ApparentFPS v1.09

[StainlessS]

ApparentFPS v1.09.

Plugin, ApparentFPS dll's for Avisynth v2.58 and v2.6, and avs26+ x64.
EDIT: zip includes full VS2008 Project files for easy re-build.

NOW Requires VS 2008 CPP Runtimes.

ApparentFPS: by StainlessS.

    Plugin dll's for Avisynth versions v2.58 and v2.6/+ (x86) and v2.6/+ x64.

    Shows underlying framerate where a clip has had many duplicates inserted, easier than counting unique frames.

    ApparentFPS(clip c,Float "DupeThresh"=0.5,Float "FrameRate"=clp.FrameRate,Int "Samples"=Int(Ceil(FrameRate)),
        \ Float "ChromaWeight"=1.0/3.0,String "Prefix"="", Bool "Show"=True,Bool "Verbose"=True, Bool "Debug"=False,
        \ Int "Mode"=0,
        \ Int "Matrix"=(c.Width>1100||c.Height>600)?3:2,    # 2=Pc.601 : 3=PC.709
        \ Int "BlkW"=64, Int "BlkH"=BlkW, Int "oLapX"=BlkW/2, Int "oLapY"=BlkH/2
        \)

    DupeThresh,   Float, Default 0.5 (Greater than 0.0), suggest 0.25 -> 1.0. FrameDifference below or equal to this is a dupe.
    FrameRate,    Float, Default clp.FrameRate.               # FrameRate of clp. Probably of no great use, just use default.
    Samples,      Int,   Default Int(Ceil(FrameRate))         # (1 or more). Min Frames to sample for underlying framerate detection. (>= FrameRate).
    ChromaWeight, Float, Default 1.0/3.0, (0.0 -> 1.0).       # Suggest 1.0/3.0 -> 1.0/2.0. (Use 0.0 for GreyScale YUV).
    Prefix,       String, Default "".                         # Prefix for returned Local vars, Default "" = None returned.
    Show,         Bool, Default True.                         # Show Info
    Verbose,      Bool, Default True.                         # Additional Info, Sequence as binary digits.
    Debug,        Bool, Default False.                        # Not of any great use.
    ### Added v1.05
    Mode          Int, default 0. (0 -> 1).
                    0) Standard RT_FrameDifference mode.
                    1) FrameMovement mode, difference is greatest difference for any BlkWxBlkH block.
    Matrix,       Default (c.Width>1100||c.Height>600)?3:2.     2)=PC.601 : 3=Pc.709. Used in Conversion of RGB to Luma-Y.
    BlkW          Int, Default 64.   (8 < BlkW, Even Only).     Block Width for Mode 1 (not used for Mode=0).
    BlkH          Int, Default BlkW. (8 < BlkH, Even Only).     Block Height for Mode 1 (not used for Mode=0).
    oLapX         Int, Default BlkW/2. (0 <= oLapX).            Horizontal block Overlap for Mode = 1.
    oLapY         Int, Default BlkH/2. (0 <= oLapY).            Vertical block Overlap for Mode = 1.
    ###

    Function samples over Samples frames, and shows instantaneous ApparentFPS and MaxApparentFPS, the real and perhaps most useful
    result is MaxApparentFPS.

    If Prefix = "" (Default) then does not return any info Local vars. When set to eg "P_" will set Local P_ApparentFPS
       and P_MaxApparentFPS as Float.
    Also returns Locals P_MinAboveDupeDif, P_MaxBelowDupeDif and P_CurrentDif which may assist in choosing DupeThresh, they are min and
    max values so far, and difference between current and previous frame. P_MinAboveDupeDif is the difference to previous frame of the
    frame which has lowest difference that what considered NOT a dupe (above DupeThresh), MaxBelowDupeDif max difference of frame that WAS
    considered a dupe (below DupeThresh). Ideally in a clip containing duplicates, there would be a distinct gap between these two
    above/below dup dif values with DupeThresh somewhere in between.
    v1.07, Also sets P_Maxndl and P_Maxndl_Frm, Max Non dupe length (up to samples length) and frame number where first occurs.
    v1.08, Also sets P_MinAboveDupeDif_Frm and P_MaxBelowDupeDif_Frm.

    !!! NOTE !!!, It may be better to set DupeThresh a little too high rather than too low, (where some motion frames will be mistaken as
    dupes), as it is likely that there will be at least one sequence in clip where there is sufficient motion to accurately set
    MaxApparentFPS.

    Can switch Off metrics (Show=False), and set eg Prefix = "P_" to return Locals P_ApparentFPS and P_MaxApparentFPS with rough
    instantaneous estimate of current underlying framerate and maximum underlying framerate detected so far.
    NOTE, ApparentFPS can range from 0.0 in static scene (can be measure of how 'active' a scene is).
    'Verbose' (Default True) shows additional info, a string of 1's and 0's showing duplicate and unique frames. Also shown in info is
    the Unique frame count for the current Sample spread (which will be reduced at either end of clip), current ApparentFPS is calculated as
    (FrameRate * UniqueFrameCount / CurrentSampleSpread), so where UniqueFrameCount == CurrentSampleSpread, the result is FrameRate, However,
    if CurrentSampleSpread is reduced due to being near clip ends, then current ApparentFPS will NOT be assigned to P_MaxApparentFPS even
    where greater than current P_MaxApparentFPS.

