Quantcast Packshot: Autoaspect - digitalFAQ.com Forums [Archives]
  #1  
06-18-2004, 07:19 PM
maurus maurus is offline
Free Member
 
Join Date: Oct 2003
Location: Madrid
Posts: 387
Thanks: 0
Thanked 0 Times in 0 Posts
Quote:
- "-noskip", "-noaspect" and "noautoexpand" are added to the commandline and "autoaspect=1" is removed for avoiding screwing up the resizing (thanks Phil for the hint)

I see in PackShot command line still autoaspect=1...

Another question. By default PackShot not add:

hqdn3d or noise settings. And it is very speed..

Are those settings importants? How about noise=3th for avoid lowers minrates < 300 kbps?


-Maurus
Reply With Quote
Someday, 12:01 PM
admin's Avatar
Site Staff / Ad Manager
 
Join Date: Dec 2002
Posts: 42
Thanks: ∞
Thanked 42 Times in 42 Posts
  #2  
06-19-2004, 05:42 AM
Dialhot Dialhot is offline
Free Member
 
Join Date: May 2003
Posts: 10,463
Thanks: 0
Thanked 0 Times in 0 Posts
hd3d is important as it is the only denoiser that you have in mencoder. To not use it is like have no filters in an avisynth script : you can understand that is better to have some
Reply With Quote
  #3  
06-19-2004, 02:00 PM
incredible incredible is offline
Free Member
 
Join Date: May 2003
Location: Germany
Posts: 3,189
Thanks: 0
Thanked 0 Times in 0 Posts
Send a message via ICQ to incredible
Well.. Autoaspect=1 is a bug!
I already mentioned that in another forum that all others like "-nosapect -noskip...." etc. is given, but I lost to delete that AR irritating line above, ... right maurus.

Also a Null Output on first pass will be give etc etc etc etc ....
... Im in coding a new vers. of packshot, ... but as I can join a bit the beginning of summer these nice things take a "little" more time.
So you just can use "Commandline only" and tweak your needed additions.

Related to noise. You can set it manually as called "Blockbuster".
At default I do keep my videos clear from noise! Why? Cause my player gots no problems when bitrate touches the 100kbit marker. But you just can rise the Denoiser to 2 ticks and BBuster to 1 and then I assume yxou got Kwags defaults of noise as I stood in contact with him according this.

Packshot is generated in a purpose that I do hope Mplayer Team will get rid of that Bitrate problem soon, so till that time we do set that noise workout manually.



But you know what! I did my two last encodings of two HQ DVD sources directly, means no HQ3d .... and that minimal "life" in the Sources "surfaces" did keep my Lowest Bitrates higher! Means Why first cleaning and then adding noise again = theoretically nonsens, practrically it MAKES sense as Optimal scripts did gave us these better results to AVOID Blocks.
But in here we wont avoid blocks as I dont get them with mencoder BUT we use that workout to avoid Bitrate drops. So the purpose is different.
BUT as every even HQ DVD comes with another type of noise we cant see that "let it as it is" generally - so my last encodings did look natural and bitrates where keeped ... even they got many dark parts. One of them was Matrix 1 and it worked.

So thats not to be seen as a result, but we "could" try to let some kind of HQ sources as they are??? We should check this.
As that "noise" workout is just a "trick" but should be useless when (i hope) soon they (mplayer team) do will get rid of that sucking bitrate problem

I just found some very old (3 Month ago or so) mencoder lines where I used NO lmin but another Type of commands which gave me perfect results with an encoding at 1 pass wich does look sexy! (Much better then just let doing 1pass using our settings now)
So I have to find a combination between last optimal settings and these old ones.

SECOND: WATCH out when applyying noise with Packshot in its state now as it "still" gots no source active pixels cropping area. So you can let generate your Commandline using "Commandline only" and change the values of the Resizing Part to the Values from FitCD crop/resize/Expand).

I got my active pixel based new resizing engine ready, ... finally and you will have the choise to determine the active pix area including x,y offsets, by this everything not needed will be cropped before and THEN the scaler incl. correct PAR calculation will bring the Image to its target related state where borders will be added finally using Expand.
Reply With Quote
  #4  
06-20-2004, 06:43 PM
maurus maurus is offline
Free Member
 
Join Date: Oct 2003
Location: Madrid
Posts: 387
Thanks: 0
Thanked 0 Times in 0 Posts
Quote:
Originally Posted by Dialhot
hd3d is important as it is the only denoiser that you have in mencoder. To not use it is like have no filters in an avisynth script : you can understand that is better to have some
Yes. but I think than a DVD source it's clear and it not have noise...

-Maurus
Reply With Quote
  #5  
06-20-2004, 08:16 PM
kwag kwag is offline
Free Member
 
Join Date: Apr 2002
Location: Puerto Rico, USA
Posts: 13,537
Thanks: 0
Thanked 0 Times in 0 Posts
Quote:
Originally Posted by maurus
Yes. but I think than a DVD source it's clear and it not have noise...

