How Reliable is DVD Media ?

[KuroNeko]

@Dasher & Debbie: jewel cases hold the disc on the center ring without touching the surfaces. Spindles stack discs on the surfaces. Every movement of the spindle, discs rub together, and any dust particles inbetween create scratches or microscratches. Dirt on top of the label side gets transferred to the read side of the next disc, and every cleaning you do can cause more scratches.

You also don't want to stack horizontal because this can warp the disc - even if a DVD does not weigh much, it can warp quite easily especially under the influence of temperature. Just a few tenths of a mm can already make some drives have problems playing them back. Storing them vertically keeps the disc nice and flat.

@Debbie: the link you provide is absolutely correct. Just buying an expensive brand is no guarantee. Always check the real manufacturer under the label. The label/big name brand just gives you a guarantee that the actual manufacturer of the disc is real i.e. no fake Taiyo Yudens for example.

@Brian: that's really harsh, losing so much data and work. Good discs isn't the only thing though and I'm sorry to say you didn't choose a good writer. Samsung is notoriously poor in the writing quality department, they haven't been making DVD writers for long and still need some experience. Same for Lite-On who made excellent CD writers at the end, they only now seem to get to grips with DVD writing. Have a look at the reviews at CDRinfo - at current prices there's no reason to stick to an inferior writer!

Neko

[LvR]

Have a look at the reviews at CDRinfo - at current prices there's no reason to stick to an inferior writer!

You have made a really sweeping statement there in your last post re the quality of burners............

Could you define what a quality burner as opposed to an inferior burner is? - and wrt your DCRInfo statement above, can you post a link to an independent comparative evaluation of DVDs with the exact same media and content burnt on different writers to justify your claim? - I don't seem to be able to find such an animal on their site.

Till such a time as you or anybody else can produce in independent test showing consistent bad "writing quality" with any and all media on a particular brand of writer, it is my opinion that its irrelevant what the general opinion is - you should be using a PI/PO error scanner to check the disk quality soon as its been created and if you get super quality scans with your particular writer and media combination then thats all you ever can do - if a "good burner" and an "inferior burner" produce the same error scans at the time of writing on the same media, what exactly else will you be looking at to determine longevity and readability of the produced DVD given you have exactly the same media?

[Brian Ellis]

I'm sorry to say you didn't choose a good writer. Samsung is notoriously poor in the writing quality department, they haven't been making DVD writers for long and still need some experience.

Sez who? They have had some excellent reviews that I consulted before purchasing. I don't know whether you are aware of it, but the model I have is actually Toshiba, presumably made under licence. This applies to all Samsung models whose number is TS-xxxx (in my case TS-225U). Anyway, I have used it almost 100 times for recording DVDs at 4x on 8x blanks (mostly DVD-R, a few DVD+R), without producing a coaster (plus a few sacrificed ones for my initial testing). I can't complain at that performance.

Actually, I did have a problem with it, but it was really a Microsoft one. With Win XP, it is an absolute pig to have the OS to see it as a UDMA device. It took me a couple of days before I found a workaround.

The main reason I chose it was a question of availability. On this island, there is very little choice. I could find only two DL burners and this one was £37 and the other was £108. None of the usual makes, such as Pioneer, are imported.

[KuroNeko]

Samsung cooperates with Toshiba but a lot of its models use different components from Toshiba. Add to that that Toshiba's DVD Writers are not all that brilliant either (poor firmware support)...

C't says Samsung has bad writing quality, CDRinfo says Samsung has bad writing quality, people on CDRinfo, CDfreaks forums say Samsung has bad writing quality, on Usenet, same story. The only good reviews I find are in magazines that haven't a clue on testing methodology. They write a few discs and if they can read them after writing they think they are good.

LvR, a quality writer will use proper writing strategies on as many media as possible, and most certainly on high quality media such as Verbatim, Maxell, Taiyo Yuden etc. CDRinfo has such comparisons. Every drive gets reviewed with the same media and a test burn is made, then verified with Plextools (the only reliable consumer test; even CDRinfo abandoned Lite-On tests with Kprobe after comparing the results to professional tools). C't uses professional test tools. While good burners write these with minimum error levels, bad writers write them with a lot to too much errors. It's seldom a hardware issue of the writers, more a firmware issue and how good and up to date the firmware is. The experience of the engineers and the amount of work/money a manufacturer is willing to spend on this (basically after-sales) determines a good brand/writer.

I agree, you have to test your drive+media, but what good is a writer that supports just a few media decently and fails on all others? This is what separates the good from the bad writers.

