Quantcast New Highly Accurate Prediction Method Reported by Jorel! - Page 2 - digitalFAQ.com Forums [Archives]
  #21  
07-06-2003, 04:37 AM
jorel jorel is offline
Invalid Email / Banned / Spammer
 
Join Date: Aug 2002
Location: Brasil - MG - third stone from the sun
Posts: 5,570
Thanks: 0
Thanked 0 Times in 0 Posts
ozjeff99
only change

GOP parameters: 1, 5823, 2, 1, 25

from the first Kwag's post:
"...if you are encoding at 23.976 fps, change the length of samples to 72.

If you are encoding PAL (25fps), change it to 75.

If encoding NTSC 29.97fps, change it to 90."


the others adjusts are fine!
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
  #22  
07-06-2003, 05:06 AM
ozjeff99 ozjeff99 is offline
Free Member
 
Join Date: May 2002
Location: Sydney, Australia
Posts: 159
Thanks: 0
Thanked 0 Times in 0 Posts
Thanks for that very quick reply Jorel :P

In the earlier posts on the other streams I noticed PAL people were getting inaccuracies with MaX N. of frames in GOP=25 and were advocating 24. Any insights on this?

Cheers
ozjeff99
Reply With Quote
  #23  
07-06-2003, 06:51 AM
jorel jorel is offline
Invalid Email / Banned / Spammer
 
Join Date: Aug 2002
Location: Brasil - MG - third stone from the sun
Posts: 5,570
Thanks: 0
Thanked 0 Times in 0 Posts
ozjeff99,

i only encode ntsc sources dvds and don't have pal for tests.
Kwag did tons of tests, much,much more than me
and he used pal sources to get preditions too!

i can't talk about pal predictions and
the references from Kwag are better than mine.



ps:
in Brasil is pal-m,seems ntsc-m(usa) in "standard-m", different from pal-g.
we encode ntsc dvds!
Reply With Quote
  #24  
07-06-2003, 08:20 AM
ozjeff99 ozjeff99 is offline
Free Member
 
Join Date: May 2002
Location: Sydney, Australia
Posts: 159
Thanks: 0
Thanked 0 Times in 0 Posts
Hi Jorel

I wanted to make sure I got this right cos this prediction seems too fast. Do I do a full prediction ie. Speed up by 1? Or do you use fast prediction?

Would help if you could post a screenshot.

Thanks mate
ozjeff99
Reply With Quote
  #25  
07-06-2003, 08:38 AM
jorel jorel is offline
Invalid Email / Banned / Spammer
 
Join Date: Aug 2002
Location: Brasil - MG - third stone from the sun
Posts: 5,570
Thanks: 0
Thanked 0 Times in 0 Posts
ozjeff99,

yes is really too fast, and i don't use fast prediction!
use like posted!

believe me,i don't know how put a picture here,seems that i need some space in the web...
i have this space(50mb) but i don't know how...
conclusion:


and more..Marco.c from portuguese forum give space in the web too!

than i'm stup twice.

i need a big tutorial from a patient friend to make my page only to put files
Reply With Quote
  #26  
07-07-2003, 12:54 AM
labomba labomba is offline
Free Member
 
Join Date: Jan 2003
Location: S.Paulo-Brasil
Posts: 185
Thanks: 0
Thanked 0 Times in 0 Posts
Kwag and Jorel, I really don't know what I'm doing wrong here.

Remember my last post : I achieved a CQ=56,210 with the July 2th script.
Well, with the latest (July 4th?), factor=1.0, sample/min=1, sample rate=72, I got now CQ=56,187
And worse, the Required Video Size: 725.840.875 and I got a final file with 677.710 k
I really don't understand. Do you think that sample/min=1 instead of 2 was my mistake (97 min movie)?

Another weird infos: The DVD2AVI (1,76) runned quickly during this project. Another thing: when I selected the d2v file on SwiftAvs, it said me to use deinterlace, even being a 23.97 FroceFilm project. I didn't deinterlace it. Could it be the problem?

