Skip to main content

Topic: Jplay - just another scam? YES IT IS! (Read 185285 times) previous topic - next topic

0 Members and 1 Guest are viewing this topic.
  • dhromed
  • [*][*][*][*][*]
Jplay - just another scam? YES IT IS!
Reply #300
I've added my own graphs from audacity (and a little help of photoshop) after earballing the volumes.

I was mistaken before; it's the jplay sample that has its phase inverted. I could then fairly accurately align them.

The excerpt and jplay samples are nearly the same. As I said, I cannot ABX them after volume leveling. The foobar sample has had some extensive work done on it.

  • Porcus
  • [*][*][*][*][*]
Jplay - just another scam? YES IT IS!
Reply #301
Well, clogging your computer with maintaining library of that size is hardly a question of playback hardware requirements.


But actually I do find it quite convenient to have the foobar2000 application open when I use it for playback. Don't think that is too much to ask. 

So that is part of the issue: I would have wanted a way to ensure that playback has sufficient priority not to stutter, period, and everything else shares the rest of the juice. On the surface of it, an application granting me that level of access to Windows' priorities, would be A Good Thing. I could even have tried JPlay hadn't it been for the price and the smell of
http://upload.wikimedia.org/wikipedia/en/6...oll_Swindle.jpg


(The Right Thing To Do would be to implement a proper database solution. fb2k is not efficient: http://www.hydrogenaudio.org/forums/index....rt=#entry811910 .)

  • db1989
  • [*][*][*][*][*]
  • Global Moderator
Jplay - just another scam? YES IT IS!
Reply #302
The excerpt and jplay samples are nearly the same. As I said, I cannot ABX them after volume leveling. The foobar sample has had some extensive work done on it.
The priority now absolutely needs to be determining the source of those introduced differences. I would like to assume they were completely unintentional, in which case, something else was confounding the signal in sabrehagen’s setup. For the record again, foobar2000 is a perfectly competent player that does not have to do anything extra to ensure unadulterated playback, and it does not demand 95 EUR for the privilege.

  • krabapple
  • [*][*][*][*][*]
Jplay - just another scam? YES IT IS!
Reply #303
I trimmed all the samples to be the same start/stop, then tried ABXing the two samples vs reference, using fooABX + replaygain


foo_abx 1.3.3 report
foobar2000 v1.2.2
2013/04/02 16:30:39

File A: 3_Copy_edit.flac
File B: Pretzel_Logic_Excerp_edit.flac

16:30:39 : Test started.
16:30:59 : 01/01  50.0%
16:31:05 : 02/02  25.0%
16:31:11 : 03/03  12.5%
16:31:17 : 04/04  6.3%
16:31:21 : 05/05  3.1%
16:31:27 : 06/06  1.6%
16:31:32 : 07/07  0.8%
16:31:35 : 08/08  0.4%
16:31:38 : 09/09  0.2%
16:31:40 : 10/10  0.1%
16:31:42 : 11/11  0.0%
16:31:46 : 12/12  0.0%
16:31:50 : 13/13  0.0%
16:31:52 : 14/14  0.0%
16:31:54 : 15/15  0.0%
16:31:57 : 16/16  0.0%
16:32:12 : Test finished.

----------
Total: 16/16 (0.0%)

Easy.

ABXing Jplay vs reference was harder...I had to focus on the very first syllables of the singing to get this score

foo_abx 1.3.3 report
foobar2000 v1.2.2
2013/04/02 16:40:40

File A: 4_Copy_edit.flac
File B: Pretzel_Logic_Excerp_edit.flac

16:40:40 : Test started.
16:41:04 : 01/01  50.0%
16:41:10 : 02/02  25.0%
16:41:19 : 03/03  12.5%
16:41:26 : 03/04  31.3%
16:41:29 : 04/05  18.8%
16:41:36 : 05/06  10.9%
16:42:06 : 06/07  6.3%
16:42:10 : 07/08  3.5%
16:42:13 : 08/09  2.0%
16:42:20 : 09/10  1.1%
16:42:24 : 10/11  0.6%
16:42:49 : 11/12  0.3%
16:43:04 : 11/13  1.1%
16:43:08 : 12/14  0.6%
16:43:14 : 13/15  0.4%
16:43:19 : 14/16  0.2%
16:43:32 : Test finished.

