I am looking to good ripping software for converting DVD to avi or
mpeg4, etc..
Easy in use, but with good tweaking options and a good quality.
Tried some ripping software, but by very much applications within
the "wood"it is hard to find out just one good.
Did not get best satisfied results till now.
E.g. AoA DVD ripper, simple but not the most sharp movie. Image is a
little soft.
http://www.dvdkit.net/dvd_ripper.htm
YASAVideoConverter
http://www.yasasoft.com/videoconverter
Very good in some options. Far more sharp movies as AoA DVD ripper.
(Especially when text is in the movie).
Has much options for tweaking. But you have to be a pro to understand
the options. No single word about those pro options within the helpfile.
I was looking the tool for "deblocking" the mosaic artifacs in some fast
changing images. Because I could not find, it is a no no application.
So I am looking for a good application, with good quality image and much
and easy tools to understand tweak in a great range of options.
- sharp
- self tweaking pixel width and height / cropping options
- deblocking mosaic artifacts (Y UV)
- deringing (Y UV)
- remove interlace options
- set output frame rate etc.
--
Vr.groet - regards, Léon Obers
With all your "wants & don't wants" Leon you should be
ordering this software to be MADE to your "specs", chum.
Not scratching around in this lowly newsgroup among us mere mortals,
expecting miracles
...."silk purse" & "sow's ear" come to mind. :-)
Velly Solly - Mijn oprechte verontschuldigingen, me.
Seems unrealistic to ask for something that powerful and
flexible, but without the ability to control it.
> I was looking the tool for "deblocking" the mosaic artifacs in some
> fast changing images. Because I could not find, it is a no no
> application.
If you actually think the artifacts can be "deblocked" then
you don't understand how very lossy compression such as
is used on DVDs actually works.
> So I am looking for a good application, with good quality image and
> much and easy tools to understand tweak in a great range of options.
Let us know if you find such a thing. I'm not holding my
breath in anticipation.
Richard Crowley wrote:
> "Léon Obers" wrote ...
>> Has much options for tweaking. But you have to be a pro to understand
>> the options.
> Seems unrealistic to ask for something that powerful and
> flexible, but without the ability to control it.
That is not unrealistic at all to ask.
There are much applications with easy to understand tools and settings
to control deep inside very complicated processes for the "engine" behind.
At least it should be normal within the help manual how to use these
controls, but for the previous YASA application not one word about it.
It is like using MSDOS commands within (or against) a Windows user
interface.
>> I was looking the tool for "deblocking" the mosaic artifacs in some
>> fast changing images. Because I could not find, it is a no no
>> application.
> If you actually think the artifacts can be "deblocked" then
> you don't understand how very lossy compression such as
> is used on DVDs actually works.
I think you don't understand the problem itself.
The images from the DVD is not the problem, they do not contain the
blocky artifacts. It is the output after ripping the DVD to another
format and size that do contain blocky artifacts only at the point of
very fast changing movie "scenes".
Typically at more "slowly" changing movie scenes (the most of it) the
problem is not there at all. Tools as "Deblocking" should be help to
avoid these problems in fast changing movie scenes only.
It should be possibly as the basic material is without errors.
For the sake of quality images for compressed output maybe less
compression only at those fast changing scenes. A smart application
should look "forward" to detect these fast changing scenes and change
the policy for converting. So a complete movie is converted and
compressed in the most efficient way for a general even quality within
the possibilities of a given general compression setting.
>> So I am looking for a good application, with good quality image and
>> much and easy tools to understand tweak in a great range of options.
> Let us know if you find such a thing. I'm not holding my
> breath in anticipation.
I hoped power users in these newsgroup could help with their experience,
but seems the only one that are posting to a question don't have
experience either like me.
YASA? You need NASA, mate. :-)
sandy58 wrote:
> With all your "wants & don't wants" Leon you should be
> ordering this software to be MADE to your "specs", chum.
Strange that in the big "wood" of all existing applications already, no
application should exist to these specs already and I should order one
to MAKE an extra application for my purpose, only for ripping a few
movie DVD's in the best way.
My friend is a moviemaker, but has no technical experience to rip his
DVD to smaller data to set his movie in the best way to the web.
See movie:
Een downloadable "trailer" als vooraankondiging voor diverse filmpjes
die eind vorig jaar is gemaakt, waarin je iets van de sfeer kunt
oppikken hoe de filmpjes van opzet zijn.
http://tinyurl.com/32kslx
Inmiddels hebben we een nieuw filmpje wat we met de beste opties op het
web willen zetten (in een eigen website).
