In the latter case the user will be prompted for each test submission about allowing or not the relevant action. Please take into account that the code used for this case is pure JavaScript and thus publicly available for viewing and evaluation by experts. So, we are completely certain that we respect our users. At least as they do respect our site themselves, along with our continuous efforts for always bringing them the best news and reviews on the market and the industry today.

The following pictures will help you adjust the security settings for the "Trusted Sites" zone under IE, after adding CdrInfo.com to the list of trusted sites.

First uncheck the requirement for Https, then add the cdrinfo.com site to the list of trusted sites.

Then, please adjust the settings as shown in the following pictures.

The following setting is the same as in the previous case.

The following settings affect only the submission of new user tests.

You might be able to find answers to some other questions on security in the FAQ section of this presentation.

---

10. Csv file format

---

Csv file format

This is the format of the test files. They are actually normal text files following a pre-specified pattern of numerals, text and punctuation.

Usually, they correspond to a simple table. This table in turn might be considered a text form of a typical database table.

A Csv file, consequently, is a series of rows (lines). In our case there are a few constrains this file should have to follow. On each line there are 3 numbers. The first corresponds to the particular time (in seconds) of each one of the following 2 measurements. The first measurement is a number representing the number of C1 errors (in the case of a tested CD) or PI errors (in the case of a tested DVD). The second number is the corresponding number of C2 or PO errors.

In some cases it is useful to have a header included in each Csv file containing an indication for each one of the following numbers, according to the column they belong.

In an original Csv file the non-integer part of time corresponds to it's normal fractional part. They do not correspond to sectors (1/75 of a second.). As a results, the error rates illustrated in the graphs of our database do not correspond to those absolute values measured by the used programs themselves. This is due to the fact of summing them up over the period of one minute for CD's and 8 minutes for DVD's. We do this for easy inclusion in our database. However, these results are completely comparable with each other, and server the intended purpose of comparing quality results from different media.

Non-integer numbers for the other two numbers per each line are due to the above mentioned summation process (rounded to 2 decimal places) performed on the original Csv file. Instead of having to show several thousand (perhaps) numbers for each test result into a single picture, something user unfriendly and database cumbersome, we only include measures averaged over 1 or 8 minutes.

Below, on the left, is a sample Csv file for the case of a test-result for a CD. On the right, the shown numbers are those appearing after shrinking the original Csv

A sample concerning the test results of DVD are shown below. To the right is shown the topmost lines of the shrinked file.Both in the previous, as well as in the present case, the measurements are averaged over1 seconds during display. During submission, however, the Csv file is further summed-up down to one result every 8 minutes.

---

11. Some suggestions for the proper submission

---

Some suggestions for the proper submission of viable Tests

Although many users seem to be aware of the proper handling of tests such as those submitted in our on-line application page (as judged by reading our forum topics), we believe adding some of our own arguments here might help both the starting and the more advanced readers on some simple issues that, despite their simplicity, might affect significantly the validity of a test submission.

Please before posting, perform each test at least twice. Use identical testing conditions (the same PC, identical running software, degree of CPU usage, etc) and try to use discs from the same batch.

Please handle discs with care. Dust or fingertips might easily produce misleading test results. Do not touch the lower surface of a disc, as this is more amenable to scratches and dust particle inclusion that will artificially produce, otherwise unjustified, errors.

Always use as recorders and readers drives that are known to perform well under regular everyday use.

Use older drives only if they keep to perform flawlessly. Please take into account that a to become defective drive shows early signs of its attitude by not recognizing some discs and/or slowing down during parts of the recording/reading process. In general, any abnormal drive behavior should be enough to keep you away from using it for submitting tests.

If you do submit them (it?s always up to you fellow!) you might find out soon that your tests deviate too much with respect to other published tests. In this case your test might seem to some readers misleading. It is a good idea to thus withdraw it or otherwise revise it by using newer or more appropriate equipment.

Please have in your mind that by submitting any tests you help both yourself and our user community towards finding out quality hardware for its needs. As more tests are available and the more reliable they are, the more helpful our application can be both to you and all the rest of us.

You have at your disposal freedom and the power you need in order your work to be seen and we encourage you to attain the best outcome.