----------
Total: 14/16 (0.2%)
  • Last Edit: 02 April, 2013, 04:46:53 PM by krabapple

  • krabapple
  • [*][*][*][*][*]
Jplay - just another scam? YES IT IS!
Reply #304
I've added my own graphs from audacity (and a little help of photoshop) after earballing the volumes.

I was mistaken before; it's the jplay sample that has its phase inverted. I could then fairly accurately align them.

The excerpt and jplay samples are nearly the same. As I said, I cannot ABX them after volume leveling. The foobar sample has had some extensive work done on it.


Pretty much the same here (though I could ABX both vs reference, the Jplay/source ABX was much harder) .  Hmm... could be I copied the wrong data over for the reference....I'm redoing it now, using excerpt- and level-matched files

(later) Yup, definitely had the wrong data in the first graph -- this new one shows Jplay to be much more like the reference than foobar in the crucial upper bass and midrange.  Of course, foobar should NOT be changing the EQ like this unless it's been set to do so...

This version of the graph starts with the files all edited in Audition so the same slice of audio is used for all three (reference, foobar and Jplay);  the files are then replaygain scanned to get the RG values, then Audition 'Frequency Analysis' scanned to get the avg level/frequency data;, then those data are dumped to Excel, then the RP values (-0.65 for the reference, +5.32 for foobar, +8.85 for Jplay)  are added to the data, and THEN the difference between sample and reference is calculated and plotted.

The corrected graph is here
http://www.hydrogenaudio.org/forums/index....st&p=829831
  • Last Edit: 02 April, 2013, 05:32:30 PM by krabapple

  • lvqcl
  • [*][*][*][*][*]
  • Developer
Jplay - just another scam? YES IT IS!
Reply #305
Also it seems that mid and side channels of 3_Copy_edit were modified differently, so it's not just an EQ.
  • Last Edit: 02 April, 2013, 05:31:38 PM by lvqcl

  • krabapple
  • [*][*][*][*][*]
Jplay - just another scam? YES IT IS!
Reply #306
Also it seems that mid and side channels of 3_Copy_edit were modified differently, so it's not just an EQ.


what do 'mid' and 'side' mean here?

  • mitchco
  • [*]
Jplay - just another scam? YES IT IS!
Reply #307
Lynx Hilo’s USB driver supports multi-client ASIO applications.  That means I can play a song in Foobar (v 1.2.4) while recording it’s bitstream in Audacity (ASIO version 1.3.13 alpha).

Following the same procedure as outlined here: JRiver Mac vs JRiver Windows Sound Quality Comparison, I was able to record Foobar, then record JPlay (v5.1), using the following settings

I have uploaded the two recorded samples, plus the original.  I edited the recorded samples so that they start and end at the same times, including the original.  The original flac was converted to wav using JRiver MC18 as JPlay can’t play flac. Note, other than inverting tracks in Audacity, nothing else was applied (i.e. no gain adjustments).

Foobar recording 60sec 33MB
JPlay recording 60sec 33MB
Original recording 60sec 33MB.

Comparing Foobar recording to the original file
Bit identical.

Comparing Foobar recording to JPlay recording
Bit identical.

For fun, comparing JRiver MC 18 on the Mac to JPlay on Windows
Bit identical.

For whatever reason, Foobar recording’s polarity was inverted compared to JPlay and JRiver (both Mac and PC versions) and the original file.  I had a look at the Foobar prefs, but could not find anywhere where this could be happening.

The JRiver Mac and Windows recordings can also be downloaded.  These are edited the same as the Foobar and JPlay recordings.
JRiver Mac recording 60sec 33MB.
Jriver Windows recording 60sec 33MB

I might try the different engines in JPlay to see what, if any, effect they have on the bitstream. 

  • db1989
  • [*][*][*][*][*]
  • Global Moderator
Jplay - just another scam? YES IT IS!
Reply #308
Thanks for the results.

