'm sorry to say that when you answer so vaguely - you should not expect anything but skepticism, and it wasn't me being the skeptic!
I appreciate it that you gave us more detailed information, so at least the owners of the DA-10 know more about the design they purchased. I don't see why Dan wouldn't answer the same way in the first place. It looks good, so why the secrecy?!
If I were the designer and my product is superior to the competition, then I would want to emphasize those differences. It's called good marketing...

 

[mindthegap]

Jon L is correct. The DAC 39, DAC 68 and DAC 192 are up sampling. Way to go Jon L!!! The DAC 38 and the DAC 60 (and I suggest you get the PV modded one since the stock DAC 60 has a hard top end no matter what you do. The PV modded one seems to take this away) are over sampling and can handle sampling frequencies to 96 kHz. The DAC 68 up samples to 192 kHz. The DAC 39 will upsample and feed any of the light DACs to 96 kHz as well as act as a DAC on its own. So imagine, upsampling a frequency and then feeding it to a DAC AH, (which is non oversampling) sounds pretty weird.

I have contacted Pacific Valve and they have agreed to adjust their web site.

If you do not want any harshness in the top end, and are not too weirded out by tubes, the DAC 68 has the sweetest high end out there. Its the SET of DACS.

 

[humanflyz]

The reason why I didn't want any tubes in my source is because I am already using a tube-amp, and my current system is already sort of tuned to have a tube-ish sound. So for the source, I'm trying to get a neutral sound so as to not make the whole system sound too tubey.

 

[slwiser]

Ori

The implication of your post is that he has something to hide that is bad. It may actually be very much otherwise. My positive take is that Mr. Lavry has some trade secrets that he has developed which are worth keeping.

Your statement is a cheap shot on the concept of the "trade secret" in business. If you had an ideal that worth anything you would want to keep it secret too. I guess we will never have to worry about your having one of those (a trade secret)....

Quote:

Originally Posted by Ori Now to the 'guts'. That was a comment I made, giving the DA-10 the benefit of the doubt, based on information from a DA-10 owner. I've seen that vague response by Dan and I'm sorry to say that when you answer so vaguely - you should not expect anything but skepticism, and it wasn't me being the skeptic! I appreciate it that you gave us more detailed information, so at least the owners of the DA-10 know more about the design they purchased. I don't see why Dan wouldn't answer the same way in the first place. It looks good, so why the secrecy?! If I were the designer and my product is superior to the competition, then I would want to emphasize those differences. It's called good marketing...

 

[mindthegap]

You can find a review of the DAC 38 on humble fi:

http://www.humble-fi.com