View Full Version : How to set Q factor?
telemike
2nd May 2003, 12:53
I want to use CCE in DVD2SVCD.
Max bitrate = 2500
Min bitrate = 500
I want to fit most movies on 2 CD's using 1-pass VBR (similar to doing CQ using TMPGenc).
How do I determine the right Q factor?
Thanks
JimBryce
5th May 2003, 08:10
Well...:(
I'm sure you're going to be told "Rule 1" :D by someone and I'll probably get blasted:scared: for telling you but...
While I can't begin to say that I totally understand "Q Factor", I believe that it is the same as what's called "Quantizer characteristics (Flat part priority)" in both v2.66* and v2.67* (someone correct me if I'm wrong, please!:confused: ) I used the following guideline from the CCE section of DVD2SVCD advanced reference (http://www.doom9.org/mpg/dvd2svcd_advref.html):
Specify the output picture by the Q. value in Q.factor and Bitrate. Q. is a concept unique to Cinema Craft Encoder. The range of the specification is 1 ~ 300. As the Q. value becomes smaller, image quality improves, and as the value becomes larger, higher compression is used. For the setting, refer to the following.
Q.factor Description:
1 ~ 40 Priority is given to image quality over compression rate
40 ~ 80 Standard setting
80 ~ 120 Priority is given to compression rate over image quality
120 ~ ? Image quality deteriorates considerably
Tip to encode quickly with great quality 2500bitrate, many CDs though, try Q.factor 3
I'm using "20" and have had very nice results. I have also turned off the other settings under "Datailed setting (btw, the spelling is exactly as written!) i.e., Low pass filter, Effect restricted vertical filter, Dithered quantization, since I use Avisynth to make any changes/clean-up to the video (again, if I've mis-understood what these mean, let me know!:confused: )
Originally posted by JimBryce
While I can't begin to say that I totally understand "Q Factor", I believe that it is the same as what's called "Quantizer characteristics (Flat part priority)" in both v2.66* and v2.67* (someone correct me if I'm wrong, please!:confused: )
No, the Q factor is an option unique to one pass VBR encoding. Telemike, you may want to look at the "D2SRoba" plugin at http://home.no.net/tylo/
Originally posted by telemike
I want to use CCE in DVD2SVCD.
Max bitrate = 2500
Min bitrate = 500
I want to fit most movies on 2 CD's using 1-pass VBR (similar to doing CQ using TMPGenc).
How do I determine the right Q factor?
Thanks
Originally posted by JimBryce
I'm sure you're going to be told "Rule 1" :D by someone and I'll probably get blasted:scared: for telling you but...
and doing a search, you would want to look for a program that FamousPerson wrote, which is a CCE 1 Pass VBR Guesser
after dropping the 1, the search for the term in italics brings up 1 thread, http://forum.doom9.org/showthread.php?s=&threadid=34521&highlight=CCE+Pass+VBR+Guesser
of course, the program is a 'guesser', not a 'calculator', so its a little harder to find than what you'd search for normally, so im not going to blast anyone, yet.
Enf...
JimBryce
6th May 2003, 02:10
From RB:
No, the Q factor is an option unique to one pass VBR encoding.
Thank you! Your reply prompted me to RTFM (again). So my answer is correct for "Q Factor" but that is completely different from "Quantizer characteristics" (just repeating to make sure I understand). I'm now looking up "Quantizer characteristics".
According to the manual, "Quantizer characteristics": "As the value becomes closer to 0, a higher bit amount is allocated to complicated images areas. As the value becomes closer to 64, a higher bit amount is allocated to flat image areas." Okay, I can follow that, but then they end the description with: "If the bitrate is relatively high, greater value may cause better result." That confuses me. Don't both states (towards 0 and towards 64) increase the bit rates? Or is this an ESL (English as a Second Language-no insult intended) and what they're trying to say is that the closer to 64, the better the quality?
As I said, I am searching this one out, but any other input you have (or anyone) would be appreciated!
This options affects how bitrate is distributes between "flat" and "complicated" areas on a per-frame basis. That is, it does not affect the global bitrate but how available bits are allocated in a single frame. I think the description is quite clear: with a lower value, you have less artifacts around sharp edges (lots of contrast) but banding and blocking effects are more likely to occur in flat, evenly coloured areas of the frame (fog, walls etc.). The effects reverse as you use a higher value. And the recommendation to use a higher value if the bitrate is high is all right because with a high bitrate there are enough bits that can be allocated to high-contrats parts of the image (there are usually not so many) anyway.
