adhoc 0 Report post Posted June 27, 2003 (edited) this thread is completely based on my understanding of how a cd works and how a cd is manufactured. while i make no claim to my current understanding being the be-all-and-end-all of cd-related knowledge, i am confident it is more (even far more) than many/some people here. PLEASE NOTE THAT THIS IS A 'WORK IN PROGRESS' type thread. [i will update it and correct any errors as i find them. it would help if you all could PM me any mistakes you might find.] some basic background info first. ERROR CORRECTION IN CDS there ARE error correction codes on any CD-DA (AUDIO cd). however, there are MORE error correction codes in a CD-ROM (DATA cd). this is because while errors may produce acceptable 'pop' 'click' or 'fuzz' sounds in the music track from an CD-DA, it is completely unacceptable for a DATA cd (eg compromised structure pattern of a compressed file = total and completely corrupted data). to compensate for this increased vulnerability CD-ROMS have additional ECCs (error correction codes) and EDCs (error detection codes). all these codes manifest themselves in the form of additional bits of data (called binary parity bits) that correct for SINGLE data bit errors (eg when a single 1 becomes a 0 or vice versa). i will now show you a VERY VERY SIMPLIFIED example of an error check (called a binary parity check) let's consider this string of TEN characters. there are 6 '1's and 4 '0's (4+6=10). 1010101011 we will add a single bit to the end (it's called a parity bit) that helps us detect errors. if an ODD number of '1's are present in the original TEN character string the parity bit is a '1'. if an EVEN number of '1's are present it is a '0'. an even number of '1's (6) are present here in the original string, thus the parity bit is '0'. here is the same string, but now with error detection capability. 1010101011 0 now let's introduce an error. let's change the first '1' to a '0'. hence now the TOTAL number of '1's is no longer EVEN (6), but ODD (5). 0010101011 0 once your drive reads the parity bit after reading the ten character string and realises that the supposed number of '1's according to the parity bit (an even amount) does not match the number read (an odd amount), it will re-read. this is probably the most basic form of error detection (AND correction, since it makes the drive re-read till it gets it right). real world error detection/correction is MUCH MORE COMPLEX. it involves 'staggered' parity bits and more than one parity bits. now i will show an example of code containing staggered parity bits. each parity bit here will correspond to the previous 10 characters and is '1' or '0' depending on 'odd' or 'even' number of '1's (like our example above). remember the parity bit is ALWAYS the LAST CHARACTER on EACH LINE [except the 6th one]. 1110001110 0 --------01110 0 1110 0 ---------------- 0 1110 0 1110 0 _------------------------ 0 1110 0 1110 0 __-------------------------------- 0 1110 0 1110 0 1110001110 0 1110 0 1110 0 1110 0 1110 0 as you can see, every parity bit is staggered 5 characters (ie 5 characters before the next block of 10 for a parity bit to check begin). this ensures that each parity bit after the first can not only check it's own 10 characters, but can also check 6 characters from the previous block of 10 characters. this system of redundancy will ensure greater reliability and accuracy when reading your info. what's the last line i hear you ask? it's the COLLAPSED form of the entire staggered parity bit structure; the top 5 lines are in reality derived from this one line - this last line is EXACTLY what your player reads. simple? thought so. in my next post i'll go over another form of error detection, or more accurately, error RECOVERY. Edited July 2, 2003 by adhoc Share this post Link to post Share on other sites
