Can you tell original file from tube amp record? - test

Which file is the original and which do you prefer

  • Apricot is the original file

    Votes: 7 46.7%

  • Avocado is the original file

    Votes: 5 33.3%

  • I prefer Apricot by listening

    Votes: 7 46.7%

  • I prefer Avocado by listening

    Votes: 7 46.7%
  • Total voters
    15
  • Poll closed .
Jump to Latest
#305
humm if i had not tried the sine test with my headphones i would have concluded that indeed my computers speakers are crap, so why are both sine files seemingly more distorted with my headphones?

and ok my headphones may also be crap but if other can hear a difference in the sines files where do i look next? soundcard?
 
#307
I agree that Direct Sound is problematic - most people try to avoid it as it can be resampling the audio file.

The older version of Foobar ABX uses Direct Sound but the newer version uses whatever output is set to, Kernel streaming, ASIO

From the new version:
"Uses core output settings on foobar2000 v1.3.5 and newer. Falls back to default DirectSound when running in older foobar2000 versions.
Usability improvements"
 
#308
turk 182 said:
humm if i had not tried the sine test with my headphones i would have concluded that indeed my computers speakers are crap, so why are both sine files seemingly more distorted with my headphones?

and ok my headphones may also be crap but if other can hear a difference in the sines files where do i look next? soundcard?
Click to expand...

What headphones, producer, type. I have some cheaper JVC and Sony headphones with distortion that doesn't allow to get the clean tone and the test doesn't pass. Sennheiser HD598 pass easily.
 
#311
Pano said:
I tested my playback chain as used for the ABX. No surprise I can't hear a difference!
Played from Foobar both sig1 and siga show H2 near -46dB. Maybe 2dB higher on siga.wave but both are close. Curious. :scratch:
Click to expand...

This makes sense and that's why I always recommend to make a loopback test of the system first. I may sound boring, but without measurements we are blind. I get this as attached when I play sig1 and siga, and couple of hours ago I showed headphone acoustic output.
Played from foobar and recorded/analyzed by Arta.

Now, it also determines our listening tests.
 

Attachments

  • mysys_sig1.png
    mysys_sig1.png
    47.7 KB · Views: 120
  • mysys_siga.png
    mysys_siga.png
    48.2 KB · Views: 119
Last edited:
#313
Thank god i was right... lol....
I felt something not so natural when the voice starts on apricot and a lack of dynamic somehow.
I felt exactly the same thing on 2 diferent setups... phones and amp...

Setup 1 - samsung galaxy s9 +audio technica studio headphones.

Setup 2 - raspberry pi+ tda1543 nos+ modulus 86 amp + jbl l830 monitors.
 
Last edited:
#314
PMA said:
@Jakob2: Thank you for being patient with my terminology inaccuracies.
Click to expand...

It is not about terminology but about discussion efficiency/habits. You raised the question if several different experiments could be summed up to calculate probability already in the blowtorch thread, i answered the question and your reaction was silence but now it appears in this thread.
You asked me some questions a day or two later in the blowtorch thread about my assessment of "sufficiently statistical ...." in different cases, i again took my time to answer it in detail, again no response, but surely it is related again to the analysis topic in this and other threads.

A couple of days ago in this thread i commented on the distinction between "being not able to reject the null-hypothesis" and "the result was nobody could detect a difference" and you asked me what i´d propose instead of p=0.5 . I asked about clarification (your response ? nothing) , i asked again and your response was pure silence, but now it appears again.

So tell me, what i should think about it?
Does raising questions without being interested in answers make any sense?

So, can I conclude, that, based on member's ABX reports posted (and assuming there were no other ABX trials of the same members remaining unpublished), we may say that a sum of the members was able to tell the difference? Or, in another way, what would be your conclusion from the results posted. Thank you.
Click to expand...

The conclusion from the combined results is that it is not so much compatible with the assumption that all results occured due to random guessing. The probability to get the observed data by chance is quite low.
So that is a hint that there might be something to detect, but as, we have no results from positive or negative controls and no measurements for the majority of the test systems (my apologies if i had missed these),it is hard to say if the (maybe existent) difference is introduced by variation of the independent variable.

Replication is needed anyway and the fundamental uncertainty wrt the systems should be addressed.
 
#315
Jakob2 said:
It is not about terminology but about discussion efficiency/habits. You raised the question if several different experiments could be summed up to calculate probability already in the blowtorch thread, i answered the question and your reaction was silence but now it appears in this thread.
.
Click to expand...

I was writing a long reply and my Internet collapsed ...

So shortly: I have asked again about summation of results because I did not know I can make a sum of results of 8 participants. To me it is not intuitive. In my previous question re results summation I was considering results of one and same person and I understood the results may be summed. Nothing personal, please.
 
#317
Jakob2 said:
The conclusion from the combined results is that it is not so much compatible with the assumption that all results occured due to random guessing. The probability to get the observed data by chance is quite low.
So that is a hint that there might be something to detect, but as, we have no results from positive or negative controls and no measurements for the majority of the test systems (my apologies if i had missed these),it is hard to say if the (maybe existent) difference is introduced by variation of the independent variable.

Replication is needed anyway and the fundamental uncertainty wrt the systems should be addressed.
Click to expand...

Thank you.
 
#318
turk 182 said:
Pano this is the sine files we're talking about, right?
Click to expand...
Yes, sine1.wav and sinea.wav

PMA said:
Pano, can you show your foobar setting (Preferences - output and DSP manager), control panel sound setting of the card, card type?
Click to expand...
See screenshots below. No DSP is active. Output is via HDMI to a Yamaha RX-V2700. Yamaha on processing bypass "Pure Direct"
 

Attachments

  • foobar output.png
    foobar output.png
    10.9 KB · Views: 111
  • foobar DSP.png
    foobar DSP.png
    10.5 KB · Views: 108
#320
mmerrill99 said:
@Pma, effectively you have done a post test calibration of the users & their systems by accident. So what does this new data tell you or what have you learned from this?
Click to expand...

My problem is that I automatically supposed that users make tests of their systems. I can see that it is not the case so I think that you know what kind of conclusions I have made for myself. On the other hand, I think it may be a positive lesson for many participants, because shared experience almost never works, one has to go through himself and then it is forever. So, if 2 - 3 users have learnt something from this exercise, to me it was worth efforts.
 
You must log in or register to reply here.
Top Bottom