A new database for media quality info (The MediaBase alpha)

vbimport

#1

Seems I’ve missed out a bit in this; I see that media quality scanning these days is a hot topic.

I came across BitBurners which needed help from the public in such area. Maybe some of us cdfreakers could be interested in participating to help out a new database for media quality information. If so, you can read more about it here:
http://forum.bitburners.com/index.php?topic=698.0


#2

MMh. Once again, a database relying on so called “quality scans” alone, no transfer rate tests, and real-world use only in comments. They don’t even ask for a standardized scanning speed for each testing drive. :rolleyes:

No statistic validity in my book, at least not to determine media quality, which cannot be told by PIE/PIF scans anyway without some real-world additional testing, the minimum being a TRT.

We have the same here at CDFreaks, with similar flaws IMO, but it’s slightly more standardized. If these guys want to re-invent the wheel, what’s the point if it’s not to do it better.

And there’s Videohelp. Heavily standardized, with the drawback that even I was discouraged with all the mandatory info. :bigsmile:

I think I’ll launch my own database :bigsmile:


#3

Actually there is :wink: :iagree:

[I]“Perform the quality scan at 4x speed CLV (Constant Linear Velocity)”[/I]

We are more than happy to have some feedback! Thank you Francksoy! This is the only way to develope this database. (I’m actually not the developer of this database but I’m one of the Admins there)

This database is in Alpha mode. So it’s not even close to final version. There is still lot of work. The main point is that this database would be users own storage for scans. Users may link their “tests” to other forums like cdfreaks. So it’s basicly your personal storage. Those scans are is safe place, easy to index and search etc. I think that is something that no other site can provide at the moment?

We don’t have the TRT scans included - yet. I’m not sure if there will be that option in the future. Some people say that those are important and others say that those are unimportant. Well I personally do think that those would be nice to have in the database too. Francksoy you need to help me to convince my colleagues :wink:


#4

Hi :slight_smile:

Sorry, I missed the @4X part.

These people are extreemly misguided. So called “quality scans” tell nothing about the actual redability of the media. I’m no saying they are useless, not in the least (I’m scanning like mad myself) but they tell about the low-level burning quality of a given disc/burner/firmware combination and that’s about it. They don’t say much about the media itself.

Media “quality” includes many other aspects, mechanical stability and reflectivity being huge culprits, and these can be caught with TRTs (mechanical: because of the higher reading speed), and real-world tests (standalone players can be picky with reflectivity), NOT with PIE/PIF scans, even less so @4X: at this speed, even discs that professional equipement (like CATS) declare as crap, look terrific in homemade scans.

The only sound and 100% legit use of scans is to compare burns. Like when you want to find the best writing speed for a given media/burner/firmware combination, the best firmware etc…
Using them to appreciate media “quality”, while possible in very narrow instances with very controlled methods, is in general unsound and plain useless. It relies on [B]no real-world evidence whatsoever[/B].

I’m not taking seriously any report or review about media or burners that’s not including at least TRTs, not anymore. This whole idea that scanning is about “media quality” and is a “one-fits-all” method to judge of burnt media is simply nonsense, close to urban legend.


#5

I know they have the best of intentions, but I have to agree with Franck here:


#6

All my tests show that Plextor quality scans (PlexTools Sum1, Sum8, Burst) can accurately show whether a disc is readable/unreadable in the same Plextor drive. So unlike quality scans in other drives, I actually trust Plextor drives to predict readability accurately.

Plextor disc quality scans (PlexTools) don’t necessarily predict whether there will be transfer slowdowns, however, and they don’t necessarily predict readability in other drives.

If anyone has experience with a disc being unreadable in a (true) Plextor drive without showing POF in the same Plextor drive, then I’d really like to hear about it and see the scans.

I would rather say: So called “quality scans” don’t tell you everything about the actual redability of the media.

Sorry for putting a spanner in the works, Franck, but there’s almost always an exception to the rule. :flower:


#7

Don’t be sorry for that, not in the least, blame it on my ignorance of Plextor scans. :flower:

I’ll certainly trust you on this, too. :iagree:

(Plextor scans are, BTW, the only officially supported “scanning feature” by a drive manufacturer, unlike scans in all other drives, as I recently explained in another thread).