adhoc 0 Report post Posted June 27, 2003 (edited) my next section, another BASIC form of error detection, or more accurately, error RECOVERY. INTERLEAVING let's say you have this information stored on your disc. ITSUCKSTOSTUDYINSINGAPORE [it's "it sucks to study in singapore" for all you mentally challenged out there.. ] oops! you scratched your disc! you've lost the first 4 alphabets of your text string! CKSTOSTUDYINSINGAPORE if that was mission critical info, you're screwed. there's no way you can guess what those first 4 characters were. your text block could be ITROCKSTOSTUDYINSINGAPORE, or even CLOCKSTOSTUDYINSINGAPORE for all you know right? :enter interleaving: interleaving is basically scrambling up your information so that any loss does not occur sequentially but is spread out all over the place. kinda like getting hit with a plank hurts less than being stabbed by a knife even though both are thrown at you with the same force; the force (damage) is spread out. here's the original text again: ITSUCKSTOSTUDYINSINGAPORE after interleaving: UTNOERPINSUSIGADYISTCSKOT oops you scratched your disc! (again - ) you've lost the first 4 characters of text! ERPINSUSIGADYISTCSKOT after decoding: ITSCKSTOSUDYINSIGAPRE much easier to decipher right? you can guess what words went missing easily now. and that about sums up the power of interleaving. now on to more hardcore stuff - ERROR PROTECTIONS AVAILABLE IN CD-DAs.. Edited July 1, 2003 by adhoc Share this post Link to post Share on other sites
adhoc 0 Report post Posted June 27, 2003 (edited) ERROR PROTECTIONS AVAILABLE IN CD-DAs ALL CD-DAs have THREE form of error correction - C1, C2 & CIRC (Cross Interleaved-Read-Solomon-Code). an example of encountering C1 and C2 error correction in your daily life would be when one uses the CD ripping program Exact Audio Copy (EAC). EAC puts heavy emphasis on the use of error detection and correction so that end users may rip PERFECT music tracks off their CDs. to quote from the EAC website, www.exactaudiocopy.de : On all CD-ROM media are at least two levels of error correction, called C1 and C2. If both fail, the output is probably not correct anymore. Most drives are not able to report if audio reads failed or not, so each block had to be read twice and be compared to make sure that everything is fine. But some newer drives are able to report if C1/C2 failed on specific samples on a read, making it possible to read only once and see if a read error occured. But there is still a problem, as some drives do not report these errors correctly, so you should test it thoroughly before trusting the results. but what if you DONT have a newer drive with what is known as 'C2' capability? what EAC then does is read and re-read and re-read (user can specify how many times it reads) a sector of info and then compares all the read versions for any differences. if there is a difference then it reports an error. hence ripping with a 'C2' enabled drive is faster as time is saved in error detection (as a sector only has to be read once to determine whether it has errors with 'C2' vs read many times and compared w/o 'C2'). IIRC almost all standalone CD-DA players have no 'C2' capability. it is safe to assume your player does not have this capability unless they specifically specify its presence. an example of a standalone player that DOES have 'C2' capability would be the NAD 541i. but then why does audio from my standalone player NOT sound as though it has errors in it when i playback the same cds? to quote again from the EAC website, Standalone CD players perform oversampling and some more error correction. Further, if the error is too big to be corrected, it will perform "error hiding". That means that the player will hide the error in a way that it is not audible to the common listener. These additional techniques are not implemented in the CD-ROM drives, thus the uncorrected data is given back. CIRC is a very very very complicated affair, and is best explained by professionals, here's a summary though. you deserve a medal if you completely understand it. even i [ ] dont. it's copied wholesale from a site i saved to my comp. The original musical signal is a waveform in time. A sample of this waveform in time is taken and "digitized" into two 16-bit words, one for the left channel and one for the right channel. For example, a single sample of the musical signal might look like: L1 = 0111 0000 1010 1000 R1 = 1100 0111 1010 1000 Six samples (six of the left and six of the right for a total of twelve) are taken to form a frame. L1 R1 L2 R2 L3 R3 L4 R4 L5 R5 L6 R6 The frame is then encoded in the form of 8-bit words. Each 16-bit audio signal turns into two 8-bit words. L1 LI R1 R1 L2 L2 R2 R2 L3 L3 R3 R3 L4 L4 R4 R4 L5 L5 R5 R5 L6 L6 R6 R6 This gives a grand total of 24 