Head-Fi.org › Forums › Equipment Forums › Dedicated Source Components › From CD to SACD how much of difference
New Posts  All Forums:Forum Nav:

From CD to SACD how much of difference - Page 5

post #61 of 161
Quote:
Originally Posted by systemerror909 View Post
For fun, I once tried recording some vinyl onto my computer varying the sampling rates between 96khz 24bit to 44.1khz 16bit. Maybe this isn't a fair comparison because my sound card is very cheap (creative external usb 24 bit), but the difference was huge.
I agree. A 16-44 copy of an LP just sounds as bad as a compact disc and nothing even vaguely resembling the original. It becomes hard, closed in and synthetic sounding. I do confess to having difficulty in being able to tell the difference between an original LP and a 24-96 copy though. I wish digital recording could go to 128 bits, since although the dynamic range is already adequate, I consistently find that increasing the bit depth improves the sonic representation of acoustic instruments and their particular positions within the 3D sound stage.
post #62 of 161
Quote:
Originally Posted by ADD View Post
I agree. A 16-44 copy of an LP just sounds as bad as a compact disc and nothing even vaguely resembling the original. It becomes hard, closed in and synthetic sounding. I do confess to having difficulty in being able to tell the difference between an original LP and a 24-96 copy though. I wish digital recording could go to 128 bits, since although the dynamic range is already adequate, I consistently find that increasing the bit depth improves the sonic representation of acoustic instruments and their particular positions within the 3D sound stage.
A 16/44.1 A/D/A stage inserted after the analog output from an turntable has been done several times, most famously Ivor Tiefenbrun was tested with this. He failed to detect when the A/D/A was inserted or not indicating that he at least could not tell the degrading effects of a 16/44.1 sampling. Others have done this and so far in blind tests nobody has been able to reliably detect the difference.
post #63 of 161
Quote:
Originally Posted by hciman77 View Post
A 16/44.1 A/D/A stage inserted after the analog output from an turntable has been done several times, most famously Ivor Tiefenbrun was tested with this. He failed to detect when the A/D/A was inserted or not indicating that he at least could not tell the degrading effects of a 16/44.1 sampling. Others have done this and so far in blind tests nobody has been able to reliably detect the difference.
Hi,

I learned long ago to trust my own ears, not those of others.
post #64 of 161
Quote:
Originally Posted by systemerror909 View Post
My other hobby is photography, and I think about sound very much like I think about photography. Recording, I imagine is very much like taking a picture. There are many ways you can place your microphone for non-optimal sound, just as there are may bad compositions/ways of lighting these compositions in photography.
That's quite true. If you are familiar with digital cameras, you're probably aware that the number of megapixels is no guarantee of good quality images. Everything depends on the judgement of the photographer when it comes to composition, the balance of light and color, and the quality of the optics. That relates in audio to the sound mixer's control of the miking, equalization and dynamics, and the fidelity of the speakers.

See ya
Steve
post #65 of 161
Quote:
Originally Posted by ADD View Post
A 16-44 copy of an LP just sounds as bad as a compact disc and nothing even vaguely resembling the original.
You probably need new capture equipment. With my rig, which is admittedly pro-grade, there is absolutely no difference between an LP and a 16/44.1 capture. It's not the format that you are hearing. It's all of the other factors involved in digitizing audio.

See ya
Steve
post #66 of 161
Quote:
Originally Posted by bigshot View Post
You probably need new capture equipment. With my rig, which is admittedly pro-grade, there is absolutely no difference between an LP and a 16/44.1 capture. It's not the format that you are hearing. It's all of the other factors involved in digitizing audio.
Perhaps you had better drop Bernie Grundman a line and tell him his 16-44 AD converters are not up to the task as well - they sound just as awful as mine.
post #67 of 161
You're talking about two different things.

The advantage of high bitrate sound is added resolution at extremely low volume levels. Dynamic range is extended LOWER, not higher. The peaks at 16 are identical to the peaks at 24.

16/44.1 is perfectly up to the task of capturing an LP with a dynamic range of perhaps 40-50dB. If you are going to record a live performance, however, high bitrates are necessary to allow headroom to raise the volume of low level sounds in the mix, without bringing up low resolution sound. You need particularly quiet mike pres and high bitrates so you can mix. If the music is alredy mixed at normal listening levels, high bitrate audio offers no advantages.

I've done the tests on this by capturing a Lincoln Mayorga D2D LP on a 24 bit protools rig and captured at standard CD quality sound. There was absolutely no audible difference between the two captures.

See ya
Steve
post #68 of 161
Quote:
Originally Posted by bigshot View Post
You're talking about two different things.

The advantage of high bitrate sound is added resolution at extremely low volume levels. Dynamic range is extended LOWER, not higher. The peaks at 16 are identical to the peaks at 24.