---

12. Measuring the quality of recorded media

---

Measuring the quality of recorded media

General Considerations

There many factors that influence the quality of a particular media recorded on a particular recorder drive. There are also many ways to identify and qualify these factors as well.

Different media are based on different chemicals and differences in the manufacturing process followed during the making and packaging of a disc might largely influence the outcoming media quality when testing them after recording. For example, the so called yellow dyes (phthalocyanine) used by some manufacturers require less laser power for recording and consequently a different recording strategy in part of the recording drive?s firmware for obtaining optimum quality. On the other hand, discs based on the same chemicals, when produced by different manufacturers have been identified to produce large deviations in quality. (We won?t name any of then here, but it is a good idea for the reader to experiment himself on the quality tests page itself!:)

Even drives from the same manufacturer based on the same model and firmware might produce different quality measurements. This is not strange in itself, all sections of traditional industry have learned to live with such discrepancies for nearly a hundred years now. The industry has also exercised the use of the so-called "statistics" for dealing with these phenomena. A large number of samples, with averaged "quality" measurements over the number of units produced offers an acceptable measure of quality in most cases.

There are many indicators of quality when dealing with CD?s and DVD?s in the labs. These are based on both mechanical and optical measurements of the surface of a recorded disc or a pressed one. Even unrecorded discs can be measured against some specific test for identifying imperfections on the ATIP or the disc geometry itself. All these measurements are of some importance to industry engineers for locating manufacturing process flows or when working towards disc and recorder improvements in research and development groups.

Most of the above measures are "analogue" in their nature. They offer some sort of a numeric value along the changes of a geometric, optical, or otherwise physical property of the medium, the recorder or some other aspect of a process. Examples of such measures is the focus/tracking deviation of the spiral of a disc with respect to the laser diode beam. They are of particular importance for drive and production machine calibration, for developing better algorithms and strategies for encoding and recording, but in the final end, these seem to be of lesser importance to end-user requirements. This might sound as a bold statement, but actually it is not. Please read below.

The final end of any "quality" measurement for a particular user is the playability or not of a particular disc on his/her player under most circumstances. Consequently, when it comes to pure user understanding of quality, of prominent importance becomes not some sort of an analogue measure in general, but the number of reading errors the drive encounters per second during playback.

These errors are either of a minor importance to overall disc readability or constitute a major and thus deterministic factor of reading quality. In the former case these errors, in the case of compact discs, are defined as C1 errors and are only indicative of the quality of the disc. The latter type of errors are called C2 errors and in the case of audio CD?s, when encountered in sufficient consecutive numbers, will most probably result in audible clicks during playback. When encountered in large quantities over a very small amount of time will most probably heard as glitches or not heard at all (muted music due to consumer player built-in algorithms).

This outcome depends on the type of a player and the playback method. For example, consumer players are self-embodied by better C2 error concealing algorithms with respect to PC drives. PC drives, on the other hand, will bypass the built-in error concealing/smoothing algorithms and will produce audible glitches when playback is done digitally (ripping or digital extraction to an outer D/A decoder/booster or an included sound card)!

As long as data CD?s (and PC backup admins) are concerned, there is also a third layer of error correction, applied only on data discs. The errors produced at this level are characterized as L3 (layer-3) errors. Encountering such an error once per disc usually means that at least one backup file is corrupted.

In the case of our tests here, we will not have to deal with such errors at all. We consider already a major event the existence of even isolated cases of C2 errors. The non-existence of errors of this type, already eliminates any possibility of L3 errors.

Implementation details

In the case of our tests, we restrict our selves to only those types of tests that are easily conducted by the average user (or, at least, by those users somewhat more technically inclined). This was the first requirement in our design board. Secondly, a number of other factors influenced our decisions and practices. Each test should be technically well recognized, widely acceptable and easily implementable by products offered in the market at reasonable prices. Since most professional equipment is out of the reach of the budget of the everyday user, we thus restricted our tests to only those offered by some program developers on market drives.

(We can very easily include professional tests if we decide to do so, and we already have included many tests we have performed during our regular drive reviews using our own expert equipment. In the future, all tests and reviews of new hardware will also be appearing in this section of "Media Quality Tests" as well.).

