c't test results, BluRay / HD-DVD burners and media

[MediumRare]

To keep things sorted, I'm starting a new thread analogous to C't test results. burners. for DVD.

The first article concerns BluRay burners only (because there aren't any HD-DVD burners available yet ), but the title is general.

G

[MediumRare]

The German computer magazine c't has 3 articles on HD-discs in the newest issue No. 15/2006, now on the newsstands in Europe. If you can get it, it's well worth looking at!!

The first article is a short introduction to the questions (particularly the what and why) addressed in the 2 longer articles.

The second article takes a look at 2 players they imported from the US (there are no home high-defintion drives currently available in Germany): Toshiba HD-A1 (HD-DVD) and Samsung BD-P1000. I won't say much about this test, other than that the technical quality of the BluRay sample movies was not that great: with the exception of one trailer, they didn't do justice to the bandwidth available. Both players had various warts and compatibility problems, but I won't go into this here.

The third article is the most interesting (at least for me ). They tested 2 BluRay burners with the presently available media types. I'll try to summarize the results in a similar style to the DVD tests.

The BluRay discs are the first comsumer optical media where the recordable discs and recorders were lauched at the same time as the players- that means (hopefully) that things like DVD booktypes won't show haunt us with compatibility problems. It also means the the rewritables (BD-RE) come with builtin defect management (like DVD-RAM) and should be more reliable than DVD or CD rewritables.

I'll say a few words about the physical construction and data structure of a BD-R because this affects what quality tests should examine. This information is also taken from the c't article. The primary source is the series of White Papers available at www.blu-raydisc.com.

The physical structure of BD-media is different from CDR and DVD, in that the information layer is only ca. 0.1 mm from the lower surface. The upper (support) layer is ca. 1.1 mm. This has several consequences for the manufacturers that can potentially affect quality:
- the lower layer is too thin for pressing and really too thick for spincoating- needs special techniques
- this lower layer is also sensitive to scratching, which is why most manufacturers use some kind of hardened coating
- the asymmetric construction means that the discs are susceptible to bowing (distortion)

The information layer consists of phase-change material (e.g. GeSb) for rewritables and (at present) various inorganic material types for write-once media. Organic pigments are in development- these will be cheaper to produce but more susceptible to temperature and humidity.

The error correction mechanisms are based on recording blocks with 64 kiB data and 14.2 kiB other codes. Special "picket" symbols aid in the localization of "burst errors" (a sequential group of errors) that normally result from scratches or spots. There are 2 specified limits for burst errors that allow the error correction to still function correctly:
- no more than 8 burst errors (Burst Count = BC) are allowed per recording block
- the sum of all burst errors in a recording block (Burst Length = BL) should not exceed 600

In addition, individual errors may occur (e.g. from poor media or faulty burning). The resulting "Random Signal Error Rate" (RSER10K) should not exceed 2 per 10000 recording blocks!

As with CD and DVD, the actual information bits are encoded in the land/pit transitions. To avoid too many sequential 1's or 0's, the information is encoded in so-called channel bits (8 data bits -> 12 channel bits). For DVD's, these vary in length between 3T and 14T (where T is the basic time unit). These are shown, for example, in the timing analysis (TA) graphs for Plextor drives and (lately) in the DVDScan jitter tests for newer LiteOn drives. For BD, variation is between 2T and 8T.

The variation in length of the pits and lands determines the jitter, which should be < 6.5%. The difference in the DC offset between short (2T) and long (8T) signals determines the asymmetry ASYM (10% to +15%), which should not vary much on a disk.

c't sent their test discs to Audiodev in Sweden who analyzed them with the new CATS B600 BDR Pro analyzer. As usual, c't combined the various measurements into a quality index (the details are not specified). As with DVD's, an index of 100 is perfect, >= 75 is very good (++), >= 50 is good (+), >= 25 satifactory (0) and >=0 is adequate (-). A negative index (--) means that uncorrectable errors occured or other signal aspects were greatly out of spec.

Now (finally ) on to the test results.

