Error Correction Tests

In the following tests we examine the error correction capabilities of Toshiba SD-R5002.

- EAC CD-R Test Disc

- Introduction

We used the testing methodology as described in EAC website. The CD-R test disc includes 5 scratches and a black triangle defect. Using the "c2extract.exe" build-in software and with the max reading speed of each drive we ripped the disc. The attached analyse.exe software used to get the necessary results.

The graph tells a lot about the abilities of the drive. The quality of the optical system (and the error correction capabilities of the firmware) can be estimated by checking the time positions the errors start, during reading. Higher wedges mean lower error correction capabilities.

The grid of the graph has the following means:

• The X position of a grid line is always a start of a new minute position on the CD (in play time, up to 74 min).
• The Y axis shows the dB (A) value of the error in the extracted file. dB (A) is a weighting/curve that is not affected much in high and very low frequencies, but mainly in medium frequencies, using an A-weighting curve.
• The graph shows a range of 6 dB (A) down to -120.0 dB (A). Each line represents a 6 dB(A) volume. (6 dB(A) louder means that the sound is two times louder)
• " For the test CD, a wedge below -60 dB (A) is quite acceptable for error correction. Anything higher than -42 dB (A) is quite critical, and you would hear that in any case. Of course that also depends on the surrounding music/noise. For example, if the rest of the song is absolutely silent, you will probably even hear a -96 dB (A) - only by using 24 bit audio files. If it is hidden between a -6 dB of music, you will probably not hear a -72 dB (A) or even -66 dB (A)...
  

- Explanation

After the analyze.exe software finishes, we get a detailed report about various errors:

- Errors total tells about the total number of wrong stereo samples. This includes areas of wrong samples that may include some correct samples.

- Errors Loudness is the number of FFT analysis done. For that the average and maximum value tells the power difference of the two signals. No acoustic model is used for this calculation, so the audible error may differ from these values.

- Error Muting gives information about error hiding. On bad errors (or bad error hiding techniques), the last error free sample is held for some time until a new valid sample is found. The number of these mutings is given as first result, following the average and maximum sample length of these mutings.

- Skips are occurring when the sync was lost and the drive had to reposition again (and start on a slightly different position). Non-accurate stream devices would loose the sync very often and smaller and bigger skips occur (thus samples are missing or extraneous). The average and maximum values denote the length in samples of such skips.

- Total Score is calculated according to the previous errors. The higher score a drive gets the better!

The C2 error graphs of EAC measure three C2 results:

• Total number of C2 errors (of 16 bit blocks).
• Number of C2 flagged 16 bit blocks that are identical to original WAV. This is number is usually not zero, as even the drive flagged samples to be wrong, they can be with some probability corrected correctly nevertheless...
• Number of differences to the original WAV that are not C2 flagged (also 16 bit blocks)

As EAC author mentions "…This third one is the most interesting figure, it will show how much blocks are going unreported by C2 error correction...The 16 bit blocks (instead of 8-bit, the native format of C2) was because of some drives interpolate a (mono) sample, etc. but only flag the part with the read error on it (but changed both parts). Thus only 16 bit blocks are checked for consistency. It is enough if one of the two bits is flagged to accept that any of the two bytes are wrong..."

- Test Results (Max Reading)

The Toshiba SD-R5002 starts reporting errors at 5 mins. The errors within the black triangle defect, reached -17.3dB in specific areas. The five scratches also produced errors but the signal/noise level was less than -60dB(A). The score is of 85.6 out of 100 (max), which is a good performance.

- ABEX TCD-721R

- Introduction

This test disc includes artificial scratches with lengths from 1.8 mm to 3.0 mm.

Using the ABEX TCD-714 as the reference disc, we created the reference.wav file that is needed for the comparison. Using the c2 extract software we compared the reference .wav file with the produced one from the defected ABEX TCD-721R.

- Test Results (Max speed)

The score for Toshiba SD-R5002 is only 55.7 points, which is a less than average performance. The total errors are more than in the previous test, as it was expected, since the disc is heavily scratched. The average level of the Loudness errors is -66.8dB(A).

- ABEX TCD-726

- Introduction

The ABEX TCD-726 includes three kinds of defects, interruption, black dots and fingerprints are prepared in this disc. The interruption defects are sized from 0.5 mm to 1.0 mm. The black dots are from 0.4 mm to 1.0 mm. Lastly, the fingerprints are from 65 to 75 micro meters.

- Test Results (Max Speed)

The total amount of errors is smaller with the TCD-726 disc. The scratches on the disc is not as heavy as in the TCD-721, and the rest defects and fingerprints seem not to produce many errors while reading. The Loudness errors were only 30 and the average level was -78.3dB(A), which is an acceptable level. However, the average amount of skip samples was high, leading to an 80.3 points score.

- Conclusion

Toshiba SD-R5002 gets an average 73.86 score out of 100 in the reading error correction tests.

- C2 Information accuracy

- Introduction

The Advanced DAE Error Correction tests are available in the latest version of Nero CD Speed. With the A-BEX TCD-721R disc, we can test the drive's C2 error information accuracy. First we created the image with the A-BEX TCD-714 test disc. Then we tested with the Advanced DAE Error Correction test, the C2 error information accuracy.

Note that the C2 accuracy test works only with drive that can report C2 errors according to the MMC standard. If the drive doesn't report C2 errors according to the MMC, but does support C2 errors, the software will give a 0 score, which is not correct, but also users couldn't use the C2 error information with any public available software (like DAE rippers).

According to the CD Speed author, the C2 accuracy should be very high (99%) to be useful. An explanation of the reported C2 errors follows:

• C2 Errors is the amount of errors reported by the drive.
• Compare Errors are the differences between the original image file and the data on the test disc.
• C2 Missed are present when the drive did not report an error while there was a difference between the image file and the test disc (ie. comparison bit to bit).

The "Quality" test score table is shown below:

• 0-50: bad
• 51-60: not so good
• 61-70: ok
• 71-80: good
• 81-90: very good
• 91-100: excellent!

- Tests Results (Max Speed)

Nero CD Speed reported that the drive missed almost the half of the C2 errors reported. As a result, the C2 accuracy is very low. Also the amount of compare errors is very high, and many sync errors occured. The quality score was also very bad with 0 score due to high sync error number (3876). While the drive got a 0 score with the CDSpeed, the EAC results are much different. This happens since according to the CDSpeed when so many sync errors occur, C2 accuracy cannot be measured.