Cookie persondata politik for GAFFA A/S
Log ind eller opret en gratis konto
Vis trådoversigt
  • 11-10-2001
    kl. 17:04

    Christian From

    24/96 vs. 24/48???

    Hvad er det der afgr hvor stort dynamisk headroom man har- er det bndbredden(16 el.24 bit),eller er det samplingsfrekvensen
    (44.1,48 el.96khz).Med andre ord;hvad har strst betydning for lydkvaliteten?
    Synes lige jeg ville hre...

    Svar på dette indlæg
  • 15-10-2001
    kl. 10:54

    Frederik Dam Sunne

    mit bud

    Hej Christian,

    Hvis du sidder med et lydkort kan du liges godt indspille i 16-bit. Sagen er at lydkort somregel har et signal stj forhold som er vrre end -80dB, derfor vil de ekstra 8-bit, ved optagelse af 24-bit, hjst sandsynlig indeholde hvid stj. mht. 48Khz kontra 44.1 Khz vil jeg nok vlge 44.1, da du jo senere skal sampleratekonvertere det frdige miks ned til 44.1 kHz, hvilket ikke er gavnligt for signalet. Personligt indspiller jeg altid i 24-bit/44.1 Khz, da jeg har en ekstern konverter som har et signal stjforhold p -121dB - derfor giver det mening at gemme de ekstra 8-bit.

    Inde under "Undervisning" -> "Musik p computer" finder du en artikel, som hedder "Bagom Digital Lyd", som mske ville vre interessant for dig at lse...

    Venlig hilsen,


    Svar på dette indlæg
  • 15-10-2001
    kl. 15:40

    Jesper Larsen

    16 vs. 24 bit - the Bit war

    En coverstory fra Keyboard magasin (USA) havde bit wars som tema artikel.
    Find June 98 p biblioteket.