Media
There are presently 5 brands (and 4 manufacturers) available or in preproduction. All come as both write-once (BD-R) and rewritable (BD-RE). The Fuji media is made by Matushita (Panasonic), Verbatim (Mitsubishi) are preproduction samples.

c't burned samples of all media with both drives (except the Verbatim: Pioneer only). The Pioneer-burned disks were measured by Audiodev, subjected to a cliamtic stabilty test with 100 hours at 80° C and 85% rel. humidity and tested again at selected radial positions. Because of the tight schedule, the Panasonic discs were also tested at selected positions only.

The results (index and grade) are summarized in the following table. All discs were burned at 2x CLV (9 MiB/s). In contrast to DVD's, c't did not specify a "mechanical index" for the unburned discs. I hope they add this information in subsequent tests.

Code: Select all

disc       manuf.          ID               Pioneer     Panasonic    Pioneer
                                                                     climatic

BD-RE25    Sony          SONY (001)         87 / ++      85 / ++      73 / +

BD-RE25    TDK           TDKBLDWBA (000)    70 / +       61 / +       72 / +

BD-RE25    Panasonic     MEI (001)         -61 / --      67 / +     -103 / --
           + Fuji

BD-RE25    Verbatim      VERBATIM0 (000)   -94 / nr        na       -266 / nr
           sample

BD-R25     Sony          SONY (001)         21 / -       67 / +       78 / ++

BD-R25     TDK           TDKBLDRBA (000)    -7 / --      84 / ++      64 / +

BD-R25     Panasonic     MEI (001)        -103 / --      58 / +     -199 / --
           + Fuji

BD-R25     Verbatim      VERBATIMa (000)   -69 / nr        na        -34 / nr
           sample

Remarks:
- In all, the quality and compatibility is very good for this early stage in the product cycle.
- The best results were attained by the Sony and TDK BD-RE.
- Most discs exceeded the jitter margin but did not show pronounced extrema in the curves.
- Bowing generally remained within spec, even after the climate test
- A tendency to increasing RSER10K with radius indicates problems with the uniformity of the applied lower layer.
- Pioneer had problenms with some discs. Firmware update will probably help.
- The Verbatim media are preproduction samples and were not graded. The poor climatic showing of the BD-RE is due to increased jitter and needs work prior to volume production
- The Panasonic drive did better than Pioneer and is able to read and write CD's as well thanks to the dual lenses in the PUH.

Drives

See the general discussion above.
There isn't much additional information here this time. There are no details available regarding DVD burns.
The Panasonic drive is only available as an OEM unit.

Code: Select all

                          Burn Quality         Reading        Noise
                    BD-R / BD-RE  DVD R/DL     DVD / CD      BD / DVD
Model +  Firmware

Pioneer     1.10       - / +        + / --       + / na      ++ / +
BDR-101A

Panasonic   B100       + / +       ++ / +        + / +       ++ / ++
SW-5582


So, that was a while in the making, but because this is the first look at this kind of drive, there is some background information that won't need repeating in the future.

Good night.

G

[Ian]

Thanks MediumRare. I always like reading these results.

It looks like the Panasonic is the better choice here.

[dolphinius_rex]

the last table says the Pioneer can read CDs but not DVDs

[MediumRare]

the last table says the Pioneer can read CDs but not DVDs

Fixed it. Thanks. I double checked the other entries and they were OK.

It looks like the Panasonic is the better choice here.

Yeah- that's fairly clear. Unfortunately it's only available as OEM. It showed up in a Sony PC (Vaio VGC-RC 204) that they looked at in that isssue, though (2500 Euros ) and it seems to be the basis for Plextor's drive, as you noted too.

G

[RJW]

Current C't number has the LG 4x DUal layer drive tested.

A short summary will be up later.
But I can say for now that 4x on verbatim and TDK disc's createdcrap burns in C't case !

[RJW]

c't 21 -2006

LG GBW-H10N was tested. (firmware GL03)

Just the blu ray disc results.

Code: Select all

disc       manuf.          ID               LG

BD-RE25    Sony          SONY (001)         51 / +     

BD-RE25    TDK           TDKBLDWBA (000)  -220 / --     

BD-RE25    Panasonic     MEI (001)          34 / O
   
BD-RE25    Verbatim      VERBATIM0 (000)   -47 / --       

BD-R25     Sony          SONY (001)         50 / +       

BD-R25     TDK           TDKBLDRBA (000)  -118 / --     

BD-R25     Panasonic     MEI (001)          60 / +   
         

BD-R25     Verbatim      VERBATIMa (000)   -58 / --     
         

Verbatim and TDK BD-R were burned at 4x the other 2 (sony, Matushita) at 2x. All BD-RE is 2x.