    The rest should be pretty straight forward.

    v1.01, Extended maximum samples from 256 to 1024 (silently limited to 1024).
        Maximum number of binary digit flags limited to 64 when Verbose=True.
        Max number of Debug Hex Values limited to 8 (ie 8 * 32 bits = 256 bit flags).

    v1.03, Added 'UniqMax' Verbose metrics indicator, maximum number of unique frames in Sample range, so far.

    NOTE, a clip that has been edited after dupe insertion may seem to jump up in MaxApparentFPS at edits, due to dupes being cut out.
       You can reduce this effect by setting eg Samples=Int(Ceil(FrameRate))*10, but it will not show any MaxApparentFPS until you have
       scanned forward Samples/2 frames (Samples are centered around current frame). It will also take a little time to start up as it
       has to sample a lot of frames at startup. NOTE, Samples is silently (no error) limited to 1024.

    v1.05.
       Added Some args, Matrix and Mode/blk args.
       Mode=1, ie RT_FrameMovement mode, measures frame differences using a Block mode, chops the frame into BlkWxBlkH blocks, and difference
       is the maximum difference for any one block. Blocks can also be OverLapping, by default block sizes / 2, so by default each frame
       is multiply sampled 4 times (ie is slower than Mode=0, but can set oLapX and oLapY to 0, for no overlaps).
       Mode=1, might be of use where there is very little movement in a sequence, but for most uses, probably best left at default Mode=0.
       Also Note, in v1.05, ChromaWeight also alters differencing method used for RGB. Where ChromaWeight=0.0, will use LumaDifferencing
       (via Matrix arg), and where ChromaWeight > 0.0, then will use (AveRedDif+AveGrnDif+AveBluDif)/3.0, for both Mode=0 and Mode=1.

    v1.07.
        Added Maxndl, Max Non dupe length and frame number where first occurs.
    v1.08.
        Added P_MinAboveDupeDif_Frm and P_MaxBelowDupeDif_Frm.
StainlessS

v1.05, Mode=1 (RT_FrameMovement Block mode) will likely require higher DupeThresh, maybe 1.0 or more (default 0.5).

See post #23 for equivalent script with ALMOST identical functionality.[docs purposes only, dont use it] http://forum.doom9.org/showthread.ph...88#post1698788


Above image, 0's show dupes, 1's non dupes, middle 0 (a bit lighter) is current frame.

Zip available via Mediafire (below this post).

[StainlessS]

EDIT: Below is early version in script, dont use it, use the plugin.

This was how v0.0 was implemented, left here in situ. See post #23 for current implementation in script.

S_ApparentFPS.avs

Function S_ApparentFPS(clip clp,Float "DupeThresh",Float "FrameRate",Int "HalfFrameRes",Float "ChromaWeight",
        \ String "Prefix", Bool "Show", Bool "Debug") {
/*
    S_ApparentFPS: by StainlessS.   Requires RT_Stats by StainlessS, GScript & Grunt Plugins (c) Gavino.
    Multi-Instance script version.

    Shows underlying framerate where a clip has had many duplicates inserted, easier than counting unique frames.

    S_ApparentFPS(clip clp,Float "DupeThresh"=0.75,Float "FrameRate"=clp.FrameRate,Int "HalfFrameRes"=ceil(FrameRate),
        \ Float "ChromaWeight"=1.0/3.0,String "Prefix"="", Bool "Show"=True, Bool "Debug"=False)

    DupeThresh,   Float, Default 0.75 (Greater than 0.0), suggest 0.5 -> 1.0. FrameDifference below this is a dupe.
    FrameRate,    Float, Default clp.FrameRate.               # FrameRate of clp. Probably of no great use, just use default.
    HalfFrameRes, Int,   Default Ceil(FrameRate) (1 or more). # Min Frames to sample for underlying framerate detection. Suggest FrameRate as min.
    ChromaWeight, Float, Default 1.0/3.0, (0.0 -> 1.0).       # YUV Only. Suggest 1.0/3.0 -> 1.0/2.0. (Use 0.0 for GreyScale YUV).
    Prefix,      String, Default "".                          # Prefix for returned Global vars, Default "" = None returned.
    Show,          Bool, Default True.
    Debug,         Bool, Default False.

    Function samples over 2*HalfFrameRes frames, but every HalfFrameRes frames it remembers frame position and Dupe Count and when it
    reaches 2*HalfFrameRes frames it swaps start position to use previous HalfFrameRes frames position and Dupe Count, this enables it to
    sample 2*HalfFrameRes frames starting with HalfFrameRes frames worth of data (See the code if that is as clear as mud).
    The function may tend to underestimate underlying framerate slightly where HalfFrameRes is at default of about FrameRate. If supplied
    with HalfFrameRes less than FrameRate, will tend to over-estimate underlying framerate.
    If Prefix = "" (Default) then does not return any info Global vars. When set to eg "S_" will set Global S_ApparentFPS
       and S_MaxApparentFPS as Float. If Multi-Instance then must give unique prefix for each instance returned Globals.

    Can switch Off metrics (Show=False), and set eg Prefix = "S_" to return Globals S_ApparentFPS and S_MaxApparentFPS with rough
    instantaneous estimate of current underlying framerate and maximum underlying framerate detected so far (perhaps of use for something).
    NOTE, ApparentFPS can range from 0.0 in static scene (can be measure of how 'active' a scene is).

    The rest should be pretty straight forward.

