Hvis du ikke kan høre forskel på Toslink og Coax ,må det enten være fordi din anlæg ikke er afslørende nok eller osse har ørene det ikke så godt mere

Ok måske der skulle have stået ,med samme nøjagtighed, istedet for hastighed. Læs evt nedenstående som er citeret fra The Lampizator`s hjemmeside ! Jeg tror sgu han har lidt styr på tingene !

Contrary to popular belief, the TOSLINK is inferior way of transmission. It has ADDITIONAL transmitter with own electronics and LED light source and corresponding received with light sensor. Its bandwidth is not fast enough to transmit the signal properly. The TOSLINK interface is NOT USING the direct laser !!! it is a cheap interface. It should be avoided whenever possible.

AT&T is bloody expensive and not popular even if very fast and accurate. The interface alone costs more than the whole transport.

The best way is working on the player to get the signal from the output balancing transformer and add a XLR female socket. The XLR cable can be made of CAT6 LAN cable with twisted pair connected to XLR jewellery of choice. This will transmit the signal in balanced mode isolated from earth and without any cable losses.

The SP/DIF

signal is a 0,5 V pp square wave. It is VERY FAST. it contains 44 kHZ of samples, each sample is a 16 bits, so it is 16 square impulses, and two channels alternating. So we send 2 x 16 = 32 squares per each sample. Plus some other information embedded like end of word, and clock and checksum. So we have roughly 35 squares times 44,1 thousand = 1,543 million pulses per second.
From Furrier's theorem we know that the square wave is an infinite sum of sine waves of odd harmonics, so if 1,5 meg is our first fundamental frequency, we  need at least 10 harmonics to be represented properly to make a square. This means that we need to send precisely the sine wave bundle which includes clean 1st, 3rd, 5th, 7th, 9th, 11th, 13th 15th and 17th and 19th harmonics
The 19th harmonics of the 1,5 meg is 30 megahertz. That is bloody fast, as fast as radio waves.
It means that there is a big demand on cable quality. What is cable quality? Simply speaking, in order NOT TO attenuate the fast signal the cable must have very low capacitance and inductance. So it must be twisted pair with physical separation of wires and good low loss dielectric. XLO way of twisting springs to mind. Twisted but quite far apart. Third wire - earth - is sent by a separate third wire or screen. The shorter cable the better. XLR made of CAT6 cable should be the best DIY option.

IS2 interface is an option which sounds superior on all equipment which has this option. It makes the data slower, the signal is split into three separate threads: actual data, clock and left-right word separation impulse signal. So we need 3 wires plus earth to send I2S but it is worth it. The SAA7220P/B chip has this format readily available, as well as all receiver chips. It is usually done by means of using computer 9 pin serial RS  socket and a computer LAN cable.

Til Kalle.dk

Billigt eller dyrt digitalkabel (Coax under 2 meter), gør ingen forskel efter min mening.......ved ikke med længere længder.

Ikke på mit anlæg til ca. 60K kunne jeg høre en sk.. til forskel.

Prøvet et til kr. 1.100,- for 1 meter og et til kr. 129,- for 1,8 meter.

Hvis du ikke kan høre forskel på Toslink og Coax ,må det enten være fordi din anlæg ikke er afslørende nok eller osse har ørene det ikke så godt mere

Ok måske der skulle have stået ,med samme nøjagtighed, istedet for hastighed. Læs evt nedenstående som er citeret fra The Lampizator`s hjemmeside ! Jeg tror sgu han har lidt styr på tingene !

Contrary to popular belief, the TOSLINK is inferior way of transmission. It has ADDITIONAL transmitter with own electronics and LED light source and corresponding received with light sensor. Its bandwidth is not fast enough to transmit the signal properly. The TOSLINK interface is NOT USING the direct laser !!! it is a cheap interface. It should be avoided whenever possible.

AT&T is bloody expensive and not popular even if very fast and accurate. The interface alone costs more than the whole transport.

The best way is working on the player to get the signal from the output balancing transformer and add a XLR female socket. The XLR cable can be made of CAT6 LAN cable with twisted pair connected to XLR jewellery of choice. This will transmit the signal in balanced mode isolated from earth and without any cable losses.