8-bit words. This is column two on the IEC 908 table. The even words are then delayed by two blocks and the resulting "word" scrambled. This delay and scramble is the first part of the interleaving process. The resulting 24 byte word (remember, it has an included two block delay -- so some symbols in this word are from blocks two blocks behind) has 4 bytes of parity added. This particular parity is called "Q" parity. Parity errors found in this part of the algorithm are called C1 errors. More on the Q parity later. Now, the resulting 24 + 4Q = 28 bytes word is interleaved. Each of the 28 bytes is delayed by a different period. Each period is an integral multiple of 4 blocks. So the first byte might be delayed by 4 blocks, the second by 8 blocks, the third by 12 blocks and so on. The interleaving spreads the word over a total of 28 x 4 = 112 blocks. The resulting 28 byte words are again subjected to a parity operation. This generates four more parity bytes called P bytes which are placed at the end of the 28 bit data word. The word is now a total of 28 + 4 = 32 bytes long. Parity errors found in this part of the algorithm are called C2 errors. More on the P parity later too. Finally, the another odd-even delay is performed -- but this time by just a single block. Both the P and Q parity bits are inverted (turning the "1s" into "0s") to assist data readout during muting. An 8-bit subcode is then added to the front end of the word. The subcode specifies such things as the total number of selections on the disk, their length, and so on. More on this later. Next the data words are converted to EFM format. EFM means Eight to Fourteen Modulation and is an incredibly clever way of reducing errors. The idea is to minimize the number of 0 to 1 and 1-0 transitions -- thus avoiding small pits. In EFM only those combinations of bits are used in which more than two but less than 10 zeros appear continuously. For example, a digital 10 given as a binary 0000 1010 is an EFM 1001 0001 0000 00 Each frame finally has a 24-bit synchronization word attached to the very front end -- (just for completeness the word is (100000000001000000000010) and each group of 14 symbols is then coupled by three merge bits. These merge bits are chosen to meet two goals: 1. No adjacent 1's from neighboring EFM encoded words Remember that there are lots of EFM words which end in "1" -- as one example, all the eight-bit binary words from 128 to 152 end in "1". Similarly, there are EFM words that start in "1". Thus, it is relatively straightforward to have to have adjacent EFM words that create adjacent "1s". For example -- binary 128 and binary 57 10000000 in EFM is 00111001 in EFM is 01001000100001 10000000001000 2. The digital sum value is kept near zero minimizing the digital sum value is just an attempt to keep the average number of "0's" and "1's" about the same. The value of +1 is assigned to the "1" states and the value of -1 is assigned to the "0" states. Then, the value of the merge bit is chosen to maintain the average near zero. SO! The final frame (which started at 6*16*2 = 192 data bits) now contains: 1 sync word 24 bits 1 subcode signal 14 bits 6*2*2*14 data bits 336 bits 8*14 parity bits 112 bits 34*3 merge bits 102 bits GRAND TOTAL 588 bits scary xia. don't worry if you didnt fully (or at all ) understand that whole chunk. neither do i - IIRC it's PhD course study material. anyway to grossly simplify the process, it involves many stages of interleaving AND binary parity checking. coincidentally, those 2 methods were covered in my first 2 threads. in my next post, i will cover how CD-DAs are MADE. Edited July 1, 2003 by adhoc Share this post Link to post Share on other sites