Having expressed the above requirements, we must state that, thus far, we are aware of only 4 such programs. PlexTools Professional, CdSpeed, UMDoctor and KProbe. It is important to point out that not all of these programs work on any available drive. In particular, PlexTools Professional works only with the Plextor Premium (523252), CdSpeed works with those drives based on the Plextor Premium chipset and the MediaTek chipsets, UMDoctor requires a Sanyo chipset with an accompanied firmware and KProbe is restricted in it?s use to only drives based on the MediaTek chipset and firmwares.

We have thus restricted our prospective drive readers to only to only a few dozens of CD and DVD readers drives. But still, we have left open the possibility of using an arbitrary recorder. This is certainly good news, as any user can find out the best disc for his recorder, albeit he will not be able to submit any tests himself!

---

13. Frequently Asked Questions

---

Frequently Asked Questions

1. Why some selections in the View Tests page are grayed out?

Depending on the grayed out option, this might happen due to several reasons.

For the case "Enhanced Recording Mode" (RecModeEnhanced), we decided, for the time being, not to offer this option publicly, in order to keep the user selections as simple as possible. We might reconsider this in the future if there is a many members? request for doing so. The same applies in the case of "Enhanced Audio Recording Mode" (RecModeAudioEnhanced).

In the case of the reading Program, we have restricted the available options (presently) to only 2: KProbe and UmDoctor, because currently these are the only programs offering text output in the form of the so-called csv file. Otherwise, each user should have to use a graphics to vector translation application (like the one included with CorelDraw, for example) for doing it. We think there is no much sense into pursuing such an endeavor and hence we have presently disabled the inclusion of tests produced by PlexTools and CdSpeed.

2. I have fixed some selections on some list-boxes and there is no way now for me to add additional tests. Can you please fix the page?

This not a problem of the page. You have just imposed restrictions on the degrees of freedom of the system that lead to tests that do not exist in our database. So, this is a problem related not to the page design itself, but rather to the use of it made by our users.

Please unfix some of the list-boxes (just try one-by-one) until you are led to cases were there are available tests in our database, among which you can make more selections.

3. I choose one particular test for viewing. It shows OK. Then, keeping the same test selection parameters, I choose the [Average] of the available tests. Instead of seeing 2 tests, I see only 1. What's the problem? The same holds when I choose [Best] or [Worst].

You have chosen parameters which lead to only 1 available test. In this case the Average is the same as the test itself. The same holds for the Best and Worst cases. So this is normal, as each curve overlaps another one. In order to be able to get distinguishable curves you will have to choose parameters which lead to more than 1 available tests!

4. I have measured the same disc with the same reader/software, but the results are different. In the first case I chose the 4x reading speed, while in the second one I chose the maximum allowed by the reader. What is the problem?

Considering the fact that the reported results are closely related to the reading quality of your drive, we could say that such results were expected. Higher reading speeds could affect the results, not always negatively, however. For this reason, we have chosen to include the reading speed selection in our test suite, in order the user to be able to identify the speed at which each disc was read at. It is up to you whether you want to compare discs tested at the same speed to keep a reference, or not.

5. Latest versions of KProbe have an "ECC" setting at the left of the main PI/PO window. Which ECC setting should I select from the available 1-10 scale?

There is not any clear answer to this, since the KProbe measuring mechanism is not clearly defined until now. Again, keeping a stable setting for your measurements could make things easier in order to have a reference when comparing the test results. In most cases, CdrInfo.com sets the ECC setting to 8.

It might be the case that this value defines the mesh of the graph, which essentially refers to how distant one measurement lays from the previous and the next ones. In this case choosing a particular scale number is irrelevant. This is due to the fact that before submitting your test results, these are necessarily summed-up per 2 minutes for database storage efficiency.