*/
    myName="S_ApparentFPS: "
    Assert(RT_FunctionExist("GScriptClip"),myName+"Essential GRunT plugin installed, http://forum.doom9.org/showthread.php?t=139337")
    Assert(RT_FunctionExist("GScript"),myName+"Essential GScript plugin installed, http://forum.doom9.org/showthread.php?t=147846")
    DupeThresh=Default(DupeThresh,0.75)                                      # (0.75) Suggest about 0.5 -> 1.0
    Assert(DupeThresh>0.0,myName+"DupeThresh Must be greater than 0.0")
    FR=Float(Default(FrameRate,clp.FrameRate))
    Assert(FR>0.0,myName+"FrameRate Must be greater than 0.0")
    HalfFrameRes=Default(HalfFrameRes,Ceil(FR))                             # (Ceil(FrameRate)), suggest about int(1.0 * framerate)
    Assert(HalfFrameRes>0,myName+"HalfFrameRes Must be greater than 0")
    ChromaWeight=Float(Default(ChromaWeight,1.0/3.0))                       # (1.0/3.0) YUV Only, suggest 1.0/3.0 -> 1.0/2.0
    Assert(ChromaWeight>=0.0 && ChromaWeight<=1.0,myName+"ChromaWeight 0.0 -> 1.0")
    Prefix=Default(Prefix,"")                                               # ("") Prefix for returned Global vars ("" = None returned)
    Show=Default(Show,True)                                                 # (True) Show on frame
    Debug=Default(Debug,False)                                              # (False) Extra debug info, not terribly useful
    FuncS="""
        Function s_ApparentFPS@@@(clip clp,DupeThresh,Float FR,Int HalfFrameRes,Float ChromaWeight,bool Show,bool Debug) {
            clp
            n = current_frame                                               # Local var
            if(Prev@@@ != n-1 || n==0) {
                # Start/Reset timer
                start_n = (n==0) ? n : n-1
                Global StartFrame@@@  = start_n
                Global StartFrame2@@@ = start_n
                Global StartTime@@@   = start_n/FR
                Global StartTime2@@@  = start_n/FR
                Global DupeCnt@@@     = 0
                Global DupeCnt2@@@    = 0
            }
            Diff = RT_FrameDifference(clp,clp,n2=n-1,ChromaWeight=ChromaWeight)
            if(Diff < DupeThresh && n > 0) {
                Global DupeCnt@@@  = DupeCnt@@@  + 1
                Global DupeCnt2@@@ = DupeCnt2@@@ + 1
            }
            CurrTime    = n / FR
            FrameCnt    = n-StartFrame@@@
            Tog1        = (FrameCnt==HalfFrameRes)
            Tog2        = (FrameCnt>=HalfFrameRes*2)
            UniqueFrameCnt  = FrameCnt - DupeCnt@@@
            TimeExpired = Max(CurrTime-StartTime@@@,0.000001)
            # Kick in only after HalfFrameRes frames done
            ApparentFPS = (FrameCnt<HalfFrameRes) ? 0.0 : UniqueFrameCnt/TimeExpired
            Global MaxApparentFPS@@@ =  ApparentFPS>MaxApparentFPS@@@ ? ApparentFPS : MaxApparentFPS@@@
            if(Show || Debug) {
                if(!Debug) {
                    RT_Subtitle("%6.3f MaxFPS=%6.3f\nDif=%.3f",ApparentFPS,MaxApparentFPS@@@,Diff)
                } Else {
                    RT_Subtitle("%d] AppFPS=%6.3f (Max=%6.3f)" +
                        \ "\nFPS=%6.3f Dif=%.3f" +
                        \ "\nStartFrm =%d StartT =%.3f\nStartFrm2=%d StartT2=%.3f" +
                        \ "\nFrameCnt =%d ExpiredT=%.3f\nDupeCnt  =%2d DupeCnt2=%d\nUniqueFrameCnt=%d\nTog1=%.1s Tog2=%.1s",
                        \ n,ApparentFPS,MaxApparentFPS@@@,
                        \ FrameCnt/TimeExpired,Diff,
                        \ StartFrame@@@,StartTime@@@,StartFrame2@@@,StartTime2@@@,FrameCnt,TimeExpired,
                        \ DupeCnt@@@,DupeCnt2@@@,UniqueFrameCnt,Tog1,Tog2)
                }
            }
            if(Tog2 || Tog1) {
                if(Tog2) {
                    Global StartFrame@@@    = StartFrame2@@@
                    Global StartTime@@@     = StartTime2@@@
                    Global DupeCnt@@@       = DupeCnt2@@@
                }
                Global StartFrame2@@@       = n
                Global StartTime2@@@        = CurrTime
                Global DupeCnt2@@@          = 0
            }
            ### Insert returned Global Vars Below if required
            @@@@
            ###
            Global Prev@@@ = n
            Return Last
        }
        Global MaxApparentFPS@@@ = 0.0  Global Prev@@@ = -666                           # Init
        ARGS = "Dupethresh,FR,HalfFrameRes,ChromaWeight,Show,Debug"
        clp.GScriptClip("s_ApparentFPS@@@(last, "+ARGS+")", local=true, args=ARGS)
    """
    RT_IncrGlobal("AFPS_InstanceNumber")
    # Are return Globals required
    IG_S = (Prefix=="") ? "" : RT_String("Global %sApparentFPS=ApparentFPS    Global %sMaxApparentFPS=MaxApparentFPS@@@",Prefix,Prefix)
    Fnd_S = RT_String("@@@@\n@@@\n")
    Rep_S = RT_String("%s\n_AFPS_%d\n",IG_S,AFPS_InstanceNumber)
    InstS = RT_StrReplaceMulti(FuncS,Fnd_S,Rep_S)
#   RT_WriteFile(RT_String("DEBUG_AFPS_%d.txt",AFPS_InstanceNumber),"%s",InstS)         # Debug script listing
    Return GScript(InstS)
}