I have to wonder whether, somewhere, someone is decrying these sorts of tests with stern admonitions that intercepting the stream in such a way is a fraudulent tactic that destroys Jplay’s ability to work its magic. Sort of a ‘Don’t put your local deity to the test’ thing… coincidentally.

  • lvqcl
  • [*][*][*][*][*]
  • Developer
Jplay - just another scam? YES IT IS!
Reply #309
what do 'mid' and 'side' mean here?

mid and side channels:

mid = (left + right)/2, side = (left - right)/2

  • lvqcl
  • [*][*][*][*][*]
  • Developer
Jplay - just another scam? YES IT IS!
Reply #310
Lynx Hilo’s USB driver supports multi-client ASIO applications.  That means I can play a song in Foobar (v 1.2.4) while recording it’s bitstream in Audacity (ASIO version 1.3.13 alpha).
[...]
For whatever reason, Foobar recording’s polarity was inverted compared to JPlay and JRiver (both Mac and PC versions) and the original file.  I had a look at the Foobar prefs, but could not find anywhere where this could be happening.


Did you use ASIO output plugin for foobar2000? Is it possible to test foobar2000 with DirectSound or WASAPI output?

  • Wombat
  • [*][*][*][*][*]
Jplay - just another scam? YES IT IS!
Reply #311
Thanks mitchco! I did read your comparison on computeraudiophile already. Nice work and amusing to read the resulting comments and weird theories why some things like MAC and Jplay still sounds better!
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

  • mitchco
  • [*]
Jplay - just another scam? YES IT IS!
Reply #312
Lynx Hilo’s USB driver supports multi-client ASIO applications.  That means I can play a song in Foobar (v 1.2.4) while recording it’s bitstream in Audacity (ASIO version 1.3.13 alpha).
[...]
For whatever reason, Foobar recording’s polarity was inverted compared to JPlay and JRiver (both Mac and PC versions) and the original file.  I had a look at the Foobar prefs, but could not find anywhere where this could be happening.


Did you use ASIO output plugin for foobar2000? Is it possible to test foobar2000 with DirectSound or WASAPI output?


I forgot to mention I used the ASIO output plugin for foobar2000.  If I have time, I will try DS.

Thanks for the results.

I have to wonder whether, somewhere, someone is decrying these sorts of tests with stern admonitions that intercepting the stream in such a way is a fraudulent tactic that destroys Jplay’s ability to work its magic. Sort of a ‘Don’t put your local deity to the test’ thing… coincidentally.


I have the Lynx Hilo patched for digital loopback where the output of the music player is going through the Hilo and routed back to the input of Audacity.  However, JPlay also supplies it's own drivers (also shows up in foobar).  Perhaps this is the magic (and possibly the JPlay engines).  I will run a few more loopback tests to see using the JPlay driver.

  • db1989
  • [*][*][*][*][*]
  • Global Moderator
Jplay - just another scam? YES IT IS!
Reply #313
However, JPlay also supplies it's own drivers (also shows up in foobar).  Perhaps this is the magic (and possibly the JPlay engines).  I will run a few more loopback tests to see using the JPlay driver.
The only ‘magic’ that’s possible is deliberate alteration of the stream, in which case JPLAY (I just realised it’s stylised using cruise-control) would be answering the title of this thread in the affirmative.

  • krabapple
  • [*][*][*][*][*]
Jplay - just another scam? YES IT IS!
Reply #314
added a third result to the graph , basically recording the output of my foobar2k setup on a work computer  (not set to be bit-perfect, it converts everything to 48kHz)

http://www.hydrogenaudio.org/forums/index....st&p=829876

  • mitchco
  • [*]
Jplay - just another scam? YES IT IS!
Reply #315
Lynx Hilo’s USB driver supports multi-client ASIO applications.  That means I can play a song in Foobar (v 1.2.4) while recording it’s bitstream in Audacity (ASIO version 1.3.13 alpha).

Following the same procedure as outlined here: JRiver Mac vs JRiver Windows Sound Quality Comparison, I was able to record Foobar, then record JPlay (v5.1), using the following settings

I have uploaded the two recorded samples, plus the original.  I edited the recorded samples so that they start and end at the same times, including the original.  The original flac was converted to wav using JRiver MC18 as JPlay can’t play flac. Note, other than inverting tracks in Audacity, nothing else was applied (i.e. no gain adjustments).

