Cogitech,
I agree that striving for low THD and high SNR is the way to go given two choices pick the better one. In a system consisting of a source, a DAC, an amp, and a speaker you have to look at the rate limiting items that contribute most to the THD and SNR and basically do a sensitivity analysis.
CD source is 16 bits or 96dB SNR
HiFi Amplifier is typically 0.1% or -60dB THD
A very good speaker will have about -50dB to -60dB THD in the main band (not bass)
These values appear to be much larger than the specs of the lowly DAC in the miniDSP and I believe one would have a hard time discerning differences between a DAC or opamp that has -104dB vs -114dB THD.
I am not saying you can't hear a difference - but I believe that difference may lie elsewhere rather than in pure THD or SNR measurements. Maybe in transients or attack, maybe in well behaved decay? The ultimate test is to see if a 20 kHz square wave can be replicated.
I agree that striving for low THD and high SNR is the way to go given two choices pick the better one. In a system consisting of a source, a DAC, an amp, and a speaker you have to look at the rate limiting items that contribute most to the THD and SNR and basically do a sensitivity analysis.
CD source is 16 bits or 96dB SNR
HiFi Amplifier is typically 0.1% or -60dB THD
A very good speaker will have about -50dB to -60dB THD in the main band (not bass)
These values appear to be much larger than the specs of the lowly DAC in the miniDSP and I believe one would have a hard time discerning differences between a DAC or opamp that has -104dB vs -114dB THD.
I am not saying you can't hear a difference - but I believe that difference may lie elsewhere rather than in pure THD or SNR measurements. Maybe in transients or attack, maybe in well behaved decay? The ultimate test is to see if a 20 kHz square wave can be replicated.
For some reason it seems to me that THD and SNR at the various stages cannot be compared individually in the way that you have done. To me, it seems that the entire chain has a cumulative effect, and that lower SNR and/or higher THD in any one component will exacerbate issues elsewhere in the chain.
But yes, perhaps the things I am hearing are not related at all to the specs. All I know from experience with 3 different DACs is that the quality of sound does correlate with the specs. This could be co-incidence, but Occam's razor...
Maybe the DAC that I built has much lower jitter. That'd be ironic!
P.S. Regardless of what DAC chip is chosen, I highly recommend a setup that allows for opamp "rolling". The different opamps really do sound quite different and it is a very fun and rewarding way to help achieve system synergy. Running a crisp, lively, fast opamp like the AD8599 or AD8066 through tubes is a great combination. If I was running class D, I might go for something a bit more rich and laid back, such as OPA2134 or OPA2107.
Last edited:
Have yet to see group delay or excess group delay results to your musings. And ringing occurs whenever anything drastic is done with I²R or FIR. Brick wall filters are some of the worst. Just look at the very top end of your sound card for example.
Bob you still there?
Yes, I am still here. Had mininal time to read BB's.
DAC's in particular and low end electronics in particular:
The IC DAC's today are better than the kilobuck DAC's of say 10 years ago. Any competent HT receiver today has decent 24/192 DAC's. The really cheap DAC's that go into laptops have noise levels below -100dB. Where the problem comes in is the outboard parts in the output filters and feedback loops. Look in a $50 DAC and all of the discrete components will be no-name parts. I got the Topping D-20 for list $200 and all the discrete parts are well known brands.
I had been using a M-Audio Ultratrack as the DAC into the miniDSP. The D-20 sounds clearer. Running the nanoDIGI through the D-20's is another step up. I am now sure that my Alpair A10P's are not fully broken in.
I do not find any adverse effects of applying what amounts to a Linkwitz transform to the subs. Pulling the 4dB 800-2K hump out of the A10P's give an odd feeling to female voice and causes the image to float a bit. This is going to be a no-no. Before I play with this some more, the A10P are going into the HT for a couple of weeks. I'll let Karen break them in with her soaps an Life Time Movies.