#8

This part is very interesting. :iagree: I personally manage my scans with my own database (and prefer to keep it this way ;)), but I know that many users would love such a feature.


#9

Francksoy,

Thanks for the feedback. There are couple of issues I would like to clarify.

Current situation is that DVD-R media quality is discussed on various forums. The information is scattered to many websites, is unorganized and not easily to be found.

I would like to see the Base as “ImageShack of DVD scans”. A place where people could upload their results, graphs, give some comments with a rating, and easily link from their favorite forum. This way the information is centralized, searchable and more easily found. As more and more data is collected, interesting queries can be made with some statistical validity - like “what is the best DVD-R brand?”. Isn’t that the noobs’ favorite question?

Regarding speed scans, my opinion is that they tell more about the reading drive than the media. A DVD-R drive is a pretty good reader, even on a bad piece of media. An OLD dvd-rom will give you a shaky transfer rate curve on any -R media. My opinion is that scans have more validity than TRT, but of course as experienced burners I/we do realize that they are not the absolute truth. But basing the data base on DVD-CD Speed output, we have some insurance about the data validity and some facts about the media - at least we know that the tester actually HAS the media at hand (or he is capable of faking the data).

VideoHelp.com database has a lot of data and with the large volumes they do have some statistical validity. The data however is totally subjective (poor/mixed/good) and anyone can feed in just about anything they see best.


#10

The best database IMHO still is InstantInfo/MediaMatch.
Its cleanliness and informativity are still unsurpassed, even if it doesn’t include any quality ratings.

However, it is outdated and has not been updated for a long time :sad:

It would be a great step forward for the CD Freaks media database if scans or photos of the packaging are added (hey, this is a real killer criteria… sorry, but without that, it doesn’t really make much sense) and the website be cleaned up a bit.
Of course, I would be happy to contribute to this :slight_smile:


#11

What you have said conflicts with your another post:

http://club.cdfreaks.com/showpost.php?p=1675324&postcount=11


#12

Not in the least. You are misinterpreting my input in the thread you link to. Please read more attentively, maybe you can learn a couple of things if you take the time to try and understand, you know? :slight_smile:

[OT]
If you want to discuss the points I make in the other thread, please feel free to join, I’m interested since if I remember well, you also like some ROM drives for testing.
[/OT]


#13

Count me in, as far as I can help.


#14

We actually did discuss this for the media section with the management, and indeed zevia did upload some great images of Blu-ray media.

I believe the outcome was that the pictures were of little use, although they looked good.

And thanks for the offers of help, I’ll keep them in mind and discuss them with the CDF management :wink:


#15

Let’s not get carried away here. Yes a TRT is THE test to perform to test disc readability at a given speed, that’s its exact purpose and I think that there should be a place for a TRT along with a quality scan. However, it is very uncommon that a good quality scan at a given speed will result in readability problems in that same drive and speed. Yes I know that there are occasionally exceptions - I don’t need you to link me to these situations, and readability at maximum speed is another story from readability at slower speeds (which is one of the advantages of higher speed scans if you want to analyze quality at high speeds). If there was little correlation between error and jitter levels and disc readability, then quality scans would serve very little purpose (where this often becomes an issue is when people think that an incredible scan means a disc with better compatibility, readability, and longer lifespan of the disc than a merely average scan, which of course is not necessarily true at all). So obviously there’s correlation between the two, however it still has to be understood that a quality scan doesn’t guarantee readability, and the odds of readability decrease when considering higher read speeds than the quality scans are performed at. So I do agree that TRTs are an important part of the equation in judging discs as that’s the only test that tells with absolute certainty readability. But let’s not ‘throw the baby out with the bathwater’ as the saying goes when considering the value that a quality scan provides. Ideally we would all perform quality scans and TRTs at multiple speeds and on multiple drives, but that’s really unrealistic and would still leave doubts for other drives, standalone playability, etc. So when you put a database like this together it’s understandable that it’s going to have its limitations and that you can’t please everyone and that people are bound to misinterpret and misuse the results that they see. But that’s not the fault of the tests themselves, the tests aren’t any less valid as long as you consider their limitations.


#16

The scans from Lite-ON DVD-ROM drives (163/5-7/16P9S) show that it is possible to predict readability issues on these drives from the digital error scans on these drives aswell (though I know DVD-ROM drives are not popular for this purpose).

