part4 (around 1min20) seems to have more "swing",preferred.
part2 (30sek) the voice is a little bit brittle or scratchy.
part3 (1min) perhaps "liveless" or so.
part1 could be near by part4,difficult to say.
part2 (30sek) the voice is a little bit brittle or scratchy.
part3 (1min) perhaps "liveless" or so.
part1 could be near by part4,difficult to say.
Thanks for giving it a listen mjf 🙂
OK I think its probably time to put this one to rest.
What you have all been listening to are three CD players, a 10 yr old £20 Lidl supermarket special, a midprice and standard Sony CDP790 from around 1990, and a MicroMega Stage 2 (four NE5534's replaced with OPA604 and AD845's and recapped a few years ago).
So which is which is which 😀
The track starts out with the Lidl portable up to around 18 seconds. Now we have a switch to the MicroMega taking us to around 46 seconds in. Next up the Sony, and that continues to around 1 min 16 or so at which point the MicroMega takes us to the track end.
I really hope these tests go some way to showing why listening is important. You had nothing to compare against, no preconceptions, no ideas about which selections should or should not be better than the other etc etc. One of the crack listening team 🙂 thought the quality steadily declined as the track went on.
I was surprised just how good the portable was tbh.
To everyone that listened... thanks 🙂
This should be a clearer picture and you will see the players listed at the left.
Directors_Cut - My Photo Gallery
Oh, and a group photo 😀
http://www.diyaudio.com/forums/gallery/showphoto.php/photo/10103
OK I think its probably time to put this one to rest.
What you have all been listening to are three CD players, a 10 yr old £20 Lidl supermarket special, a midprice and standard Sony CDP790 from around 1990, and a MicroMega Stage 2 (four NE5534's replaced with OPA604 and AD845's and recapped a few years ago).
So which is which is which 😀
The track starts out with the Lidl portable up to around 18 seconds. Now we have a switch to the MicroMega taking us to around 46 seconds in. Next up the Sony, and that continues to around 1 min 16 or so at which point the MicroMega takes us to the track end.
I really hope these tests go some way to showing why listening is important. You had nothing to compare against, no preconceptions, no ideas about which selections should or should not be better than the other etc etc. One of the crack listening team 🙂 thought the quality steadily declined as the track went on.
I was surprised just how good the portable was tbh.
To everyone that listened... thanks 🙂
This should be a clearer picture and you will see the players listed at the left.
Directors_Cut - My Photo Gallery
Oh, and a group photo 😀
http://www.diyaudio.com/forums/gallery/showphoto.php/photo/10103
it is funny..........
i preferred part4 and did not like part2.but it was the same cd player ( "high end" micromega).
i preferred part4 and did not like part2.but it was the same cd player ( "high end" micromega).
There is no answer to that 🙂 I suppose if you were listening for real, at home, then you would try several different discs between players and come to a choice that way. Maybe hearing one "type" of sound directly after another colours the perception in some way. Like eating some type of food after another, both can be good on their own, but one followed by another, not good.
Dunno... can't answer that one.
Strange. Even with poor situation I tried to do a listen and my result was strange.
0-20 sec : standard normal sound, better in instrument sound than the one at 60' (the Sony).
20-44 sec : the best, typical high end sound with slow decay and lower freq response.
44-120 sec where instruments ended: not much to say but instruments sound are sub standard.
120 - where the vocal started: horrible! The vocal is unnatural and fatiguing. The worst part.
Well the vocal at the end is very demanding so not a fair comparison I guess.
0-20 sec : standard normal sound, better in instrument sound than the one at 60' (the Sony).
20-44 sec : the best, typical high end sound with slow decay and lower freq response.
44-120 sec where instruments ended: not much to say but instruments sound are sub standard.
120 - where the vocal started: horrible! The vocal is unnatural and fatiguing. The worst part.
Well the vocal at the end is very demanding so not a fair comparison I guess.
I think what we can draw from this is that the recording is what makes the difference. I think she moved around a little as she was recording this and her changes in position altered the way she sounded more than any differences the players put on us. If there were any audible differences they were extremely small.
Having said this the bit that I noticed the most was around 30 seconds between phrases she sounds a little like she's started singing through a sock after than before and that was through the same CD player.
120 seconds where she starts is just the way she sounds, she's singing with more oomph and her sound hardens up considerably because of it. She doesn't have a particularly melifluous voice anyway, not that I am particularly familiar with NC, but I wont be inclined to become more familiar after this either.
There's a very good reason why I build things to specification or on engineering principles, rather than boutique rubbish, because I simply know that I wont hear a difference. I am also psychologically happy when I know something is performing, objectively, in the way in which it should. And I know that that makes a difference to how I think something sounds.
Having said this the bit that I noticed the most was around 30 seconds between phrases she sounds a little like she's started singing through a sock after than before and that was through the same CD player.