It would be unrealistic to expect Richard Crowley to
understand Motion Estimation and Deblocking. He gets
hung-up on his phobia of the lossyness of temporal
compression, and can never realize that it can be done
well, without the disaster he fears so much.
But he is right, that you can't expect to effectively address
complex technical issues with a simplified interface. You
don't have to be a pro, but you do need to understand the
effect of the "tweaks"/settings. One way to get better
performance from the default settings, is to pay more for
the program.
You also might consider that there is not a lot of
commercial incentive to develop such a program, at least
not for any major player. Not with the law suites from
the movie industry.
You could break up what you want into separate
processes and acquire a better encoding program like
TMPGEnc 4.0 XPress, which goes to some lengths to
make the process a little easier. But that wouldn't be the
integrated software you seem to be asking for. (Generally
speaking the more integrated; the less technically capable
the parts.)
Luck;
Ken
>Tried some ripping software, but by very much applications within
>the "wood"it is hard to find out just one good.
>Did not get best satisfied results till now.
>
>YASAVideoConverter
>http://www.yasasoft.com/videoconverter
>
>Very good in some options. Far more sharp movies as AoA DVD ripper.
>(Especially when text is in the movie).
>Has much options for tweaking. But you have to be a pro to understand
>the options. No single word about those pro options within the helpfile.
>I was looking the tool for "deblocking" the mosaic artifacs in some fast
>changing images. Because I could not find, it is a no no application.
>
>So I am looking for a good application, with good quality image and much
>and easy tools to understand tweak in a great range of options.
>- sharp
>- self tweaking pixel width and height / cropping options
>- deblocking mosaic artifacts (Y UV)
>- deringing (Y UV)
>- remove interlace options
>- set output frame rate etc.
It seems you already have what you need to do the job.
All you have to do is learn how to use it.
Ken Maltby wrote:
> "Léon Obers" <mail....@invalid.invalid> wrote in message
>>Richard Crowley wrote:
>>>"Léon Obers" wrote ...
[...]
>>>If you actually think the artifacts can be "deblocked" then
>>>you don't understand how very lossy compression such as
>>>is used on DVDs actually works.
[...]
> It would be unrealistic to expect Richard Crowley to
> understand Motion Estimation and Deblocking. He gets
> hung-up on his phobia of the lossyness of temporal
> compression, and can never realize that it can be done
> well, without the disaster he fears so much.
It seems you no him better than me ;-)
> But he is right, that you can't expect to effectively address
> complex technical issues with a simplified interface. You
> don't have to be a pro, but you do need to understand the
> effect of the "tweaks"/settings.
OK, but by lack of a simple glossery within a helpfile
(not ONE singel word about the next settings within the helpfile),
nor to find at their converting tutorials on their website......
http://www.yasasoft.com/tutorials/tutorials.htm
......the next is hardly to understand:
http://www.fotograaf.cc/ftp/YASA.jpg
in opposite to this:
http://www.fotograaf.cc/ftp/AoA.jpg
If you can only give a little hint which one of the most above
screenshot from YASA is for "deblocking", I should be satisfied.
(The above application has an overall more sharp conversion than AoA)
> One way to get better
> performance from the default settings, is to pay more for
> the program.
Both applications I mentioned are full working applications.
> You also might consider that there is not a lot of
> commercial incentive to develop such a program, at least
> not for any major player. Not with the law suites from
> the movie industry.
Strange, with the every year growing bandwidth of internet, the
distribution of film over the internet shall be as normal as "still"
images (pictures) already has conquer the world already now.
Less to do with law suites from movie industry, evrything with simple
home movie, business movies, tutorial movies, web advertising movies.
> You could break up what you want into separate
> processes and acquire a better encoding program like
> TMPGEnc 4.0 XPress, which goes to some lengths to
> make the process a little easier.
Thanks for the option to look for.
I am trying the demo now. Bulky software. Although a good helpfile.
(Learned what I maybe can use within the other software what I have
already). Have to see further for extra options.
> But that wouldn't be the
> integrated software you seem to be asking for.
At least not the demo version. Looking to the helpfiles the demo version
has less options for tweaking. The first "dirty" conversions what I
made, I was not impressed by the quality, from the results what I have
already with the other applications.
Nobody "no"s me better than Ken!
Richard Crowley wrote:
OK (LOL)
Apologise, English is not my mother tongue (Holland --> Dutch language),
so I easily make mistakes writing and understanding English.
