PI/PIF ... does it really matter?

vbimport

#1

Salutations,

Has anyone ever had a problem regarding PI/PIF tests showing “bad” results?
The erasure codes should take care of the errors shouldn’t they?

I’m just wondering if this PI/PIF is nonsense voodoo or if it really, actuall, does something helpful.

I have never, ever, done any kind of PI/PIF test. And I have never had a bad burn or coaster. The media works like the day I burned it years later. It seems like a non-issue to me.

I almost always buy the cheapest media I can find too. Again, never a problem. It just seems to me this is like people buying ferraris because they say a porsche doesn’t drive right all the time …

In my experience, the bottom of the barrel media is the same as the top of the line in that I’m able to use it years after it was burned with no side effects.


#2

Mods,

Please don’t move this thread. It is on topic for the NEC forum even though NEC doesn’t support it (yet).

This is in response to that poll about asking NEC for PI/PIF testing.

Is this really something that helps out? Or is it a futile waste of time?

If I don’t get any replies in here, I’ll ask in the LiteOn forum. I want this group’s opinion because they are probably a little less biased. I’ve heard people buy LiteOn’s specifically for their ability to do this. I don’t expect to get an unbiased answer when it was one of their buying criteria.

I wouldn’t ask a painter if his/her new paint brush made the painting “more alive” because they will probably just be justifying the purchase without being objective. Their mind was made up when they bought it.

Yay, more analogies. I think by the end of the month, I’ll stick an ostrich farmer in there somewhere … I’ll find a way.


#3

If you haven’t had any coasters and all your media are readable years after you burn them, then good for you!

Not everybody is this lucky and some of us want to feel confident that what we burn can be read back in other drives and at a later date. It is not a workable solution to burn a DVD, wait five years and then test it in some other drives, so instead we check the readability right away by doing Read Transfer tests and PIE/PIF tests; if it only barely works now, chances are that it could be unreadable the next time we need the data on the disc.

Interpreting PIE/PIF scans and Read Transfer graphs is not an exact science, but it provides a lot more information than just a Verification passed/failed and especially if compared to no verification at all.

If you trust the data you write to CD/DVD now, without doing any tests, that is your choice, and you could very well turn out to have chosen right by not spending time testing. OTOH if some of your important CD/DVDs turn out to be unreadable or corrupted when you need them in two years, then maybe it would have been better for you to spend the time to check the media when it was burned.

It is like a car insurance: If nothing ever happens to your car, the insurance you paid was a waste of money, but if something does happen, you’ll be glad you have the insurance.

I perform a Focus Error / Tracking Error test on the CD/DVD media i buy, using PlexTools and my Plextor PX-712A recorder, and also a PIE (Sum8) test and a PIF (Sum1) test for DVD media, or a C1/C2 test for CD media. I don’t do this for every disc, but for one disc in every pack I buy.

For every disc I burn I let Nero do a verification to ensure that all the data can be read back; this takes up to twice as long as just burning the disc, but it also lets me know when something bad happens. Over the years I have made at least a couple of handful of bad burns, and I would have regretted it later if I hadn’t done the verification at burn time.

Since I have both a NEC ND-3500AG and a Plextor PX-712A, I can live without PIE/PIF scanning abilities in the NEC drive, but it would definitely be an improvement if NEC implemented that feature.

For most “normal” users, who would consider the act of performing and interpreting a PIE/PIF scan as some sort of cruel and unusual punishment, I would suggest that they simply let the burning application do an automatic verification of the burned disc - ESPECIALLY when burning or archiving data.

I’ll stop rambling now… :slight_smile:


#4

Indeed.

Not everybody is this lucky and some of us want to feel confident that what we burn can be read back in other drives and at a later date. It is not a workable solution to burn a DVD, wait five years and then test it in some other drives, so instead we check the readability right away by doing Read Transfer tests and PIE/PIF tests; if it only barely works now, chances are that it could be unreadable the next time we need the data on the disc.

But my question is, do the tests actually prevent anything? Sure, you can show me a pretty graph and say the errors are “low” but are the DVDs burned with ugly graphs just as resilient?

