johniboy

Very nice experiment. There is just one caveat I see. The software audio-diffmaker MAY in fact dilute some the difference between tracks which are only present in the time domain. Here is their paper: http://www.libinst.com/AES%20Audio%20Differencing%20Paper.pdf This sentence that me nervous: "In the time domain, a digitally recorded track can be easily shifted only by whole samples. But if it is transformed into the frequency domain, delays can be easily varied by any amount by adjusting the phase of each component an increment proportional to its frequency. For this reason, trial and error iteration to optimize delay compensation to fine values is done in the frequency domain." If I understand this correctly, the software iterates through different parameters to adjust and compensate for shifts in the time domain. This would inevitably discard all of the differences that may be present merely by overfitting. I may be wrong but a more conservative alignment approach may lead to results which better match reality. Probably not an easy task. Just some thoughts.

It would also be great to know the upgrade policy for Sonata. JRiver asks for something between $20 and $30 for each upgrade. Also, which version of JRiver is Sonata based on? 16? 17? Are all bug fixes transferred to Sonata?

Nice review!<br /> <br /> Just to make sure I got this right:<br /> Will Sonata only tag the classical albums for me if I rip a CD? Or will I also have access to the tagging functionality / databases if I already have my flac files?<br /> <br /> I am asking because I am already using Jriver and dBpoweramp and the main reason for buying this software (for me) would be the additional 18 fields together with the automatic tagging functionality. If I have to tag all my albums by hand OR re-rip all my CDs, it might not be worth it (for me).<br /> <br />

and that is why i would like to have a client-server software that is better than jriver! you can put as much buffer in the system as you like. this should take care of the latency issues. at the end of the day, this would very similar to the linn ds system with the exception that i could optimize the hardware at will. you see no advantage? i think there is a huge advantage to have all the dirty stuff: hdds, music management, etc far away from the dac. this setup: server -> client -> dac is equal: dirty -> cleaner -> very clean. in this setup the client could even be a small ARM based computer with minimalistic OSS.

i control the playback via android phone (or tablet) using the gizmo app which controls jriver on the server. the server sends the required data to the client. see here: http://www.jriver.com/thenet.html i know, it sounds a bit complicated but its not that bad. and it enables me to have the actual data far away from the hifi equipment. have also a look here: http://wiki.jriver.com/index.php/Library_Server maybe it explains it a bit better.

the server sends the data to the client via network. the client has a player software that outputs to an internal soundcard or to an external dac? again, this is my setup already working with jriver running on server and client. i just wish the client side server could be a bit more like jplay...

sorry, i missed that you want to use a macbook. then you definitely need an external soundcard like the weiss int202. it is probably one of the best options out there. don´t be fooled, the elgar is still a very good and will compete with many modern "high-end" dacs!

now i know where you are going. had to read your post a couple of times to get it... not a bad idea once windows 8 tablets come out (next year!). i was thinking about a somewhat different route. 1. server: hdds with all the music + a software for the library management similar to what jriver does. this will be a pretty "messy" machine with all the hard disks, processor intensive tasks. 2. client: small computer without any moving parts. clean power. usb or spdif out to the dac. 3. controller: tablet to control the server, select a song or album, which is then sent to the client for playback. I have this setup with jriver running on the server and on the client. my controller runs gizmo. but actually it would be much nicer if the client would not need jriver but could run jplay instead. this setup is similar to a NAS - player setup with the exception that the NAS runs a full blown operating system and allows to run a decent library management software with a great ui.

I have an Elgar Plus and i need two aes wires to get beyond 96k. If 96k is sufficient, one wire will do, either aes, spdif or optical. Just to put an additional option out there: Have you considered using a good soundcard like RME? I use my Elgar that way and it works like a charm.

this setup is exactly what I was thinking of. client - server. the client can be extremely slim with a clean connection to the dac (either via usb or spdif) while the server would do the "number crunching". the server would run jriver, the client Jplay. the best of both worlds. :-)

Very interesting piece of software, indeed. And it gets pretty close to what I am envisioning as best solution: very god user interface and best possible SQ. No idea whether it is possible but could you envision a system comprising of 2 computers, one for managing the music database and having the user interface and one which is responsible for the playing the music. The first computer could be positioned in a basement and connected to a network router. The hardware can be standard and you could use normal HDDs. The software could be something similar to J.River, which in my opinion has a fantastic user interface and music management system. You could select the music you want to play via phone or tablet and hit enter. Then the required information would be transferred to a second computer that runs something like JPlay. No interface, best SQ. The hardware could be tweaked for best sound, a fanless setup that could be located at your hifi rig. What do you think? Technically possible?

so, is it worth downloading?

I was interested in these boxes ever since BUT they are incredibly expensive. Have a look at this PCI breakout box: http://www.magma.com/2slot.asp At almost 1300$ the value of the product quickly diminishes...

thats the destiny of innovators... but once we got it up and running everyone will envy us and will want to have one. so you are saying MPD can load a title into memory before playing it? i do not have any experience with MPD servers. i will have to try. i like the development i am seeing on the client side, though.

see here: http://www.engadget.com/2011/09/02/archos-80-g9-and-101-g9-hands-on-video/