which do you guys use

[techdude]

hi guy i have a question i am using dvd2svcd program with tmpgenc plus to convert my dvd's to svcd and would like to know what's best to use constant quality or 2 pass vbr and what's the best for motion search precision. thanks for all the help

[Nick]

Hmmm. These "what's best" ones are tough to answer as what's best for me might not be best for you. (nb I think there is a forum rule requesting such things are not asked for this very reason)

I suggest 2-pass vbr above CQ as file size prediction is more reliable. Hence you tend to get the number of cd's you specified and all of them full - so maximum available bitrate is used.

CQ can be a bit hit and miss on this score.

I tend not to use tmpgenc over much so I'm no expert on Motion Search Precicion. All I know is that the higher the quality the slower it is so you have to find a balance between encoding speed and quality that you yourself are happy with. Start with Normal and see what that's like and work from there.

I'm afraid it's testing time because no-one can find the balance that works best for you except you yourself.

Sorry to be of little help but it ain't a simple question so there ain't a simple answer!

[UltimateDBZ]

For largely the same reasons has Nick has already mentioned, I too would recommend you use the 2-Pass VBR method. But then, I also recommend you use CCE :p

[jsoto]

Hi techdude,

I have been using TMPGEnc during a long time (before change to CCE, which is the best option, as Ultimate recommends). In my experience, CQ mode gets better quality than 2-VBR. And you have to wait less time.

About motion search precision, I've done some tests (with the same avisynth scripts and the same movies) and I was not able to distingish the quality differences between High and Highest. You can see the effects in encoding time in my post inthis thread (http://forum.doom9.org/showthread.php?s=&threadid=53740)

Anyway, as Nick has pointed out, using CQ mode the resulting size is not accurated. Many times I have had to reencode the video selecting manually a lower CQ...

Hope it helps.