VideoHelp Forum

Try DVDFab and download streaming video, copy, convert or make Blu-rays,DVDs! Download free trial !
+ Reply to Thread
Results 1 to 13 of 13
Thread
  1. Member bmwracer's Avatar
    Join Date
    Sep 2003
    Location
    Driver's Seat
    Search Comp PM
    Hi all,

    Been reading a lot of good things about HC Encoder, but I haven't tried it our yet.

    Has anyone tried the various profiles (Fast, Normal, Best) and compared results?

    Thanks in advance.
    Frank Zappa: "People wouldn't know a good movie if it smacked 'em in the face."
    Quote Quote  
  2. Member MysticE's Avatar
    Join Date
    Nov 2003
    Location
    United States
    Search Comp PM
    Best is best.
    Quote Quote  
  3. Member bmwracer's Avatar
    Join Date
    Sep 2003
    Location
    Driver's Seat
    Search Comp PM
    Uh, yeah.

    Does that mean 100 vs. 95?

    100 vs. 75?

    100 vs. 50?

    How much better is "Best" over the other profiles?
    Frank Zappa: "People wouldn't know a good movie if it smacked 'em in the face."
    Quote Quote  
  4. Member AlanHK's Avatar
    Join Date
    Apr 2006
    Location
    Hong Kong
    Search Comp PM
    I only tested this briefly. "Best" wasn't much slower than "Fast", so I just always use "Best".

    If you want a quicker encode, you can use the Constant Q single pass mode. In this the final file size is unpredictable exactly. I use this when I want to watch something once and just burn to a DVDRW.

    Hank, the author of HC, can be found posting on Doom9
    Quote Quote  
  5. Member
    Join Date
    Dec 2004
    Location
    Triptonia
    Search Comp PM
    Yes 'best' is best.

    how do you propose to compare/measure quality?

    as for speed,
    i'd put the difference between 'best' and 'normal' at around 25%,
    that's a significant amount,
    and if you take my estimate of roughly an 8% quality diff,
    it's more than a fair tradeoff to switch 'best' quality for 'normal' speed,
    and one might say, that if you don't plan on milking every single bit,
    'normal' is the profile to go.

    'fast' probably offers an equal 8% quality drop over 'normal', for a 8% speed boost.
    not so interestng, unless speed takes the highest priority.

    HC is fast!

    with quality a concern though,
    you should really do a few tests of your own,
    rate quality in your own way,
    and report back with any findings,
    for future reference.

    will also give you a chance to see if i've been pulling all these figures out of a hat


    tripp
    Quote Quote  
  6. Member bmwracer's Avatar
    Join Date
    Sep 2003
    Location
    Driver's Seat
    Search Comp PM
    Ran various encodes using different profilesin HC and didn't see much of a variation between them...

    I did, however, notice that all profiles exaggerated the moire pattern in various scenes with repetitive patterns like the brick walls of a building... The original source has some moire as well, but it's considerably less obvious.

    Using the encoder in TMPGEnc, I did not see an exaggeration of the moire pattern.
    Frank Zappa: "People wouldn't know a good movie if it smacked 'em in the face."
    Quote Quote  
  7. Member AlanHK's Avatar
    Join Date
    Apr 2006
    Location
    Hong Kong
    Search Comp PM
    Originally Posted by bmwracer
    Ran various encodes using different profilesin HC and didn't see much of a variation between them...

    I did, however, notice that all profiles exaggerated the moire pattern in various scenes with repetitive patterns like the brick walls of a building... The original source has some moire as well, but it's considerably less obvious.

    Using the encoder in TMPGEnc, I did not see an exaggeration of the moire pattern.
    Was that both using the same AVS file?

    In HC there are a lot of matrices you can choose, aside from the default "mpeg". This may have an influence on things like moire -- I've never gone into that myself. I've never found any explanation of what each matrix is supposed to do, other than the usually cryptic filename.
    Quote Quote  
  8. In my experience with the encoder (using the GUI), BEST was not much slower than FAST in version 0.22. Version 0.21 took about an hour longer on a near two hour project (resized to 720x480, telecide, slowed down to 23.976) but you still need 0.21 for some projects (such as the aforementioned one) where 0.22 did something strange (still have no idea what is was). Hopefully 0.23 will be fast and correct whatever problem cropped up in 0.22.
    Quote Quote  
  9. Member AlanHK's Avatar
    Join Date
    Apr 2006
    Location
    Hong Kong
    Search Comp PM
    Originally Posted by ecc
    I where 0.22 did something strange (still have no idea what is was).
    There have been a few bugfix releases since then, though not yet an official .23.
    Check the Doom9 thread.
    Quote Quote  
  10. Member bmwracer's Avatar
    Join Date
    Sep 2003
    Location
    Driver's Seat
    Search Comp PM
    Originally Posted by AlanHK
    Originally Posted by bmwracer
    Ran various encodes using different profilesin HC and didn't see much of a variation between them...

    I did, however, notice that all profiles exaggerated the moire pattern in various scenes with repetitive patterns like the brick walls of a building... The original source has some moire as well, but it's considerably less obvious.

    Using the encoder in TMPGEnc, I did not see an exaggeration of the moire pattern.
    Was that both using the same AVS file?
    Yup.

    Though I'm using FAVC as a front end for HCenc.

    In HC there are a lot of matrices you can choose, aside from the default "mpeg". This may have an influence on things like moire -- I've never gone into that myself. I've never found any explanation of what each matrix is supposed to do, other than the usually cryptic filename.
    Unfortunately, I don't have access to the matrices using FAVC.

    EDIT: found the Matrix box... I'll try 'em out and report back .
    Frank Zappa: "People wouldn't know a good movie if it smacked 'em in the face."
    Quote Quote  
  11. Member
    Join Date
    Dec 2004
    Location
    Triptonia
    Search Comp PM
    Originally Posted by bmwracer
    The original source has some moire as well, but it's considerably less obvious.
    sample?
    Quote Quote  
  12. Member bmwracer's Avatar
    Join Date
    Sep 2003
    Location
    Driver's Seat
    Search Comp PM
    Originally Posted by 45tripp
    Originally Posted by bmwracer
    The original source has some moire as well, but it's considerably less obvious.
    sample?
    I was afraid you'd say that.
    Frank Zappa: "People wouldn't know a good movie if it smacked 'em in the face."
    Quote Quote  
  13. Member bmwracer's Avatar
    Join Date
    Sep 2003
    Location
    Driver's Seat
    Search Comp PM
    Pretty much tried all the various setting combinations (VBR, CQ, Predictive, Fast, Normal, Best) using FAVC and I still get that distracting moire pattern.

    I run TMPGEnc 2.5 in Constant Quality mode and the moire is no worse than that from the original source and the encoding doesn't appear to be that much slower than HCEnc.

    Guess I'm sticking with TMPGEnc for now...
    Frank Zappa: "People wouldn't know a good movie if it smacked 'em in the face."
    Quote Quote  



Similar Threads