Welcome to Doom9's Forum, THE in-place to be for everyone interested in DVD conversion. Before you start posting please read the forum rules. By posting to this forum you agree to abide by the rules. |
19th June 2005, 07:31 | #1 | Link |
Doom9 Member
Join Date: Sep 2003
Location: Australia
Posts: 210
|
HC Encoder & Matrices (Request For Sticky)
Hi guys,
First of all, big thanks to Hank315 and all testers for providing this community with such a good MPEG2 encoder. I currently use my favorite encoder Canopus for all my DV encoding, for my personal DVD backups I've recently used HC v0.15 and I was very surprised with the quality. However, reading the forum there is no clear information on which matrices should be used with which source. Except this clarification: 1."MPEG" - For higher bitrates 2. "QLB" - good for mid to lower bitrates 3. "HC" - For very low bitrates (unverified) It would be nice if we can get some info about ALL of the matrices included in HC and when/how to use them. Also BIAS settings should be explained as well. Hank mentioned that he got some very good results with BIAS 30, but he didn't explain on what source, DVD, DV.... Looking forward to test this encoder as it's output is simply amazing. Here's some info about some matrices that I found: QLB v1.2 ---- . first seen in QuEnc(Mpeg 1/2 encoder) . matrix designed by dragongodz Soulhunters v3 (MPEG-EX) HBR ---- . Its just the high bitrate version of the standard MPEG4 matrix... . It was made to encode movies @ 1024x576 pixels -> 1x DVD (4.5GB) !!! Soulhunters v6 (HVSv2) LBR & v8 (HVSv3) LBR ---- . They are mid/low bitrate CQM's... . Made for the usually 1 or 2 CD encodes !!! SxGz (DVD v3) ---- . MPEG2 single/dual layer DVD matrix . High bitrate matrix SxGz (HB v3) ---- . MPEG4 1 DVD matrix . High bitrate matrix SxGz (LB v4) ---- . MPEG4 1-2 CD matrix . Low bitrate matrix SxGz (MB v7) ---- . MPEG4 2-3 CD matrix . Medium bitrate matrix SxGz (SVCD v3) ---- . MPEG2 SVCD/½ DVD matrix . Low bitrate matrix Y.A.C.Q.M. v2 (ANIME) ---- . It was made for mid/high bitrate anime encodes... . It works best with clean content (~2500kbps @ 640x480) !!! Here's link to some matrices(if not all)Matrices Regards, DSP8000 Last edited by DSP8000; 19th June 2005 at 08:19. |
19th June 2005, 08:47 | #4 | Link |
Doom9 Member
Join Date: Sep 2003
Location: Australia
Posts: 210
|
Suggestion For Sticky
Matrices should be in different gorups like:
Low Bitrate 2000~4000 Medium/High 4000~6000 High/ExtraHigh 6000~8000 Motion Based High,Low Source Based CG,Anime,DV,DVD... Creator info & recommendation Matrices DVD Compliant Matrices VCD/SVCD Compliant Also BIAS Settings explanation(as in manual, maybe a bit more) If anyone has anything else to add, please do. Regards, DSP8000 |
19th June 2005, 16:41 | #5 | Link |
Gold (Au) Member
Join Date: Apr 2005
Location: USA
Posts: 832
|
Here is a link to more Matrices.
I agree that having some explanation of general use would be helpful. Are you using them with DVD Rebuilder Pro v.93.2, by the way? Regards,
__________________
*********************************************************** Don't assume anything...Verify! It's as easy as 3.14159265 The Problem is Usually found between the Chair and the Keyboard! |
20th June 2005, 00:40 | #6 | Link | |
Registered User
Join Date: Mar 2005
Posts: 36
|
Quote:
HC is classified as 'for very low' bitrates... but I assume this would be 2000~4000... going a class lower with mpeg2 wouldnt be reasonable. hmm so it seems for encoding for example 8 xvid 40min-episodes to 1 DVD the "HC" matrix could be advised. I didn't even know there was a matrix with this name! I have not been copying any dvds/xvids2dvd since HC came out... just because I first wanted to know what settings were recommended for what situation.. Last edited by zilexa; 20th June 2005 at 00:44. |
|
20th June 2005, 03:31 | #7 | Link |
Doom9 Member
Join Date: Sep 2003
Location: Australia
Posts: 210
|
@feedback,
I'm using the free version of DVD-RB, but I'll purchase it when it comes out v1.00. Yesterday I did some testing with QLB2,AUTOQ2,MPEG on a certain movie & definately the matrices make big difference.I can't coment till I test all of them.I suggest to everyone to do the same thing coz, what is good,better,perfect for my eyes may not be for someone else.Also different matrices perform different @ given AVG bitrate 2000,4000,6000. I'm also testing the so called "Free CCE trial encoder - QMat...." DSP8000 |
20th June 2005, 12:42 | #9 | Link |
....