-Maurus
Not necessarily
You should see my DVD of "Back to the Future"

-kwag
Reply With Quote
  #6  
06-21-2004, 03:22 AM
Dialhot Dialhot is offline
Free Member
 
Join Date: May 2003
Posts: 10,463
Thanks: 0
Thanked 0 Times in 0 Posts
Quote:
Originally Posted by kwag
You should see my DVD of "Back to the Future"
And as I experienced, even VERY clean DVD have noise !
("we were soldiers", PAL edition).

In fact the trick we use to avoid DCT blocks by adding a little (unvisible) noise is used also by the industry
Reply With Quote
  #7  
06-21-2004, 03:41 AM
incredible incredible is offline
Free Member
 
Join Date: May 2003
Location: Germany
Posts: 3,189
Thanks: 0
Thanked 0 Times in 0 Posts
Send a message via ICQ to incredible
Well, as "they" shoot their material on classic Film, its clear that a film by nature gots an amount of noise. And If your eyes are 100% correct, then you can see that dark scenes often do got more noise cause of more light sensible film material, its like when doing fotos using ISO 100 (normal) and ISO 400 (more dark environments). The ISO 400 is more light sensible but will result in more grain (hard to notice buts its there).
Thats also (IMHO) why you Phil, didnt see that grain first on that orig state.
As I do remember you had to switch the source to a more bright state, wasn't it? I do assume that this also is Film Material related

Thats why also all my captures did result in not lower than 400kbit bitrate drops as I never do "plain" my caps by denoising them into a surface-dead state.
Thereare also some DVDs around resulted out of bad "dups" from film material, like the "BlackRain" example showed in the "How Funny" Thread at optimal scripts forum - most of my captures do look better in their orig state than this "blackRain" example.
Reply With Quote
  #8  
06-21-2004, 03:50 AM
Dialhot Dialhot is offline
Free Member
 
Join Date: May 2003
Posts: 10,463
Thanks: 0
Thanked 0 Times in 0 Posts
Quote:
Originally Posted by incredible
Thats also (IMHO) why you Phil, didnt see that grain first on that orig state.As I do remember you had to switch the source to a more bright state, wasn't it? I do assume that this also is Film Material related
I don't know what you mean by "bright state" but non, I didn't do anything on the source. I just have to use an avs script for this movie as mencoder as totally unable to treat it decently !

And in my eyes thus noise has nothing to do with argentic noise that I used to because I took photo in the theaters, during the movie (so with no flash), using 1000 or even 1600 ISO.
Reply With Quote
  #9  
06-21-2004, 09:07 AM
kwag kwag is offline
Free Member
 
Join Date: Apr 2002
Location: Puerto Rico, USA
Posts: 13,537
Thanks: 0
Thanked 0 Times in 0 Posts
Quote:
Originally Posted by Dialhot
using 1000 or even 1600 ISO.
Off Topic:
Ahh, I guess I beat you all guys
Back in 1980, I was experimenting with Illford AcuFine Film Developers, and I was doing "Push Processing" of B/W Film (Tri-x Pan I believe it was) up to ASA 6,000
I did many night shots with just moonlight, that you could have sweared they were taken during the day, until you started to look for details

-kwag
Reply With Quote
  #10  
06-21-2004, 09:17 AM
Dialhot Dialhot is offline
Free Member
 
Join Date: May 2003
Posts: 10,463
Thanks: 0
Thanked 0 Times in 0 Posts
Quote:
Originally Posted by kwag
I did many night shots with just moonlight, that you could have sweared they were taken during the day, until you started to look for details
In fact (you probably know that) BW paper with high ASA are current but 1600 was COLOR film (from fuji). The only drawback was the greenish tint if badly treated in the labs

But 6000... I don't even know about a camera taht can handle it .
Reply With Quote
  #11  
06-21-2004, 09:40 AM
kwag kwag is offline
Free Member
 
Join Date: Apr 2002
Location: Puerto Rico, USA
Posts: 13,537
Thanks: 0
Thanked 0 Times in 0 Posts
Quote:
Originally Posted by Dialhot

But 6000... I don't even know about a camera taht can handle it .
That's why it's called "Push Processing"
I would take the pictures based on calculations (can't remember now my exact procedure), and the development was done like if you had taken at ASA 6,000.
The basic is that you would use a colder developer ( several degrees colder than normal), and longer time in the tank ( way longer. Like 15 to 20 minutes, instead of 5 to 7 or so.)
The colder chemical would cause a very slow development, reducing the grain in the film, and the longer time would correctly develop the normally "underexposed" film
These are very old tricks, and I had a lot of fun at the time. Once you learn to do it correctly, you can do MANY things that are normally considered impossible in "regular" photography.
As a matter of fact, I recall push processing regular color film (ASA 400) to 1,600, at the time when ASA 1,600 didn't exist
So you could (in theory) take the curent ASA 1,600 color film and push it up to about 3,200 to 6,400 with only a small increase of grain.
I haven't done that in 20 years, but many people do that.
Guess I started early pushing film to the limits, could have been the KFILM at that time
Here are some methods and references: http://www.ilford.com/html/us_englis...processing.pdf
http://www.kodak.com/US/en/motion/su...ing/push.shtml

-kwag
Reply With Quote
  #12  
06-25-2004, 03:09 AM
Fluffbutt Fluffbutt is offline
Free Member
 
Join Date: Apr 2004
Posts: 189
Thanks: 0
Thanked 0 Times in 0 Posts
Kwag, have you tried colour process B&W film yet.
Ilford XP2 Ultra, rated at 400, and can be shot at anything from 50 - ~12000 (if camera will do it), ON THE SAME ROLL.

