Having some hesitancy about the quality of BDXL burns from my last posting, I set out to see if I could do some quality scans. After all, the discs themselves are really only rated to 4x and while the Pioneer will do 6x on all of them and the LG would even go to 8x on the Verbatim discs (but only 4x on the Riteks), higher speeds often mean lower burn quality as it is more demanding on the mechanics of the drive/disc and optics in terms of tracking, focus and signal bandwidth. On the flip-side, the large capacity of the discs means that higher speeds would really be appreciated and the reduced size of the pits/increased linear density should mean that the same speeds are achieved at lower rotation speeds compared to ordinary Blu-ray, being a bit more gentle on the mechanics (e.g. 6x BDXL would rotate at ~4.5x Blu-ray) so perhaps we might be able to get away with this.
So I dug out my one and only other BDXL drive in my possession, a Samsung SE-506CB Portable Blu-ray Writer that I had picked up from the Central Coast Amateur Radio Club Field Day back in 2020. This drive has been sitting in my collection for emergencies … so time to put it to use.
The Drive
The drive is definitely no spring-chicken, but it’s likely not seen much use either given that I purchased it for a very reasonable price, complete in its original box. It can be dated thanks to the reference to Android Honeycomb on the rear and the fact that Blu-ray 3D and XL were the “new” features of the era.
It also supports up to Windows 8, which was an amazing flop for its time. But such drives rarely care about such trivial matters …
Inside the box, we have the drive, still complete with its protective wrap on the top, a USB cable, a quick installation guide and a software CD-ROM.
The drive itself has rubber feet on the bottom, along with all the drive details including the fact it shipped with firmware TS00 and it is from April 2014 which would be about four years after BDXL first hit the consumer market. This drive was Made in Philippines.
The front has a very sloped fascia and the rear has a USB mini-B socket.
The drive is supplied with a single-headed cable which is one of the claimed drive features. Considering the age of the drive, it seems odd that a drive carrying a 5V 1.3A rating would even dare to run off a single port, but perhaps it has some peak-current mitigation capacity. While I ran mine off a charging USB port (1.5A rated), I have tried some double-headed cables from old external hard drives as well.
The included CD-ROM has the usual Cyberlink suspects, not that I really care to install such software nowadays.
Under the Covers
Taking two screws out of the bottom allows the lid to slide off, revealing what looks like a laptop slimline drive, but with a twist.
Four more screws later and the whole unit is extracted from its shell. But do you see something missing?
There’s no bridge-board! The drive is USB internally! That’s the first time I’ve seen this happen on an optical drive. I’ve seen it on external hard drives and that’s always been a troubling thing.
Curious features include the holes in the tray – I wonder if that makes it quieter, or if it’s just for a minor weight reduction.
A close-up of the pickup and its lenses.
“Calibrating” the Scanner
If I’m going to use the drive as a scanner, it has to be able to do quality scans and it should report values that are somewhat similar to other drives that I know to be “reliable”. For this, I picked a dusty-burned (i.e. with spikes) cheap BD-R from my limited collection of jewel-cased discs. This one was a Moser Baer India made PHILIP-R4-000 solder under the Tevion (Aldi) brand which was burned by the former-favourite Pioneer BDR-209DBK (RIP) in 2015 at 4x.
Scanning this using QPxtool with the LiteOn iHBS312, it wasn’t possible to select 4x CLV, so I settled for 4x CAV which results in this graph that shows errors tailing upwards as the disc spins faster, suggesting the errors are likely focus/tracking related. Note the default QPxtool scale is logarithmic – while it took a little getting used to at first, I do like it.
For fun, using the same drive at 8x, we can see how these errors seem to be quite a bit more magnified. This is why burning at high speeds can be problematic – if the disc is warping or wobbling, it’s impossible to write well when you can’t even read it well.
This is evidenced by the LiteOn’s TRT curve which shows slowdowns and fall-back towards the outer edge. Because this disc contains a variety of errors, it’s a good candidate for checking or “calibrating” the TSST slim-drive.
As slim-drives have less stable mechanics in general and can’t achieve CLV at higher speeds, I opted to test at 2x instead. This is gentle on the drive and the disc, at the consequence of taking longer. This shows the drive can return similar trends in terms of the spikes being at the same place, but also suggests this drive is not as good at reading the disc in general, seeing around 5x the average LDC and 2.5x the average BIS errors (just eyeballing it) compared to the desktop drive. As a result, while the drive can scan, we probably would expect higher error rates.