120 seconds where she starts is just the way she sounds, she's singing with more oomph and her sound hardens up considerably because of it. She doesn't have a particularly melifluous voice anyway, not that I am particularly familiar with NC, but I wont be inclined to become more familiar after this either.
There's a very good reason why I build things to specification or on engineering principles, rather than boutique rubbish, because I simply know that I wont hear a difference. I am also psychologically happy when I know something is performing, objectively, in the way in which it should. And I know that that makes a difference to how I think something sounds.
Btw, the sound of portable player is standard. One bit DAC can sound smooth without mistake but that's all you get, no more. I can hear potential with micro mega on second part, this is usually due to better opamp. That's why I prefer to see schematic than listening to electronics. All need upgrade/mod.
But the last part was really unacceptable. I don't know why. What opamp is used? Is it too expensive to use 5532?
But the last part was really unacceptable. I don't know why. What opamp is used? Is it too expensive to use 5532?
Jay, until the file structure was posted you were not able to give any listening impression, and you evaded listening comments on both files. Now the results are public its easy to pick the choices you want. I can only draw my own conclusions from that.
All the players were untouched during the test, The file for each player was recorded as a complete "take" and all the editing done later. The MicroMega is actually a bitstream DAC (twin TDA1305T's)and the opamps for the MicroMega are OPA604 and AD845 as mentioned in post #103
The Sony uses JRC5532's and the Lidl I'm unsure, probably 4558's or similar.
Yes, picking suitable recordings for these kind of tests is difficult. Its impossible to please everyone on musical content.
I wonder if the AD process has any bearing on that. The picture shows things "getting a bit busy" amplitude wise. Maybe the first three sections are more representative and if the Lidl player had been used for the last it would have struggled.
Hmmm 🙂 So you buy the generic Multicomp caps from CPC then... given a choice of course 😀
Actually after you asked me my result and I said I couldn't play the file, I went to listening thru my iPod QuickTime using my regular speaker. I came back with a note on my hand but I saw your post. So that's the true result, no right or wrong here. I was actually surprised how others prefer the last part, which to me was a complete mess. This is exactly the point why I prefer schematic when choosing electronics. There are too many possibilities. Small change to capacitor and potential DUT will show its class
First comments were rather off the cuff.
I was in unusually sensitive listening form. Reflecting on my own flip comments put bug in ear to listen again. I dropped cue points in while listening. Opening piano is missing something in low end, then rather abruptly becomes fuller and more natural. A recurring theme.
Then looking deeper with Cool Edit, spectrum above 20kHz cuts in and out. Enough data also exists in track to clearly see HP filtering of capacitor(s?).
Close listening reveals switching clicks. Some of these seem to be when a microphone is switched.
19.2kHz spike is right channel only and was introduced somewhere in original production recording chain. Similar defects crop up in various recordings.
Second track has less bass content overall, this could be HP of original, or strictly Mooly's caps used throughout.
Mooly's mastering skills grew a jump with doing all work at higher sample rate. Perhaps small cross fades were used too. Bass guitar is fuzzy electric sound where phase is already junked up.
Modulation of D/A A/D chain with music may be picked out as changes to sound character such as above mention of cymbals.
Here is shot of a tentative mix point in Rythmm and Soul track:
Instrumentation and voice don't change in above segment, but something does. Cymbals become grainy.
To do this right for test track requires recording source with no components other than interconnect of D/A to A/D. This and test conditions are sampled up at 32bit, normalized in levels, mixed with cross fades, and rendered back down to fs for preferred playback. Devil is in the details.
Goal here is getting single recording where regions are test patches. File difference and spectral analysis are obscured, and only choice is listening.
----------
Group delay smearing of IIR HP filters is most apparent with cleanly recorded acoustic instruments and voice. It is easily habituated to, thus all pass filters formed by most crossovers using passive components, or DSP counterparts is largely ignored. Differentiating between brands of these filters is only apparent when for engineering reasons they introduce tragic flaws deviating from the mathematical models that represent them.
Comparison of filter v no filter by ear is relatively easy, even with corner frequency set <15Hz.
Square waves with perfectly defined phase and amplitude of spectral components readily reveal differences too.
Once coupling caps are used with high enough value, difference of cap/no cap becomes less apparent.
Burden of difference then falls on woofer system. Lumped model capacitance of woofers include plenty of capacitance and inductance leading to group delay. Baffle be it open, closed, or variously ported with mounted driver defines the system.
When system is held constant, swapping of caps, and of other electronics may be differentiated.