Bob
DAC's in particular and low end electronics in particular:
The IC DAC's today are better than the kilobuck DAC's of say 10 years ago. Any competent HT receiver today has decent 24/192 DAC's. The really cheap DAC's that go into laptops have noise levels below -100dB. Where the problem comes in is the outboard parts in the output filters and feedback loops. Look in a $50 DAC and all of the discrete components will be no-name parts. I got the Topping D-20 for list $200 and all the discrete parts are well known brands.
I had been using a M-Audio Ultratrack as the DAC into the miniDSP. The D-20 sounds clearer. Running the nanoDIGI through the D-20's is another step up. I am now sure that my Alpair A10P's are not fully broken in.
I do not find any adverse effects of applying what amounts to a Linkwitz transform to the subs. Pulling the 4dB 800-2K hump out of the A10P's give an odd feeling to female voice and causes the image to float a bit. This is going to be a no-no. Before I play with this some more, the A10P are going into the HT for a couple of weeks. I'll let Karen break them in with her soaps an Life Time Movies.
Bob
Last edited:
If you think the PCM7193 and OPA2134 combo in the D-20 sound good, I just found this unit for about $60 that may work. Probably will not have top of the line name brand components for caps and resistors and has no USB input but for nanoDIGI, I think spdif coax is all that is needed right?
Online Shop hifidiy.net aune dac mini pcm1793 SPDIF Coaxia Optical DIR900 OPA2134 Finished|Aliexpress Mobile
Online Shop hifidiy.net aune dac mini pcm1793 SPDIF Coaxia Optical DIR900 OPA2134 Finished|Aliexpress Mobile
There are a few options with that combination of chips, actually. MUSE Mini, SMSL SD-793II, etc. All in the $60-$70 range including shipping. As Bob mentioned, the quality of the rest of the components matter. What a lot of people do is buy the cheapest one they can find with the combination of chips they want and then replace key components (power caps, output filter caps/inductors, etc.) with quality stuff. It ends up considerably cheaper and soldering is fun! But this approach is not for everyone. The Topping stuff seems to strike a pretty good value balance for those who just want to unpack it and plug it in.
What I am saying is the chip portion of most DAC's is good enough. Where the difference comes is in the external parts in the filters and feedback loops. The difference between the cheap DAC's and moderate priced ones is parts quality and makes a significant difference in SQ. I make no claim that the D-20 is the best in the price range, just that it seems adequate for the job.
Same with the Topping TP-60 amp I use. It is the best of the price range that I have come across. Again, probably because of the periferal parts. I would like to try one of the newer TI amps, but I don't do kits. When I come across one in the $200-500 range, I'll give it a try.
Bob
Same with the Topping TP-60 amp I use. It is the best of the price range that I have come across. Again, probably because of the periferal parts. I would like to try one of the newer TI amps, but I don't do kits. When I come across one in the $200-500 range, I'll give it a try.
Bob
The same quality differences that you mention about peripheral parts applies to the opamps. In a highly resolving system, you will hear the difference, both in quality and "qualities". The OPA2134 in your D-20 DACs is a very good opamp. Mostly neutral to slightly dark, laid back, warm, very easy to listen to, but with mediocre soundstage, loose bass, and slightly veiled highs. Swap that out for an AD8066 (for example) and you will notice bigger, faster bass with more punch, extremely clean, detailed highs (without any harshness), less midrange shout, and a deeper, wider soundstage.
I am not saying this to be a pest or to say your DACs are no good. That is the furthest thing from my intentions. I am saying it because you seem to be the kind of guy that would truly appreciate the differences that I am talking about. I am also not saying that the AD8066 is the right opamp for you or your DAC or both. But wouldn't you like the opportunity to try a bunch and decide for yourself?
(If the answer to my last question is yes, then a very simple modification to your DACs would provide you the option of trying many different opamps - heck, you could even choose one flavour of opamp for bass and another flavour for the rest of the range - very cool!)
Last edited:
I've been running a D20 for a month or so now in a 3rd system, and first impression was that it did not embarrass itself at all when compared to the onboard DAC / tube buffered output of my 10yr old Jolida (JD603?)
