Thanx both Inc and Rui. I see I'll have to update my tools.

If there's any trouble in the code, I'm sure you'll fix it in a second, Inc.

Good job.

Inc, I tried now with DGIndex 1.43, and here is the first lines in my d2v:

It's still recognized as 16:9 by d2vReader-0.02b, so I'm afraid it's not an old version DGIndex issue. :(

Hey, Doc.

Im very busy these days and Im going to Nizza from sunday till thursday, so Ill "enter" a d2v file, changing the AR line to 4:3 and see what happens.

So main bug fixing wont be possible before next week/thursday. :)

Andrej, I hope you travel on a nice free weekend.

Have a nice time!.

"See" you when you come back "home" :lol: .

Some Bugs have been fixed

Version 0.03b
http://home.arcor.de/packshot/d2vReader_0.03b.exe

Read the changelog in the first post.

Hi Andrej,

Downloaded!
Title still says 0.01, though :)

-kwag

it IS 0.03 8O Ill fix the name immediately :)

Hi Inc,
so you're back. First congratulations: I don't know if you're a basketball fan, but germany has just beat spain in EuroBasket. It was a nice game WE played better and deserved to win... but this is basketball.

Well, related with last update, it doesn't work well for me now. Even when I keep a copy of dgdecode both in d2vreader and in avisynth plugin folders, it complains about dgdecode missing, and doesn't preview nor crop. Frame count 1234567890.

I can see the generated script, and when I preview the script it plays properly. But I can get anything else to work... :(

Im still here, the trip starts tomorrow morning, thats why I had some time to get into some bugs.

Many people did report that there where bugs finding the location of the used dll for decoding d2v files. So I got a hint to parse the registry and there the value in "Avisnth" which shows the correct plugins folder location on your system.

BUT! I do see that maybe some users did move that avisynth fodler after installation to a diff. location, which makes the registry string senseless.

So I did at least make the Errormessage outputting now WHERE the Avisynth plugins/dgdecode.dll "should" exist as read out of the users registry.

Please do upload 0.03b again:
http://home.arcor.de/packshot/d2vReader_0.03b.exe
... as the message will be more informative about the plugins path now. ;) No other chages have been made so I didnt release a higher number.
When the Error appears, do make a screenshot and post it in here.
Then I'll see whats recognised as your avisynths plugins folder.

Thanks for your tests doc!

THEN when you gave me the screenshot from the erormessage of 0.03b you can do the next step downloading and testing 0.04b:

In 0.04b I did implement a check routine that if the registry does output a path which doesnt contain the dll, then at least the location of the apps. current folder will be used (as it had to).

http://home.arcor.de/packshot/d2vReader_0.04b.exe

Well Inc, this is the first time I upload a screenshot (thanks ImageShack for hosting it), I hope I won't do anything wrong. Here it is:
http://www.digitalfaq.com/archives/error.gif

As you can guess, in this path:
I've got an updated copy of DGDecode.dll.
I see your path lacks a "\" between plugind and DGDecode. I don't know if it's of any importance...

And finally, v.0.04b worked OK. It also shows correctly that source is 4:3 AR.
8)

Inc, every film I backup I test your nice tool, and compare output with that of FitCD (in case FitCD is to be considered as a reference...)

I got different results in a film. I don't know if it's important at all.
With d2vreader:
With FitCD:
Both were tuned the same way: accurate cropping round to 4; 2 blocks of overscan/resize; resize macroblock optimized round to 16; anamorph source and destination,...
I took film area from your tool: 684x576

Both options look quite similar, so I don't think it's of an importance. But I post in case it helps you.

Cheers.

shh did a Trick ("seems" like a cropping barrier).
As you see in "my" calcs the height gets cropped more to fit the MOD16 at width and height finally. So FitCD does "disrespect" the MOD16 at least at the Width to keep as much mixels as possible in the height. THATS why imho for shure both outputs dont look similair, as FitCD does apply right an left borders of 24 each when selecting 2 Blocks of Overscan.

So its your decision what you do prefer.

Ill implement in the Appl. a Preferences Option where a user can choose a "HQ Crop/Resizing" including a less interpolated (via resizing) result or a "forced for compression at its best" result which for shure will result in more resizing and therefore in more interpolation --- and thats what most peolpe are used to do (Mencalc, Moviestacker, GripFit, FitCD) which makes sources blurrier.

Update...

http://home.arcor.de/packshot/d2vReader_0.05b.exe

See changelog in first post for details.

:ole: :D

-kwag

This weekend is updating weekend :lol:

New Release:
http://home.arcor.de/packshot/d2vReader_0.06b.exe

- Resizing core obtained some "check" modes for more userfriendly results
- Generic Par Mode can now be choosen above under "Settings" (more settings will be added and able to be safed as ini soon).