[Ian]

I'm surprised they didn't try 50GB BD-R discs. Oh wait.. the GBW-H10N doesn't support them.

[MediumRare]

It's a good thing someone else reads c't! I just haven't found the time to post this information.

I've been very busy at work and have been putting most of my spare time into polishing my display tool for KProbe and DVDScan data (almost ready to go public now).

G

[MediumRare]

The latest issue of c't, No. 2/2007, has a look at the Blu-ray burners from LiteOn (LH-2B1S) and Philips (SDP7000P), which are listed at 749 €.

The hardware is identical, the firmware differs in details, e.g. Philips reads audio CDs with Cactus Datashield protection whereas LiteOn has problems. Other than this, the article only reported one set of reaults for both drives (so the summary "tables" below are somewhat anemic ).

The drives have a SATA interface and a PUH designed by Philips with 3 lasers and a single lens (Panasonic uses 2). The single lens reduces costs, but entails compromises in the quality of CD burns- the tested TY media showed roughly twice the BLER other burners had.

The drives read and write sinlge layer BD-R and BD-RE media at 2x. However, burning dual layer BD-R50 is not supported and Philips couldn't say if a future firmware update would enable this.

The BD media results consist of some tabulated measurements and the derived c't quality index shown below:

Code: Select all

                             BD-R25                BD-RE25
Drive                    Sony      TDK          Sony      TDK

Philips SDP7000P        42 / 0    80 / ++      57 / +    47 / 0
LiteOn LH-2B1S

Drive ratings are regrouped a bit from the first survey:

Code: Select all

                       Burn Quality        Reading           Noise
                      BD / DVD / CD     BD / DVD / CD     BD / DVD / CD
Model +  Firmware

Philips     BPBK       + /  +  / -       + /  0  / 0      ++ / ++  / +   
SDP7000P

= LiteOn    AL0S
LH-2B1S

DVD+-R media tested was Sony, TY and Verbatim at 16x, Verbatim DVD+R DL (8x rated) at 4x and Ricoh 8x +RW. As in previous Blu-ray tests, no details are given for CD or DVD burns. Results were good except for the Verbatim -R (don't use) and elevated error rates in the inner part of the +RW.

Half the article was devoted to the playback software (Cyberlink PowerDVD 6.6 BD), which needs some work.

The drives gave better results than the Pioneer BDR-101A or LG GBW-H10N, but (because of the lack of BD-R50 support) couldn't match the Panasonic SW-5582 (=Sony BWU-100A) or derived Plextor PX-B900A.

G

[[buck]]

Thanks as always, MediumRare. Seems like the Panasonic drive is still the way to go.

[Ian]

The drives gave better results than the Pioneer BDR-101A or LG GBW-H10N, but (because of the lack of BD-R50 support) couldn't match the Panasonic SW-5582 (=Sony BWU-100A) or derived Plextor PX-B900A.

I was under the understanding that the LH-2B1S supported BD-R50 but wrote to it at only 1x. Are you saying that it doesn't support it at all?

[MediumRare]

Are you saying that it doesn't support it at all?

Yes. That's what the article said (I don't have one - I hope you do or will soon).

Die Laufwerke können allerdings keine zweilagige Medien mit 50 GByte brennen

That leaves it open if they can even read them- the article wasn't clear on that point.

I can try to ask the author about that.

G

[frank1]

CDRinfo published on Friday, July 06, 2007 this article about
"Blu-Ray Writing Quality Tests"
http://www.cdrinfo.com/Sections/Reviews ... 8&PageId=0

Here for example the measurements for Plextor PX-B900A 2x & Verbatim BD-R:
http://www.cdrinfo.com/Sections/Reviews ... &PageId=15

Unfortunaly I am unable to explain these tests and comment them.
I noticed this in the conclusion:

For now, our first test with Blu-Ray media brought to the surface a serious parameter related to Blu-Ray discs:
The necessity for very careful handling and storage

[Ian]

CDRinfo published on Friday, July 06, 2007 thisd article about
"Blu-Ray Writing Quality Tests"

Looks like another glorified Datarius advertisement.