adhoc 0 Report post Posted June 27, 2003 (edited) firstly, a CD-DA (aka AUDIO cd) is a 'sandwich' composed of, from bottom to top, a clear piece of injection molded polycarbonate plastic, an aluminium sheet, an acrylic layer and finally your printed label. for all those 'special edition' cds, the aluminium layer is replaced by gold. hence the 'golden' colour. the total thicknes of a cd is always 1.2mm. if you dont believe me you can go measure yourself. good luck. here's a picture of what a cd looks like in cross section. note the difference between a 'bump' and a 'valley' - only 125nm! ( ) to give you all an idea of scale, a typical hair from our head is 2000 times thicker! (30 microns) [ ]: all cds work by having a laser beam shone at it and the reflections picked up by a lens that converts this information into a useable form. when the beam meets a BUMP, it is DEFLECTED AWAY FROM THE LENS [hence not picked up]. when it meets a VALLEY it is REFLECTED BACK INTO THE LENS. ok now, hands up how many of you think that in a CD-DA the ALUMINIUM is 'pressed' (ala vinyl) by a 'master disc' and then stuck to a clear polycarbonate base? well, all of you that did are WRONG! it is the POLYCARBONATE is pressed with a master disk. the aluminium (or gold) is then 'sputtered' onto it to act as a reflective base. this reflective layer is then covered by a protective acrylic layer which can have designs printed onto it. this is where issues of quality control can come in. these are the 2 most common ways a cd can be of 'lesser quality': a. a shorter shelf life. there is a condition known as 'rot' that afflicts cds. this occurs when the reflective layer in the cd seperates from the polycarbonate base. the cd is thus destroyed irrecoverably. this can occur for many reasons. the acrylic layer may not be bonded well to the reflective layer. or the reflective layer may not be bonded well to the polycarbonate layer. this is why it is recommended that you do not subject your cds to extremes of temperature- the expansion/contraction as your cd heats/cools, and the fact that being made of different materials the different layers expand/contract at different rates will, over an extended period of time, increase the likelihood that the layers will seperate and 'rot' will occur. UPDATE 27/2/04 - chemical reactions can also destroy cds. cds can be made from polycarbonates that degrade when exposed to certain compounds. for example, the independent classical label hyperion once had a fiasco on it's hands when the polycarbonate used in it's discs reacted with sulphur compounds found in the printed liner notes to slowly destroy their cds. it's called 'bronzing' in case you're wondering, and it's symptoms are a bronze coloured stain spreading in from the edges of the cd. b. having a modified (deviating from original) data track. this will be covered in detail later. oh i see. then what about hybrid AUDIO CDs like the SACD-CD combo?? how do they fit 2 different formats [sACD & NORMAL CD-DA] on ONE disc? 'hybrid' cds, or cds with 2 differerent formats are relatively uncommon, and you may not even know even if you have one. it's easy actually how they fit 2 different formats onto a disc, they simply have 2 reflective layers. TWO REFLECTIVE LAYERS??! yes, 2 reflective layers. remember i mentioned earlier about special edition cds being made with gold reflective layers instead of aluminium layers? there's something else special about gold layers besides the nicer colour: they're semi-reflective. what does that mean? it means that a laser beam can either bounce off it [aka read that layer] OR focus through it. which means you can have another reflective layer underneath. like this: remember that the reader is simply a beam of laser light. it can be focused to different 'depths'. hence you can have a single disk with double data layers. do note also that these 2 layers can be ANYTHING. one could be dvd-audio [24bit/192khz] and the other normal audio in surround sound mode [16bit/44.1khz in 5.1 surround]. or HDCD. or anything you wish it to be. UPDATE 27/2/04: this entire section on hybrids and focussing is only partially correct. the laser in a redbook-only player can ONLY focus on the redbook layer in a SACD because the wavelength of the redbook-only laser (it uses a red laser with a relatively long wavelength) cannot even 'see' the pits and bumps of the sacd layer. if something is less than half the size of a wavelength of a particular length, it cannot be detected using that particular wavelength. hence the sacd layer on a hybrid disc is in essence 'invisible' to that red laser. BUT on a sacd player, the focus of the laser does change. a bit OT, but some dual layer MOVIE DVDS have both layers the SAME format with each layer being HALF THE MOVIE. we ALL KNOW lasers read information in a spiral pattern from near the center of the disc to the sides. what these DVDs do is have the laser read the FIRST HALF FROM