It was a double pun. If you have read this newsgroup much,
you will know that Ken has a much higher opinion of temporally-
compressed codecs than I do and he wastes no opportunity
reminding everyone of our difference in worldviews.
Your English is infinitely better than my Dutch. :-))
In my defense, I was attempting to observe that it is not possible
to eliminate ALL artifacts of heavy, lossy compression, no matter
how much computing power you throw at it. (Hence the term:
"lossy"). In some very "busy" scenes in a movie, the compression
can't keep up even up to the max limit of the format. That is just the
way they had to design it for compatibility with all DVD players, etc.
OTOH, I am surprised that ANY ripping software is so shoddy that
it can't recover at least as much of the image as any common DVD
player does on the fly. That seems inexcusable.
Richard Crowley wrote:
> "Léon Obers" wrote ...
> In my defense, I was attempting to observe that it is not possible
> to eliminate ALL artifacts of heavy, lossy compression, no matter
> how much computing power you throw at it. (Hence the term:
> "lossy"). In some very "busy" scenes in a movie, the compression
> can't keep up even up to the max limit of the format. That is just the
> way they had to design it for compatibility with all DVD players, etc.
As a photographer I am not used to film / movie, but in basic image
compression used in film or pictures are in accordance with each other.
As for film, an extra difficulty to overcome with fast followed changes
in scene.
In several years interpolation techniques has developed at a high level.
That's why you can find so many "standards" and codecs. Every new method
and approach is to be used in advantage of the user.
As for images for photography the emphasis is more in interpolation
techniques to avoid artifacts by converting RAW data from a Bayer
pattern (the same for camcorder with only one imager) and by lowering
noise levels at more high ISO ratings, than compression techniques.
Fact is that in accordance with film and photography the negative
aspects of image recording has to be lowered as much as possible.
One approach does a better job than the other.
(Thats why I am comparing some of the applications).
Me myself am in digital photography business from about 1993 and I am
much surprised by the big quality steps which are made in last years,
thanks too, to many good interpolation techniques used in "building up"
a new image from a more basic "bad" recorded image.
So, I think you and I do accept fully the advantage of using
"deblocking" tools as a usefull tool, and every other "restoration"
technique to rebuild "destroyed" images.
Did read some scientific articles in past, with amazing techniques to
rebuild new images out of nearly “nothing” by a method of “what to
expect” in what the image should be found out from a scene, rather than
a "dry" calculation only.
> OTOH, I am surprised that ANY ripping software is so shoddy that
> it can't recover at least as much of the image as any common DVD
> player does on the fly. That seems inexcusable.
I don't know if it is. I want to make a smaller copy for using on the
web. Looking high or low, it is more bad as the original already of the
fact it is more small. Just try to find a balance between best options
in quality by more or less not to big data files.
But for our job I shall aks the base "mother" material of the movies to
be ripped. That is a better quality (HD) as the DVD version to use as
base for ripping.
Thanks for your input.
Right. But don't forget that the codec parameters for video DVDs was
frozen over 10 years ago and can't be "improved" because of the massive
installed base of players. Whole new formats (HD, Blu-Ray, etc.) are
required to advance beyond the standards set for DVD-Video back in
the mid-1990s.
The problem you have had, still have, and will always have is an
intellectual disconnect: People cannot and do not perceive isomorphically
all information in the physically analyzable world. Lossy is a terrible,
biased terms, since it incorrectly presumes that physical measurements
define relaity.
Lossy physically CAN BE lossless perceptually. Get used to it.
I have no problem with lossy compression. I use it all the time.
(DV uses 5:1 lossy spatial compression, for example.)
My objection is the massively lossy compression used by the
low-end crowd. If they can live with it, good for them. It drives
the rest of us batty. Even the mild 5:1 spatial compression of
DV fails on some program material. But overall, it is worth the
cost/benefit tradeoff.
Mine is a purely technical issue. You can take your psychological
and philosophical babble over to some metaphysical newsgroup.
I do not disagree and therefore withdraw my previous comment with apology.
PS
As to this comment, technical and scientific issues are, and have been
throughout history--and certainly with Einstein and Heisnberg--psychological
and philosophical!!!!
There is NOTHING metaphysical about such a discussion.
Why do so many technical discussions in Usenet end up as flame wars
between intellectual snobs?
Is that a metaphysical question or an existential one?
Because people take disagreement about technical
facts as personal attacks.
Well, I admit that I would take it personally if the
topic at issue involved something to do with my motherboard.
Who, if he has any sense of honor, would not defend his
mother,...er...,board?