Is it normal for PIE errors on DVDs to increase over time?

[Ronaldus]

A while ago I got a BenQ 1620 drive and started saving scans of various DVD -R and +R media burns. I did some scans of the same discs 1 month later and noticed a disturbing trend (this happened on every disc I checked):

For example, the first scan of a YUDEN000T02 +R media written at 4x had:
PIE average = 12, PIE maximum = 22, PIF average = 0.03, PIF maximum = 5.

One month later the same disc scanned in the same BenQ drive:
PIE average = 31, PIE maximum = 58, PIF average = 0.06, PIF maximum = 7.

Basically, the PIE and PIF roughly doubled in 1 month.

Do recorded media typically show an immediate increase in PIE after they are burned? Does the PIE keep increasing every month until it exceeds 280 and cause PIFs which eventually cause POs and then the disc dies?

In other words, are the increased PIE errors I'm observing normal or did I get bad TY media or is the dye "settling down" after being burned? I'm afraid it might be a sign that even high quality Taiyo Yuden discs like these start to rot immediately and most of my discs will be lucky to survive more than a few years.

[lordsmurf]

Scans are voodoo. You could have scanned it 10 time within 10 days and gotten 10 different values. Scanning is mostly a theory or a guideline, it's not any kind of "end all, be all" kind of test. The best way to verify media is with 3-4 different kinds of tests.

[Ronaldus]

Lord Smurf, you may not think much of scans, but I think they are important. Given that I can rescan the same disc multiple times on the same day and get very similar results, I think that proves the scans are valid.

My issue is why the scans vary so much only a month later. I was wondering if anyone else has seen this, especially on TY media. Unfortunately I haven't had enough time to determine if it's only TY media or something else (such as heat because unfortunately the discs are stored around 86F during the summer).

[lordsmurf]

You fail to remember that the conditions of the drive and the surface conditions of the media can also play a factor here. This is why it's nothing more than voodoo, the condition for testing are not controlled.

[non-linear]

The only "scan" I think actually matters is a CRC Read Error Test. As long as you can complete that test without any read errors, then all the data blocks on the disc are readable and the data is intact.

[QuaiBoy]

The only "scan" I think actually matters is a CRC Read Error Test. As long as you can complete that test without any read errors, then all the data blocks on the disc are readable and the data is intact.

That's fine for data, but not for video. If a drive has to slow down to do error correction on a sector, that will show in playback. Also, a data/video disc might pass the CRC test and be "borderline"... lots of errors. A single scratch might cause a data loss.
I like the CRC test as well as a transfer rate test. The transfer rate test will show slowdowns due to the drive having to perform error correction. If the transfer rate is junk (rare with TYG02/G05s), then I test with Kprobe for entertainment purposes and usually scrap the disc.

-Evan-

[SCDVD]

You fail to remember that the conditions of the drive and the surface conditions of the media can also play a factor here. This is why it's nothing more than voodoo, the condition for testing are not controlled.

This is very true. IF the person doing disc scans is knowledgeable enough to do the scans in a properly controlled manner, the scans would be much more repeatable. The problem is that there are so many variables that affect a scan, which the person doing the scans isn't aware of, that the scans are not repeatable. Two of the biggest variables are temperature of the drive doing the scans and a damaged bottom surface on the disc. You can see the effect of drive temperature yourself by taking a disc and scanning it in a cold drive. When I say cold, I mean leave it off overnight and turn it on in the morning and immediately do a scan. Then on the same system and drive, burn several discs to heat up the drive and then repeat the scan on the original disc. You will see a change in the scan results. You can crosscheck the scan by repeating the scan on the same disc the next day or days on a cold drive and you will see that the scan resembles the other cold drive scans. So, which scan are you going to believe?

Another issue is the condition of the disc bottom surface. It is astonishing how badly some people treat discs. Fingerprints, dirt and scratches significantly affect scan results, yet some people seem oblivious to this and abuse discs and then wonder why the media is "so bad"!

Yet another issue is the condition of the drive used for scanning. Is it nearing the end of the life expectancy of the laser? Is the laser lens dusty? These variables will definitely affect disc scans.

A blood pressure tester is very useful in medically qualified hands. An untrained person may get some useful information from taking blood pressure, but if they don't know the impact of varying conditions such as intense exercise, lying down, standing up etc., they may jump to some wrong conclusions.