S_ApparentFPS_Client.avs

Import("S_ApparentFPS.avs")

#Avisource("D:\AVS\Test2.avi")
Avisource("D:\SCDump\CAP1.avi")
#Avisource("WebCam.avi")
KillAudio

DUPTHRESH_S     = 1.0       # Script version

DUPTHRESH_P     = 1.0       # Plug version

A = s_ApparentFPS(Dupethresh=DUPTHRESH_S,Prefix="S_",Show=false)
B =   ApparentFPS(Dupethresh=DUPTHRESH_P,Prefix="P_",Show=false)

RT_Graphlink(A,B)   #  Ensure both clips processed [about same as Echo() in v2.6Alpha5]

ScriptClip("""
    RT_SubTitle("%d] %6.3f : Max=%6.3f (Script)",current_frame,S_ApparentFPS,S_MaxApparentFPS,align=0)
    RT_SubTitle("%d] %6.3f : Max=%6.3f (Plug)"  ,current_frame,P_ApparentFPS,P_MaxApparentFPS,align=1)
""",After_Frame=True)

Assumefps(250.0)

Produces pretty much identical results to plugin.

[turbojet]

This seems interesting, I'll give it a try, it would be awesome if it performed ivtc. Is this a step towards auto ivtc? MadVR has had this for awhile now and it's pretty accurate.

[StainlessS]

Is this a step towards auto ivtc?

Nope, purely to save counting unique frames for decimating of dupes via eg multidecimate (or MDec2, YV12/YUY2/RGB, YV24 etc).
The script version is exactly how it is implemented.

EDIT: Globals are returned for whatever use people can make of them (Locals for the plug version).
(I live in PAL land and have little need for anything NTSC related).
I guess with a bit of bobbing it might be used to detect if IVTC were required, I look forward to seeing what use you can make of it.

EDIT:

The script version is exactly how it is implemented.

GetFrame() from source, what an almost exact conversion to plug might look like.

PVideoFrame __stdcall ApparentFPS::GetFrame(int n, IScriptEnvironment* env) {
    n = (n<0) ? 0 : (n>= num_frames) ? num_frames - 1 : n;      // Range limit n
    PVideoFrame dst = child->GetFrame(n, env);
    if(Show||Debug) env->MakeWritable(&dst);

    if(Prev_n != n-1 || n==0) {
        // Start/Reset
        const int start_n = (n == 0) ? 0 : n-1;
        StartFrame2 = StartFrame    = start_n;
        StartTime2  = StartTime     = start_n/FrameRate;
        DupeCnt2    = DupeCnt       = 0;
    }

    AVSValue aFpsArgs[5]={child,child,n,n-1,ChromaWeight};
    double Diff = RT_FrameDifference_Lo(AVSValue(aFpsArgs,5),env);

    if(Diff < DupeThresh && n > 0) {
        ++DupeCnt;
        ++DupeCnt2;
    }
    const double CurrTime   = n / FrameRate;
    const int FrameCnt      = n-StartFrame;
    const bool Tog1         = (FrameCnt==HalfFrameRes);
    const bool Tog2         = (FrameCnt>=HalfFrameRes*2);
    const int UniqueFrameCnt= FrameCnt - DupeCnt;
    const double TimeExpired= max(CurrTime-StartTime,0.000001);
    // Only kick in after HalfFrameRes frames done
    const double ApparentFPS    = (FrameCnt<HalfFrameRes) ? 0.0 : UniqueFrameCnt/TimeExpired;
    MaxApparentFPS              = (ApparentFPS>MaxApparentFPS) ? ApparentFPS : MaxApparentFPS;

    if(Show || Debug) {
        if(!Debug) {
            DrawFStr(dst,0,0,"%d] %6.3f MaxFPS=%6.3f\nDif=%.3f",n,ApparentFPS,MaxApparentFPS,Diff);
        } else {
            DrawFStr(dst,0,0,
                "%d] AppFPS=%6.3f (Max=%6.3f)"
                "\nFPS=%6.3f Dif=%.3f"
                "\nStartFrm =%d StartT =%.3f\nStartFrm2=%d StartT2=%.3f"
                "\nFrameCnt =%d ExpiredT=%.3f\nDupeCnt  =%2d DupeCnt2=%d\nUniqueFrmCnt=%d\nTog1=%s Tog2=%s",
                n,ApparentFPS,MaxApparentFPS,
                FrameCnt/TimeExpired,Diff,
                StartFrame,StartTime,StartFrame2,StartTime2,FrameCnt,TimeExpired,
                DupeCnt,DupeCnt2,UniqueFrameCnt,Tog1?"T":"F",Tog2?"T":"F");
        }
    }

    if(Tog2||Tog1) {
        // FrameCnt cannot be 0
        if(Tog2) {
            StartFrame    = StartFrame2;
            StartTime     = StartTime2;
            DupeCnt       = DupeCnt2;
        }
        StartFrame2       = n;
        StartTime2        = CurrTime;
        DupeCnt2          = 0;
    }