The SP/DIF

signal is a 0,5 V pp square wave. It is VERY FAST. it contains 44 kHZ of samples, each sample is a 16 bits, so it is 16 square impulses, and two channels alternating. So we send 2 x 16 = 32 squares per each sample. Plus some other information embedded like end of word, and clock and checksum. So we have roughly 35 squares times 44,1 thousand = 1,543 million pulses per second.
From Furrier's theorem we know that the square wave is an infinite sum of sine waves of odd harmonics, so if 1,5 meg is our first fundamental frequency, we  need at least 10 harmonics to be represented properly to make a square. This means that we need to send precisely the sine wave bundle which includes clean 1st, 3rd, 5th, 7th, 9th, 11th, 13th 15th and 17th and 19th harmonics
The 19th harmonics of the 1,5 meg is 30 megahertz. That is bloody fast, as fast as radio waves.
It means that there is a big demand on cable quality. What is cable quality? Simply speaking, in order NOT TO attenuate the fast signal the cable must have very low capacitance and inductance. So it must be twisted pair with physical separation of wires and good low loss dielectric. XLO way of twisting springs to mind. Twisted but quite far apart. Third wire - earth - is sent by a separate third wire or screen. The shorter cable the better. XLR made of CAT6 cable should be the best DIY option.

IS2 interface is an option which sounds superior on all equipment which has this option. It makes the data slower, the signal is split into three separate threads: actual data, clock and left-right word separation impulse signal. So we need 3 wires plus earth to send I2S but it is worth it. The SAA7220P/B chip has this format readily available, as well as all receiver chips. It is usually done by means of using computer 9 pin serial RS  socket and a computer LAN cable.

sshd skrev: Der er muligvis forskel på implementation i billige produkter og dem som kalder sig high-end uden at være det. Jeg har imidlertid lavet en test med to computere, TOSLINK -> TOSLINK. Men da de stod med 11 meters mellemrum måtte jeg bruge to kabler, et 10 meter plactickabel til 89 kr og et 1½ meter noname kabel købt for 150 kr for 12 år siden. Kablerne blev så loopet igennem en lige så gammel DAC fra Denon. Resultat: Samme bits blev modtaget som afsendt (bitperfekt). Jeg gentod testen tre gange. Så det kan sagtens laves ordentligt! Personligt foretrækker jeg TOSLINK, da det giver elektrisk adskillelse mellem signalkilde (computer) og pre-amp.

Der er tydeligvis forskel på bits !!

I was making a false assumption, that the CD mechanism picks pretty much all the data. Only SOMETIMES, occasionally when the laser beam hits a scratch or a blemish - the missing data is interpolated.
WRONG !
In computer world, all the data is always perfect, otherwise our bank statements and excel spreadsheets would go bananas. If data is missing on CD ROM - the computer recreates it perfectly from checksum, or if it can't - it displays error message. Or a blue screen of death for those unfortunate folks who did not yet switch to MAC.

The CD playback has no time, no computing power and no means (checksum) to recreate missing data. So it averages the missing sample by looking at the nearest good one. This is obviously not the real true information.
If it was just one sample every now and then, say one per minute, that would be OKAY. But what if the CD misses 30 or 50% of data?
It still plays, I mean if the interpolation works, there will be sound even with 90% of missing data. We need at least two healthy teeth to fix the dentures with 30 false ones.

So the sonic signature of a transport must be a combination of jitter, squareness AND PERCENTAGE OF INTERPOLATED DATA. That answers why Lens does not make all transports sound the same as it should well do.
A burned CD ROM with a text file is THE SAME whether it was created on a IBM, HP or Mac. It has NO HISTORY of the "quality".

That whole speculation, assuming it makes sense, and it does - to me at least, also answers why transports sound different depending on the table, the vibrations, the green pen trick, or the LED light addition. Or say - why different BRAND of CD-ROM blank media sound different. Why is Verbatim better than a noname.

My conclusion is: The amount of interpolation must be HUGE, it is a secret of the industry but they read only a certain percentage of the data, rest is ADDED by the solomon code CIRC chip

Jeg orker ikke oversætte

golfdane skrev: Meningen er vel også tydelig. Jeg har en Meat Loaf plade, som jeg fik af en onkel. Den ser ud som om den har set mere "battle action" end Meat Loaf selv, og den spiller nu så godt som Meat Loaf nu engang kan lyde i de drev jeg har haft den i. Men, alle forsøg på at rippe denne CD på en computer, har hidtil været fejlbehæftede, trods flittig brug af Slipol og knofedt. Kun hvis alle checks slås fra, kan jeg rippe de fleste tracks (der er 2 stædige tracks tilbage).