As I'm now doing backups to DVD-R, I set this value to 30 and always had excellent results.
JimBryce
7th May 2003, 02:50
:)Thanks RB, for the reply!:)
To ensure that I understand what you (and CCE! :D ) are saying, it's not that the total amount of bits change, rather the distribution of the bits available.
Assuming the effect is linear (what is that simple these days? :confused: ) then a value of "32" would have the bits distributed "evenly" between the complicated areas and the flat image areas. Since you're working with DVD-R, you have sufficient bits available to get good results at either extreme, whereas since I'm working with SVCD I might want to go to around 20 (17~22 - need to experiment with it) to "help" the more complicated areas since the flat areas will have a greater tolerance for less bits. Correct?
While I "have" you here, RB, and you seem pretty knowledgable about video in general and CCE in particular, let me ask you another question.
I've been doing whatever "cleaning" the picture needs through Avisynth and its filters and judging the results before encoding. If I have the result I want, then I should disable the "Datailed setting"* (Low pass filter, Effect restricted vertical filter, and Dithered quantization) in the "Quality Settings" of CCE, else risk more changes beyond what I'm already expecting. Especially if I'm already where I want to be(?).
Thanks again for your help, RB and my apologies to telemike if we got slightly off track from his original post, but I think this has been beneficial and in the "spirit" :D of the original...
* ...gotta love ESL... If you knew that people who speak 2 languages are bilingual and people who speak 3 languages are tri-lingual, what do you call someone who speaks one language? ...American!
Yes, you are correct in both your assumptions.
Originally posted by JimBryce
...gotta love ESL... If you knew that people who speak 2 languages are bilingual and people who speak 3 languages are tri-lingual, what do you call someone who speaks one language? ...American!
LOL :) But don't you use the word "Kindergarten", BTW? It's a German word.
Holomatrix
7th May 2003, 22:32
Originally posted by telemike
I want to use CCE in DVD2SVCD.
Max bitrate = 2500
Min bitrate = 500
I want to fit most movies on 2 CD's using 1-pass VBR (similar to doing CQ using TMPGenc).
How do I determine the right Q factor?
Thanks
Are you planing on having CCE create the vaf? if yes, then why don't you just select multi-pass , pass= 1. Then you don't need to worry about Q, let the bitrate do it's thing. if no, then nevermind what I said.
telemike
12th May 2003, 17:45
Thanks-
I am just going to use a Q factor of 5 after testing some different settings. I prefer to have my bitrates set to 2200 max, 300 min for SVCD's.
Ran 4 test runs using DVD2SVCD on a DVD music video that is about 7 minutes long.
CCE 2.50ran twice as fast as TMPGenc 2.51
CCE Q5 = 121,341KB
CCE Q40 = 115,420KB (semi-blocky)
CCE Q80 = 102,885KB (blocky - vcd quality)
TMPG CQ80 = 125,072KB
All settings were same except for Q value for CCE
Max bitrate = 2500
Min bitrate = 300
SVCD 480 X 480 mpeg-2
Holomatrix
12th May 2003, 17:52
Originally posted by telemike
Thanks-
I am just going to use a Q factor of 5 after testing some different settings. I prefer to have my bitrates set to 2200 max, 300 min for SVCD's.
Ran 4 test runs using DVD2SVCD on a DVD music video that is about 7 minutes long.
CCE 2.50ran twice as fast as TMPGenc 2.51
CCE Q5 = 121,341KB
CCE Q40 = 115,420KB (semi-blocky)
CCE Q80 = 102,885KB (blocky - vcd quality)
TMPG CQ80 = 125,072KB
All settings were same except for Q value for CCE
Max bitrate = 2500
Min bitrate = 300
SVCD 480 X 480 mpeg-2
Was the CCE Q5 encode better quality than the TMPEG CQ80 encode?
telemike
12th May 2003, 20:47
I would say it was roughly the same for TMPGenc CQ80 and CCE Q5.
vBulletin® v3.8.11, Copyright ©2000-2026, vBulletin Solutions Inc.