JLP music player

[cvrle59]

I find a very interesting story from Barry Diament on Naim forum. I guess his name is recognized by most of the members here, if not just google it. A lot here matches what Gordon is all about with MQn and JLP, "less is better".

 

"I've always felt that what is "good", "better" or "best" depends entirely on exactly what one seeks.

The goal might truth to the source or it might be what someone may just like. Personally, I think both are equally valid as I'd never argue with whatever brings anyone their listening pleasure.

 

But if the goal is truth to the source, my experience has been that well ripped files (in raw PCM format such as .aif or .wav) invariably are truer to the CD master than any CD pressing played back via any transport.

 

Ever since I created my first CD master, early in 1983, and heard the results that came back from the replication facility, I realized that CDs made at different plants (and often on different lines at the same plant) all sound different from each other and *none* sounds indistinguishable from the master used to make it. The differences, in my experience, range from subtle to not-at-all subtle, with focus and fine detail being lost to varying degrees.

 

I've always found a slow-burned CD-R to get closer to , if still not indistinguishable from, the CD master. But when any decent pressing is properly extracted to hard drive (in raw PCM format such as .aif or .wav), I've found the differences go away. This is what has me so excited about playback from the computer: for the very first time in my experience, the listener can get the sound of the master, unaltered.

 

I had a related experience not too long ago when comparing an SHM pressing with its plain vanilla counterpart, both created from the same CD master. At first, when listening via my transport, I was certain I was hearing two different masterings, so different did the drums, cymbals and piano sound. I would have bet money I was hearing two different EQs created by two different engineers. I thought it was a trick to make SHM, which I took to be just another way of pressing a CD, sound better than it really was. Much to my surprise, when I extracted both the SHM and CD versions to computer hard drive and played them from there, they were indistinguishable.

 

What accounts for the differences? I'm can only speculate. Playback from the disc involves several processes. The player must track the spiral of pits, maintain focus on the laser, read the pits, decode the 8:14 modulation by which the binary data is encoded to create the nine different length pits on a CD, perform any necessary error correction, and convert the binary stream to analog audio, all in real time. While playback from hard drive in a computer certainly involves some processes too, it seems to me there is less to do overall and my experience has been when there is less to do, better results are achieved. Again, just speculation on my part. (I'm always open to learning something new.) What I do know is that I've never heard CD playback from any transport/DAC combination, regardless of price that sounds indistinguishable from listening to the CD master, while playback from the computer routinely does sound indistinguishable from the master.

 

So again, if someone feels playback from the CD is "better", I wouldn't argue. They may simply like the resulting sound. But my ears tell me it doesn't sound the same as the master sent to the replication facility by the mastering house. Me? I'd rather hear the sound of the master.

Just my perspective, of course.

 

Best regards,

Barry"

[cvrle59]

I haven't tried as Service yet, but there is a bit of difference between the guide from the link you pointed out and what v25 would take. You can not use driver # as that guide shows. You need use full name like we used to do.

[cvrle59]

Yap...I figured it out...thanks.

I don't want to run too fast, but it does sound pretty good so far!

[cvrle59]

Could you explain this " profile guided optimisation" little bit, please. I'm not sure if these are new features or something that we were able to do through web browser.

[cvrle59]

managed to get squeezeserver running as a service, need to change the properties of squeezesvc.exe to run as admin in the compatability tab, also set it to run in reduced color mode and run in 640x480. Also set these for squeezesvr.exe.

I've done these, but I still don't understand how you made it to run as service. I can only set it up inside that panel to run on start-up, that doesn't mean it is a service?

[cvrle59]

I've done these, but I still don't understand how you made it to run as service. I can only set it up inside that panel to run on start-up, that doesn't mean it is a service?

Got it...no worries!

[cvrle59]

Both running now as services. V27 is the best SqueezeLite so far. I noticed that SqueezeLite is sitll running even if you stop service. I guess you need to kill it in Task Manager if you change version or some of the parameters associated with it, then restart the service.

[cvrle59]

I made mistake a few times being too fast in my conclusion, so this time I wouldn't comment much except, that I would advice all the "big guns" ( guys with good equipment and ears) to install LMS and SqueezeLite as services and to give a try to V27. The rest will be self-explanatory.

[cvrle59]

2channelaudio,

I assume that you are fine now, but anyway I wanted to answer your first question. I didn't see LMS in service list until I rebooted computer.

[cvrle59]

It's frustrating how much lms and squeezelite is affecting the sq, I guess the ideal would be to load the music into ram and then suspend the lms and squeezelite decode/stream etc threads. Think the next thing to do is a standalone MQn type version.

I know that you won't stop until you squeeze that "last drop of blood" from LMS+SqueezeLite, but V27 is the first version that it reminds me a lot to MQn being so relaxed, but detailed at the same time, with excellent definitions throughout frequency range.

[cvrle59]

I listened to 21/28 for some time then I switched back to MQn. The JLp is getting closer, but MQn is still the king, at least in my system.

[cvrle59]

Thank you! Will try it once I get home...

 

Plus, my wife will be able to use Squeezer for her 'oh my god how can you listen to that crappy 128k mp3s and not realize it is burning your ears' kind of music.

I call those people lucky once. They don't need to spend tons of money to enjoy music...

[cvrle59]

V30 has a problem on my system. I can see it playing, but no sound out of speakers. I am running -a 23 -b 1000:2000000. I tried something extremly different like -b 2000000:2000000, still the same.

[cvrle59]

Got it...I wasn't expecting it takes for so long.

[cvrle59]

Got it...I wasn't expecting it takes for so long.

I was able to play one album (24/192), so once I switched to something else, it stopped working at all. It doesn't matter if 16/44 or 24/192.

[cvrle59]

The same happened to me with v30-v31. Went back to v29, works like a charm!

 

EDIT: increased buffers and now it plays. Can it be that the file does not fit into the buffer?

What is your -a and -b that it works?

[cvrle59]

-a 10 -b 1000:2000000

Mine is -a 23 -b 1000:2000000 (tried -b 2000:2000000 too), something is wrong. I did play one album and that's it. I stopped the services, tried different things, no success...back to MQn.

[cvrle59]

I am "JLp V30/31 handicapped". I doesn't work on my system at all...sorry Jonathan.

[cvrle59]

you do have to wait a few seconds for the music to start with these versions, don't know if that's the issue.

Even 5 min doesn't help. I am afraid that there is more to it (system specifics), sorry. I tried any possible scenario with different buffers sizes too.

[cvrle59]

Yes, I did, a few times, they are both running as services.

[cvrle59]

Have you tried different -a values like 25, 250, 500? It worked for me.

I changed 1, 10 and 23. I will try higher tonight, as you mentioned.