    Her er lidt fra

    We've all seen lab tests comparing the sound quality of 16-, 20-, and 24-bit converters, in which "golden-eared" listeners give us the lowdown on the differences they hear between the various bit depths. You may even have come across a test or two comparing various sample rates. While highly controlled lab tests are always a great thing, and it's nice to know what a golden-eared engineer can hear, the question always remains: Are the differences audible in the real world? This is the question that Keyboard magazine set out to answer.
    While our tests were controlled, our intention was never to produce a "lab-approved" body of data. Rather, we wanted to make some real-world recordings, in real-world situations, then compare the results when heard by a variety of listeners over a number of different sound systems. Most importantly, we wanted to test whether or not the differences were still audible when a final 16-bit/44.1kHz CD had been produced from the tracks.
    Here's the approach we used: We brought several acoustic musicians (two different acoustic guitarists, a violinist, and female vocalist) into Brilliant Studios in San Francisco. We recorded them using Neumann U87 and AKG C460 microphones through A Range mic preamps, directly into a Sonic Solutions system using dB Technologies dB 44-96 converters at 16-bit/ 44.1kHz, 24-bit/44.1kHz, and 24-bit/88.2kHz (96kHz sample rates require two sample-rate-conversions in Sonic Solutions to bring them to 44.1kHz; we opted to use 88.2kHz which requires only one conversion). Once the tracks had been recorded, we converted some of them to 16-bit/44.1kHz, and burned them to an audio CD. Would our listeners still be able to hear any differences after the final CD had been produced?
    The listening tests consisted of two parts: The first took place during the tracking session. The listeners were played the raw tracks, at their original sample rates and resolutions, directly from the Sonic Solutions system, through dB Technologies D/A converters and into Meyer HD-1 monitors. The results varied. Some listeners had difficulty in discerning any difference between 16 and 24 bits of resolution. In fact, when the same track was repeated, listeners would frequently change their minds.
    The difference between 44.1kHz and 88.2kHz was more apparent. Several of us could immediately hear the difference when the higher sample rate was played.
    One interesting point: Several of the listeners preferred the sound of the 16-bit/44.1kHz tracks. Their contention was that the lower resolution tracks were warmer and smoother, and had a presence that was not heard in the higher resolution tracks. They felt that the higher resolution tracks had a thinner, more brittle sound. Other listeners disagreed, pointing out the more open and detailed sound of the higher resolution.
    After the first listening sessions were completed, we converted the 24-bit/44.1kHz tracks to 16-bit/44.1kHz for burning to CD. As it turned out, we were unable to convert the 24-bit/ 88.2kHz file to 16-bit/44.1kHz resolution (the Sonic Solutions conversion algorithm for this resolution had not been completed at the time, and no hardware solution was available).
    We then took the 16-bit/44.1kHz version and the converted 24-bit/44.1kHz version into Keyboard's studio and chopped them up into identical 30-second segments using Digidesign Pro Tools. CDs consisting of 13 of these segments and the two original full-length tracks were created. These CDs were then distributed to the listeners along with an instruction sheet/questionnaire.
    The results: While there were audible differences between the segments originally tracked at different resolutions, it was not always apparent to the listeners which version they were hearing. When reviewing the questionnaires, we discovered that some listeners were marking the segments they felt "sounded" better as the 24-bit/44.1kHz tracks. In fact, the segments they subjectively described as "sounding better" were often those that were tracked at 16-bit/44.1khz.
    Another interesting fact: Some listeners had great difficulty discerning the resolution of the 30-second segments, but no problem picking out the resolution when the full-length tracks were played. For other listeners the opposite was true.
    The differences heard also varied depending on the monitoring systems used by the listeners. In Keyboard's main studio, tracks were played from a Pro Tools system with an 888/24 I/O (playing 16-bit files) through Event 20/20 powered monitors. Listeners had no difficulty in hearing a difference. The same was also true when we used Genelec and Dynaudio Acoustics monitors. With home stereo systems, over various headphones, in cars, and in other environments, the results were less conclusive.
    The general comments: As in the listening tests conducted at the tracking sessions, most people felt that the segments originally tracked at 24-bit were more open, spacious, and detailed than those tracked at 16-bit. There was disagreement on whether or not this was always a good thing. Some listeners felt that the higher resolution revealed as many negatives (performance glitches, microphone artifacts, etc.) as it did positives. And, as in the first listening tests, some felt that 16 bits simply sounded better. Others disagreed, preferring the extra detail and openness captured by the higher resolution.
    With the thirteen 30-second segments, listeners were able to discern the correct resolution about 65% of the time. With the two full-length tracks, listeners correctly discerned the resolution about 43% of the time. It should be noted that virtually all the listeners could detect differences between the segments and resolutions. They were less successful at deciding which resolution they were actually hearing.
    What does this all mean? First, based on our tests, there is an audible difference between tracks recorded at different resolutions and sample rates -- even after they have been converted to 16-bit/44.1kHz format. But listeners couldn't agree about whether or not higher resolutions sound better.
    The comment that was made several times during the tests was, "24-bit definitely adds something, but I'm not sure I'd want or need it for my style of music." In particular, those involved with dance and electronica were unimpressed with higher resolutions. Those whose main concern was acoustic recording felt there were significant benefits to using a higher resolution. "For acoustic music I'd definitely want the extra resolution 24-bit provides." All of the listeners agreed that higher resolutions could be put to good use during mastering or critical live recording applications.
    [Editor's Note -- These tests would not have been possible without the untiring support and assistance of Rolf Hartley and the staff at Sonic Solutions. Dave Van Hoy and Randall Sanderson at Advanced Systems Group of Oakland, California, went far beyond the call of duty to make these tests a success. Without the assistance, generosity, and expertise of our Sonic Solutions operators, Dave Tinsley and Margaret Schock of San Rafael Studios in San Rafael, California, these tests simply would not have happened. Thanks are due dB Technologies for providing converters, and to Jim Andrews and Brilliant Studios in San Francisco for booking us in on extremely short notice.
    Special thanks are also owed our fine musicians: Becky Lloyd, Joe Paquin, and Tom Rigney, all of whom turned in outstanding performances. And let's not forget our panel of listeners: Marvin Sanders, Mark Vail, Greg Rule, Mitch Gallagher (Keyboard editors), John Krogh (Music and Computers editor), Andy Ellis (Guitar Player editor), Jim Andrews (Brilliant Studios), and Billy Gould (Faith No More).]

    Svar på dette indlæg

Skriv svar

Log ind eller opret en gratis konto for at tilføje indlæ til forumet.

Om Soundcheck | Annoncer | Kontakt | Jeg er lavet af og