Process as normal, no special 'push/pull' needed, adjustments can be done in darkroom or minilab (I've done mine on minilab (I used to be a minilab operator), and results are good).

Grain results:

@ 50 = about 200 colour
@400 = about 125 B&W
@800 = about 400 B&W
@6400 = about 800 colour (fuji press 800)
@12000 = arty 'golfball' grain
__________________
|
Meeow!
Reply With Quote
  #13  
06-25-2004, 11:57 AM
kwag kwag is offline
Free Member
 
Join Date: Apr 2002
Location: Puerto Rico, USA
Posts: 13,537
Thanks: 0
Thanked 0 Times in 0 Posts
Quote:
Originally Posted by Fluffbutt
Kwag, have you tried colour process B&W film yet.
Ilford XP2 Ultra, rated at 400, and can be shot at anything from 50 - ~12000 (if camera will do it), ON THE SAME ROLL.
Quote:

Process as normal, no special 'push/pull' needed, adjustments can be done in darkroom or minilab (I've done mine on minilab (I used to be a minilab operator), and results are good).

Grain results:

@ 50 = about 200 colour
@400 = about 125 B&W
@800 = about 400 B&W
@6400 = about 800 colour (fuji press 800)
@12000 = arty 'golfball' grain
You're making me interested in photography again
It's been many years since I developed film, but I did it so long that I would pick it up again in a matter of hours
I'll read about this
I love push processing and I even had a bunch of B/W books from Kodak (Don't know where they are now )
I even got involved developing "Cibachrome", so I could do color slides directly to color print. It was SOOOO cool
That was a LONG time ago

Thanks,
-kwag
Reply With Quote
  #14  
06-25-2004, 11:43 PM
Fluffbutt Fluffbutt is offline
Free Member
 
Join Date: Apr 2004
Posts: 189
Thanks: 0
Thanked 0 Times in 0 Posts
Hehe I never got the hang of colour processing - I just did it on the minilab. Got quite good at B&W process tho.

Note that XP2 is a colour process B&W film - you need colour chemicals to dev it.

Xp2 is a great film (the ultra is the newer vesion released about 2002/3).
It's actually possible to do seperate shots, 50, 400, 3200, one after the other, and the film responds in the dev machine sort of automagically.

Any minor dev errors you correct as you print (with a good minilab).

It does have its problems tho - Fuji Delta 100 is a sharp crystal clear film (and rated at 160 is an excellent portrait film). By comparison, XP2 is a softer less contrasty film.
If you need clarity and sharpness, XP2 is not the answer.
If you want a soft (not blurry, soft) image for landscapes/flowers/nudes, XP2 is great.

http://www.ilford.com/html/us_english/pdf/XP2SGB_QX.pdf

I did a fabulous (even if I say so myself ) picture of a nude girl on a pool table @ 6400 on XP2 (indoor, evening, natural light).
It came out as an artistic, almost 'painted', image. She was very happy..
(No not THAT happy, is was a job after all )
__________________
|
Meeow!
Reply With Quote
  #15  
06-25-2004, 11:43 PM
Fluffbutt Fluffbutt is offline
Free Member
 
Join Date: Apr 2004
Posts: 189
Thanks: 0
Thanked 0 Times in 0 Posts
<edit> Ahh POO!!, Sorry Kwag, accidental double click.

(My mouse is playing up - I have to hammer the right button to make it work, and the left does two clicks itself. Time for a new one...)
__________________
|
Meeow!
Reply With Quote
Reply




Similar Threads
Thread Thread Starter Forum Replies Last Post
PackShot: MakeAVis Error Aielman Video Encoding and Conversion 2 10-30-2004 02:42 PM
PackShot suggestions kwag Video Encoding and Conversion 106 07-28-2004 04:47 PM
KDVD: Has anyone used packshot ? The Untouchable Video Encoding and Conversion 7 07-07-2004 01:51 PM
Using packshot and audio ? bigggt Video Encoding and Conversion 5 06-04-2004 09:54 PM
MencodeMe and Packshot bigggt Video Encoding and Conversion 5 05-19-2004 06:29 PM

Thread Tools



 
All times are GMT -5. The time now is 06:33 AM  —  vBulletin © Jelsoft Enterprises Ltd