Is it normal that the speed of scanning DVD-ROM discs in the DW1620 has all kind of spikes in it? I do not get those spikes when scanning DVDR discs.
Edit: I am referring to the variations in the read speed, not the PI/PO levels.
(And yes, I did do a search for this topic but I found nothing.)
This is with CDSpeed 3.61.
Even pressed cds have PI and PO failures, scratches as well as dust and fingerprints can cause higher spikes.
I’m sorry, I did not explain myself very well. I was asking about the variations in the read speed, not the PI/PO levels.
You’ll notice that the reading curve is not smooth at all.
Jucius_Maximus,
The reading speed curve on the graph you posted is characteristic of a dual layer (DL) DVD. Most commercial DVD movies are produced this way, which gives them almost double the capacity of a single layer (SL) DVD. The reading starts at the inside and gradually increases as it approaches the outer part of the disc. On a DL disc, when the outer limit of the first layer is reached, the reading is switched to the second layer. The second layer reading continues at full speed from the outer disc and reduces speed as it reads backwards to the inner part of the the disc.
The disc that you tested appears to be just marginally over the SL capacity limit. Most DL DVDs hold considerably more data when they are full. A full DL disc that uses all capacity will have a faster maximum read speed because it reads all the way to the outside of the disc.
If you test read a SL DVD, it will have a straight line ascending read speed as it reads one way to the outer disc and stops. 
I wan’t referring to the nature of reverse-spiral DL discs.
I was referring to the way the reading speed graph looks like a comb, indicating that it slowed down every few hundred MB and then returned to the previous speed.
When I scan on a Pioneer or Plextor, there is no ‘comb’ effect, just a smooth read all the way through. For example, if you look at the attached graph, the curve is smooth on both layers (except for a minor glitch.) Even when I had a Liteon 1633S installed, its Disc Quality tests had a smooth reading speed graph, as opposed to what the DW1620 returns.
Why does the DW1620 behave like this when scanning? Is it normal or is mine misconfigured or defective?
It is caused by scanning at Maximum speed using CD-DVD Speed. I think it may be the way the author is getting the drive to reach max speed.
You are comparing apples and oranges. First of all, the scan done with the Pioneer DVD-ROM is just a simple Transfer Rate test. The test run on the BenQ is the Disc Quality Test, which operates in a different mode. If you run the Transfer Rate test on your BenQ, you will get a similar graph without the descending lines.
As to why the descending lines on the Disc Quality Test, some BenQ drives seem to do this and some don’t. This was reported in the NEC forum testing beta firmware and using BenQ 1620s for testing PI errors. Dee-27, the moderation and thread author has two BenQ 1620s, one of which does the lines and the other doesn’t. Both drives perform normally in her opinion. That was her answer to another BenQ owner who reported the lines to her and asked if he had a problem.
My drive usually does one downward line near the beginning of the Disc Quality Test, but in testing a commercial DL disc to check the results I had three of these downware glitches. I don’t know the reason for it, but I would disregard it. If your drive is O.K. otherwise, I would consider this behavior normal.
If you want straight lines, run the Transfer Rate test. 
I realise that it is apples to oranges, I was simply trying to show the kind of transfer rate curve I expected in the disc quality test. I got the same kind of smooth graph in the Liteon 1633S disc quality test, but I can’t post a picture because I don’t have that drive installed right now. Therefore I posted the closest thing to that I could generate, which was the transfer test graph.
Though your second two paragraphs have answered my question, thank you.
You’re welcome.
The latest CD Speed v.3.61 allows the BenQ to be tested in 12x and 16x in the Disc Quality test. When I test the same DL disc from which I got 3 spikes in my earlier test at 8x, I get spikes throughout the burn just as you have experienced.
If you want to see lots of these spikes, go to the 16x testing at Nero CD/DVD Speed V3.61 released
It seems it’s not limited to test @max : some people reported even those “dips” when doing a scan at 8x and did not get the dips at 4x or 12x.
My own Benq 1620 B7T9 shows this dips only when doing a quality scan using CDSpeed 3.61 at 12x or at 16x, no dips at 4 or 8x.
My Benq drive works perfect in all ways, even the best quality burner I own out a bunch of over 10 DVD-writers I own. 
We’ll here’s an interesting development.
When I made the first scan, the drive was secondary master on a Silicon Image 680 ATA133 card using multi word DMA 2.
I moved it to secondary slave on my motherboard’s IDE controller, so now it’s on Ultra DMA mode.
Now I only get 1 dip on the first layer and the same dips on the second layer.
Go figure!
Your drive’s overall performance would have been limited (poor) under Multiword DMA Mode 2. It has to run in Ultra DMA Mode 2 to allow its maximum speed capability and minimum CPU utilization.
I’m glad you found out about this and fixed it.
That would be because it was on the end of a 26 inch Antec Cobra IDE cable. The ATA133 card refused to run it on UDMA. (Though I had a Liteon 1633S there before and it was running on UDMA 4 with the same cable!)
Unfortunately my mobo (Asus A7N8X Deluxe) will not run these things at anything higher than UDMA unless I install the NVIDIA IDE S/W driver, which is known to mess up other things. So I will eventually have to upgrade to a different mobo.