Foobar recording 60sec 33MB
JPlay recording 60sec 33MB
Original recording 60sec 33MB.

Comparing Foobar recording to the original file
Bit identical.

Comparing Foobar recording to JPlay recording
Bit identical.

For fun, comparing JRiver MC 18 on the Mac to JPlay on Windows
Bit identical.

For whatever reason, Foobar recording’s polarity was inverted compared to JPlay and JRiver (both Mac and PC versions) and the original file.  I had a look at the Foobar prefs, but could not find anywhere where this could be happening.

The JRiver Mac and Windows recordings can also be downloaded.  These are edited the same as the Foobar and JPlay recordings.
JRiver Mac recording 60sec 33MB.
Jriver Windows recording 60sec 33MB

I might try the different engines in JPlay to see what, if any, effect they have on the bitstream.


Testing continued...

I changed the JPlay engine from Beach to River and recorded in Audacity.

Comparing JPlay Beach engine to JPlay River engine
Bit identical.

Next up using the JPlay driver in Foobar and another recording in Audacity.

Comparing Foobar using JPlay driver to the orginal recording
Bit identical.

Finally, using the JPlay driver in JRiver and another recording.

Comparing JRiver MC 18 using JPlay driver to the orginal recording
Bit identical.

The recorded samples:

JPlay with Beach engine 60sec 33MB
Foobar with JPlay driver 60sec 33MB
JRiver with JPlay driver 60sec 33MB


  • bennetng
  • [*][*][*][*][*]
Jplay - just another scam? YES IT IS!
Reply #316
My previous attempt (after tweaking my comodo firewall) to test Jplay also showed bit-perfect result when using Jplay's ASIO mode, but not bit-perfect with KS mode. KS mode will truncate 24-bit signal to 16-bit without dithering, other than this KS is basically same as ASIO, I mean, I cannot observe any other processing including volume, EQ, stereo treatment and so on.

That's why I want sabrehagen to test his system using my signal. Even though he is not recording in digital, the differences of Jplay vs foobar should not be so big.

  • Nessuno
  • [*][*][*][*]
Jplay - just another scam? YES IT IS!
Reply #317
But actually I do find it quite convenient to have the foobar2000 application open when I use it for playback. Don't think that is too much to ask.

Well, I don't use Fb2K myself and don't know how it internally works, but if this feature actually overloads a CPU (whichever!) so much to cause issues with his own playing function, I think it's time for you to switch to a more properly designed piece of software! The like of Jplay, for example...
... I live by long distance.

  • sabrehagen
  • [*][*]
Jplay - just another scam? YES IT IS!
Reply #318
Hi bennetng,

Sure, I'd like to test it again. If, like people's analysis shows here, my version of foobar is altering the sound, I would like to configure it so that it doesn't do so, and to an ABX test again.

sabrehagen,

I would like to assist you to do a recording test, please provide the information below:
1. Your OS version
2. The recording software you are using

Quote
I was using a fresh install of foobar, having never used the software in my life before. No DSP, no nothing, just a stock install. I connected the output of my DAC to the line in on my on board sound card and recorded it there.

3. Does it mean onboard sound digital output> DAC digital input> DAC analog output > onboard sound analog input?


As for your questions:
1. Windows 7 Professional 64-bit
2. I used Audio DiffMaker
3. Yes, exactly.

I will definitely conduct these tests, however I am in the middle of exams, so I will try to get it done in the next couple of days.

Lynx Hilo’s USB driver supports multi-client ASIO applications.  That means I can play a song in Foobar (v 1.2.4) while recording it’s bitstream in Audacity (ASIO version 1.3.13 alpha).

Following the same procedure as outlined here: JRiver Mac vs JRiver Windows Sound Quality Comparison, I was able to record Foobar, then record JPlay (v5.1), using the following settings

I have uploaded the two recorded samples, plus the original.  I edited the recorded samples so that they start and end at the same times, including the original.  The original flac was converted to wav using JRiver MC18 as JPlay can’t play flac. Note, other than inverting tracks in Audacity, nothing else was applied (i.e. no gain adjustments).