There's no side by side comparison on CDRinfo; you have to open each drive's review and put the same media scans side by side.

Top writers currently are (always from a certain firmware version; all writers need a bit of maturation, although better brands take less time i.e. have more frequent firmware updates) Pioneer, NEC, LG, to a certain extent Plextor, Philips/BenQ. Asus uses Pioneer OEM, also very good. Notoriously poor: Samsung, Lite-On (although their latest model seems better), Sony (Lite-On OEM), to a lesser extent: Toshiba, Aopen, MSI.

Brian, I really would advise you to find someone with a Plextor drive. Ask them to scan your current burns with Plextools. Check if your Samsung does a good write on those particular media. I dont' know anything about your local marketplace, but surely you can import from internet stores? Or are taxes on imports that high?

Neko

[LvR]

Well - in order to come back to the original question of the reliability/durability of written DVD media:

Its a function of how good Ito PI/PO errors the original disk is you produced and the longterm stability of the media. Personally I cannot see everybody going off and buying a Plextor just to have the bling-bling of PlexToolPro and the rumored ultimate compatibility of all media in all parts of the world - Plextor is not the only drive manufacturer that are able to make writers that can produce good DVDs on well-known brands of media so I cannot see why everybody must start importing specific drives

We live in a real world with real media brand availability issues and writer brand availability issues. I can see the only real realistic approach being to optimize your media and writer combination by using ANY scanner - be it KProbe/Nero/DVDInfoPro/PlexToolsPro etc - once you know what gives YOU the best possible burns with the lowest number of errors, you stick to that brand in order to ensure best longterm stability - eg - if you can consistently produce PI/PO scans below 10/2 while testing with KProbe and NEro and DVDInfoPro I really doubt that Plextools is going to find the absolute numbers very much different and as such I would say you are producing good quality disks...................

Brian - if you can it would be interesting to see a KProbe scan of one of your produced disks for example.

[KuroNeko]

Plextools Pro is indeed more flashy interface than worth the price, but all Plextor burners come with standard Plextools, which IS worthwhile. I don't use my Plextor to burn btw, I use it only to scan. My Pioneers handle the burning much better.

I don't want to start a war here on Plextools vs Kprobe. Anyone who is willing to learn more about the subject can check the comments in C't of results with testing methods on consumer drives vs professional testtools, or visit the earlier mentioned forums. The fact that all current testresults published there stick to Plextools should give you a hint though

As for real world... no drives or discs here that I like. Doesn't mean I have to buy them anyway and be happy with junk. It's a small world and credit cards, internet and UPS make sure I get the stuff I need at very acceptable prices.

Or would you start editing on a VIA board just because no-one sells SiS or Intel at your location?

Neko

[xortam]

My main concern is when you have recognized good writers burning verified high quality DVDs, yet you land up with bad DVDs that only show up after several years of sitting on the shelf (properly stored). You can buy a quality DVD from a recognized quality company and there are still batch differences due to various manufacturing plants and dye lots. You can inspect the media for manufacturing codes but you still don't know the quality of that particular dye lot. There is no guarantee that the verified quality DVDs will have all its data intact in say 10 years. I think you have to refresh your data by re-burning onto new media on occasion and/or storing the data onto the latest generation of high quality storage medium. I also think you have to maintain your precious original analog source material because encodings are always improving and you may want to re-encode that material to the latest highest quality digital standards (encodings and media).

[Debbie]

xortam,

I agree with you 100%. I find a dvd media that gives me a good burn, then I go out and buy again, a few days later, of that very same brand (same dye as far as I can see) and I get duds. The only good thing is that the store takes back what I don't burn.

I also agree that perhaps the best thing is to reburn at the first hint of failiour. I have managed to salvage about 30 DVDs (the origin of this thread) using DVD copy software (DVDFab Express) reading from the very same burner that burned them. No other DVD reader or writer could read from the media without failing. All the media have come from the same spindle and all played well when originaly burned.

Debbie

[LvR]

Debbie

All the media have come from the same spindle and all played well when originally burned.

Thats the reason for my initial post in this thread. You may be able to produce a "playable disk" Ito just sticking it in a DVD player and having the nice pictures come off it, but in reality you may be sitting on a time-bomb because your disk may be marginal and "just make the cut" at the time of production when looking at error rates.

If you look at how Verbatim defines life expectancies and do durability testing Ito error rates, you have to conclude that the best you can do is to find the lowest error rate scans from your particular burner/media choice combination and then simply hope for the best when wondering about the stability of the media and possibly listen to people when they say a particular brand of disk doesn't seem to survive real-life too well even though it may have been perfect (Ito error rates) initially.