Erasure codes are pretty good at what they do. I’ve read in FAQs that you’re more likely to damage the disk by writing on it/etc than having the media fail by itself.

All disks are subject to the material break downs. Do these “clean” PI/PIF disks actually last longer than the dirty disks?

Actually, I don’t run tests because I just reburn. :slight_smile: Better to have 2 copies than 1 old copy that I’m unsure of.

I guess I can sum up my concerns like this: Are my disks more likely to be corrupt from something related to PI/PIF or environmental/material? If it’s the latter, then I’m just wasting my time with PI/PIF.

Interpreting PIE/PIF scans and Read Transfer graphs is not an exact science, but it provides a lot more information than just a Verification passed/failed and especially if compared to no verification at all.

That’s the part that worries me. I was reading through the PI/PO thread from my RFC: Call for a new FAQ post. In there, the people were talking about disks that could possibly be read and others that would “probably” fail.

But if these tests are accurate, you should know 100% whether the erasure codes are going to work. It’s all deterministic and there’s no chance involved. There are no tornado codes in DVDs so it either has to work or has to not work. I should be able to tell by the areas that fail.

Actually, I don’t run tests because I just reburn. :slight_smile: Better to have 2 copies than 1 old copy that I’m unsure of.

If you trust the data you write to CD/DVD now, without doing any tests, that is your choice, and you could very well turn out to have chosen right by not spending time testing. OTOH if some of your important CD/DVDs turn out to be unreadable or corrupted when you need them in two years, then maybe it would have been better for you to spend the time to check the media when it was burned.

Yes, but does it actually matter? I know you can say that the “best” media should last longer/less errors, but is that actually what happens in practice?

It is like a car insurance: If nothing ever happens to your car, the insurance you paid was a waste of money, but if something does happen, you’ll be glad you have the insurance.

Yes, but car insurance pays dividens. I can’t find information on whether this extra pricey media will actually ever pay off.

For most “normal” users, who would consider the act of performing and interpreting a PIE/PIF scan as some sort of cruel and unusual punishment, I would suggest that they simply let the burning application do an automatic verification of the burned disc - ESPECIALLY when burning or archiving data.

I’m hardly a normal user. :slight_smile: I’m just curious if this is something I should be worried about since there’s nothing like it in Linux.

I’ll stop rambling now… :slight_smile:

Why? I’ve been doing it all my life. :slight_smile:


#5

I also test a couple of discs in every spindle I buy and I have gotten a couple of bad spindles that were returned… 1 of sony08d1 dvd-r that had a bad spindle but others I bought at the same time were fine…

and one yuden0000t02 that was bad fuji branded but the others again were fine…

without this you wouldn’t know and be bitching about your media or driver being bad…


#6

Just burn a DL from Ritek (media code RitekD01 : Ridata, Traxdata DVD+R DL 8.5Gb…) and you will see the need of such a test…


#7

I run a number of different test on a regular basis, some quick and some more thorough to see how burns are going. I have found stacks of media that burned flawlessly and then turned bad part way through the stack. I have also seen rips that were bad and showed up through testing. I have seen a very good indication that the PI/PIF errors I am reading are a reliable indicator of readability. All in all, I find that reliance on these tests has saved me a lot of annoyance, money, and time. Theoretical considerations aside, they work for the purposes I expected. I am sold.


#8

The pi/pif does tells the quality of the disc. I am a liteon 1633s owner. I burn my opto. disc at 8x with 1633s, the pi/pif is very high with the Kprobe. I put the disc into my desktop player, the disc skip so much which is unwatchable. I re-burn the disc in 4x and the scan the disc again and the pi/pif scan shows a much low number, and it plays in the desktop player smoothly. Right now, I am sanning my wedding and my kids video, if the they have high pi/pif, I will re-burn them. I want to keep those precious memory as long as I can. To answer the question, it is WORTH to do the pi/pif scan for your important movie eventhough it taks time.