    if(*Prefix != '\0') {                                // Most of this could have been prepped in the constructor :(
        int pfixlen=strlen(Prefix);
        if(pfixlen>128) pfixlen=128;
        char bf[128+32];
        memcpy(bf,Prefix,pfixlen);
        char *d=bf+pfixlen;
        const char *p,*afps="ApparentFPS";
        for(p=afps;*d++=*p++;);         // strcat variable name part
        AVSValue var = GetVar(env,bf);
        env->SetVar(var.Defined() ? bf : env->SaveString(bf),ApparentFPS);
        d = bf+pfixlen;
        const char *afpsmx="MaxApparentFPS";
        for(p=afpsmx;*d++=*p++;);       // strcat variable name part
        var = GetVar(env,bf);
        env->SetVar(var.Defined() ? bf : env->SaveString(bf),MaxApparentFPS);
    }

    Prev_n=n;
    return dst;
}

[StainlessS]

turboJet,

I only have a single (good quality) NTSC FILM sample (about 16 mins rescued from a now totally corrupt DVD),

tried this: (EDIT: With 'Honour Pulldown Flags' in DGIndex, ie Not 'Forced Film').

MPeg2Source("NTSC_Top_FORCEFILM-ReturnOfTheKing_NTSC_FORCEFILM.d2v")
Robocrop(WMod=4)

Bob
ApparentFPS(Dupethresh=2.5)

Assumefps(250.0)

Showed MaxApparentFPS as 23.976FPS
(after full 16min scan, initial 23.976FPS detected at frame 121 of 57668 bobbed frames, ie 2* HalfFrameRes + 1 frames, frame 0 does not count [EDIT: we count frame 0 as neither a dupe not unique]).

EDIT: Fairly high Dupethresh to see both bobbed frames belonging to single progressive frame as same, detects
bobbed output @ 59.94FPS as 23.976 MaxApparentFPS.

EDIT: Script version detects as 23.977FPS, probably due to lower precision Float, I'm guessin at this line:

CurrTime    = n / FR

[johnmeyer]

Nope, purely to save counting unique frames for decimating of dupes via eg multidecimate (or MDec2, YV12/YUY2/RGB, YV24 etc).

I can think of several situations where this is going to be useful.

Now, if I can just remember that it exists the next time I need it ...

Thanks!!

[turbojet]

I tried it on a few and it detected 23.976 at 24.0*, 25 at 26.* and 29.97 30.0 fps on limited testing. Which is accurate enough for scanning like autocrop/robocrop to determine what it should be ivtc'd to. It's not possible currently to get this info to pass on to call an ivtc filter, is it something that might be considered in the future?

[StainlessS]

So far as I know, TFM uses 'd2v' file to auto fix IVTC on MPEG2. (think it also outputs some kind of logfile, dont know if any
decimation info in there). EDIT: d2v has line at end eg something like "100% FILM", can use RT_Stats to get that line of text.
EDIT: Is eg "FINISHED 100.00% FILM".
From RT_Stats doc:

RT_ReadTxtFromFile(String ,Int "Lines"=0,Int "Start"=0)
 Non-clip function.
 String Filename, Name of text file to load into a string.
 Lines=0=unlimited. Set to number of leading lines in text file to load, eg 1 = load only the first line of text file.
 The return string is n/l ie Chr(10) separated, and carriage returns are removed from the returned string.
 If source file was missing newline on very last line, it will append a newline so that all lines are similarly formatted.
 v1.03, Added Start arg default=0=very first line (relative 0). Would have been nice to have start and lines in reverse
 order but implemented as above to not break scripts.
 Throws an error if your requested Start is >= to the number of lines in the file, or zero len file.
 To fetch the last line of a text file, use eg Start = RT_FileQueryLines(Filename) - 1 (Start is zero relative).
 You could eg get the last line of a d2v file which might look like this:- "FINISHED  100.00% VIDEO"

You could write a 'PreScan' script to scan part of a file and return results, and do with that whatever you want.
I dont have that much/many samples, knowledge nor inclination to do so myself, GScript could come in handy here.

something like:

Function fn(clip c) {
    c
    GSCript("""
        z = bob.ApparentFPS(Dupethresh=2.5,Prefix="P_",Show=False)
        FC=z.FrameCount
        cnt=min(FC,1000)
        for(i=0,cnt-1) {
            RT_YankChain(z,n=i) # force process frame i
        }
        return P_MaxApparentFPS
    """)
}

completely untested (would probably need to be a bit more involved than above with some kind of decision made based
on both original framerate and apparent framerate).

EDIT: I used RoboCrop(), just to remove any borders that may 'water down' metrics and possibly necessitate adjustment of DupeThresh.

[johnmeyer]

TDecimate doc is here:

TDecimate Documenation

Decimation is set using two parameters:

cycle
This is the group of frames, prior to decimation, which contain the duplicates. If you have a complicated decimation pattern, you need to specify a large enough "cycle" number so that you include all the pattern variations, but only until those patterns repeat. For instance, if the video has one repeat in the first group of five frames, but then has one repeat in the next four, and then this same thing happens with the next group of five, and the following group of four, you need to specify a "cycle" of nine frames. For me, this is always the most difficult parameter to determine.

cycleR
This is how many frames you want to decimate in every "cycle." In my example above, this would be 2.

So, the challenge would be to be able to determine what "cycle" to use. Once that is determined, I think it should be easy to compute CycleR, using this new function, and pass that to TDecimate.