Foobar recording 60sec 33MB
JPlay recording 60sec 33MB
Original recording 60sec 33MB.

Comparing Foobar recording to the original file
Bit identical.

Comparing Foobar recording to JPlay recording
Bit identical.

For fun, comparing JRiver MC 18 on the Mac to JPlay on Windows
Bit identical.

For whatever reason, Foobar recording’s polarity was inverted compared to JPlay and JRiver (both Mac and PC versions) and the original file.  I had a look at the Foobar prefs, but could not find anywhere where this could be happening.

The JRiver Mac and Windows recordings can also be downloaded.  These are edited the same as the Foobar and JPlay recordings.
JRiver Mac recording 60sec 33MB.
Jriver Windows recording 60sec 33MB

I might try the different engines in JPlay to see what, if any, effect they have on the bitstream.


Testing continued...

I changed the JPlay engine from Beach to River and recorded in Audacity.

Comparing JPlay Beach engine to JPlay River engine
Bit identical.

Next up using the JPlay driver in Foobar and another recording in Audacity.

Comparing Foobar using JPlay driver to the orginal recording
Bit identical.

Finally, using the JPlay driver in JRiver and another recording.

Comparing JRiver MC 18 using JPlay driver to the orginal recording
Bit identical.

The recorded samples:

JPlay with Beach engine 60sec 33MB
Foobar with JPlay driver 60sec 33MB
JRiver with JPlay driver 60sec 33MB


The fact that you are getting bit-perfect matches is very exciting. Thank you for doing these tests, it is really appreciated. This means one of two things to me; 1. JPLAY actually does what it claims, and delivers audio 'better', resulting in a better overall sound, or 2. I am not hearing correct sound out of foobar and VLC leading me to believe JPLAY is actually better, when infact, it is the only correct source.

  • bennetng
  • [*][*][*][*][*]
Jplay - just another scam? YES IT IS!
Reply #319
Lynx Hilo’s USB driver supports multi-client ASIO applications.  That means I can play a song in Foobar (v 1.2.4) while recording it’s bitstream in Audacity (ASIO version 1.3.13 alpha).
[...]
For whatever reason, Foobar recording’s polarity was inverted compared to JPlay and JRiver (both Mac and PC versions) and the original file.  I had a look at the Foobar prefs, but could not find anywhere where this could be happening.


Did you use ASIO output plugin for foobar2000? Is it possible to test foobar2000 with DirectSound or WASAPI output?


http://www.hydrogenaudio.org/forums/index....st&p=829184
But my previous test with video illustration showed foobar+ASIO with bit-perfect result without polarity inversion...

  • hlloyge
  • [*][*][*][*][*]
Jplay - just another scam? YES IT IS!
Reply #320
The fact that you are getting bit-perfect matches is very exciting. Thank you for doing these tests, it is really appreciated. This means one of two things to me; 1. JPLAY actually does what it claims, and delivers audio 'better', resulting in a better overall sound, or 2. I am not hearing correct sound out of foobar and VLC leading me to believe JPLAY is actually better, when infact, it is the only correct source.

I am sorry, but from his test results it is obvious that Foobar2000 and JPlay do the same thing, except that you pay 100 dollars more for JPlay. Now, if you analyse your playback chain, it should be pretty obvious where the problem is - of course, if you know what you are doing.
But then again, if you enter such discussion as this, I would expect you know what are you doing, playback chain as simple as this is the very basic of PC audio...

  • bennetng
  • [*][*][*][*][*]
Jplay - just another scam? YES IT IS!
Reply #321
Hi bennetng,

Sure, I'd like to test it again. If, like people's analysis shows here, my version of foobar is altering the sound, I would like to configure it so that it doesn't do so, and to an ABX test again.