[MediumRare]

More than a year after the first Blu-Ray recorders, the latest issue of c't (No. 20/2007, available now in Europe) has a look at the first HD-DVD burner available: the Toshiba SD-L902A.

This drive is atypical, in that it's a slim line (notebook) drive and only available in some top of the line Toshiba notebooks- you'd be hard pressed to get one without that "packaging". Furthermore, it's a 1x burner and takes over 56 min for a SL disc (15 GiB) and more than 114 min. for a DL disc (28 GiB).

The only media presently available comes from Verbatim (both SL and DL).

For this first test, I'll add a few words about signal processing, limits and quality criteria (more information is in the c't article). These discs are based on regular DVD's, so a lot of items will appear familiar.

• The length of an ECC block is 64 KiB (vs. 32 KiB for regular DVD's) in order to distribute the information over a greater track length (reduces susceptibility to scratches).
• The error correction mechanism is again similar, with PI and PO codes. Limits are PI Sum 4 <= 280 and PIF <= 5.
• The Channel Bits resulting from the 8 to 12 modulation vary from 2T to 11T (DVD: 3T to 14T, BluRay; 2T to 8T). There are only 54 valid channel bit patterns.
• The signal quality suffers from the greater separation of the PUH from the information layer (in comparison to BluRay) and has a high noise level. Hence, instead of directly determinig a bit pattern, the decoder compares the signal to all valid channel bit patterns and picks the one with greatest probability. Invalid bit patterns are flagged. The Simulated bit Error Rate SbER should be < 5x10-5.
• There are no limits for Jitter (measured values were > 25%). Instead, limits are given for the Partial Response Signal to Noise Ratio: PRSNR > 15%.
• The signal asymmetry Asym2T gives the ratio of the reflectivity of 2T and 11T marks and should not vary by more than 10%.

c't sent the burned discs to Audiodev in Sweden for analysis. Please note that official calibration discs are not yet available for HD-DVD drives and the test drive could not report POFs.

As usual, c't publishes a subset of the measurements and reports their own weighted quality index (details not specified). An index of 100 is perfect, >= 75 is very good (++), >= 50 is good (+), >= 25 satisfactory (0) and >=0 is adequate (-). A negative index (--) means that uncorrectable errors occured or other signal aspects were greatly out of spec.

Although there are only 2 media types and 1 drive in this test, I've set up tables with the results to allow better comparisons with subsequent tests (if any become available).

Media
With only one burner, it's not really possible to separate media properties from the drive properties. No data is given for the physical properties (planarity etc.). A "mechanical index" or information to climatic stability is missing as well. I hope c't adds these in the future if more media or drives become available.

Code: Select all

disc          manuf.           Quality   PI Sum 4 / PIF     PSNR
               ID                            SbER         min / ave

HD-DVD-R     Verbatim (MCC)   -328 / --     559 / 85    13.5% / 24.2%
             Verbatim 1101                  4.51 E-4   

HD-DVD-R DL  Verbatim (MCC)   -787 / --    1030 / 60    11.4% / 16.0%
             Verbatim 2101                  1.24 E-3   
             (MCC)


Drives
The first table shows how the drive coped with various media types. The first line shows th c't quality index and grade, the second line shows the rated media speed and the actual speed that the Toshiba realized for these supported media types.

Code: Select all

              HD-DVD-R   HD-DVD-R DL   DVD+R DL     DVD-R       CD-R
              Verbatim    Verbatim     Verbatim   Taiyo Yuden   TY
Toshiba     
SD-L902A     -328 / --    -787 / --    28 / O       70 / +      84 / ++
               1x / 1x      1x / 1x    8x / 2.4x   16x / 4x    48x / 16x


The drive rating summary is grouped as for the BluRay drives:

Code: Select all

                       Burn Quality        Reading           Noise
                    HD- / DVD / CD    HD- / DVD / CD    HD- / DVD / CD
Model +  Firmware