#9

Welcome on board rcm168 :wink:
You are just doing the right thing… But don´t forget to test your burned media on one or more standalone DVD players as well… :cool:

A quality scan test can tell alot, but never the whole picture. :smiley:

No matter what nec3500question tries to fool you in, the old truth still stands, IMHO.
Crap in --> crap out! :bigsmile:

BTW, I have never seen princo giving a life-time-warranty. (Not that I believe my Verbatim burns will last that long, but I can already see my one year princo discs deteriorate.)


#10

I’m not trying to fool anyone in. Given the choice of having a PI/PIF test and not, I would prefer to have the test.

I’m just wondering how much I’m missing out by not having it. Is it essential?

Crap in --> crap out! :bigsmile:

Here’s my point as far as cheap vs. expensive DVD media: It’s like car tires for DVD media. The cheap media ones are the type that you’ll find on cheap cars: narrow and not that much tread. Someone tells me that I should get racing tires (thicker and better tread) but in the end … they both allow the car to move forward. So mission accomplished either way, do you need the racing tires? :slight_smile:

As far as PI/PIF, I’m at a loss for why no one in Linux complains about this problem. Certainly it’s not operating system specific. So what’s going on? You guys have had problems that you tested with PI/PIF that manifested themselves in the final output. I’ve never heard of having a problem like that in Linux … it’s either a good burn or a bad burn based on if there are buffer overruns.

This isn’t a win vs. lin discussion. I’m just wondering if we (lin) are too trusting in the media and the output. Should we be looking into getting some sort of PI/PIF/PO testing too?


#11

Topic says; PI/PIF … does it really matter?

My answer is YES, but because you are not even able to make a quality test, I´m surprised you ever started this thread…:confused:

A good long read-in (on topic) and a visit to audiodev would be helpful in your case, me thinks. :wink:


#12

dual post, please remove.


#13

The question about the usefulness of PI/PIF testing in general has been addressed before. Traditionally, PI/PIF testing issues don’t go in the recording hardware forum, but the media forum (or, nowadays, a specific section of the media forum). I think that place will be more appropriate. I don’t see anything NEC-specific here except for the desire to get an unbiased answer, which you should get in the media forum. So thread moved…

CATS is more useful for testing physical disc quality than for burn quality. Various perspectives on CATS and its meaningfulness (or lack thereof) have been presented in the past in various threads and can be found through a search.


#14

c64, my link to audiodev was only because topic starter stated this here ;

And to my knowledge, this is about “physical disc quality”… :smiley: lol.

Have a nice evening gentlemans.


#15

Fair enough. The things that got my attention were PI/PIF testing, which implies burn quality testing, hence my response.


#16

Theoretically a disc that already has PIE or POE errors on it will be more likely to get a PIF or POF when quality deteriorates due to wear, scratches, sunlight or whatever; some of the bits on the disc are already corrupted and there’s only so much that ECC (Error Correcting Codes) can handle, so the likelihood of a sector becoming unreadable is higher if you start out with many parity errors!

Perhaps. I have personally managed to make several archived CDs almost unreadable by gluing a label to them. :frowning: (they were unredable in the NEC 3500 but not in my laptop drive. :confused: )

For important stuff I do this as well; I burn two DVDs of different types on two different DVD burners, so that a batch of unreliable media will only affect one of the copies, and so that in the event of one drive going bad, the burn from the other drive should work.

The problems add up, so if you start with a shaky burn, the chance of ending up with a corrupted or unreadable disc increases.

The tests are not accurate in the sense that they are not reproducible; a scan ONLY shows how the disc was seen with that drive, at that reading speed, at that point in time. If you read it again on the same drive immediately after, you will not get the exact same results, and if you change the reading speed or the reading drive, the resulting scan can be even further from the first scan. You could even have a disc that reads perfectly in one drive and is unreadable in another.

