SOHW-1633S v LH-18A1P

I bought a new drive, LH-18A1P, to replace an old one, SOHW-1633S, that was failing to read, and since I have some valuable archive data I decided to scan my DVD library with my new drive. Bad news: PIE and PIF off the chart, several coastered. I reburned everything using onhand media, mostly Memorex +RW (RITEK 004) and Maxell +R (CMC MAG 01), however PIE and PIF remained high (typical PIE max:600 average:400, PIF max:25 average:10.) I tried 2x, 4x, 8x,12x speeds with GLOC, GLOF and GLOG firmware, but results remained unpredictable and many I had to reburn more than once to get something acceptable. I checked Smartburn, DMA, there’s nothing else on the IDE bus, cables and connectors are clean & tight, no multitasking during burns, etc… :doh:

Later I recalled a college friend said he couldn’t afford to replace his DVD drive, so he took it apart, blew it out with canned air, reassembled it, and it worked great. On a whim I tried it, and the results were amazing. Using my old drive, I’ve rescanned much of my recently reburned media and I’m getting awesome results, although sometimes there’s a spike or two that show up in scan statistics as “repaired glitches”. I returned the new drive (it was also an October 2006 pre-updated spindle model that was very loud) and in the future I’ll buy TY media online, but some questions remain:

Can someone more experienced infer anything about what was the problem with the new drive by comparing the scans, or if you disagree with my evaluation, please tell me why?

What is a ‘glitch’, how does it affect data integrity, why does the glitch PIF spike not show up in PIE, and how is it repaired?

Are DVD cleaner discs like Allsop effective/safe, and would that help eliminate glitches?

Thanks, Jake

My bad, the first scan from a disk burned on a different day. I don’t see how to swap uploaded images, and all the new/old have similar characteristics, so it’s just the same. Everytime I compare these it surprises me–PIE/PIF are reduced about 80% with the (cleaned out) old drive, and with the exception of the razor thin glitches are far more consistent.

Above I was trying to A-B the new/old drives, but this tends to make my new drive look better than it is because of how many DVDs I rejected. For a better perspective of the bad scans with my new drive, here are a few of my bad rejects. While these graphs show a right skew, many also had a left or center skew. Sometimes the bad areas of the graph stood out like columns. Slowing the drive down improved scans, but I expect 16x media to be able to be read reliably at 8x in an 18x drive.

I found one answer for the definition part of the glitch question:
“Glitches” are typically drive command signals that can appear to evaluation software as error spikes. Some software will remove these spikes and not count them as errors. These types of command “errors” will appear in the same place on every disc tested, an indication that they are not related to the disc at all but come from the drive.

That suggests the way to “repair” a glitch is to ignore it (as scan statistics report, although not mention in the FAQ), however ignoring something is not usually considered repairing it.

While most of my DVD/CD’s have very low repaired glitches, one particular CD has more than usual. (see attached)