CENTER TO SIDE, then the SECOND HALF OF THE MOVIE FROM THE SIDE TO CENTER [after changing the laser's focal length to focus on the second layer]. get it? the laser reads from center to side, then from side to center in one smooth 'in-out' motion. um, 'in-out' reminds me of something else. anyway, coincidentally this is why movie dvds have a storage capacity of 8.5gb (in 2 layers) while dvd-recordables only have 4.7gb. no burner can burn TWO different layers due to the difference in method from traditional 'pressing'. a laser in 'burn' mode does not discriminate between layers. it simply burns [decomposes light sensitive dye]. an indepth explanation of burning will be covered IN THE NEXT POST. just now you spoke of quality control during the manufacture of cds, what was that all about? i believe in general asian-made cds are inferior. why? 1. it is more likely that a cd from the states/europe is of higher quality simply because as much as we hate to admit it, consumers there are far more discerning. and lawsuit-happy too. in addition most of our music comes from the states/europe. this is due to the undeniable fact that most of the studios/artistes are located there. it is also a given fact that states/europe based recording companies are definitely far more wealthy than asian companies. just look at Sony paying Mariah Carey $80 million to buy her out of a contract or Sony advancing Michael Jackson $100 million for an album that made less than $18 million in sales. in case you were wondering, i doubt that regional branches of such companies (eg sony asia) have the financial clout to rival their parent companies. it is economically possible feasible and likely that parent companies are far more able to use better and more technologically advanced equipment to press their discs. to use an analogy, the riches of most countries are concentrated in the capital. 2. do you notice that 'asian editions' usually take a longer to appear than the original imports? 'asian editions are typically pressed by asian factories owned by asian branches of the original parent company. [eg sony asia vs sony] IIRC asian 'master discs' are copies of the original 'master disc'. this may lead to a loss of quality. remember that cds are read by a laser skimming across a 'BUMP' (it's a 'BUMP' from the laser's perspective) or flat area and a lens consequently picking up a reflection (when it REFLECTS off a 'valley') or not (when the beam is DEFLECTED off to another direction by a 'BUMP'). i am going into pure speculation, but i would expect copies of the 'original' master disc to have slightly less protruding bumps. this would lead to less 'protruding' bumps on a polycarbonate disc pressed from these copied master discs. this would increase the likelihood that the lens reading a beam that DID NOT DEFLECT ENOUGH off a shallower bump and hence interpreting a bump as a flat. no amount of error correction can help this. 3. pirated editions. such cds are common in pasar malams. in fact, usually all cds in pasar malams are pirated. these are almost always copied from original cds to a computer and then burned onto CDRs. it is a accepted fact that most CDRs cannot match a pressed cd for quality. why? in my next post, i'll explain HOW A CD IS BURNED AND WHY A BURNED CDR USUALLY CANNOT MATCH A PRESSED CD FOR QUALITY. Edited February 27, 2004 by adhoc Share this post Link to post Share on other sites
adhoc 0 Report post Posted June 27, 2003 (edited) A CDR works by having a data track burned into it by a cd burner. CD burners, DVD burners, LD burners, VCD burners, etc etc. ALL burners WORK THE SAME WAY. But how many of you know exactly how a burner works? thought so. HOW ANY BURNER 'BURNS'. [we will choose to look at cd burners here ONLY for simplicity's sake. however, the burning method is the same for optical discs, be they dvd, hdcd, sacd, vcd etc etc] a cd burner has one laser with TWO modes. a strong mode to burn and a weak mode to read. a BURNED CDR (or CDRW) differs from a normal cd in that there are NO bumps. there are only 'dark' and 'translucent' areas. the dark absorb light, the translucent areas reflect it. comparing to a traditional pressed disc, dark = bumps, light = flat. a CDR is composed of an bonded acrylic-aluminium layer and a flat polycarbonate disc template sandwiching a layer of dye