Yes, but what drive is being used to perform the TRT and at what speed?
Even TRT is quite drive & Speed dependant, and is not neccessarily an infallible test.


#17

I’m suprised that even you didn’t entirely get my point, [B]scoobiedoobie[/B]. :slight_smile:

One of the driving ideas behind these guy’s project is to testify of media quality, and have a rating system for it.

As we should all know by now, 4X scans in Liteys or Benqs (Plextor 2X scans being an expection :wink: - a :flower: for [B]Drage[/B]) don’t attest of media quality.

> Comparisons with CATS have shown that many discs can look good in homemade scans @4X or @8X, yet appear out-of-specs in professional testing equipment.

> There has been many examples recenty on this board that discs showing good PIE/PIF scans at “standard” speeds can nevertheless have a troubled reading curve in a TRT.
BTW you mention that these cases are scarse, but that’s probably because you don’t deal with lower quality media. That’s what I’ve learned: [B]lower quality media’s first flaw is, in general, poor mechanical properties, leading to reading issues at high speed near the end of the disc, where the disc’s tilt and stability is the most important.[/B] This true nature doesn’t necessarily show in “standard” scans, actually it’s rare that it does.

So if they really build this database with only PIE/PIF scans @4X, [B]we’ll have another misleading source on the net making people wrongly think that they get the best media if they choose the best looking PIE/PIF figures in 4X scans, which is as sound as choosing the “best” car by choosing the fastest engine.[/B]

“Good” media is about good [I]real-world performance[/I], not good-looking homemade scans (with maybe some exceptions with very controlled methods including scanning at different speeds in different drives, if that). The only trustable real-world performance test existing for us, poor end-users, is the transfer rate test. Dismissing it when building such a project is IMO foolish, and will confuse even further thousands of users on the net.

[B]Then people in the know about testing will feel very, very lonely and helpless when facing the misconceptions of these thousands zealots of the “quality scan” dogma.
[/B]
I don’t want that to happen, personally. Things are already bad enough. :doh:

Do you know that the very people who originally created the “media tests section” here at CDfreaks, who happen to be among those who introduced PIE/PIF homemade scanning to the general public, have had second thoughts about it, and wondered if it was not a mistake considering all the consecutive misconceptions and unsound uses?

I won’t rest my case easily on this one. :disagree: The last thing we need is another source of what I consider as [B]misinformation[/B].


#18

I’m currently experimenting with a 16P1S, it’s kinda promising. :iagree: but that is beside the point.

Yes, but what drive is being used to perform the TRT and at what speed?
Even TRT is quite drive & Speed dependant, and is not neccessarily an infallible test.
That’s a good point of course. Nevertheless, though both test (PIE/PIF/(jitter) and TRT) are relative, one is real-world (TRT), the other is not (PIE/PIF). Same difference as consulting the specs of a car on one hand, and driving it on the other hand.

The fist quality of a disc is that the data can be retrieved easily without reading errors and, if possible, without slowdowns. A PIE/PIF scan @4X leaves you with [I]conjectures[/I], a TRT gives you a [I]proof[/I] (it can be fully retrieved [I]at least in this very drive[/I], or not).


#19

Same goes for PIE/PIF scans. So I don’t get the point. Unless you select a few drives for PIE/PIF scans and don’t let people post scans, for instance, from NEC or PIONEER drives, you certainly don’t avoid the issue, not in the least.

But basing the data base on DVD-CD Speed output, we have some insurance about the data validity and some facts about the media - at least we know that the tester actually HAS the media at hand (or he is capable of faking the data).
You have a point here. :slight_smile: But if you followed my above points with an open mind, you should realize that maybe things will be even worse in terms of distorted information.

That’s actually my preoccupation, and why I’m so “slating” about it (sorry for my approximative english). I guess we are, somewhere, on the same wavelenght, but we happen to consider this issue from different points of view. I spend most of my time here at CDFreaks showing people that they’ve been misinformed about so-called “quality scans”, and trying to get them to understand testing better, so I’m personally afraid that I’ll have to retire from media discussion with newbies in the future, and stick with discussions only with people “in the know”. That would be sad.


#20

[B]Francksoy:[/B]

Are you referring to this post by [B]rdgrimes[/B] (where is he now? :frowning: )

http://club.cdfreaks.com/showpost.php?p=1245551&postcount=8