sexta-feira, 27 de novembro de 2009

LP versus CD. E o resumo deste Blog.

Trata-se de trecho um texto da Doutora Christine Tham. O texto integral pode ser lido em:


VINIL = Espectro de freqüências de 7hz até o padrão de 48 kHz, isto na gravação; porém, o vinil por conta própria, ultrapassa em transientes indo até 100 kHz ou mais. E seu sinal não é fragmentado pela numerização de 0 e 1 e pelos algoritmos que não conseguem criar transientes num CD.

SACD, HD AAC, XRCD, SIGMA DELTA EM AUDIO, AAC, DVD-A e esse monte mídias numerizadas = Elas levam um corte em 48 KHz. E pronto, não se fala mais nisso. E a numerização (Vulgo digitalização), faz com que um "coralista" (Coral de Sinfônica ou só Coral), um coralista se pareça em voz com o outro.

CD = Esse é o primo pobre que leva um corte em 2.050 Khz e não se fala mais em nada. Tudo que passaria disso, é jogado no lixo.

Nota: Falam que não escutamos acima de 15 Khz... Pois bem: As freqüências "tem que subir" além dos 2.050 Khz do CD; e 48 kHz das mídias-de-software (DVD-A, SACD, XRCD...) e em transientes LIVRES para formarem os harmônicos pares que irão compor o encorpamento musical. Não se os ouve, mas eles são imprescindíveis na formação da cadeia dos Harmônicos de Fourier, Formantes de instrumentos e timbre de vozes. 