It is not 100% accurate… but it is still better than closing your eyes and hoping for the best. :wink:

If the extra pricey media means that you can still read your only backup of the blueprints to the Imperial Death Star in five years, then the price has paid off. If on the other hand the pricey media means that your copy of the movie Gigli is still readable in five years, it means that you should DEFINITELY have bought cheap Princo media and burned it at 16x with no verification, and cleansed it with sandpaper every week while storing it in direct sunlight. :wink:

Are you sure there isn’t a mode in Emacs that will let you do it? (Only partly joking… I think)

And I’ll bet you love a good discussion! :stuck_out_tongue:


#17

True.

Perhaps. I have personally managed to make several archived CDs almost unreadable by gluing a label to them. :frowning: (they were unredable in the NEC 3500 but not in my laptop drive. :confused: )

I still use sharpie markers. I’ve seen conflicting stories on whether or not it is bad for the disk. I’m sticking with sharpie because I’m too lazy to go buy more markers. :slight_smile:

For important stuff I do this as well; I burn two DVDs of different types on two different DVD burners, so that a batch of unreliable media will only affect one of the copies, and so that in the event of one drive going bad, the burn from the other drive should work.

I have a lot of data that I consider important so I can’t really burn 2 DVD copies. But I do keep the really important stuff on CD, DVD, and a hard drive.

The tests are not accurate in the sense that they are not reproducible; a scan ONLY shows how the disc was seen with that drive, at that reading speed, at that point in time. If you read it again on the same drive immediately after, you will not get the exact same results, and if you change the reading speed or the reading drive, the resulting scan can be even further from the first scan. You could even have a disc that reads perfectly in one drive and is unreadable in another.

Yeah, I realize it’s only 1 scan and it is dependent on a few factors.

It is not 100% accurate… but it is still better than closing your eyes and hoping for the best. :wink:

Right now my eyes are open and I’m hoping for the best. :slight_smile: But still … there’s no software to do it in Linux.

If the extra pricey media means that you can still read your only backup of the blueprints to the Imperial Death Star in five years, then the price has paid off. If on the other hand the pricey media means that your copy of the movie Gigli is still readable in five years, it means that you should DEFINITELY have bought cheap Princo media and burned it at 16x with no verification, and cleansed it with sandpaper every week while storing it in direct sunlight. :wink:

I would use princo for both of them. Does anyone, honestly, respect Lucas these days? :slight_smile:

Are you sure there isn’t a mode in Emacs that will let you do it? (Only partly joking… I think)

This is getting a little off topic … but that never stopped me before. =)

Emacs has this cool new feature in the CVS version (or soon in there) that allows you to have “multi-tty” support. This means that you have one emacs session running and you can connect to it from the console or X. Before, you had one or the other.

So yes … this means that as long as you keep that Emacs process in the background, when you start a new Emacs “shell” it is faster than vi. Oh yes, no reason not to use Emacs now. :slight_smile:

And I’ll bet you love a good discussion! :stuck_out_tongue:

Of course.


#18

Fair enough. It did belong here.


#19

Well I didn’t mean a copy of the Star Wars movies, I meant ACTUAL blueprints of the Imperial Death Star. :stuck_out_tongue:


#20

A lot of people seem to be basing opinions about error scanning on some pretty incomplete knowledge of what is being measured in an error scan. The notion that “burn quality” is being measured is only partly accurate. What is being measured is quite simple, it’s how well the disc can be read in the scanning drive, expressed in terms of how “busy” the error correction scheme is while reading the disc. Lots of things can be inferred by this measurement, but very few things can be known for certain. Once you have scanned a pile of discs, you get to know how your scanning drive reacts to various things, and you learn to recognise the good and the not-so-good discs. But, things like burn quality and longevity can only be inferred from an error scan…based on an understanding and comparisons with other discs. One scan does not make a bad disc, or bad media or a bad burner.

Those of us who have been scanning CDR’s long before DVD came along know that high error rates almost always mean that a disc is far more likely to fail at some point, at least in some reading drives. We have no reason to think that DVD’s are any different in this regard, but they have not been around long enough to know. Most likely, the higher speed discs will be the most suseptable to failure, but this too remains to be seen. All thermal reactive dyes are potentially unstable, especially if they are poorly burned to start.