LaBomba
Reply With Quote
  #27  
07-07-2003, 01:08 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
Hi labomba,

If you are using AviSynth 2.52, you should be using DVD2AVI version 1.77.3 Not 1.76.
Also, if your file is >=95% FILM, you should have set "Force FILM" in DVD2AVI. Did you

-kwag
Reply With Quote
  #28  
07-07-2003, 01:22 AM
ovg64 ovg64 is offline
Free Member
 
Join Date: Jan 2003
Location: Puerto Rico
Posts: 423
Thanks: 0
Thanked 0 Times in 0 Posts
Send a message via MSN to ovg64
Yes it looks like he use force film, however i think what is happening
is what has been happening ever sence avi2.5, the non action film go over the mark and the action films fall short of the mark, thats just my opinion. There is no such thing as an accurate prediction now but as long as i come up close enough im happy.
Reply With Quote
  #29  
07-07-2003, 01:32 AM
labomba labomba is offline
Free Member
 
Join Date: Jan 2003
Location: S.Paulo-Brasil
Posts: 185
Thanks: 0
Thanked 0 Times in 0 Posts
Yes, Kwag, I used ForceFilm option, but my DVD2AVI version is 1,76.

I will try the 1.77 version, after I find it. Do you know if this version is at the download thread?

LaBomba

Edited. Don't worry. I already have that version. Let me install and try again.
Reply With Quote
  #30  
07-07-2003, 06:41 AM
ozjeff99 ozjeff99 is offline
Free Member
 
Join Date: May 2002
Location: Sydney, Australia
Posts: 159
Thanks: 0
Thanked 0 Times in 0 Posts
Hi Jorel

Unfortunately the method did not work for me . Gave me CQ 57 with 900Mb total .mpg file size. I think the Speed up by 10 is too high. When I reduced to Speed up by 5 the predicted CQ dropped to 50.52 which is very close to Full sampler (with GOP 24, samples length 72) . Will try full encode with this CQ. So I'll know in about 8hrs.
Thanks
ozjeff99
Reply With Quote
  #31  
07-07-2003, 06:46 AM
jorel jorel is offline
Invalid Email / Banned / Spammer
 
Join Date: Aug 2002
Location: Brasil - MG - third stone from the sun
Posts: 5,570
Thanks: 0
Thanked 0 Times in 0 Posts
what is the resolution choosed and the total time of the movie
just curious about the prediction that you got.
Reply With Quote
  #32  
07-07-2003, 11:20 AM
ozjeff99 ozjeff99 is offline
Free Member
 
Join Date: May 2002
Location: Sydney, Australia
Posts: 159
Thanks: 0
Thanked 0 Times in 0 Posts
Move was Star Wars II (130min). Resolution is at 480x576 (the only kvcdx3 version my player can handle). I'm going to use the new script just posted and keep trying till I get it right.
ozjeff99
Reply With Quote
  #33  
07-07-2003, 02:47 PM
jorel jorel is offline
Invalid Email / Banned / Spammer
 
Join Date: Aug 2002
Location: Brasil - MG - third stone from the sun
Posts: 5,570
Thanks: 0
Thanked 0 Times in 0 Posts
ok ozjeff99,
don't know....i never did like you and it's just my opinion...
i think that 130 minutes in 1 cdr using resolution 480x576
and with only CQ 50,xx .....the result is not too good.
why don't 2 cds with incredible CQ and quality with 480x576?

did you see the quality in your first encode with CQ57 - 900mb
if is not good,means that with less space and less CQ
the quality will worse CQ50,xx in 1cdr 80
Reply With Quote
  #34  
07-07-2003, 10:45 PM
Bilal Bilal is offline
Free Member
 
Join Date: Jun 2003
Posts: 73
Thanks: 0
Thanked 0 Times in 0 Posts
why did you never like him what did he ever do to you j/k
Reply With Quote
  #35  
07-08-2003, 02:59 AM
labomba labomba is offline
Free Member
 
Join Date: Jan 2003
Location: S.Paulo-Brasil
Posts: 185
Thanks: 0
Thanked 0 Times in 0 Posts
Hey guys!
Great news: I finally found the problem with my prediction error, well, i guess so
My DVD2AVI was 1.7.6 version, my MPEG2DEC3 was NIC's 1.00 version, OK with that D2A version, but not good to the new prediction tricks.