16/44.1 is perfectly up to the task of capturing an LP with a dynamic range of perhaps 40-50dB. If you are going to record a live performance, however, high bitrates are necessary to allow headroom to raise the volume of low level sounds in the mix, without bringing up low resolution sound. You need particularly quiet mike pres and high bitrates so you can mix. If the music is alredy mixed at normal listening levels, high bitrate audio offers no advantages.

I've done the tests on this by capturing a Lincoln Mayorga D2D LP on a 24 bit protools rig and captured at standard CD quality sound. There was absolutely no audible difference between the two captures.

See ya
Steve
Maybe your sacd player doesnt have the resolution to reveal the difference. I see you use a Philips 963sa. I had a stock 963sa and the sound was at best mediocre and resolution/ detail retrival was not very good. The player is quite nice though if you mod the output stage and even better with an aftermarket clock. I have all my players modded with atleast zap filters .... and the difference is very noticeable. You should give a zap filter a try. The zap filter is one of the best bang for the buck upgrades you can do in my experience.
post #69 of 161
Quote:
Originally Posted by ADD View Post
Hi,

I learned long ago to trust my own ears, not those of others.
Good to hear that. But in a public debate like this you have to learn to make others trust your ears and perceptions. Blind tests will go a long way.


Regards,

L.
post #70 of 161
Quote:
Originally Posted by bigshot View Post
I've done the tests on this by capturing a Lincoln Mayorga D2D LP on a 24 bit protools rig and captured at standard CD quality sound. There was absolutely no audible difference between the two captures.
Steve,

Honestly, you've tried my patience long enough. If you really think that I am going to believe any of the rediculous balony you spout on these forums regarding the quality of 16-44 digital, then you are as naive and self-centred as your hearing is bad.

You seem to have some belief that the results of your own personal tests become a defacto paridigm for the hearing standards of the remainder of the human race. As far as you are concerned, if you can't tell the differences then no one else on the planet can.

So, please, enough of the patronising garbage and talking down to everyone here as if you are the God of Audio and that if they don't agree with you that they are wrong and you have to be right.

I should have put you on my "ignore list" long before now, but the previous amusement caused by your self-absorbed ramblings has now become more an irritation than anything else. Then again, anyone with a screen-name of "bigshot" is most likely up themselves, so it's not really surprise that you do what you do on this and other forums around the net.

You can continue to insult others and talk down to them if you wish, but you are done with me.
post #71 of 161
Quote:
Originally Posted by Leporello View Post
Good to hear that. But in a public debate like this you have to learn to make others trust your ears and perceptions. Blind tests will go a long way.
I don't think blind tests would achieve anything at all. People will always come up with ifs and buts regardless. If I were to score 100% in being able to identify the difference between SACD and CD or vinyl versus a 16-44 copy with classical music, then people would just start blaming the inadequacy of the equipment or any other number of factors.

Afterall, the fact that I have already scored 100% in blind testing of 16-44 versus 24-48 clearly counts for nothing, since the Guinness Book of Records were not there on the day to record this seemingly super human achievement
post #72 of 161
Quote:
Originally Posted by ADD View Post
IAfter all, the fact that I have already scored 100% in blind testing of 16-44 versus 24-48 clearly counts for nothing, since the Guinness Book of Records were not there on the day to record this seemingly super human achievement
Intriguing, can you tell us how this was done ? - that is an honest question by the way , not a wind-up.
post #73 of 161
Hi,

I used two commercial HDAD releases made by Classic Records as well as the Mendelssohn Scottish Symphony on Linn records. The HDAD releases are the Stravinsky Rite of Spring (Everest) and Pictures at an Exhibition (also on Everest).

The HDAD releases were 35mm recordings made in the early 1960s. The Linn recording is a modern fully digital recording. In each HDAD release are two discs, each with identical program content. The difference between the discs is that the HDAD version contains both a 24-192 and 24-96 track, whilst the CD is just the redbook version. Both versions were converted to digital simultaneously during commercial mastering at Grundman studios using two seperate analogue to digital converters concurrently - one for the 24-192 version and another for the 16-44 version. That way, neither the final 24-96 version nor the 16-44 version suffered the ill-effects of non-integer downsampling. I then downsampled the 24-96 tracks on the HDAD disks using r8Brain Pro and created a DVD audio disk at 24 bit 48 Khz that was playable on an ordinary DVD player. This facilitated a direct comparison between the redbook disk and my 24-48 track of the same using the exact same equipment the whole time.

As for the Linn records release, it is available in both resolutions (24-96 and 16-44) from Linn's website, so I actually purchased both versions. Again, I created a redbook CD and DVD audio disk that could be played in the exact same player.

I then had an accomplice switch disks behind my back so to speak, and deliberately asked them to cheat at will so to speak by not necessarily alternating at all (i.e opening the disk tray and "swapping" disk but actually putting the same one back in again).