that is affected by a strong laser beam (the write mode of the burner's laser). like this: WHEN WRITING TO A DISC, the laser alternates between strong (to burn a translucent area by decomposing dye) and weak (to leave the area dark) as the '1's and '0's come along.. a cd can be burned at different speeds. how? BURNING SPEEDS a cd can be burned at different speeds (the 'x'- rating) depending on the strength of it's BURN MODE laser. when a cd is burned at 1x (2 to the power of 0), it takes 80 mins to burn a 80 mins CD (80mins = 700mb). when a cd is burned at 2x (2 to the power of 1), divide 80mins by 2 to find 40mins needed to burn a 80mins CD. when a cd is burned at 4x (2 to the power of 2), divide 80mins by 4 to find 20mins needed to burn a 80mins CD. continue increasing the power of 2 to get 4x, 8x, 16x, 32x, etc etc. divide the length of the cd (60mins, 74mins, 80mins or 90mins) by the 'x' rating (2x, 4x, 8x, etc etc) to get an estimate for the time needed to burn your cd. so a 80min cd burned at a 52x speed would take: 80/52 = ~1.5mins. this is close to the real-world time taken (~2mins). so why is the real world time longer? because of other factors such as the time taken for the cd to spin up, error checks, buffering etc etc MAX burn speed of a burner is determined by 2 things: 1. the strength of the laser a stronger laser needs less time to 'burn' a translucent area. 2. and the speed at which it can alternate between strong and weak. the faster the response time, the faster a cd can literally burn '1's and '0's. BUT SOMETIMES, MY BURNER CAN GO UP TO 52X MAX BUT I CAN ONLY BURN 40X MAX!! this is because the eventual burning speed is not only determined by the BURNER's max speed, but by the MEDIUM's (eg CDR, CDRW) max compatible burning speed. HUH? the medium can determine the maximum burning speed simply by the quality of it's dye. ?? if the dye on the medium is of LOW QUALITY, it will take a LONGER BURST OF STRONG LASER LIGHT to DECOMPOSE IT. cd-burners autodetect the maximum burn speed of cd by bouncing a laser pulse of set intensity off the disc. based on the different amount of light that returns (i'm not sure how exactly) the burner can determine the maximum safe burning speed for that particular disc. IF you force a burner to burn a disc that can be burnt at 40x max at 52x, you will likely end up with an unusable disc as not enough light energy was absorbed by the dye to fully decompose it when the laser was trying to make a translucent (light-coloured) area. CDRWs are ALWAYS burned slower as the dye used in CDRWs is different. the CDRW dye REQUIRES A GREATER AMOUNT OF LASER to be decomposed. hence in this way overall burn speed can be affected by the medium type and quality. But then, why are CDRs inferior sonically to pressed discs? it's simple. since the translucent flat areas in a CDR are NOT as reflective as the flat aluminium in a pressed disc due to the extra layer of dye. it is easier for a translucent (light) area to be read as a dark area if not enough of the beam gets reflected back into the lens to be recognised as a translucent (light) area. AND/OR more than enough light gets reflected off a dark area and the lens reads it as a light area. hence you have far more errors. hence you have inferior audio. ALSO, if the thickness of the layer of dye varies across the disc (as in common in cheap CDRs) Then why do black cdrs have superior audio? my take on this phenomenon is due to the colour, black. this entire section may or may not be true. it's all just speculation. hence the term: 'my take', aka my theory. we all know black absorbs the most light and heat of any colour. my guess is that the black dye is thus MORE decomposed by the strong mode laser to give a LIGHTER translucent area than any other colour. hence the very light translucent area contrasts very strongly with the black bits. the blacks are black and do not reflect, or reflect the least. the translucents are the lightest and reflect back the most light of any CDR. hence we have less mistakes. hence we have better audio. hopefully this theory of mine has some truth in it. another possible reason could just simply be that the particular black dye used in black CDRs is 'better'. in my next post, i'll cover the differences between DIGITAL and ANALOGUE sound and show you WHY THE "DIGITAL REVOLUTION" is a whole LOAD OF CROCK. Edited July 26, 2003 by adhoc Share this post Link to post Share on other sites