[nic2k4]

SCDVD has good points. You might want to add the storage factor. Discs need to be stored in dry cool dark place. The organic dye used in DVDs is sensitive to UV light, moisture and heat; placing your DVD rack in front of a sunlight window is not helping. The best place to store your discs is in its case, in a closet.

It's true, no two scans will be identical, but they will be similar. I would never think the error rates could double and that would be normal. If anything, I would suspect the drive or the disc is going bad. The only way to rule out the disc, would require scanning with another drive, but don't expect identical results, just similar (unless you scan with the same model drive). You can easily check the drive by burning another disc on the same media (preferably from the same batch) and compare the results with the old discs.

[Hawseman]

Quality scanning is not totally voodoo. The same arguments above can be made with all types of media testing.

Scans are true feedback of how a drive reads your disk. Since most burner manufacturers make ROM drives for DVD standalone players, it's nice to know how they "see" burned media. Agreed, the results will vary from uncontrollable circumstances, but where are these controls for CRC or transfer testing? It would be nice if all standalones and storage areas were kept at a constant temperature and humidity, but that's just not real-world. Lab conditions would be terrific, if all my clients lived in labs. If Ronaldus's media continues to degrade, he'll have a chance to save them. A luxury that he wouldn't have with CRC, pass-fail testing. That's what makes scanning valuable.

SCDVD has some good, noteworthy points (no need for me to repeat them). IMO, my BenQ is known for scanning inconsistencies. If you look at the design of the 1640 as compared to the 1620 you'll see some cooling mods have been added. The 1620 performs much better (burning/scanning) after a cold start. My Liteons' don't show such tendencies....although my NEC shows slight differences in very warm, humid weather.

I have rarely seen the error double unless the media has been handled a lot (micro scratches and dirt). Your scan is still good, and the transfer/CRC test shouldn't show any issues.....so just keep an eye on it. If it continues to degrade, use some different media.

Scanning is not exact (yes, somewhat voodoo), so it shouldn't be micro-analyzed. "Excellent, very good, good, fair, questionable, coaster" is about as far as I take it.

I use multiple scanning drives and never scan with a drive that I've burned with (it's like asking a cook if his cooking is good).

[Ronaldus]

I store my discs in a dark place away from the sunlight where they do not get touched by anyone else. Therefore, fingerprints, scratches, and dirty surfaces are non-existent issue on my scans. The only possible environment issue I have with my disc storage is that it stays in the 82-86 F range during the summer and in the 70-76 F range during winter. I know that's not the ideal range, but it's the best I can do without constantly running the AC. I'm hoping 86F is not bad enough to kill my discs quickly.

I did several more scans, including ones immediately after a cold start. I found that the average PIE/PIF counts increase as the BenQ drive warms up, but not by a huge amount (i.e. maybe 10-20 PIE higher and maybe 2 PIF higher on average)... less than 25% change. That accounts for some of the variation I saw, but not for most of it.

I re-scanned several discs I'd scanned about a month ago by various manufacturers to see if there was much change. The results I found were surprising. I'm not sure if I have enough samples to make a definite conclusion.

This is what I found:
- YUDEN000T02 burned at 4x on my BenQ and scanned on the BenQ hadn't changed much in the last month.
- YUDEN000T02 burned at 2x or 4x on my Pioneer A06 or A07 and scanned on the BenQ had more than doubled their PIE/PIF counts in the last month.
- YUDEN000T02 burned at 2x or 4x on my A06 and A07 back in January which I hadn't scanned before usually had PIE counts averaging 600-1200. Interestingly, the PIFs looked OK and there were no PO failures.
- RITEKG03, MXLRG03, RICOHJPNR01, TDKG02 burned at various speeds on the BenQ, A06, or A07 all had PIE/PIF scans within 25% of their previous scans... no significant changes.

My conclusion is either that:
(1) YUDEN000T02 media is poor and degrades quickly.
(2) The Pioneer A06 and A07, even with latest firmwares 1.09 and 1.21, do very poor burns on YUDEN000T02. What's weird is that the A06 and A07 burns initially start with PIE counts averaging 10 (very good), a month later they average 40, and if my discs from January are an example then 7 months later the PIE goes up to 600-1200. It may not be a linear degrading, but it's pretty fast.

Is there a logical explanation why a BenQ burn would be stable and a Pioneer burn on the same media would quickly degrade?