6. What is the acceptable levels of PI/PO and C1/C2 error rates?

1. Parity Inner and Parity Outer Errors for DVD (ECC=8):

• DVD PI<280/sec (standard)
• DVD PI<<280/sec (good quality)
• DVD PO = 0

2. C1 and C2 errors for CD:

C1=E11+E21+E31

C2=E12, E22 and E32

• E22- Approaching uncorrectable error
• E32- Uncorrectable
• CD E22 and E32 = 0

BLER (Block Error Rate – CD only) = E11 + E21 + E31 per second averaged over ten seconds.

One or more bad bits in any frame

• CD BLER< 220/sec (standard)
• CD BLER<< 220/sec (Good quality)

6. KProbe offers 3 different options regarding the address on the CD and DVD. The MSF, LBA and Disc size. What should I choose for measuring my discs?

In general, the "Disc Size" option will work with your CD and DVD media. You should choose this option in order the software to automatically measure the whole disc inserted to your reader. (Please also choose to fill the disc with data completely during recording.)

Generally for a CD:

M:S:F (Minutes:Seconds:Frames) is the original unit used for addressing the positions of a CD. A frame is actually a sector and the unit corresponds to 75 frames in 1 second.

LBA (Logical Block Addressing) is also known as sector of a CD. In the MMC specification document where it is described, there is a defined range for the mapping of MSF to LBA values.

Sectors of the CD are defined as signed long integer values to account for negative values. Here's the relationship:

Positive values are mapped:

MMC LBA:0~403964 to MSF: 00:02:00~89:59:74

and negative values are mapped accordingly:

Below 2 secs (MSF), ranging from 00:00:00 to 00:01:74 (MSF) are represented as MMCLBA: -150 to -1.

The range from 90:00:00 to 99:59:74 (MSF) are represented as MMC LBA: -45150 to -151.

The range of MMC LBA is a linear range with the integer values starting from -45150 to 403964. However, it is not linear comparing with Red Book Addressing (MSF) because it wraps around.

Logical Block Addressing are the mapped integer numbers from CD Red Book Addressing to MSF. It is also known as sectors of a CD. Other people may have different interpretations for LBA, or simpler:

The starting sector is -150 and ending sector is 449849, which correlates directly to MSF: 00:00:00 to 99:59:74. This removes the wrapping as in MMCLBA representation and the transformation will be linear when comparing with the Red Book Addressing (MSF).

---

14. Glossary of Terms

---

Glossary Terms

Below is a list of terms (in alphabetical order) that appear in this guide and might be of help to our readers. Please ask us to add more terms by submitting the form here.

ECC

Error Correction Code

MSB

Most Significant Byte

LSB

Least Significant Byte

Channel bit (DVD)

The elements by which, after modulation, the binary values ZERO and ONE are represented on the disk by pits.

Physical sector number (DVD)

A serial number allocated to physical sectors on the disk.

Track

A 360° turn of a continuous spiral.

Track pitch

The distance between the centrelines of a pair of adjacent physical tracks, measured in radial direction.

DVD Disc Types

Type A

Consists of a substrate, a single recorded layer and a dummy substrate. The recorded layer can be accessed from one side only. The nominal capacity is 4,7 Gbytes.

Type B

Consist of two substrates, and two recorded layers. From one side of the disk, only one of these recorded layers can be accessed. The nominal capacity is 9,4 Gbytes.

Type C

Consists of a substrate, a dummy substrate and two recorded layers with a spacer between them. Both recorded layers can be accessed from one side only. The nominal capacity is 8,5 Gbytes.

Type D

Consists of two substrates, each having two recorded layers with a spacer between these two recorded layers. From one side of the disk, only one pair of recorded layers can be accessed. The nominal capacity is 17,0 Gbytes.

Track modes

Tracks can be recorded in two different modes called Parallel Track Path (PTP) and Opposite Track Path (OTP). In practice, the lengths of the Data Zones of both layers are independent from each other.

In PTP mode, tracks are read from the inside diameter of the Information Zone to its outside diameter, this applies to both Layer 0 and Layer 1 for Types C and D. On both layers, the track spiral is turning from the inside to the outside.

In OTP mode, tracks are read starting on Layer 0 at the inner diameter of the Information Zone, continuing on Layer 1 from the outer diameter to the inner diameter. Thus, there is a Middle Zone at the outer diameter on both layers, see figure 5b. The track spiral is turning from the inside to the outside on Layer 0 and in the reverse direction on Layer 1.