[turbojet]

d2v parameter only helps when pulldown is used, which is common in big studio dvd's but not all that common in tv, it's only for 29 -> 23 too.

On a 23.976 telecined source: The script's return value was not a video clip, (Is a float, 24.169378).
On a 25 fps telecined source that's time compressed: The script's return value was not a video clip, (Is a float, 33.300034).

The latter is concerning, this same file when scrubbing is around 26-28.

How to take this info and apply these rules to TDecimate(cycle=n)
if fps 22-24.5 n = 5
if fps 24.5-28 n = 6
else omit TDecimate

[StainlessS]

As I said, I dont know that much about it.

The script's return value was not a video clip

Was not intended to be full script, just something to start from and play with. Could do some kind of testing inside of the for loop
to judge whether we hit maximum (eg ApparentFPS varied much below MaxApparentFPS but then hit exact MaxApparentFPS, AGAIN,
if so exit loop early, probably need a bit more thought).

EDIT:

It's 23.976 fps so that's correct, how can I write the fps into a text file from cli?

There are a couple of examples somewhere on-site using MediaInfo (CMD Line version) and RT_Call() or CallCmd() funcs.

Think I've gotta bow out soon, before I fall down, exhausted.

[turbojet]

Ya I never expected it to output video and it doesn't matter if it does. I can send you some samples if you want to work with this more.

Those will all report 29.97 which is the source fps. I'm trying to get apparentFPS into a text file to determine what to put in TDecimate, if any. Or better yet calling the correct TDecimate in the script, which would make it an auto ivtc function. However with significant blending srestore usually performs better.

Sleep well, this can wait. It's something I've been waiting for years for and I was shocked when it was mostly realized in MadVR.

[StainlessS]

I can send you some samples

Please no, I love not having to play with stuff like that

You can output whatever you wish using eg

    RT_WriteFile(Filename,"Int=%d Float=%f String=%s NewLine=\nI'm On a New Line",42,42.0,"42")

You can also read in lines of a text file, or complete text file, or create list of files of eg "*.BMP|TIFF|JPG" etc using RT_Stats funcs.
(I think currently something like 150 functions)

You could also use a temp RT_Stats DBase file to return multiple info from a function, caller supplies DB name, function creates and fills DB
and returns a status code with existing DB. [caller deals with results and then RT_FileDelete(DB)].

Some stuff MediaInfo related:

RT_Call(String Cmd,bool "Hide"=false,bool "Debug"=false)
 Run an external program using supplied cmd command line, hiding the console window if Hide==True.
 Debug, If true, send some debug info to debugview including return code from called process.
 DOS Filenames should be enclosed in quotes to avoid the system misinterpreting spaces in filenamess, see RT_QuoteStr().
 Returns 0 if process successfully started, otherwise non zero (ie 1).
  Example to write Aspect Ratio in form "1.333" to file D:\OUT\MEDINFO.TXT using command line version of MediaInfo.
   RT_Call(RT_QuoteStr("D:\TEST\MediaInfo.Exe") + " --LogFile=" + RT_Quotestr("D:\OUT\MEDINFO.TXT") + \
       " --Inform=Video;%DisplayAspectRatio% " + RT_Quotestr("D:\VID\1.mpg"))

demo: http://forum.doom9.org/showthread.ph...47#post1589747

EDIT: And CallCmd() plug here: http://forum.doom9.org/showthread.ph...hlight=callcmd

[StainlessS]

Turbo,
Here, output from the NTSC FILM VOB, via MediaInfo NTSC.VOB > NTSC.txt

General
Complete name                            : ntsc.vob
Format                                   : MPEG-PS
File size                                : 320 MiB
Duration                                 : 16mn 2s
Overall bit rate mode                    : Variable
Overall bit rate                         : 2 789 Kbps

Video
ID                                       : 224 (0xE0)
Format                                   : MPEG Video
Format version                           : Version 2
Format profile                           : Main@Main
Format settings, BVOP                    : Yes
Format settings, Matrix                  : Custom
Format settings, GOP                     : M=3, N=12
Duration                                 : 16mn 2s
Bit rate mode                            : Variable
Bit rate                                 : 2 286 Kbps
Maximum bit rate                         : 9 802 Kbps
Width                                    : 720 pixels
Height                                   : 480 pixels
Display aspect ratio                     : 16:9
Frame rate                               : 23.976 fps
Standard                                 : NTSC
Color space                              : YUV
Chroma subsampling                       : 4:2:0
Bit depth                                : 8 bits
Scan type                                : Progressive
Scan order                               : 2:3 Pulldown
Compression mode                         : Lossy
Bits/(Pixel*Frame)                       : 0.276
Stream size                              : 262 MiB (82%)

Audio
ID                                       : 189 (0xBD)-128 (0x80)
Format                                   : AC-3
Format/Info                              : Audio Coding 3
Mode extension                           : CM (complete main)
Format settings, Endianness              : Big
Muxing mode                              : DVD-Video
Duration                                 : 16mn 2s
Bit rate mode                            : Constant
Bit rate                                 : 448 Kbps
Channel(s)                               : 6 channels
Channel positions                        : Front: L C R, Side: L R, LFE
Sampling rate                            : 48.0 KHz
Bit depth                                : 16 bits
Compression mode                         : Lossy
Delay relative to video                  : -717ms
Stream size                              : 51.4 MiB (16%)