[Flying dutchman]

I would also like to remark that DVD+/- RW are more reliable than DVD+/-R.
Or so I've read in a test report in the CT magazine.
The principle of the dye is completely different (it's a phase change dye), it consists
of more layers than the "normal" +/-R dyes and it's more resistant to UV and heat.

However, they improve with age... The dye gets its optimum quality after roughly
10 burns. So one has to completely overwrite the disks a couple of times before
entrusting anything valuable to them.


As an additional life insurance, I would buy a writer that supports testing software such as Kprobe so one can actually measure PI/PO errors. Make sure the disks match the writer!

[Brian Ellis]

I'm afraid CT mag may not be aware that there is no dye in RW media, unlike R media. RW media rely on a phenomenon called phase change in an evaporated metal alloy. This relates to the crystalline structure of the alloy, which, if I remember correctly, is ternary. If heated above a certain temperature and cooled, the crystals take on one given structure. If then heated to a given temperature lower than the first one, then there is a different crystal structure formed which assumes a different optical reflectivity. To erase, it needs to be taken again to the original hotter temperature.

This kind of thing is common with many alloys. For example, a gold-tin binary alloy can have any of 7 different phase structures at room temperature, depending on the alloy composition and the temperature history. Annealing and hardening steel is another example of phase changing.

[Debbie]

Brian,

Top marks in chemistry & physics. Just joking mate. Very interesting indeed!

Debbie

[Flying dutchman]

...but do you agree with their reliability claim (rw being superior to r)?

[Brian Ellis]

...but do you agree with their reliability claim (rw being superior to r)?

Maybe aye, maybe och aye. Translated: neither for nor against. Translated, I haven't a clue.

With the dye based ±Rs, the worst that can happen is that the non-polymerised parts become polymerised. This could happen by any one or more of three mechanisms: light, heat or free radicals. To explain the latter, when you go to the dentist and he puts in a polymer filling, heavily loaded with titanium dioxide, he zaps it with a UV source to initiate the polymerisation. This hardens only the stuff on the surface, as the UV cannot reach the interior of the cavity with sufficient intensity to trigger the polymerisation, but, as it hardens, it releases an OH radical which can trigger the polymerisation of an adjacent monomer or, rather, prepolymer molecule. This starts a chain reaction which can go through the whole mass, even with no light reaching it. On a DVD, it is not inconceivable that something similar could happen in time, although it is less likely a) because the polymer is presumably not engineered to do that and b) the polymerisation process is more thermal than photonic (the wavelength of the laser used for burning creates heat to trigger the reaction. That having been said, all monomers and prepolymers can polymerise spontaneously in time. The possibility of this happening would be reduced by good edge-sealing of the disc.

I speculate that Blu-Ray discs would be less stable than current ones, as the shorter the wavelength to trigger desired polymerisation, the easier it is to trigger spontaneous polymerisation, assuming the technology uses a similar principle.

The ±RW discs depend on metallic phase change. This is not necessarily stable, either. It depends on the temperature differential required to effect changes of the two phases and a host of other factors, such as mechanical strain. To illustrate this, in a former life, I had to make metallographic microsections of solder joints, to view the different phases. Briefly, these were cast into a plastic, cut with a fine diamond saw and polished. The surface of the solder was stressed by the final polishing and resultant strain caused phase changes, lasting over several days, which were visible by examination of the crystal structure under a microscope. I don't know which alloy is used in discs nor its phase diagram or its susceptibility to stress, but I could imagine that thermal or hygrometric expansion/contraction of the polycarbonate could create some stress on the metallic layer, creating the possibility of phase change.

All this is pure speculation, of course, but I'm trying to illustrate that both types of disc could possibly be unsuitable for ultra-long term archival purposes. Which is better, I don't know. All I do know is that you cannot tell by looking at a disc whether it's good, bad or indifferent. For example, some say gold discs are better than silver ones: it does not make the slightest difference what the reflective surface is.

[Flying dutchman]

A word of warning: never use paper CD labels on a DVD!

I had problems with DVD+R's that initially played very well, but refused to play after a few months. It was very puzzling. Until I discovered that the paper labels I used were to blame.

After removing those (I soaked the DVD's for an hour in hand warm water with soap, peeled the labels off and then rubbed the glue remains away with salad oil) they worked perfectly again!

I assume the problem arose because paper responds differently to warmth than the plastic used in the DVD's (the dvd's will deform in the player).