Back in February I was running short on DVD-RW discs and rather than buying more for temporary use I decided I'd pick up a small spindle of Maxell 16X discs to do some temporary burns (much cheaper than DVD-RW) on my Pioneer 633.
Although they were just temporary discs, I did my usual quality checking on them and found that these RITEKF1 discs burned very poorly in with the Pio 633. No matter, I needed them only temporarily anyway. But 3 months later I still hadn't finished with them and when I rechecked the burn quality it was way worse - bad enough that I actually re-copied them to a new set of 16X DVD-R discs (Verbatim MCC 03RG20) that burned much better.
I hung onto the RITEKF1 discs just to see what would happen them over the longer term, and now it's been 6 months and they're continuing to deteriorate, albeit at a slower rate than they did initially.
Here's an animation of three quality scans, the first right after the initial burn with the next two being 3 and 6 months later:
In the first 3 months, the total PIE error count increased by a whopping 216%. In the next three months, the rate increased by about a further 25%. Although this is only one disc, I actually burned a couple of dozen RITEKF1 discs that have all showed the same pattern of degradation. So there really is some sort of fairly serious and systematic "fade" going on with these discs (although whether "fade" is a good term for it or not is perhaps debateable).
Note that even now, when the disc shows PIE error rates above the maximum 280 specified by the DVD standard, the errors are still correctable and in fact the disc plays perfectly well. But if the deterioration continues then at some point the data will become unrecoverable. With this kind of deterioration it's very easy to imagine a bad burn that initially had error rates at around 1000 or more which increase over the first few months to a point where the disc can no longer be played. This would match the "disc fade" scenarios that a lot of folks have complained about.
I usually throw out discs whose initial burns are as bad as this one was, so I don't know if this is typical of poorly burned discs. But I can tell you that none of my other discs, including a few hundred RITEKG05's (which lot of folks have complained about), have shown any significant degradation from their initial burn quality in over 18 months (so far). From this I'm led to suspect that well-burned discs are less susceptible to this kind of degradation.
If nothing else, this will hopefully reinforce the idea that just playing your discs after burning them is really no guarantee of anything. As I've mentioned before, a playable disc may still be "just barely readable", and any degradation, particularly the kind seen here, could easily render the data irretrievable. The only way to be sure is to check the burn quality using your PC.
Edit - noted that it's been an 18-month track record for my good burns so far