"Spectral and Dynamics Comparisons of LP's versus digital formats (Part 4 of an on-going series of comparisons).Yes, you have heard all the arguments before, and you are probably sick and tired of it. LP versus digital is a "religious war" that has been played out by various audiophiles ever since the CD format was introduced in the early 1980's."Vinylphiles" claim that CD's do not sound as good as LP's, period. CD's appear to sound "harsh", "unlistenable", "lacking in dynamics", plus a myriad of other "faults." Some vinylphiles even extend this to all digital formats, including the new Super Audio CD and DVD-Audio high resolution formats, whereas others believe the higher resolution formats either equal or at least get closer to the "superior" sound of LP. "Digiphiles" on the other hand laugh at LP's pitiful dynamic range, surface noise, pop and crackle, harmonic distortion, and various other limitations to do with the ability of either the cutting head to master difficult signals onto disc, and the ability of stylii to track them without "jumping." So, are there any evidence to support these claims? Can both parties be right? I was interested to find out if there are any objective evidence that I can gather using my sound card (Audiotrak Prodigy 7.1) and my very old copy of Cool Edit. I am particularly intrigued by vinylphile claims that LPs mastered from a digital recording sound better than that exact same digital recording on CD. Plus, recording an LP onto CD yields most of the benefits of the original LP, and still sound better than the commercially pressed CD.The ApproachI took a few albums from my personal collection that I have on both LP and at least one digital format (CD, SACD, or DVD-A).I selected the following songs from the following albums: Main Titles from the original motion picture soundtrack to Chariots of Fire (Vangelis) o LP (Polydor 2383 602) 1981 Australian PolyGram pressing, purchased second hand o CD (Polydor 800 020-2) 1984 Polyram made in Germany, purchased new, Mick's Blessings from Café Bleu (The Style Council) o LP (Polydor 817 535-1) 1984 Australian PolyGram pressing, purchased new o Original CD release (Polydor 817 535-2) 1984 PolyGram made in Germany, purchased new o "Digitally Remastered" CD re-release (Polydor 557 915-2) 2000 Universal made in the EU, purchased new. What's New from What's New (Linda Ronstadt and The Nelson Riddle Orchestra) o LP (Asylum 9 60260) 1983 USA WEA pressing, purchased second hand o DVD-Audio (Elektra/Asylum/Rhino 8122-78341-9 begin_of_the_skype_highlighting            8122-78341-9      end_of_the_skype_highlighting) 2002 Warner Strategic Marketing made in Germany, purchased new; Genesis Ch.1.V.32 from I Robot (The Alan Parsons Project); LP (Arista Code 304 AL.7002) 1977; Australian EMI pressing, purchased second hand o DVD-Audio (Classic HDAD 2003) 2004 Classic USA release, purchased new; More Than This from Avalon (Roxy Music); LP (EG 2311 154) 1982 Australian PolyGram pressing, purchased second hand o Hybrid SACD (Virgin ROXYSACD 9) 2003 Virgin made in Holland, purchased new Notice that most of the LPs are second hand. LPs deteriorate with every play, so this comparison is potentially biased against LPs.What I did was cleaned all the LPs using a VPI HW16.5 vacuum based record cleaning machine (using VPI record cleaning fluid, VPI brush and a domestic lint brush - don't ask why, but it works!) to ensure that I minimize surface noise and crackle, and take out any dirt and grime from the second hand LPs.
I recorded all the tracks on the LP's using my system (Dynavector DV-20xL cartridge, Rega P3 turntable, Dynavector P-75 phono stage) via the analog outputs of my Denon AVC-A1SE+ amplifier. I used my HTPC and n-Track Studio to do the recordings in stereo at 96 kHz sampling rate and 24-bit resolution.For the CD's, I ripped the actual digital information (at 44.1kHz/16-bit resolution) directly from the discs using Exact Audio Copy. In some cases, I also recorded them as reproduced by my Sony SCD-XA777ES player via the analog outputs of my amp. I recorded SACD and DVD-A tracks using the analog outputs (via my amp) of the Sony SCD-XA777ES and Panasonic DVD-RP82 players respectively. Linear PCM 96kHz/24-bit tracks encoded as DVD-Video content were recorded using the analog outputs (via my amp) Panasonic DVD-RP82, but also digitally ripped from the disc using DVD Decrypter. All wave files were then analysed using Cool Edit Pro. The results are ... interesting ... Main Titles from Chariots of Fire Let's look at some statistics first, comparing: The digitally ripped copy of the track (using Exact Audio Copy); the analog recording of the SCD-XA777ES playing the track; the recording of the LP The following statistics were obtained using the "Analyze Statistics" function of Cool Edit (which produced separate results for Left and Right channels, which were then averaged): Each CD and LP.

The CD and DSD recordings have such similar values it's scary (apart from slightly higher peaks for the DSD recording). I suspect the CD layer is a downconversion of the DSD master. Notice once again the relative dynamics superiority of LP over the digital formats.Looking at the digital rip of the CD reveals a typical over-compressed and heavily manipulated pop recording: Which the recording of the CD faithfully captures: And of course the DSD version as well: The LP recording shows that the compression is intentional and present in the original master: The spectral view of the recording of the CD layer: Interestingly, the DSD spectral view shows extended frequencies: Unfortunately, though, not on the LP: I suspect the LP is derived from a low resolution PCM digital master. ConclusionsIt appears that the vinylphile claims are not as outrageous as they seem: LP's do have a usable dynamic range far greater than the measured dynamic range would suggest, and LP's consistently have higher relative dynamics over digital formats. But it is also true that LPs have higher distortion levels which translate to ultrasonic frequency harmonics. The question is: is the higher relative dynamics of LP's an indication of higher accuracy, or are LP's exaggerating transients and dynamics? I'm not sure, and I would welcome comments. If LP's have higher distortion and are exaggerating dynamics, it may explain why the apparent "benefits" of LP's translate even into LP recordings, and potentially explain why LP's of digital recordings sound better than their CD equivalents.
© 2004. Christine Tham.

Tendo lido o texto todo em inglês da supracitada autora, passo então a escrever aqui o que entendi sobre seu tão primoroso e detalhado teste. (Em construção, aguarde...)

E-mail para