The D30 looks even more interesting if you could use an analog line input and level control, but is a new model and wasn't available at the time.
The D30 looks even more interesting if you could use an analog line input and level control, but is a new model and wasn't available at the time.
Once upon a time I could see well enough with the naked eye to do 1/200" pitch 6 layer PCB repair work. Today absolutely require a microscope
At one time I was very leery about getting the soldering iron out, and I am still no expert, but the procedure is simpler than you may be thinking.
The OPA2134 in the D-20 is a DIP8 package soldered "through hole" (not SMD), so it really is just a matter of heating each of the 8 welds till your wick sucks up the solder, then pop the chip out, place a DIP8 socket in there and solder the 8 pins. Do the heating and wicking from the bottom of the board as much as possible to give you more room to maneuver.
From then on, you'd just have to carefully pop the opamp out of the socket and install the next one.
If this is still something you'd rather not do, but you would like to have done, send me a PM and I suspect we can work something out. I am no pro, but I can handle that job in about 30 minutes and I'd do it mostly for the fun and for the opportunity to hear your impressions on different opamps.
This DIP8 socket comes with a very nice opamp
http://www.ebay.com/itm/1PCS-LT1364...mplifier_Parts_Components&hash=item33698a0cea
Last edited:
They only box 2.1 amps - just leave the sub unused if you only need stereo. It is setup for fullrange on the stereo.
2.1 HIFI digital amplifier board TPA3116D2 HIFI beyond TPA3123 TDA2030 LM1875-in Consumer Electronics on Aliexpress.com
I truly like your objecive approach to all things you do, but I can assure you that this is a very simplified view of what is happening.
Ok, you guys talked me into it. My next electronics purchase will probably be a DAC. Here is one with the PCM1793 and OPA2134 plus headphone amp, swapable op amp 8-pin DIP socket, nice aluminum case, Elna & Nichicon caps, etc. $57. Looks like a fun bargain
Free shipping SD 793II Mini DAC DIR9001 + PCM1793 + OPA2134 Coax/Optical Input + Headphone Amp-in Consumer Electronics on Aliexpress.com
Free shipping SD 793II Mini DAC DIR9001 + PCM1793 + OPA2134 Coax/Optical Input + Headphone Amp-in Consumer Electronics on Aliexpress.com
Would go for a good implementation of the ESS9023, like this.
USB Digital Analog Converter
Well expleined and thoroughly tested.
http://www.jdslabs.com/products/39/odac-objectivedac/
Or, high quality version of ess9018, like found in the oppo105. This is a phenominal piece of equpiment and has bested any DAC i have heard to date, including Peachtree, Peter Daniels NOS Dac, CS4398 with transformer output, and various others that were in SACd players or DVD players.
USB Digital Analog Converter
Well expleined and thoroughly tested.
http://www.jdslabs.com/products/39/odac-objectivedac/
Or, high quality version of ess9018, like found in the oppo105. This is a phenominal piece of equpiment and has bested any DAC i have heard to date, including Peachtree, Peter Daniels NOS Dac, CS4398 with transformer output, and various others that were in SACd players or DVD players.
Can vouch for the ODAC being a very good USB DAC... There are others available that use the Tenor USB chip and ESS9023 DAC, however, can't say that they perform objectively as well as the ODAC. NwAvGuy put in a lot of hard work to get it performing just so, same goes for the O2 headphone amp
If you want stupidly high levels of objective performance look at the 'The Wire' series of Headphone amps and amps from member opc.
Still in the process of making my woofer speaker enclosures, but have had my miniSharc > Curryman DAC's running for a few day's now and am not disappointed by what I've heard so far through the FH3s.
Paul
If you want stupidly high levels of objective performance look at the 'The Wire' series of Headphone amps and amps from member opc.
Still in the process of making my woofer speaker enclosures, but have had my miniSharc > Curryman DAC's running for a few day's now and am not disappointed by what I've heard so far through the FH3s.
Paul
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.