Text #1
ID                                       : 189 (0xBD)-32 (0x20)
Format                                   : RLE
Format/Info                              : Run-length encoding
Muxing mode                              : DVD-Video
Duration                                 : 15mn 50s
Delay relative to video                  : 2s 269ms

Text #2
ID                                       : 189 (0xBD)-33 (0x21)
Format                                   : RLE
Format/Info                              : Run-length encoding
Muxing mode                              : DVD-Video
Duration                                 : 15mn 50s
Delay relative to video                  : 2s 269ms

Text #3
ID                                       : 189 (0xBD)-34 (0x22)
Format                                   : RLE
Format/Info                              : Run-length encoding
Muxing mode                              : DVD-Video
Duration                                 : 15mn 35s
Delay relative to video                  : 2s 302ms

Text #4
ID                                       : 189 (0xBD)-35 (0x23)
Format                                   : RLE
Format/Info                              : Run-length encoding
Muxing mode                              : DVD-Video
Duration                                 : 15mn 35s
Delay relative to video                  : 2s 302ms

Text #5
ID                                       : 189 (0xBD)-36 (0x24)
Format                                   : RLE
Format/Info                              : Run-length encoding
Muxing mode                              : DVD-Video
Duration                                 : 15mn 35s
Delay relative to video                  : 2s 302ms

Text #6
ID                                       : 224 (0xE0)-CC3
Format                                   : EIA-608
Muxing mode                              : DVD-Video
Muxing mode, more info                   : Muxed in Video #1
Bit rate mode                            : Constant
Stream size                              : 0.00 Byte (0%)

Menu

Do your weird samples give any extra useful info ?

(I think Mediainfo gives more detail if given an IFO rather than VOB, I dont have the IFO, DVD corrupt).

EDIT: Output from ffprobe

[mpeg2video @ 032b0760] Invalid frame dimensions 0x0.
    Last message repeated 10 times
Input #0, mpeg, from 'ntsc.vob':
  Duration: 00:16:02.68, start: 1758.473567, bitrate: 2788 kb/s
    Stream #0:0[0x1e0]: Video: mpeg2video (Main), yuv420p(tv), 720x480 [SAR 32:7 DAR 16:9], max. 9801 kb/s, 29.83 fps, 59.94 tbr, 90k tbn, 59.94 tbc
    Stream #0:1[0x80]: Audio: ac3, 48000 Hz, 5.1(side), fltp, 448 kb/s
    Stream #0:2[0x1bf]: Data: dvd_nav_packet
    Stream #0:3[0x20]: Subtitle: dvd_subtitle
    Stream #0:4[0x21]: Subtitle: dvd_subtitle
    Stream #0:5[0x22]: Subtitle: dvd_subtitle
    Stream #0:6[0x23]: Subtitle: dvd_subtitle
    Stream #0:7[0x24]: Subtitle: dvd_subtitle

with weird 29.83 fps

EDIT: VOB probably went through DVD shrink without compression but with some streams removed.

[StainlessS]

Treat ApparentFPS() definition as currently transient, just had a daft idea and want to try it out.
May drop HalfFrameRes arg altogether.
Gonna try scan ahead(& behind) a number of frames to get better current estimate, maybe about 15 ahead,+ 15 behind + current.

[StainlessS]

OK, got this, need some guinea pigs (or a few small hamsters will do I suppose).

Function    Test(clip clp,Float "DupeThresh",Float "FrameRate",Float "ChromaWeight", String "Prefix", Bool "Show", Bool "Debug") {
    myName="s_ApparentFPS: "
    Assert(RT_FunctionExist("GScriptClip"),myName+"Essential GRunT plugin installed, http://forum.doom9.org/showthread.php?t=139337")
    Assert(RT_FunctionExist("GScript"),myName+"Essential GScript plugin installed, http://forum.doom9.org/showthread.php?t=147846")
    DupeThresh=Default(DupeThresh,0.5)                                      # (1.0) Suggest 1.0 (clean) down to 0.5 (lossy compressed)
    Assert(DupeThresh>0.0,myName+"DupeThresh Must be greater than 0.0")
    FR=Float(Default(FrameRate,clp.FrameRate))
    Assert(FR>0,myName+"FrameRate Must be greater than 0.0")
    ChromaWeight=Float(Default(ChromaWeight,1.0/3.0))                       # (1.0/3.0) YUV Only, suggest 1.0/3.0 -> 1.0/2.0
    Assert(ChromaWeight>=0.0 && ChromaWeight<=1.0,myName+"ChromaWeight 0.0 -> 1.0")
    Prefix=Default(Prefix,"")
    Show=Default(Show,True)
    Debug=Default(Debug,False)
    FuncS="""
        Function ApparentFPS@@@(clip clp,DupeThresh,Float FR,Float ChromaWeight,bool Show,bool Debug) {
            clp
            n = current_frame                                           # Local var
            LastFrame = FrameCount - 1
            if(Prev@@@ != n-1 || n==0) {
                # Start/Reset timer
                Prior_n = n - 1                                         # As if at previous n position (ie pretend previous iteration)
                BITS = 0
                SFrm = max(Prior_n - 15,1)                              # Frame 0 has no prior frame to compare with
                EFrm = min(Prior_n + 15,LastFrame)
                for(i= SFrm,EFrm) {
                    Dif = RT_FrameDifference(clp,clp,n=i,n2=i-1,ChromaWeight=ChromaWeight)
                    if(Dif >= Dupethresh) {
                        BITS = RT_BitSet(BITS,15-(i-Prior_n))
                    }
                }
                Global BITS@@@ = BITS
            }
            BITS=BITS@@@                                                # Local var, from previous iteration
            BITS=BitAnd(RT_BitLSL(BITS,1),$7FFFFFFE)                    # shift over and discard oldest bit (and clear bit zero new bit)

            if(n + 15 <= LastFrame) {
                Dif = RT_FrameDifference(clp,clp,n=n+15,n2=n+15-1,ChromaWeight=ChromaWeight)
                if(Dif >= Dupethresh) {
                    BITS = BITS + 1                                     # Bit zero was clear
                }
            }

