Romy the Cat wrote:

Thos who do not know how audio sound during the Super Electricity Days they do not know. I do know and I wonder: are the Super Electricity Days have own after-mark. I kind of pissed that it did not hit me early as I might wait for net Super Electricity Day who know how long. So, what I would like to do is to do a recording during a Super Electricity Day. Let play an PL and output the preamp output to the Pacific’s 88K. Then in a few days what the electricity is regular to do the same recording of the same LP to another 88K file. Then to compare two 88K files. There is a remote chance that it will be difference between those two files, and if it be then I would like to see what the difference would be.  That might be very interesting experiment. The Cat

Romy the Cat wrote:

Those who do not know how audio sound during the Super Electricity Days they do not know. I do know and I wonder: are the Super Electricity Days have own after-mark. I kind of pissed that it did not hit me early as I might wait for net Super Electricity Day who know how long. So, what I would like to do is to do a recording during a Super Electricity Day. Let play an PL and output the preamp output to the Pacific’s 88K. Then in a few days what the electricity is regular to do the same recording of the same LP to another 88K file. Then to compare two 88K files. There is a remote chance that it will be difference between those two files, and if it be then I would like to see what the difference would be.  That might be very interesting experiment.

Ok, I did it and the results are superbly illustrative I did not even expect that it will be so obvious. Also, it is a first attempt to look at the objective data about sound of electricity that I ever seen. I am sharing all data for you; you make your own research and conclusions.  First of all what it is.  I made 2 weeks ago a recording. I did not use LP as I initially intended but decided to use a CD as it gives more stable, repeatable result. So, it was a very good electricity day (not the Super Electricity Day but still very good day) and I recorded a file from my CD player. Today was bad electricity sound and I recorded the same file again. The path was the following:

CEC TL0 -> Bidat DAC -> Placette Preamp -> Tape Out -> Pacific Microsonic II in Master Mode ->Linx 16 -> Wavelab 6 at 88/24.

There was no editing or processing of files done of any kind and there was no change of equipment of any kind done. Pacific was working for a couple hours before recording. All equipment, including DAW was running from one single PP2000.

Below are both files recorded during bad and god electricity days. The file sound dramatically different.

http://www.mediafire.com/?jt88ts40q33u00r

http://www.mediafire.com/?qaljp2m1a294yt2

The visual comparing of the file also indicate that difference is VERY dramatic. The file above is from good electricity day and below is from bad day.

 

Here is the 3D diference from 20Hz to 20.000Hz

 

The data above proves that it is perfectly possible to develop a sonic text that will objectively, without any BS or excuses for deafness or ignorance to measure SONIC performance of electricity. The test might be loaded from USB flash card and measure how suitable given electivity for Sound.

manisandher wrote:

Romy, this is very interesting indeed. I'm assuming that the sonic difference between these two files is in line with the sonic difference you hear between bad and good 'electricity' days when you're just listening to music, with no recording chain involved? If this is indeed the case, then what you've done here is astonishing - a clear indication (to me at least) that what we're calling 'good/bad electricity days' are a bigger determining factor in the final sound than probably the hardware itself. Surely the root cause must now be explored. You've mentioned before that these differences exist even when the PP2000 is in battery mode (i.e. totally isolated from the AC mains). So can we confidenty eliminate electricity from the equation? Although I don't want to propose a strong hypothesis right now, can I just say that I seem to experience a similiar (although somewhat smaller, I think) effect by using a 'Schumann Resonator'. Maybe I should repeat your experiment with the SR on and off - I suspect the results might bear a striking resemblance to yours...

Bill wrote:

Could it be that the really great recordings of the 50's and 60's compared to those today were due to the fact that they had more "great electricity days to record.

Bill wrote:

I do believe if these experiments are continued, it would present a great paper for the AES. It certainly gives clear evidence that electricity quality does make a difference in the recording chain which they have been disputing for years.

Romy the Cat wrote:

The visual comparing of the file also indicate that difference is VERY dramatic. The file above is from good electricity day and below is from bad day.

Can you point out where and how in the picture the two files look very dramatically different? Because they look pretty much the same to me.

Romy the Cat wrote:

Here is the 3D diference from 20Hz to 20.000Hz

Did you do an audio file comparison on the two files posted? Unfortunately one should not do that, because the files are not synchronized/time-aligned. For example, look at this close-up starting at 1m52s:

It is clear that the waveforms for good and bad are identical, but look at how the one for the good electricity (the top one) is late (shifted to the right) compared to the other, by around 2.8ms. This means that a raw diff is meaningless.