Headphone.com has numerous measurements, revealing basic inherent depth of phase shift, and raw frequency response issues with drivers:
Sennheiser HD650, 50Hz square wave:
[http://graphs.headphone.com/graphCompare.php?graphType=0&graphID[]=856&scale=30
Etymotic ER-45-4S, 50Hz square wave:
http://graphs.headphone.com/graphCompare.php?graphType=0&graphID[]=746&scale=30
And one more: HiFiMAN-HE500:
http://www.headphone.com/buildAGraph.php?graphType=0&graphID[]=3244
The Sennheiser gets it up, but can't keep it up. Waveform falls rapidly to zero well before 1/2 cycle.
Headphone.com needs to make a living, and not surprisingly, they don't provide detailed measurement of less expensive gear.
I've got a $10 pair of Skullcandy ear buds. On closed coupler similar to ear canal in length, this is 50Hz square wave result:
HiFiMAN and Etymotic are in one class, and Sennheiser and $10 buds in another.
I've listened to the Sennheiser and HiFiMAN. HiFiMAN smokes the Sennheiser in one drag.
With electrolytics I buy to a specification, but for things like 0.1u ceramic decouplers, I tend to go with the cheapest per 100 in the right case style and voltage rating. I don't put lytics in the signal path unless called for (ie the values are large and either take up too much space, or too much wallet).
Barleywater, Pavel and others... I know I have said this before but I am in awe of the skills you guys have in looking and working with these type of files. This magic number of 19.2kHz intrigues me. I know one of my PC's lays it down on recordings... I've seen it in the Supertramp disc, and I think Pavel might have mentioned it as sometimes showing on discs. But what it is it ?
"Mooly's mastering skills grew a jump with doing all work at higher sample rate. Perhaps small cross fades were used too. Bass guitar is fuzzy electric sound where phase is already junked up."
😀 I wish ! For cutting and joining the tracks I simply relied on getting the start points the same by looking in Audacity and then cutting the silence before the first audio appeared, and for the later cutting trying to pick a spot where it might go unnoticed. Again squinting at the screen and trying to get a silent join around a zero crossing. Not easy.
"Mooly's mastering skills grew a jump with doing all work at higher sample rate. Perhaps small cross fades were used too. Bass guitar is fuzzy electric sound where phase is already junked up."
😀 I wish ! For cutting and joining the tracks I simply relied on getting the start points the same by looking in Audacity and then cutting the silence before the first audio appeared, and for the later cutting trying to pick a spot where it might go unnoticed. Again squinting at the screen and trying to get a silent join around a zero crossing. Not easy.
You've stayed on extremely polite side, with this comment ....
Personally, I give almost no attention/importance to ex-post expressed impressions.
Of course, one should listen to the same passage through the different sources to do an accurate comparison, but that can't really be done here, 🙂. I'm the one who found the quality to be best at the beginning, and worst at the end - but then, I am the chap who says the price of the equipment only vaguely relates to the perceived auditory quality.
The beginning of the track is relatively straightforward in the mix, not as testing so to speak - but here the vocals, and the work on the drums are most appealing.
The beginning of the track is relatively straightforward in the mix, not as testing so to speak - but here the vocals, and the work on the drums are most appealing.
I didn't post any comments in the thread because I had been told what was going on (with the first track) revealed to me by mooly in PM. Before he told me I had made my initial comments regarding the vocals sounding distorted when they peaked, but had decided that was my cheap headphones rather than the track.
So I knew that there were going to be multiple tracks spliced together in the second test, just not what or where.
I had a quick listen and commented to Mooly in PM that I thought that there was a change of track around 49 seconds. This was my PM.
Now I didn't get the transition point exactly (3 seconds late) but perhaps it took my brain 3 seconds to fully register that it had indeed changed 🙂 I was picking up on an additional edginess in the singers voice.. raspy or slightly ragged.
So for me the "obvious" difference was between the "good" player and the Sony, but only when changing from good to bad...
Tony.
So I knew that there were going to be multiple tracks spliced together in the second test, just not what or where.
I had a quick listen and commented to Mooly in PM that I thought that there was a change of track around 49 seconds. This was my PM.
wintermute said:
Now I didn't get the transition point exactly (3 seconds late) but perhaps it took my brain 3 seconds to fully register that it had indeed changed 🙂 I was picking up on an additional edginess in the singers voice.. raspy or slightly ragged.
So for me the "obvious" difference was between the "good" player and the Sony, but only when changing from good to bad...
Tony.
What's good about these exercises is that we're getting a group of people "training" themselves up to be more sensitive to tonal variations, 😉 ...
I would be struggling to pick the precise changeover points in the 2nd track - I listen to an overall 'sense' of the sound, the 'texture' of it, and it takes a significant bit of time to build up a feeling of the sound - by which time the transition is long gone ...
I would be struggling to pick the precise changeover points in the 2nd track - I listen to an overall 'sense' of the sound, the 'texture' of it, and it takes a significant bit of time to build up a feeling of the sound - by which time the transition is long gone ...
- Status
- Not open for further replies.