Testing the Burns
Burn #1 – HiDisc TL BD-R @ 6x Pioneer BDR-X13JBK
The first test, a TRT, failed at the first layer switch. This is not a great start.
The error values are already quite marginal on the first layer, very poor on the second and downright unreadable on the third to the point of losing tracking towards the end. This is perhaps not unexpected as that’s where the “hole” in the disc is. But this indicates this BDXL drive can’t read this disc even past the end of the first layer.
Burn #2 – Verbatim TL BD-R @ 6x Pioneer BDR-X13JBK
The Verbatim without the defect didn’t fare much better, now showing TRT slowdowns at the end of the first layer and an error changing to the second.
The error curve shows elevated errors towards the end of L0, elevated errors at the beginning of L1 reducing as the burn progresses, but L2 is just looking very bad all-round. I suspect the drive is very picky, but usually a picky drive is good because it’s sensitive to defects. But in this case, it seems the drive isn’t useful at all for BDXL.
Burn #3 – Ritek TL BD-R @ 6x Pioneer BDR-X13JBK
With the Ritek, we manage to just get L0 and L1 read but an error changing to L2. The change from L0 to L1 is quite bad and there is a significant slow-down patch across most of L1.
The error rates seem marginal on L0, poor on L1 and quite bad on L2 with another loss of tracking towards the end. I think this drive is just simply incompetent at reading BDXL discs.
Burn #4 – Sony QL BD-R @ 6x Pioneer BDR-X13JBK
On the Sony QL discs, three layers read out with nearly no speed dips at all. Perhaps the lighter colour means higher reflectivity (despite the translucent appearance) and the drive is getting a better signal.
The error rates are marginal on L0, poor-ish on L1 and L2 but surprisingly consistent, with L3 reading quite badly.
Burn #5 – HiDisc TL BD-R @ 2x Pioneer BDR-X13JBK
It seems a reduction of burn speed to 2x didn’t fix this either. Erroring out at the L1 layer change.
The error rates are definitely improved, but not by enough. That “curve” in the error rates on L1 is one of the downsides to multi-layer media – they often have issues at layer-change possibly because of disc geometry (e.g. spacer layer) and focus/tracking issues.
Burn #6 – Ritek TL BD-R @ 2x Pioneer BDR-X13JBK
The Ritek at 2x also did not better its record in terms of amount read, but the slow patch on L1 is reduced.
Again, there is an elevated error around the layer change point, but errors are slightly lower than at 6x.
Burn #7 – HiDisc TL BD-R @ 2x LG BH16NS55
Curiously, using the LG burner at 2x did result in a TRT that allowed the Verbatim-made disc to read out L1 and a small part of L2.
The error profiles seem quite similar to the Pioneer 2x effort, but the fact more appeared to be readable was astonishing. But this drive is zero-for-seven so far in terms of discs fully read …
Burn #8 – Sony QL BD-R @ 2x Pioneer BDR-X13JBK
It was only the eighth disc, a Sony QL burned at 2x, that managed to read in its entirety. The final L3 was read out at 2x, suggesting the drive was still really struggling.
The quality graphs seem to suggest L0 and L1 are good-to-marginal, L2 is poor and L3 is quite bad, but still far from the worst compared to the other burns.
Burn #9 – Sony TL BD-RE @ 2x Pioneer BDR-X13JBK
The BD-RE disc also proved to be no better, being problematic to read as well.
On its first burn, error rates suggest a very marginal L0, bad L1 and problems with tracking L2.
Burn #10 – Sony TL BD-RE @ 2x LG BH16NS55
When rewritten by the LG, the readback got even worse …
Now, L0 seems to be poor, L1 is atrocious and L2 is untrackable by the drive resulting in a lot of error spikes.
Can It Read or Write BDXL?
Given the poor scanning performance, it’s interesting to ask the question whether the drive should even be considered a BDXL drive. It seems totally incapable of doing anything right, reading only one of the discs completely.
In terms of reading, it’s pretty hopeless. The drive barely even read a single disc, but when it recognised them, frequently it would fail to read very important filesystem sectors resulting in the disc appearing to be empty to the OS.