Generic par is related to the "Keep ITU...." Option.
Means with that u can fix a bad par source as known from FitCD.
Sometimes even DVDs are digitized at 702 or 704 and encoded at 720.
That bug also often happens when noobs do take a 704 source and do a 720 directly-resized out of it also vice versa 720 to 704.

So 720 encoded using a generic PAR would end up in a correct picture IF it would be just resized in its width on the TV o 768 PAL or 640 NTSC, ... but no SAP will do that. Thats why this ITU Box is included in FitCD and in this appl.

Update:

http://home.arcor.de/packshot/SizeMe_0.07b.exe

see changelog in the first post.

mpeg4 target size supported!

Thanks, Incredible
 

Hi Andrej !! Thanks for the continuous upgrade, you're superb!! One information: BJ_ from BDVD Team has a tool also named "Size Me". :o

JFYI.

Cya!

@Incredible

How about calling it "Super Size Me" :D

:lol:

That is not the same that "Me? Super size!"

:rotf: :rotf: :rotf:

:)

@ Danpos
Whats the purpose of BJs Tool? I guess a Bitratecalculator?

So name ideas are welcome - except "fit" in the name as this would be too concurrecy-offensive like ;)

Here an example of a full d2v anamorph 720x576 brought to 1040x432 PAR1:1 (mpeg4) non-anamorph.


The Active Pixels of the Source are 720x432 so to 1040x432 means only width is resized. Error is 0.13% means "Peanuts".

http://www.digitalfaq.com/archives/error.gif

http://www.digitalfaq.com/archives/error.gif

http://www.digitalfaq.com/archives/error.gif

:lol: :lol: :lol:

AspectMaster
AspectMatic :arrow: ;)
AspectTagger
MovieTailor
ScreenTailor
HWSizer (for Heigth, Width resizer)
AspectMolder
IncrediMage

-kwag

My vote.
 

My vote go to IncredImage (sounds good for me :D ).

Cya!

As said I had the idea of T-Made as a shortcut for taylor made.
And now as Karl also suggests the name xxxx-Taylor,
TMade or ImageTaylor will maybe a nice name.


I got it!!!(maybe) :lol:

PARanoia or PARanoiac

:lol: :lol: :lol:

I liked too ! :lol:

Cya!

and what about

IncrediSized

[OT]
I understood what PAR1:1 means when MPEG4 resizing, after looking at pictures you posted. But I have a doubt: if your MP4 encoding is intended to be seen on PC then PAR1:1 is OK. But if you want to see it later on TV: should we encode it 4:3?. Even more, does MP4 accept 16:9 flags to play it afterwards on TV as anamorph?.
Sorry me if too much OT. But your guilty: since your post I've been testing a lot H264/x264, AAC and MP4, and I'm finding it really interesting. I think I'll post my experience in little time.
[/OT]

and IncrediPAR?

It will be "PARanoia" ... nice, as it points to that crazy confusing resizing and AR math headache issue. :lol:

Yup, also known mpeg2 PARs/ARs are very efficient for watching on PC as all softwarePlayers do access the AR flag within the streamheader. As said. Non PAR1:1 encodings even on mpeg4 do mean a significant compression factor by keeping FULL height details.

And thats the biiiig misunderstanding related to mpeg4 in the past.
Shure many mpeg4 addicted ones did tell you "Oh my mpeg4 DVD rip got less artifacts then your mpeg2 one at teh same final size" - BUT these mpeg4 ones where mostly encoded at smaller heights like 640x480 or so and finally these have to be bumped up more when watching at fullscreen!

I could encode at 384x288 with NO artifacts at minimal space, but the blurrier picture when watching at fullscreen will be the anti-goal.

But beside that "newbie"-wrong-approach, these newer mpeg4 codecs DO have a very big potential. Its the ecndoing routine and also that they can determine their own standards where a DVD has to be DVD compilant if you wont get into any risk---- point.

haha, thats a nice Q. And thats also why I dont go into the store and get the Nero compatible mp4 player that fast, as almost NOOO www-review does handle the AR/non-PAR1:1 capabilities which for me are obligatory when doing encodes and watching later on TV.

So wait for a very skilled review or buy and try where if it wont work you can sell it well on ebay afterwards --- Ill be your review listener, granted! 8) :)

Also the interlaced aspect is a clear point IF "that" SAP is able to play back interlaced YV12 contend fawlessly - so thats also a point.