I'll do more tests and post. I bet nobody else has encountered this strange YUDEN000T02 phenomenon. The timing is interesting as I just bought several YUDEN000T02 Fuji color media packs at Staples recent sales. I should return them if YUDEN000T02 is unstable. At minimum I've learned only to burn this media on the BenQ, but I'd like to know why.

[Ronaldus]

I agree with others' comments that scanning on the drive that the media was burned on is not wise. I learned that from the floppy disk days where an out of alignment drive would produce beautiful results on that drive, but not work on other drives.

I've been wanting to get another error-scanning drive, but I'd prefer not to get another BenQ, especially a 1620, just so I have some variation in the hardware I'm using. However, it seems that Lite-Ons are not well liked any more, Pioneers still can't do PIE/PIF scans, and NECs have limited and less accurate error scanning. So what brand/model of drive should I get for my 2nd error-scanning drive?

I have become a big fan of error scanning as it helps steer me away from bad media, bad media-burner combinations, and it gives me early warnings about media likely to die. For instance, I probably need to redo all those YUDEN000T02 burns from the Pioneer drives where the PIE is way over 600. What a sad waste of discs!

[SCDVD]

Is there a logical explanation why a BenQ burn would be stable and a Pioneer burn on the same media would quickly degrade?

One possibility is that the drive manufacturers quickly abandon the firmware upgrades on "older" burners. If the media that you are using is not in the drive's firmware table, it will burn it using a default write strategy. About the only thing that you know for sure is that this default write strategy is NOT optimized for the unsupported media. If the write strategy is not correct for the media that you are using, it could be the best media in the world and you couldn't prove it in your burner. If a burner is a couple of years old, a significant percentage of the media currently on the market will not even be in the drive's firmware table. I strongly suspect that accounts for some of the "I have used Media brand ABC with xoxoxox media code and it works great in my Belchfire 101 burner but the new media from Media brand ABC is junk" comments we get.

[Shocker Milwaukee]

FWIW
Same disc - notice the dates...





After almost a year, I'm picking up a slight spike, but still a good scan.
The disc is clean & scratch-free.

[SCDVD]

Great scans. The small difference between them could also be because you scaned them on different drives. The first was done on a LiteOn 812 and the second one on a LiteOn 832. Drive to drive variation is often more than you see here.

[Ronaldus]

Scan of a disc burned on the Pioneer A06 on May 20th and scanned on the BenQ on June 5th:


Scan of the same disc on the same BenQ drive on August 5th:


Notice how the PIE doubled, especially near the end in only 2 months. To me this is proof positive of a degrading disc and beyond what can be accounted for by the drive simply warming up. So far I've only seen it on TY media. I'll keep doing more tests.

I'd buy the argument that the Pioneer A06 and A07 weren't optimized for the latest firmware except the first burn looks pretty good. You'd think if it started out good, it would stay good.

[Ronaldus]

It looks like nobody knows why the PIE is increasing on my TY Pioneer burns, but not my BenQ burns. I guess I'll just have to keep monitoring.

In the meantime, I'd like to get a 2nd drive for PI/PO error scanning drive. I don't want to get another BenQ 1620 just for variety. I hear that Lite-Ons are not well liked any more, Pioneers still can't do PIE/PIF scans, and NEC 35xx models have limited and less accurate error scanning.

Could someone recommend the next best error scanning drive that isn't a BenQ 1620? For instance, should I get a BenQ 1625 or 1640 or a NEC 3520 or 3540?

[Hawseman]

Personally, I'd suggest the Liteon 1693. It's a very solid scanner, ripper and reader......burning is improved, but still not to the quality of the Pioneer, NEC, LG or BenQ.

[Shocker Milwaukee]

Great scans. The small difference between them could also be because you scaned them on different drives. The first was done on a LiteOn 812 and the second one on a LiteOn 832. Drive to drive variation is often more than you see here.

Same drive - new firmware.

[chas0039]

Could someone recommend the next best error scanning drive that isn't a BenQ 1620? For instance, should I get a BenQ 1625 or 1640 or a NEC 3520 or 3540?

Any of the BenQ drives will do a decent job scanning but any of the Liteon drives will also do a good job and will scan the PIF errors at more frequent intervals. I also would recommend the 1693. It is good at scanning and unlike earlier Lieon burners, it is a very good burner on quality media.

NEC drives can be made to scan but they leave a lot to be desired and should not be relied on.