Join Date: May 2002
Location: Australia
Posts: 2,797
|
the bitrates for groups is inherently flawed because you then have to take in to consideration the type of footage being encoded, is it interlaced, is it high motion, is it animation or cg or live, is it grainy or clean ?
rail grinder - try here Code:
http://www.rapidshare.de/
__________________
Narrator: And of course, with the birth of the artist came the inevitable afterbirth - the critic. (History of the World part 1) |
21st June 2005, 03:16 | #10 | Link |
Registered User
Join Date: Nov 2003
Posts: 21
|
ok so here goes. the files are currently being hosted on my company's webserver. if the traffic gets too high i'll remove them temporarily or host them on my personal server.
anyways so here goes ... the comparisons are done on an anime (Samurai Champloo) intro. i've included screenshots of what i thought would be relevant segments. if you have any objections to my choices, by all means correct me . in each file you'll find the comparisons for each segment and a short "preview" clip just so you guys know what kind of motion is happening in each segment. i based the comparisons on: - the original source - hc 0.14 (cce smooth - cg) - hc 0.15 (cce smooth - cg) - hc 0.15 (default mpeg) the scenes are: - low motion - text -> slow scrolling text with a slow scrolling tapestry [here] - low motion - clean lines -> very low motion scene with hard lines and solid colours [here] - high motion -> scrolling backgrounds, fast animation and some solid colours [here] - high motion - blocky -> super fast scrolling scene with macroblocks in the source [here] i'll let you guys/gals be the judge but to me it looks like hc acually made things look a whole lot sharper. as for ini files and configuration, here they are: hc 0.15 (default - mpeg) Code:
*infile d:\....\test.avs *outfile d:\....\test.m2v *bitrate 4000 *maxbitrate 9000 *profile best *aspect 4:3 *bias 30 Code:
*infile d:\....\test.avs *outfile d:\....\test.m2v *bitrate 4000 *maxbitrate 9000 *profile best *aspect 4:3 *bias 30 *custommatrix Code:
*infile d:\....\test.avs *outfile d:\....\test.m2v *bitrate 4000 *maxbitrate 9000 *profile best *aspect 4:3 *custommatrix Code:
[Options] SkinVersion=3 RemoveDTS=1 QTrellis=1 QKVCDOpt=0 QuEncHQ=1 QuEncScene=1 Mode=2 ReduceOpt=3 HalfD1=00 Convert_16_9=00 DisableInterlace=00 idct7Opt=1 DECOMBFlag=1 Completed=16 LogFile=1 OneClick=1 max_bitrate=7125 TargetSectors=2260000 CCE=3 QSinglePass=0 [Paths] QuEnc=C:\Program Files\DVD-RBv091Free\HC_015\HCbatch_015.exe MPEG2DEC=C:\Program Files\DVD-RBv091Free\MPEG2Dec3dg.dll DECOMB=C:\Program Files\AviSynth 2.5\plugins\Decomb521.dll Source=C:\SC3_PREP\ Working=H:\DVD WORKING\ Output=H:\DVD WORKING\ CCENEW=C:\Program Files\Custom Technology\CCE SP\cctspt.exe [Audio] Selected=11 [Subpictures] Selected=11 [CCEOptions] VBR_bias=30 Quality_prec=24 eclPasses=10 |
10th July 2005, 22:48 | #12 | Link | |
Gold (Au) Member
Join Date: Apr 2005
Location: USA
Posts: 832
|
Quote:
If you go to this thread you will find an app. that will at least give you the chance to determine what is the best Matrix for your particular DVD. It is to be used with DVD Rebuilder Pro. You will be able to observe the effects of different Matrices with a short segment rather than having to encode the whole mpeg2 first. I am encoding the same short segment with various Matrices to judge what is best for me. Maybe some posters could post their own observations about the different Matrices. It is worth repeating that you can also download a Matrix zip folder Here. There is a different Matrix zip folder Here. Regards,
__________________
*********************************************************** Don't assume anything...Verify! It's as easy as 3.14159265 The Problem is Usually found between the Chair and the Keyboard! |
|
Thread Tools | Search this Thread |
Display Modes | |
|
|