            SFrm    = max(n - 15,1)
            EFrm    = min(n + 15,LastFrame)
            Samples = EFrm - SFrm + 1                                   # Valid samples

            Unique = RT_BitSetCount(BITS)

            ApparentFPS = Unique * FR / Samples

            Global MaxApparentFPS@@@ =  (Samples!=31) ? MaxApparentFPS@@@ : max(ApparentFPS,MaxApparentFPS@@@)

            if(Debug || Show) {
                Dif = (n==0) ? 0.0 : RT_FrameDifference(clp,clp,n2=n-1,ChromaWeight=ChromaWeight)
                LS=RT_NumberString(RT_BitLSR(BITS,16),2,15)
                MS=RT_BitTst(BITS,15) ? "1":"0"
                RS=RT_NumberString(RT_BitAnd(BITS,$7FFF),2,15)
                RT_Subtitle("AppFPS=%6.3f MaxFPS=%6.3f\nDif=%.3f (Set=%2d Samples=%2d)\n\aR%s\a-%s\aR%s\a-",
                    \ ApparentFPS,MaxApparentFPS@@@,Dif,Unique,Samples,LS,MS,RS)
            }
            ### Insert returned Global Vars Below if required
            @@@@
            ###
            Global BITS@@@ = BITS
            Global Prev@@@ = n
            Return Last
        }
        Global BITS@@@=0  Global MaxApparentFPS@@@ = 0.0  Global Prev@@@ = -666
        ARGS = "Dupethresh,FR,ChromaWeight,Show,Debug"
        clp.GScriptClip("ApparentFPS@@@(last, "+ARGS+")", local=true, args=ARGS)
    """
    RT_IncrGlobal("AFPS_InstanceNumber")
    # Are return Globals required
    IG_S = (Prefix=="") ? "" : RT_String("Global %sApparentFPS=ApparentFPS    Global %sMaxApparentFPS=MaxApparentFPS@@@",Prefix,Prefix)
    Fnd_S = RT_String("@@@@\n@@@\n")
    Rep_S = RT_String("%s\n_AFPS_%d\n",IG_S,AFPS_InstanceNumber)
    InstS = RT_StrReplaceMulti(FuncS,Fnd_S,Rep_S)
    RT_WriteFile(RT_String("DEBUG_AFPS_%d.txt",AFPS_InstanceNumber),"%s",InstS)
    Return GScript(InstS)
}

#Avisource("D:\AVS\Test2.avi")
#Avisource("D:\SCDump\CAP1.avi")
Avisource("WebCam.avi")
KillAudio
#StackHorizontal(Last,Last)
Test()

AssumeFPS(250.0)

Will be better instantaneous ApparentFPS but is taken from sample of 31 frames so will likely be out some at MaxApparentFPS.

[turbojet]

Mediainfo reads 29.97i or 59.94p on all tv I just checked, the rare one that uses pulldown may state 23.976p but I don't have one at the moment. I don't think it will be any help.

Is the latest function relevant to what I'm trying to do?

[StainlessS]

Originally Posted by turbojet

Mediainfo reads 29.97i or 59.94p on all tv I just checked, the rare one that uses pulldown may state 23.976p but I don't have one at the moment.

Thank you.
The one I posted said:

Frame rate                               : 23.976 fps
Scan type                                : Progressive
Scan order                               : 2:3 Pulldown

Perhaps old version.

Originally Posted by turbojet

Is the latest function relevant to what I'm trying to do?

Hope so. Is replacement for previous [just different name ie Test()].
BUT, currently trying to make it sample up to 64 frames, even numbers probably better for 30fps (eg 30 or 60) and 25 or 50 for PAL. If even then will sample 1 less after current frame than before current frame.

Do you see yourself more as a large guinea pig, or a small hamster ? (stay away from Peru either way).

[turbojet]

Tried test() on a 23.976 telecined source and it bounce around between 22 and 27 while scrubbing, max fps was 29. How can I get an average fps of the file after quick scanning?

[StainlessS]

OK, thanks.
Probably need to do more samples, only doing 31 samples with that try (current + 15 either side, [limited 32 bits in Int]).
Not suitable for use with the bob function thing, not enough samples (old version did 2*framerate samples (2*2*framerate for bob thing).
The hi-lited binary digit is the current frame, is it showing the sequence you expect ? (if not, adjust Dupethresh)

I'll try to get 64 sample script going today (forget bob thing for now), and if I get it working OK, will attempt plug at 128 sample max using 64 bit int's.

It's not intended for scrubbing around, continuous play only, and the only numeric of real value is the MaxApparentFPS.

OK, I give in, can you up a sample of about 30-60 secs compressed (Not HD), with varying motion would be ok, at least 2 seconds of continuous motion at some point.
EDIT: And can you say what sequence I should expect from the sample, ta.