I was listening via my Firestone Cute Beyond amplifier (with OPA2107 op amp) and HD555 headphones. It was not difficult to tell the differences between the disks - the sound of the violins and to a lesser extent woodwinds is a dead give away just for starters, as well as the sounds of instruments with extreme high frequency energy such as cymbals.

The reason I conducted the 24-48 experiment was because I wanted to test my belief that I was incapable of discerning the difference between 24-48 and the original 24-96. The test seemed to bear this out - I did not get the 24-48 / 24-96 comparisons right, as opposed to the 16-44 and 24-48 comparisons.

But as I said earlier, it strikes me now that no one can necessarily be satisfied even if public blind testing was carried out. People would just say, well, if you had used this component or that component instead, then you would not have noticed any difference.

Unlike some, however, I am fully accepting of the fact that there are many, many people on the planet with far better hearing than mine, and I have no reason to doubt that there are people out there who could tell the differences between higher resolutions that completely escape me. I am also fully accepting of the fact that if I was blind tested in electronic / popular music, that I would fail to tell any differences at all between CD, SACD and propbably even the highest bitrate ogg. It's only in orchestral music that I can hear the difference and I have maintained this position all along.

I think some of the reason I may have little trouble with classical music is that in my life as a violinist, I became intimately acquainted with violin tone (almost obsesseed). To my ears, it takes 24-48 to get a violin to sound "right" and at 16-44 it sounds synthetic. Violins also sound "right" on vinyl, which is why I went to vinyl as my primary source.
post #74 of 161
violins are tough to reproduce........very tough

Isn't there a school of thought which says that recorded classical music is a tougher sell today than in the pre-digital past, because violins and other strings are so harsh and shrieky on standard CDs?

Wasn't SACD etc. supposed to help fix that?
post #75 of 161
I have heard that too. It doesn't surprise me. Of course having been a violinist I have known enough of them as well. They are very particular and picky people when it comes to sound and I don't blame them They stand quite uniquely in the world of musicians (and perhaps most other professions) in that the tools of their trade will regualry exceed the size of their mortgages. Many end up using instruments on loan because understandably they don't have the $200,000 - $4 million to pay for them.

I was brought up in the analogue era and when I studied the violin we referenced professional recordings on LP and open reel tape, plus of course live performances. The first CD I ever bought was one of Itzhak Perlman playing the Beethoven Spring Sonata. Whilst my jaw dropped at the silent noise floor and complete absence of wow and flutter, his violin did not sound anything like it was supposed to I then thought back to the recordings of youth, such as the late 1950s recordings of Oistrakh and Schneiderhan on DG. Now those sounded like violins. Of course, CD has come a long way, but still I hanker for the sound I know should come from stringed instruments but doesn't ever seem to do so.

Having been party to more live performances than I can remember, plus hearing my teacher play her Carcassi at short range, CD has never ever been able to come close to that sort of sound, but analogue does, whether it be vinyl or open reel.

I will make one concession to digital though in terms of string sound. Nowadays, most players use synthetic strings, and in my opinion these synthetic strings have done the same to string sound that digital has done to recording. So we get a double-wammy effect. In the old days, analogue got a helping hand in my view, because the players really only had three main choices - Eudoxa, Olive or Kaplan - all wound gut. There were others, but those were the main ones for non-Baroque players. Those older strings are warmer, have more complex overtones, respond differently and require a more subtle bowing technique. Listen to old recordings by players such as Menuhin and Szeryng and the differences the strings made were quite clear. Nowadays, Anne Fontanella is probably the best example of a living player who can emulate the sound signatures of the players of yesteryear.

What has surprised me, however, is the good results DAT can provide in reocrding violin sound. The closeness to the CD standard would suggest otherwise, however in my experience the result is more than the sum of the parts. I could live reasonably happily with DAT if it were more robust, but I guess my main use for it would have been LP archival.

Incidentally, I found a very good high resolution recording of the violin on the net. That company High Definition Tape Transfers has Edith Peinemann playing the Dvorak - it's an open reel tape transcribed to 24-96. I'm not convinced about the equalisation curve, but the sound on that comes much closer to the real thing than CD.

Anyway, as to the question as to whether SACD address the problem, it certainly helps alot with regards to whole sections (i.e the sound of the first and second violins as a group, the violas, cellos, etc), but I have not yet had an opportunity to hear a solo violinist on SACD yet. I would be very interested indeed to hear one.

I do have a two track 24-96 version of the Prokofiev second Concerto on Linn Records, but I must confess I don't feel it was much better than the redbook version, even though I could hear differences between them. I certainly have to agree that the influence of the delivery format comes further down the chain of importance than getting the mics and balance right to begin with (in other words, I would stil much rather listen to a near perfectly produced CD than a mediocre SACD, DVD-A or HDAD).
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Dedicated Source Components
Head-Fi.org › Forums › Equipment Forums › Dedicated Source Components › From CD to SACD how much of difference