Not wanting to sacrifice a real write-once BDXL disc, I tried with the BDXL BD-RE disc to see if it would burn. Curiously, it wouldn’t erase the disc, hanging spinning up and down constantly. Writing wouldn’t work either … it hung writing the lead-in. It didn’t matter if I used a double-headed cable or not. Furthermore, there were also anomalies in the sense the drive reported itself capable of writing at 4x to a BD-RE BDXL disc which is just simply not part of the standard.
There’s definitely no shortage of posts of people with problems writing BDXL and ordinary BDs on this drive. As a result, it just seems like it is a dud product that somehow got onto the market.
Firmware Update & Hidden Secrets
In case you were wondering, yes, I actually did update the firmware just out of pure frustration and found no difference – some of the tests above were done with the updated TS02 firmware.
The TSDNWIN tool ran just fine and confirmed the update. Unfortunately, in doing so, I’ve probably destroyed the only copy of TS00 firmware I ever had in my possession as I failed to investigate a means of backup prior to executing the update.
The actual download is a self-extracting .exe with the raw firmware binary in full view. Some interesting discoveries trudging through that include:
I’m not sure what they blieve (sic) in but it’s hopefully not this drives’ BDXL capabilities.
It seems the drive may have an MT1939 or related Mediatek chipset.
No idea why Moai, Easter Island, Thomas and Yoyo have anything to do with this …
Not sure if this is by accident, but there’s a low of sadfaces here … and “MTK INC_UNI_20070801”.
This section suggests it’s an MT1956. Not sure which it is – probably won’t know for sure unless I take apart the drive.
Conclusion
It seems to me that the Samsung (TSST) SE-506CB external BDXL drive is just a bad drive. It’s not just me either – there’s plenty of people complaining about issues even on dual-layer discs at the layer break, not reading discs, having laser failures, etc. In my case, my drive was a second-hand unit but it has seemingly worked just fine on all the discs up to DL that I’ve thrown at it for reading (as I don’t trust slimline drives to make a good write).
As the only BDXL capable drive in my possession that natively seems to support quality scanning, I pressed it into service to scan discs only for the drive to reveal that it’s just not good at reading BDXLs. I’m not convinced that the burns themselves are bad – just the drive is not able to read them. Perhaps this is because it’s based around an old SoC platform that’s not optimised for BDXL and perhaps it’s also because it may be an “earlier” drive.
Nevertheless, relative differences would still seem to suggest that slower burns do produce higher quality, if only by relative comparison. The paler colour of the Sony QL discs may contribute to increased readability. But unfortunately, with no absolute point of reference, the results are somewhat lacking in significance.
I don’t think this is cause for alarm, but does give me pause to remember – if you have BDXL discs but no access to a BDXL drive to read them back with, then your data may be as good as gone. In my ownership of Blu-ray drives, I’ve suffered more laser failures than I did in my DVD era. While I’m not sure if this is still the case, blue lasers definitely had lower lifetimes at the outset … so now I’m starting to wonder given the longer burn and read times whether “consuming” these drives would be a potential worry. But that’s just one of those thoughts that quickly vanishes once one remembers just how much BDXL media costs …
Join me in the next installment (tomorrow) as I try to resolve this ambiguous situation by doing something I said I wouldn’t do …
Big rate of christianity among koreans, not surprised the firmware dev wanted to “blieve”, but maybe for this drive to read bdxl properly, you do need divine intervention !
I’m not religious, but I did pray that it would work …
– Gough
Considering the massive quality control fail on the scratched disc, and the flaky drive, that doesn’t bode well for using BDXL for backups. It reminds me of how the quality of floppy diskettes was deteriorating in the late 90s/early 2000s…
Regarding the holes in the tray, could they be a means of shifting away some kind of mechanical resonance?
I’d have to say that manufacturing QC is probably just fine – it’s more likely that HiDisc were obtaining “low-grade” quality-control-failed discs and selling them off under their brand. Remember the day of “overprint” DVD-Rs and CD-Rs? This is likely a similar situation – these would’ve never been let out on the streets with Verbatim on the outside. That being said, I suspect the Samsung “BDXL” drive might have just been a case of “cashing in” on the “new tech” by pushing an older drive design or optical pick-up unit past its limits to get to the one or two new layers.
An interesting thought regarding the holes … maybe that could serve a function like that. There isn’t much in the way of counterbalancing on slimline drives either, so your guess is probably as good as mine.
– Gough