At du kan spille den i en almindelig cd-afspiller er nok fordi den ikke brokker sig ved læsefejl. Den kører bare videre, og gætter på de bits der mangler. Derfor hører du ikke deciderede fejllæsninger. Det lyder bare ikke korrekt.

golfdane skrev: Meningen er vel også tydelig. Jeg har en Meat Loaf plade, som jeg fik af en onkel. Den ser ud som om den har set mere "battle action" end Meat Loaf selv, og den spiller nu så godt som Meat Loaf nu engang kan lyde i de drev jeg har haft den i. Men, alle forsøg på at rippe denne CD på en computer, har hidtil været fejlbehæftede, trods flittig brug af Slipol og knofedt. Kun hvis alle checks slås fra, kan jeg rippe de fleste tracks (der er 2 stædige tracks tilbage).

At du kan spille den i en almindelig cd-afspiller er nok fordi den ikke brokker sig ved læsefejl. Den kører bare videre, og gætter på de bits der mangler. Derfor hører du ikke deciderede fejllæsninger. Det lyder bare ikke korrekt.

erlingt skrev: Full_blow skrev: golfdane skrev: Meningen er vel også tydelig. Jeg har en Meat Loaf plade, som jeg fik af en onkel. Den ser ud som om den har set mere "battle action" end Meat Loaf selv, og den spiller nu så godt som Meat Loaf nu engang kan lyde i de drev jeg har haft den i. Men, alle forsøg på at rippe denne CD på en computer, har hidtil været fejlbehæftede, trods flittig brug af Slipol og knofedt. Kun hvis alle checks slås fra, kan jeg rippe de fleste tracks (der er 2 stædige tracks tilbage). At du kan spille den i en almindelig cd-afspiller er nok fordi den ikke brokker sig ved læsefejl. Den kører bare videre, og gætter på de bits der mangler. Derfor hører du ikke deciderede fejllæsninger. Det lyder bare ikke korrekt. Helt præcist - det er lige det faktum, som "Gonz" lange engelske citat påpeger. En CD afspiller gætter hele tiden løbende på manglende bits, da den jo afspiller musikken i real time; hvorimod en computer har tiden og processorkragten til at rekonstruere alle bits helt perfekt efter de checksummer, som også findes på CD'en. Derfor oplever vi at CD afspillere og medier lyder forskelligt, der er simpelthen forkel på hvor mange "gæt" der løbende laves, og hvor godt disse interpolationer udregnes. Det spændende tankespin herfra er jo, at så må en computer afspilning af en bit korrekt lossless ripning jo faktisk være den perfekte afspilning. Idet man i denne forbindelse har buffer og computerkraft til at sikre, at det er de korrekte bits, der videresendes til DAC'en. En afspilning via en computer, squeezebox, Sonos box m. fl. er den helt korrekt datamæssigt, hvorimod afspilning via en CD afspller er gætværk og interpolation

Ja det ville jo være super, hvis der nogle der har et setup, hvor denne løsning bare lyder så meget bedre end afspilning via en cd afspiller, af god kvalitet ! Jeg har i omgangskredsen lyttet til flere forskellige kombinationer og den bedste lyd via pc indtil videre har været via Firewire og en RME Firewire DAC ! Det lød på niveau med analog afspilning via Copland CDA 289 ! Til gengæld kunne det slet ikke hamle op med en Accuphase DP-67 ! Har osse prøvet med Stello DA220 mkII som osse lyder ok !

Hvis der skulle være nogle derude på forum, som har et klart bud på en væsentlig lydforbedring via PC uden at det koster ½ million ville det være super, hvis i ville dele lidt info med os andre !

Netop købt en Music Hall Dac 25.2 koster kr. 4.490,- - spiller fremragende med super stram bas/mellemtone - altså en rigtig god dynamik. Med USB selvf.

Contrary to popular belief, the TOSLINK is inferior way of transmission. It has ADDITIONAL transmitter with own electronics and LED light source and corresponding received with light sensor. Its bandwidth is not fast enough to transmit the signal properly. The TOSLINK interface is NOT USING the direct laser !!! it is a cheap interface. It should be avoided whenever possible.