MediaTek chipset bag and quality scans



On CDR-Zone.COM you can find article “Verbatim 16x DVD-R review”

and on the last page the author wrote:

…While looking at the results from the Disc Quality tests you may notice rather alarmingly low scores on some results and may be wondering how come CDR-Zone.COM are giving the Verbatim 16x DVD-R media a good rating. This can best be explained by first explaining how CD Speed comes to its Quaity Score. It is based soley on maximum reported PIF errors. So if single spikes similar to what have been reported by the Liteon-It 1653S whilst scanning are found the Disc Quality score really does take a beating. Although the Quality score has to be based on something, it does not take into account the known bug in the MediaTek chipset used by the this particular Liteon and hopefully the transfer and playback tests are proof enough that these single spikes really do not effect the final recording. If you ignore this ‘bug’ and remove these spikes from the equation, the Quality scores would be in the 90’s…

I have 1653S and would like to know:
-how this bag affects quality scans
-are scans reliable with this and other 3S Family models
-are 1S-5/2S Family and 3S-7 Family better for quality scans

1S-5/2S Family

3S Family

3S-7 Family



@ miro68
“how this bag affects quality scans” - This ‘bug’ refers to single spikes in kprobe and CD-DVD quality scans that appear where a drive decouples and recouples after a write-speed-shift or a (W)OPC recalibration. These single spikes can be discounts as they do not affect the readability/playability of the disc. They happen in all Liteon drives which have Z-CLV write strategies where decoupling and recoupling of the laser takes place. I have even seen some CD-DVD quality scans of NEC burns that exhibit this ‘bug’ so I don’t know if it is only Liteons which exhibit this attribute. So, the quick answer,“no, they do not affect the quality of the burn, just the scan”.

“are scans reliable with this and other 3S Family models” - yes scans are ‘reliable’ to the point of that the drive is scanning/seeing what it sees at the time of the scan. Some drives scan/see better than others. These scans are just a generalization of the burn quality without using expensive machinery to judge disc burn quality.

“are 1S-5/2S Family and 3S-7 Family better for quality scans” - No one family is better than the other IMHO. There have been several cases of poor-scanning 411/811s, but in general, all should scan with generally equal ability. However, it is best to compare scans done with the same drive and firmware to eliminate differences between drives as no two are the same…in general…:).

Scanning is definitely not an exact science, particularly when using dvdrw drives versus expensive machinery designed for scanning only. if you are interested in more info, have a look at Help understanding CD/DVD Speed’s “Score” , CDSpeed v3 Disc Quality Test and The Big KProbe2 Thread


Thanks please for this fine explanation.

I though that this bug affects 1653S drive as CDR-Zone.COM wrote at their review.
(…known bug in the MediaTek chipset used by the this particular Liteon…)

Now it’s more clear for me. If I understood it I choosed one picture and mark regions where these spikes are.

Because of bad quality scans on Verbatim -R x8 (MMC 02RG20) and other
media I returned my 1653S to the shop and have to wait up to 2 weeks
for new one. Now I have an option to choose between 1653S and 1673S.

And now I don’t know which drive to take. Which is better?


@ miro68
Yes your graph with the arrows is exactly what they are talking about, and you can see the ‘quality score’ of 64 is greatly affected by these spikes, yet that is a great burn! I would score that about a 93 on the ‘please’ scale:).


Don’t assume that because one reviewer calls these spikes a “bug” that it’s not going to affect playback. These spikes are real, and can cause playback glitches in some of the more picky players. With other players, and particularly PC drives, they seem harmless. The occurance of these spikes seems to be related to burn speed, they can occur on discs burned in a variety of burners. Whether a player can cope with them is directly related to it’s error correction abilities.


Platinium -R x8, rec. x8
Three scans of the same disk recorded on Lite-On 1653S CS09
What do you think about these scans?
Three single spikes are shown by KProbe not by CD-DVD Speed.


CDSpeed is designed to delete certain types of spikes. Kprobe can also be set to ignore some single spikes, but they each work a bit differently. Not all of these sort of spikes are caused by WOPC functions. Some can be due to chipset reading glitches, some are not. Most are “true” PI/PO errors.


When testing recorded disks I noticed that media recorded on my BenQ 1620
looks better while sanning them in BenQ drive than in Lite-On 1653S.
On the other hand disks from Lite-on have less errors scanning them in 1653S
than in BenQ.