I changed the DVD2AVI to version 1.7.7.4 and also to version 1.7.7.3, and when using the newest script (July 6th), a error occurred. Looking the TOK 0.0.5.2 logs, all info about resolution, frames and time was wrong.

So, I found the Boulder's thread about a new NIC's version of MPEG2DEC3, the 1.08 version.

All problem of prediction solved!
Also, I don't have any issue with gripcrop and other issues that others guys are having now. So, I suggest, IMHO:
DVD2AVI = version 1.7.7.3
MPEG2DEC3 = Nic's version 1.08
AVS = version 2.5.2

Another thing: the CQ raised from 56,186 to 57,109

The last one, and probably the most important: I only burned the first try of that movie ( "Heaven",97min,704x480,128kaudio,1CD80min ), made prior to the new prediction tricks, old optimal script (July 2th) and a CQ=56,186.

The quality was ... AMAZING!!!! You can't imagine how perplexed I was. Only 56 CQ value

IMHO, we are living the most important moment of this site!

I really need a new emoticon here : of APPLAUSES

Thanks, guys!

LaBomba
Reply With Quote
  #36  
07-08-2003, 07:23 AM
black prince black prince is offline
Free Member
 
Join Date: Jul 2002
Posts: 1,224
Thanks: 0
Thanked 0 Times in 0 Posts
@ALL,

http://nic.dnsalias.com/MPEG2Dec3v108.zip

-bp
Reply With Quote
  #37  
07-08-2003, 09:05 PM
vhelp vhelp is offline
Free Member
 
Join Date: Jan 2003
Posts: 1,009
Thanks: 0
Thanked 0 Times in 0 Posts
@ BP..

Thanks !!

Enjoy..
-vhelp
Reply With Quote
  #38  
07-08-2003, 09:08 PM
vhelp vhelp is offline
Free Member
 
Join Date: Jan 2003
Posts: 1,009
Thanks: 0
Thanked 0 Times in 0 Posts
hi labomba..

If you are refering to kDVD ??

Yes, I've ben testing CQ w/ just 50, and am amazed (at least when I do'em)

They do look great though, don't they ??
-vhelp
Reply With Quote
  #39  
07-09-2003, 10:41 AM
labomba labomba is offline
Free Member
 
Join Date: Jan 2003
Location: S.Paulo-Brasil
Posts: 185
Thanks: 0
Thanked 0 Times in 0 Posts
Hi vHelp

I'm talking about KVCD (MPEG1 300-2500 bitrate!!!)

I don't know how to post a sample of that, but it's really amazing. It's very impressive get 97min movie @704x480 on just 1CD of 80min. And wasting lot of bytes because my image file has 788mb ! So, it could be up to 825mb !! I know, I could try to increase the CQ value, but when I use the new script (July 6th) my CQ raised to 57.106, but my mpg file (prior VCDeasy step) raised only 13mb, so I presume that I have some space on my final image file yet

LaBomba
Reply With Quote
  #40  
07-14-2003, 09:14 PM
jorel jorel is offline
Invalid Email / Banned / Spammer
 
Join Date: Aug 2002
Location: Brasil - MG - third stone from the sun
Posts: 5,570
Thanks: 0
Thanked 0 Times in 0 Posts
someone got good news results or
bad news results with this prediction way

please,...all hints, errors, goods...all results are welcome.
the target is find the best way!

thanks.

Reply With Quote
Reply




Similar Threads
Thread Thread Starter Forum Replies Last Post
CQ prediction method WOWIEGURL Avisynth Scripting 13 03-12-2005 04:56 PM
File size prediction sometimes not accurate using vmesquita? rs008f Video Encoding and Conversion 4 09-07-2003 05:09 AM
Faster prediction method ARnet_tenRA Avisynth Scripting 19 04-12-2003 09:11 AM
New 10 % prediction method ? hedix Avisynth Scripting 9 03-04-2003 07:27 AM
File Size Prediction Dosen't Seem To Be Accurate Paul0889 Avisynth Scripting 2 12-31-2002 03:50 PM

Thread Tools



 
All times are GMT -5. The time now is 06:32 AM  —  vBulletin Jelsoft Enterprises Ltd