I know this is a little off the topic from the original thread....

What are everybody's opinions in regards to DVD-A and SACD??

I needed to replace my old Sony DVD player and debated on what to purchase. Part of the debate was deciding between DVD-A and SACD. I'm kind of anti-Sony, and at the time it seemed that there were more DVD-A titles available, so I went with a DVD-A compatible player...the Denon DVM 4800.

This is my first changer, and I'm not sure I like it.

From lurking around here a bit, it seems that SACD is a little more popular with the members here. Maybe I should go ahead and get rid of the Denon and go with one of the new Pioneer's???

What are some of your opinions between these two formats??

fly guy

Personally, I think SACD and DVD-A are not worth the expenditure for the upgrade. I've read many reviews from audiophiles, and many think that currently, there is little to be gained by these formats. This is an interesting link on the subject.

http://www.audioholics.com/techtips/specsformats/dvd_sacd1a.php

Others believe the wheel has been reinvented. The best sound I've heard to date has come from very expensive 2 channel speakers driven with a quality amplifier and a quality CD player. Namely, Thiel or wilson speakers driven with a Levison amp at a local store.

bp

You might want to wait for awhile for a 2nd generation combination DVD/DVD-A/SACD player. Pioneer makes one now, but personally, I'm waiting.

In my own opinion (Bosso, please step out of the room for a bit...thanks) is that the main draw to either of these formats is the ability to play multichannel material. This can be either awsome or so-so of an experience depending on the mix of the material you are listening to - it varies. For classical, jazz, and some other acoustic music, in all honesty, I can get pretty much the same effect by feeding my surrounds with a studio reverb fed off the stereo fronts. Believe it or not, this is exactly how it is done on some movies with the music track: surround channels are sometimes left up to the mixing engineers because they can integrate it better with the other sounds in the track. With other types of music, you are at the mercy of the taste of the person mixing the tracks. Personally, I don't like guitars coming from behind me - but sometimes it gets done extremely well.

As far as the supposed increased resolution of 24 bit and 96K, well, I don't think you would be able to really tell if it were well done in 16 bit and 44.1k. The differences are _extremely_ small, to inaudible. There is no such thing as a 'real 24 bit' analog to digital converter; 144db of dynamic range is just beyond the capability of any practical electronic circuit in use today. The most that can be reliably captured is about 18-20 bits - the lower bits are just recording electronic noise. Also, to put a digital to analog converter in a consumer product with _actual_ resolution of 24 bits would cost, well, more than you could imagine, if it could be done at all.

The increased sample rate? well, that too, in my opinion is pretty much wasted. I sure as hell can't hear as high as 46,000 hZ, can you? Can you speakers reproduce that high? Can your power amps pass that high a frequency?

In a mastering situation, it can certainly make sense to use 24 bit resolution, but only because most of the internal math done by the DSPs is done at that bit depth, and multiple DSP operations would pile up rounding errors if the word length were reduced for every operation.

Personally, I think they should have used the 50,000hZ sampling rate used by the original Soundstream digital recorder as a standard, and called it a day; 44.1 is cutting it a bit close to the audible band.

Bosso - you can come back into the room now

[This message has been edited by soundhound (edited October 27, 2002).]

i chose sacd because i believe sony will win this one. at issue in the format war is the choice between sony/phillips' dsd (i believe it stands for digital stream data) and hi-res pcm (pulse code modulation).

dsd is a 1 bit system. the most glaring advantage is that dsd can be converted to ANY digital format. indeed, current sacd players convert to pcm @ 24/96 multichannel and 24/192 stereo, making the 2 formats identical.

currently, there are studios around the world with priceless master analog tapes that are deteriorating to the point of absolute ruin. they are being and will all be converted to a digital format if only to save them from extinction.

now...if you own a master and you want to archive it digitally, if you choose pcm at the current highest res available, once you do that, it cannot ever be re-converted to a higher res when that option comes along in the future.

if you choose the dsd 1 bit system to archive, it can be re-converted to any digital scheme at any time in the future. sony's engineers (and these are the guys who have been on the cutting edge of recording hardware for more than 30 years) are so convinced that this is the smart way to proceed (and time is of the essence do to the 'once the tape is gone, it's gone' situation), they have convinced sony/phillips to donate the expensive equipment (with strings attached i'm sure) to studios for archiving purposes.