PO

Parity (of the) Outer (code)

PI

Parity (of the) Inner (code)

DVD ECC Blocks

ECC of DVD block is a self-contained block of data and error correction codes that grouped into a sequential series of 16 DVD sectors.

An ECC Block is formed by arranging 16 consecutive Scrambled Frames in an array of 192 rows of 172 bytes each. To each of the 172 columns, 16 bytes of Parity of Outer Code are added, then, to each of the resulting 208 rows, 10 byte of Parity of Inner Code are added. Thus a complete ECC Block comprises 208 rows of 182 bytes each.

PI/PO

Type of measured errors for DVD's. The I corresponds to "inner", the O to "outer". This is a distinction being made when each sector along with the error correction data is presented in a tabular form. The I's and O's then correspond to the 2 different directions of the 2-dimension Euclidean space (plane).

Recording Frames

Sixteen Recording Frames shall be obtained by interleaving one of the 16 PO rows at a time after every 12 rows of an ECC Block Thus the 37,856 bytes of an ECC Block are re-arranged into 16 Recording Frames of 2,366 bytes. Each Recording Frame consists of an array of 13 rows of 182 bytes.

Random errors (DVD)

A byte error occurs when one or more bits in a byte have a wrong value, as compared to their original recorded value. A row of an ECC Block that has at least 1 byte in error constitutes a PI error. If a row of an ECC Block contains more than 5 erroneous bytes, the row is said to be “PI-uncorrectable”. During playback after the initial recording, the errors as detected by the error correction system shall meet the following requirements:

- in any 8 consecutive ECC Blocks the total number of PI errors before correction shall not exceed 280,

- in any ECC Block the number of PI-uncorrectable rows should not exceed 4.

Jitter (DVD)

Jitter is the standard deviation s of the time variation of the digitized data passed through the equalizer. The jitter of the leading and trailing edges is measured to the PLL clock and normalized by the Channel bit clock period.

Sector

In case of CD media, “Sector” refers to the data contained in one frame. In the CD-ROM standard document the term block is used for this unit. In the case of DVD media, “Sector” is the smallest user addressable part of media. The user data contained within a sector is 2048 bytes.

Physical Sectors (DVD)

The structure of a Physical Sector shall consist of 13 rows, each comprising two Sync Frames. A Sync Frame shall consist of a SYNC Code from table 4 and 1,456 Channel bits representing the first,respectively the second 91 8-bit bytes of a row of a Recording Frame. The first row of the Recording Frame is represented by the first row of the Physical Sector, the second by the second, and so on. Recording shall start with the first Sync Frame of the first row, followed by the second Sync Frame of that row, and so on row-by-row.

Data Frames (DVD)

A Data Frame shall consist of 2 064 bytes arranged in an array of 12 rows each containing 172 bytes. The first row shall start with three fields, called Identification Data (ID), the check bytes of the ID Error Detection Code (IED), and Copyright Management Information (CPR_MAI), followed by 160 Main Data bytes. The next 10 rows shall each contain 172 Main Data bytes, and the last row shall contain 168 Main Data bytes followed by four bytes for recording the check bits of an Error Detection Code (EDC). The 2,048 Main Data bytes are identified as D0 to D2047.

Reed-Solomon code

An error detection and/or correction code for the correction of errors.

BLER

Block Error Rate. The total number of C1 errors per 10 consecutive sectors. It should be below 200, for perfectly readable disc.

C1/C2

Type of errors during reading a recordable or pressed disc. It is evaluated per sector and presented (usually) averaged or summed-up over a number of sectors. (This gives rise to the various BLER measurements.)

CD-R

A glorious recordable disc:)

DVD-R

The recordable DVD Forum defined format. The term might also refer to the media or a drive capable of handling (recording) these.

DVD-RW

Re-writable format/disc/..., according to the DVD Forum specifications.

KProbe

A program for detecting reading errors that works on drives (readers, recorders) based on the MediaTek chipset.

LBA

Logical Block Address. The disc is measured (addressed) according to 1/75 of the second intervals (sector, frame), in 1x speed.