adhoc 0 Report post Posted June 27, 2003 (edited) despite the incredible amounts of money one can spend on digital equipment, frequently one hears that analogue sound is still better. is it true? YES. WHY??! well, here we go! why ANALOGUE sound just sounds SO much more better than DIGITAL sound.. digital sound is composed of TWO components. sampling rate and complexity. sampling rate is expressed in hz. cds are 44100hz (44.1Khz). dvd-audio is 96000hz (96khz) i believe. hz is a unit for frequency. this means that cds are sampled 44100 times a second and dvd-audio is sampled 96000 times a second. complexity is expressed in bits. cds are 16bit. dvd-audio is 24 bit. let's examine sampling rate first. NOTE: i have intentionally LEFT OUT the entire Y-AXIS on the graph [thus in turn skipping the part on resolution of the y-axis]. this is for simplicity's sake. after my previous posts, i doubt you all would want to have some more mental exercise. here's a very simple sine wave, the raw analogue wave we want to convert to digital: we shall apply a constant sampling rate to it. the red lines represent each time the wave is sampled. when a wave is sampled, the intersection points of the wave and the red lines are noted. these points of intersection have been represented by red dots: the location of these dots are coded to binary and stored. THIS is the ESSENCE of Analogue-Digital Conversion the cartesian positions of these dots are stored in binary form. this is what they would look like as plain dots over/under an axis: when decoded back to analogue, this is the wave that comes out your speakers: NOT EXACTLY A SINE WAVE EH? finally an overlap picture to highlight the differences between the original analogue wave and the one that comes off your cd. this difference is known as the SAMPLING ERROR: these were done with a simple sine wave. how many of us listen to sine waves? ah, thought so. imagine the increased sampling error if a more complex waveform was used.. while this series of pictures were way oversimplified it still does not detract from the fact that the 'DIGITAL REVOLUTION' is a huge pile of BS for us audiophiles. nothing in digital can compare to analogue for sheer resemblence to the actual performance. not even dvd-audio with it's fancy 5.1 sound. unless of course you like 5.1 surround inaccuracy. no matter how high the sampling rate there is still going to be a discreptancy when the reproduced wave is compare to the original wave. and remember, one can ALWAYS have a CRAP digital-analog-convertor. this adds another entire range of possiblilties on how digital sound can sound worse than analogue sound. but what about upsampling? i paid $100 more for it - doesnt that help the sound?! well, yes, but only to a limited extent. upsampling is a noble attempt to make up for the information that is lost while converting from analogue to digital. it basically involves using intelligent algorithmns (a set of mathematical "rules" applied to an input. Generally used to describe a section of computer code which performs a specific function) to 'predict' what the original analogue wave would look like in those areas which are not sampled. in the context of my series of pictures, it very simply means adding more 'dots' between those 'dots' that are aleady present. like this: then what about bits? to cut a long story short, to accomodate a higher amount of information that typically comes with higher sampling rates and higher graduations in the detection of pitch, a higher bit (eg 24 vs 16) is required to properly code the information satisfactorily in binary (digital). i apologise for the skimpiness of this explanation. i'm not totally sure (for once) about bits here. perhaps if you could refer to my previous post on how a cd is encoded you might see something that i dont. for an explanation on the part i missed out concerning the y-axis (with nicer pictures), go here: http://entertainment.howstuffworks.com/ana...og-digital3.htm i hope you've enjoyed reading this small series of articles as i did writing them. if you can find any mistakes, PLEASE DO PM ME! - aaron aka adhoc - Edited July 26, 2003 by adhoc Share this post Link to post Share on other sites