Toshiba     TU02     -- /  0  / ++      - /  -  / 0       + /  -  / +
SD-L902A


Remarks:
- The reflectivity and Asym2T values were good, but the Signal/noise ratio was < 15% in the inner region of the discs. The SbER was more than 10x or 20x (DL) the limit in this region.
- The Audiodev test drive couldn't read the second layer of the DL disc.
- The burner was able to read these discs, but an external Xbox 360 drive choked on them.
- According to the checklist, CD-Speed can report PI/PO for the drive.
- As usual, the c't article has more information on reading properties (e.g. error correction and audio copy protection schemes).
- The listed price in Germany is 3350 € (including a Qosmio G40-11L notebook ).

Frankly, this drive is less than impressive- slow, expensive, poor quality burns and hard to get. And this more than a year after BluRay burners became readily available!

Unless something drastic happens, HD-DVD burners are dead. Maybe that's another reason some studios are plugging this format exclusively?

G

[Ian]

Thanks for the info MediumRare. It's much appreciated!

[*] The signal quality suffers from the greater separation of the PUH from the information layer (in comparison to BluRay) and has a high noise level.

And yet HD DVD backers dump on Blu-ray due to their data layer being closer to the surface. I wonder if Blu-ray backers considered this when developing the format.

I agree about the drive being less than impressive. The fact that you can't buy one without getting a laptop makes it even more of a deal killer.

[dolphinius_rex]

Did anyone else notice that the firmware listed as being used on the Toshiba SD-L902A is not quite right looking? Both RPC1.org and www.cdr.cz list completely unrelated firmwares for that drive.

And yeah, I agree with everything MediumRare said for sure. It's taken so long to even get *THIS* much from HD DVD-Rs... and this is still just 1x recording, and this is TOSHIBA'S drive, the INVENTOR of HD DVD.

I mean, if THEY can't get it right... and after delaying it for so long.... and with *VERBATIM* media.... is there any hope at all of it being fixed?

[MediumRare]

Did anyone else notice that the firmware listed as being used on the Toshiba SD-L902A is not quite right looking? Both RPC1.org and www.cdr.cz list completely unrelated firmwares for that drive.

I don't know exactly what it means, but rpc1.org lists the TF31 and TF32 firmware for a Toshiba TS-L902A. If I google for that drive, it only shows up in connection with this firmware.

OTOH, a search for Toshiba SD-L902A turns up lots of references to the drive itself.

Are you sure it's the same drive?

G

[MediumRare]

c't had a short look at the LiteOn DH-4O1S BD-ROM drive in the latest issue (05/08 ). The drive itself isn't that exciting because it's not the fastest reader and can't write any discs.

However, the Mediatek MT1929NBE chipset used and firmware CP51 together allow quality tests for BD media (as well as CDs and DVDs) with Nero CD-DVD-Speed 4.4.7.15. It reports
- the Log Distance Code (LDC), the error count in a 64 kiB User Data Block (mean < 13 )
- the Burst Indicating Subcode (BIS) which counts burst errors longer than 40 bytes (max < 8 ).

So again LiteOn is leading the way in providing access to quality scans by home users. I hope that similar capabilities will be implemented in other drives, esp. burners, by other manufacturers as well.

G

[Ian]

Thanks for the heads up on the write quality testing. I didn't realize the drive could do that.

[MediumRare]

Do you have one of these drives? I'd be really interested in seeing one of these scans.

G

[Ian]

Here you go. It hangs at 99% so there are still some issues that Deppe needs to work out.

I have never done any BD quality testing so I don't know what LDC or BIS values are considered "good".

[MediumRare]

Thank you!
Looks like you've got some relink points or similar critical positions there. Does it play OK?

I don't know what LDC or BIS values are considered "good".

I don't think any of us has much experience with this. According to c't the average LDC should be < 13 and the maximum BIS should be < 8. I had a look in the publically available documentation at the Blu-Ray site, e.g. White Paper: BD R - Physical Specifications but couldn't find any information there.

G

[Ian]

That's the BD-R disc I use for transfer rate tests. It reads fine on the drive.

Those LDC and BIS levels seem a bit low, at least for this drive. I also tested a BD-ROM disc and while the levels were better than the BD-R, they were considerably higher than what c't states.