dsd is actually an incredibly simple format. i like that. less manipulation=better sound. always has, always will. i understand also, that all that is needed to convert dsd is a high quality, high pass (like about 70khz) filter.

as far as what a human can hear (more like perceive), sony/phillips conducted extensive double blind tests. short story...you'll know the difference.

on the subject of what resolution ends up at your ear, currently, soundhound is correct. i love multichannel for this obvious and much overlooked reason: (though the soundstage and effects options being endless as a palate is also appealing to any artist from a purely artistic standpoint).

for many years you have had only 2 channels into which to 'place' all of the information contained in your musical piece. saturation of those two channels available headroom caused the invention of compressors, limiters and the like. they work, but they do as their names imply. they compress and they limit.

you now have 6 discrete channels into which to 'place' your information. each channel now has much less to handle, with more headroom than ever before. the simple bottom line is astoundingly better clarity and, with proper speaker placement, a soundstage you have to be 'surrounded' by to appreciate.

btw, the audioholics articles sited above never call into question anything i've just said. they only point out the problems of lack of standardization and limits due to greed and piracy issues. i choose to ignore the downside and look for the few mixes by producers who get it and artists who work with them.

borrow a player, pay $20 bucks for spyro gyra's 'in modern times' multichannel sacd, tweak your system's levels, placement and eq and LISTEN. if you don't hear a quantum leap forward in quality over stereo cd, let me know. short of doing that...it's all just academic discussion.

soundhound...you should never have brought that soapbox out...it's addictive. sorry all for the long-winded rant.

"tweak your system's levels, placement and eq and LISTEN."

And for me, therein lies the problem. My surrounds are fixed regards to distance. All this tweaking should be done in the digital domain, IMO. Until that's available SACD and DVD Audio are non starters for me.

As an aside, we already had multi-channel delivery systems (DD 5.1 and DTS). Why didn't these take off as audio only media, dispensing with the need for more formats?

Yeah, me too. But I'm also enjoying some DVD-V compatible tracks on DVD-Audio disks.



That's pretty much my experience too. My opinion, based on observation and some technical knowledge, is that implementing a D->A->D playback system that has minimal distortion below 20kHz and a 22kHz Nyquist frequency is a tough job. I would have rather seen a 50-60 kHz sample rate, not because it would be a better system, but simply because it would be easier to implement the system to uniformly high quality levels.

I firmly believe that CD got off to a bad start simply due to poor implementations [which may still be happening] and in reality can be a perfectly fine sounding format.

Indeed - the first CD players made by Philips, and maybe some others OEMd by them actually played back in _14 bit_ because a good 16 bit converter couldn't be made at a realistic price point. And the 'truncation' wasn't done gracefully on these players, to put it mildly! The 'brick wall filter' at 20khZ was implemented by a big, honking, ugly _analog_ low pass filter that had god-knows-how-much phase shift. Record companies just grabbed the "EQ master" which was equalized for LP cutting and transferred them to CD. Of course these sounded like crap. Microphones had a big peak of around 5-10 db at around 6-10khZ, again to make the sound register on LPs. Mix enough of these into stereo, and you get some pretty 'glarey', 'hard' and 'brittle' sound. This was covered up by LPs, but came through all too well on CD. Techniques now are much more natural.

You guys never cease to amaze me. I can come here and read your posts and learn more than I can spending hours combing the web or reading other forums. Your posts are not regurgitations of magazine articles, they are thoughts and insights from those "in the industry". Please continue to post here. I benefit greatly from your knowledge and intelligent opinions.

Bill Polley

Bosso: I know of no reason why PCM cannot be converted to other encoding formats as DSD can. Could you elaborate on that point?

Additionally, if you can mathematically convert DSD to PCM without a loss of fidelity, then what does that say about the claimed superiority of DSD over PCM?

I'm not claiming that PCM is "better" than DSD, I'm simply looking for some more information about the relative differences of the two. My personal gut feeling would be that at the end of the day, it doesn't really make much difference which encoding format was used.

------------------
Matthew J. Hill
matt@idsi.net

In general, I think it's always better to minimize *any* amount of conversions to an audio/video signal. There's always the chance that signal quality will degrade. (I still even have my doubts as to how well upsampling say, 16/44.1 LPCM to 24/192 can be done. "Upsampling" isn't the problem, it's the "re-sampling" from 44.1 to 48 or from 177.6 to 192 kHz where the problem comes in, for example.)

And especially when any type of PCM format is involved, where jitter *does* come into play.