MSF (CD)

MSF Address (Minute/Second/Frame) The physical address, expressed as a sector count relative to either the beginning of the medium (absolute) or to the beginning of the current track (relative). As defined by the CD standards, each F field unit is one sector, each S field unit is 75 F field units, each M field unit is 60 S field units. Valid contents of F fields are binary values from 0 through 74. Valid contents of S fields are binary values from 0 through 59. Valid contents of M fields are binary values from 0 through 74.

For example 63m:45s:70f. An alternative disc addressing scheme. It is essentially the same with the LBA method, in case you divide LBA by 75 (the number of frames per second) and then transform the seconds to minutes by dividing by 60. In the literature there is an arbitration of a 2 second track lead-in.

UmDoctor

A program by Sanyo for estimating and writing down disc reading errors (in the form of C1/C2 for CD's and PI/Unc-PI for DVD's). It works on drives based on Sanyo chipsets. In case of DVD, the software measures PI/Unc-PIU errors on each 8 ECC.

Groove

A trench-like feature of the disk, applied before the recording of any information, and used to define the track location. The groove is located nearer to the entrance surface than the land. The recording is made on the centre of the groove.

Land

The area between the grooves.

CIRC

Cross Interleaved Reed-Solomon Code (CIRC) is the error detection and correction technique used within small frames of CD audio or data. The CIRC bytes are present in all CD-ROM data modes. The error correction procedure which uses the CIRC bytes is referred to as the CIRC based algorithm. In most CD-ROM drives, this function is implemented in hardware.

Frame (CD)

A sector on CD media. Also the F field unit of a MSF CD address. The smallest addressable unit in the main channel.

DVD Control Area

The DVD Control area is comprised of 192 ECC blocks in the Lead-in Area of a DVD medium. The content of 16 sectors in each block is repeated 192 times. This area contains information concerning disc.

Lead-in Area

The CD Lead-in area is the area on a CD-ROM disk preceding track one. The area contains the TOC data and precedes each program area. The main channel in the lead-in area contains audio or data null information. This area is coded as track zero but is not directly addressable via the command set. The Q sub-channel in this area is coded with the table of contents information.

The DVD Lead-in area is the area comprising physical sectors 1.2 mm wide or more adjacent to the inside of the Data area. The area contains the Control data and precedes the Data area.

Lead-out Area

The CD Lead-in area is the area on a CD-ROM disk beyond the last information t rack. The main channel in the leadout area contains audio or data null information. This area is coded as track AAbcd but is not directly addressable via the command set. The READ CD-ROM CAPACITY data is the first logical block address of this area minus one.

The DVD Lead-out area is the area comprising physical sectors 1.0 mm wide or more adjacent to the outside of the data area in single layered disc for PTP (Parallel Track Path) disc, or area comprising physical sectors 1.2 mm wide or more adjacent to the inside of the data area in layer 1 of OTP (Opposite Track Path) disc.

DVD Reference Code

The DVD Reference code is comprised of 2 ECC blocks (32 sectors) in the Lead-in Area and used for the adjustment of equalizer system of the drive hardware.

---

15. Programming decisions

---

Programming decisions for developing the Media Quality Tests

During the preparation of this section of our site, we understood that the user interface would play a significant role in the satisfaction of our readers while they would be involved in querying our database for quality tests.

From many respects pure html is inadequate in handling highly interactive scenarios as that we felt our solution represented . We had thus have to make some decisions in advance.

Instead of having each user press a dozen of buttons on scattered pages for finding the tests he wanted to find, we felt we had to offer him a seemingly simple approach of the feel and taste of an interactive (windows) program.

All options should be available under a single page. All interactions with the page controls (buttons, list-boxes, etc) should not distract the user attention by imposing postbacks to the server, at least not in a viewable way:) Taking a look at both the initial and final state of the main Test Page (see below), shows that we have succeeded into devising the illusion that the user is actually working on a "pure" (windows) program. In fact, any such program, in reality, would be as fast as our Dhtml code, not a single second faster!

In the above picture (showing the selections leading to a particular test) we see that the user interface has been kept constant throughout all user interactions with the page.