If you will be independant from ANY manufacturer or whatever - then you should playback via a HTPC as there you can choose by yourself WHICH decoder is used and WHICH softwareplayer is used ;)


An example: do crop a 720x576 2.35:1 source to 702x432, resize it to 1024x432 PAR 1:1 (Nero cinema profile compatible!) almost like in the pics above, apply LimitedSharpen() using correct values, encode it using AVC mp4 (DVD-5 target final size - 1€ per Media *lol) and watch it via RGB! connection from a HTPC via a good beamer and the details will kiss your eyes ;)

And that sounds PARfect :lol:
:cool:

-kwag

ccooool! :)

Update, see changelog.

Its now called PARanoia.
PARfect is cool but forces the program "to be perfect" imho ;) And thats no application I know of.

OK Inc, already downloaded.

I did a test with previous version 0.07, and again compared with FitCD (I find it difficult to consider it a reference... but for your info).
The source is the same as my previous posts, 720x576 AR 4:3.
Borders as detected by PARanaoia (they won't think we're all pananoic, will they?) are left=7, top=5, width=706, height=566. This figures I fed FitCD.
I asked to resize to PAR1:1 (nice idea that of overscan greyed out when PAR1:1 selected. I see you're in everything).
SizeMe resizing:
FitCD resizing:
You can see there's a difference. Again I don't know if it's that important. I'd rather resize just stretching it horizontally, but I don't know what's the better result with these different results...

[OT](yes again, sorry) :oops:
I'm thinking it would be nice to open an AVC thread (and AVC can also stand for Advanced Video Conversion :lol: , so I do think it could be interesting to treat other video encoding formats. But as I don't know if people in general and mods in particular would be interested, I'm just waiting somebody else doing so).
Your answer clarified me a lot... that there's little clear about this issue. How is it to be done with mpeg4 formats until now, what would be the correct way in order to see them later well (correct AR) in current SAP that support this format? Would this be appliable also to AVC-MP4?.
OK, you're right, that's enough OT by now.
[/OT]

Incredible,

Wow look what I started...lol anyway im glad you looked into what the problem was, too bad i cant get the new version and try it out until the morning...but i cant wait....

thanx again everyone for the help............

Hi,

I-m trying to convert a 25 fps PAL resolution DVD to 23,976 NTSC resolution KDVD, and in this case, PARanoiac is givin me this script:
The resie part is perfect ¿also fitcd gives me the same resizing?, but the borders are way to huge. Should be (8, 16, 8, 16)

This is what I choose:
http://www.digitalfaq.com/archives/i.../2005/10/1.jpg
Thanks to ImageShack for Free Image Hosting

Thanks,
Fabrice

The suggestion from PARanoia is correct. As its a coomon PAR 1:1 width for SAPs. This is made via a build-in-barrier.
Now do check anamorph to nonanamorph and .... the width increases dramatically. But still correct as the appl. then assumpts you want to end up in a full width PAR 1:1

But no problem. If the Image for mpeg4 is too large or too small, just use the height-resizing spin gadget and hit up'n down and you will see that you also can endup in other Widths if wanted by keeping wifth and height at the MOD you choosed for resizing.

I'll add an option in the settings so a user can choose between SAP optimized or PC optimized mpeg4 resizing. Then also a anamorph to non anamorph choice wont overflow a 720's width.



AVC "afaik" means AdvacedVideoCodec btw.

Related to non PAR 1:1 AVC or ASP mpeg4 codecs playing back fawlessly on a given SAP, .... well as said ... you have to watch out for reviews or try by your own. the mpeg4 support via SAPs is in very young stage, and so I would wait or ... as said above buy at least a one which has been proofed to play back Nero AVS mp4s - the rest depends on luck.

@ Fabrice

I see youre still using 0.07 please update as many user calculations did fail due the avsscript window was still open when they went again into the tabs of the main window.

But anyhow this addborder-result is odd. So Ill get into it this evening.

Thanks for the Info.

Well, I prefer to keep in the safe for SAP side, and follow PARanoia settings. I can't check anamorph to nonanamorph in this case (sure it happens since the film is nonanamorph...)

I tried this way to compare width when height was 560, but I can make height lower than 528 but not higher than 528. It stops when I arrive to this figure. Some internal limit to avoid a width greater than 720?.

:wink:
Thanx, but I already new. I just was "playing" with the sense of letters here. Look at top of your window. It says:
www.kvcd.net
Advanced Video Conversion
That's why I said we can talk here about AVC, because it can also be Advanced Video Conversion, apart from it's original sense Advanced Video Codec... just a joke.
I was asking about those that play in their compatible SAP their mpeg4 (divx or xvid). Do they encode 1:1 or 4:3 to play later on TV?.

1:1. SPAs consider all MPEG4 as 1:1 and you have distortion if you encode them 4:3.