Thanks. I have another question and request as well. Jplay requires an ASIO-compatible audio interface to work but you were saying you are using onboard sound. As far as I know, onboard sound usually has no native ASIO driver, but it can be solved by installing a third party ASIO driver. If you are doing so, which third party ASIO driver are you using? Or if your onboard sound has a native ASIO driver, what is the name of your onboard device? You should be able to find the name of your device in Jplay's settings, if you are not sure, please post some screenshots.

Another request is, please go to
http://www.foobar2000.org/components
and install "ASIO support" and "WASAPI output support" and configure foobar to output using ASIO and WASAPI, then perform the recording test again. You may have more than one ASIO or WASAPI output device to choose, if yes, please test all of them, if you can choose different output bit-depth, test all of them. If you encounter error or unable to playback my test signal under some combinations, for example, some bit-depth doesn't work, can't play one of the two files and so on, please report them and post the error messages.

You may need to familiarize yourself with foobar's configuration dialogs, please be patient, don't rush.

  • sabrehagen
  • [*][*]
Jplay - just another scam? YES IT IS!
Reply #322
The fact that you are getting bit-perfect matches is very exciting. Thank you for doing these tests, it is really appreciated. This means one of two things to me; 1. JPLAY actually does what it claims, and delivers audio 'better', resulting in a better overall sound, or 2. I am not hearing correct sound out of foobar and VLC leading me to believe JPLAY is actually better, when infact, it is the only correct source.

I am sorry, but from his test results it is obvious that Foobar2000 and JPlay do the same thing, except that you pay 100 dollars more for JPlay. Now, if you analyse your playback chain, it should be pretty obvious where the problem is - of course, if you know what you are doing.
But then again, if you enter such discussion as this, I would expect you know what are you doing, playback chain as simple as this is the very basic of PC audio...

The reason it excites me, is I can actually hear the difference. That said, I have been hearing a difference in an evidently different output, as proved by people's analysis. Note, that I have yet to pay for JPLAY, I am still only using the trial.

Well, it doesn't seem all that hard to analyse the chain, but there seem to be so few points it can change. My current chain is file > software > DAC > headphones. Foobar has all DSP turned off, and my audio control panels have all DSP turned off. Therefore I am having trouble working out where the issue could arise. I think I saw a link a few posts back regarding how to get unadulterated output from foobar, so I shall have to search back for that and follow the guide.

  • sabrehagen
  • [*][*]
Jplay - just another scam? YES IT IS!
Reply #323
Hi bennetng,

Sure, I'd like to test it again. If, like people's analysis shows here, my version of foobar is altering the sound, I would like to configure it so that it doesn't do so, and to an ABX test again.


Thanks. I have another question and request as well. Jplay requires an ASIO-compatible audio interface to work but you were saying you are using onboard sound. As far as I know, onboard sound usually has no native ASIO driver, but it can be solved by installing a third party ASIO driver. If you are doing so, which third party ASIO driver are you using? Or if your onboard sound has a native ASIO driver, what is the name of your onboard device? You should be able to find the name of your device in Jplay's settings, if you are not sure, please post some screenshots.

Another request is, please go to
http://www.foobar2000.org/components
and install "ASIO support" and "WASAPI output support" and configure foobar to output using ASIO and WASAPI, then perform the recording test again. You may have more than one ASIO or WASAPI output device to choose, if yes, please test all of them, if you can choose different output bit-depth, test all of them. If you encounter error or unable to playback my test signal under some combinations, for example, some bit-depth doesn't work, can't play one of the two files and so on, please report them and post the error messages.

You may need to familiarize yourself with foobar's configuration dialogs, please be patient, don't rush.


I have a Realtek ALC892 chipset. I will install ASIO4ALL now, to ensure I have asio. I have been using WASAPI so far. I have installed those components as per your request.

  • Porcus
  • [*][*][*][*][*]
Jplay - just another scam? YES IT IS!
Reply #324
Foobar has all DSP turned off, and my audio control panels have all DSP turned off. Therefore I am having trouble working out where the issue could arise.


Looking for simple explanations: Could it be that your computer has some EQ applied in the Windows mixer? Win7 offers per-device tone controls. If such are applied, then the signal will differ from a path that bypasses the mixer altogether.
  • Last Edit: 03 April, 2013, 05:36:41 AM by Porcus