I tried to remove some samples from one file to "fix" the time offset. I was able to reduce the diff amplitude already, but it seems the offset varies with time which makes it so much more difficult to get the real diff between the two signals.

Diff for the original files:

Diff for the files with reduced time offset:

(Scale is the same)

In particular, look at how the large diff near the end (from 2m23s onwards) almost disappears.

Romy the Cat wrote:

The data above proves that it is perfectly possible to develop a sonic text that will objectively, without any BS or excuses for deafness or ignorance to measure SONIC performance of electricity. The test might be loaded from USB flash card and measure how suitable given electivity for Sound.

Well, maybe this can be done. I am sure that bad sound from electricity can be measured and shown. But I do not think the pictures posted so far do that.

I am pretty sure that normal playback variations can explain differences in the time domain that we observe. We should try to remove that variability to be able to analyze what happens in other domains (amplitude, frequency?). However, that could be very difficult (if not outright impossible) and in doing so we risk losing information about time garbage introduced by bad electricity (jitter? hopefully time variability from playback is slow, so removing it should keep electricity-induced jitter).

I'll try getting better results and will post if anything interesting comes up.

Romy the Cat wrote:

You absolutely incorrectly, in my view, look at the data. The files of cause do not have the same start mark. Why shell they? I did not subtract one from another. You did in phase the files and did the subtraction but what the subtraction shows? You see the delta dose not indicate anything that we as listeners understand.  [...] I did not run subtraction but instead I run global picks extraction into 3D. This way the difference in timing marks is irrelevant.

Romy the Cat wrote:

What I did was zooming the file to the point where the resolution is high, the difference between the file is visible but the resolution still present the note or a phase in it’s completeness. Then you can see how the difference relates to Sound.

Romy the Cat wrote:

Lx, this is very interesting. At the initial picture I posted you do can clearly see the difference in dynamic range. When I did what you did and went to higher resolution then you are right - the difference disappears. Dose the Wivelab is so screwed that it deforms the presentation in respect to graphic resolution. Then what the DC came from. You are right it does have DC but where it come from?! The most important I stat the files do sound drastically different, it can’t be explained by DC and we need to see objective data differences. I agree that the experiment need to be doe again with more methodologically clean condition, the DC at least shell not be there. The Cat

• It makes the smaller signal look less dynamic; and as it happened, it was the bad signal that got the smaller rendering and looked worse, which seemed natural
  
• It completely changes the way a very packed signal is drawn on screen, and changes how the signal looks like; I would not fault wavelab, this would happen for nearly all rendering software and is due to the pixel size: resolution on today's screens is quite good, but far from being enough in this case

• I would try to repeat the measures several times with bad electricity, just to see how much variation we get from the measuring process only (or from normal variation from playback)
• I would use a signal with greater dynamics or louder so as to minimize the influence of noise from the recording equipment
  
• I would use a signal with more HF (there is almost none here)
  

N-set wrote:

Sorry for being naive, but may gathering spectral fingerprints of the electricity during bad and good days give something? Perhaps after N samples some pattern might emerge, although I guess nobody knows what to look for.

N-set wrote:

Also, is that certain that the bad interference comes solely via the lines and not (partially) via the air? I remember reading once an interview with a guy commercially converting listening rooms into faraday cages by placing copper rods.

Lx_ wrote:

As to the DC offset, I find it strange too obviously. However as it quickly disappears I am not too worried about it. Maybe this indicates a problem in the soundcard when starting the recording.

steverino wrote:

I'm not surprised at your disappointing results as I have experienced the same with a variety of power filtration or generating devices. Your surmise is correct that such devices work much more easily with AV equipment or computers than they do with audio. I have an Audience power filtration unit which was unacceptable with my 2 pure audio systems (although HP swore by them) but works well on my TV and its associated stereo system as well as on my computer (audio). Similarly, I have a PS Audio power cable unacceptable with the audio systems but works like a charm with the computer. I did not like the early PP for both sonic and practical reasons. The only device which I have found to work with multiple units in the chain are the PS Audio Ultimate Outlets (discontinued). They can be found used at modest cost but they sell quickly. You could try them with the other units. The lack of any real improvement in batteries over the years is becoming more frustrating with the increasing garbage in the AC and in the surrounding environment.