HELP! handbrakecli is not taking my hqdn3d settings..

[RogerTango]

handbrakecli 1.4.2

Its a damn shame the GUI cannot puke out the command line of settings used in a CLI environment,
would make EVERYONE's life easier!

I am trying to set hqdn3d to 1:1:1:3:3:3, but every combination Ive tried does not reflect in the HB encoding output..

I have tried:

--hqdn3d custom y-spatial=1:cb-spatial=1:cr-spatial=1:y-temporal=3:cb-temporal=3:cr-temporal=3

--hqdn3d custom: y-spatial=1:cb-spatial=1:cr-spatial=1:y-temporal=3:cb-temporal=3:cr-temporal=3

--hqdn3d y-spatial=1:cb-spatial=1:cr-spatial=1:y-temporal=3:cb-temporal=3:cr-temporal=3 <- THIS is the one I would have expected to work

and of course,

--hqdn3d custom 1:1:1:3:3:3

--hqdn3d 1:1:1:3:3:3 <- this is ignored as if it was never called

The DOS window is showing the settings being used, and it looks like its just ignoring my directive and using a default.



[Attachment 62287 - Click to enlarge]


This is from the help output of my used version:

-8, --hqdn3d[=string] Denoise video with hqdn3d filter
Presets:
ultralight
light
medium
strong
Custom Format:
y-spatial=y:cb-spatial=c:cr-spatial=c:
y-temporal=y:cb-temporal=c:cr-temporal=c
Default:
y-spatial=3:cb-spatial=2:cr-spatial=2:
y-temporal=2:cb-temporal=3:cr-temporal=3

Any insight would be great, thanks for your time!

[davexnet]

The log shows this. Perhaps that's enough of a clue

[RogerTango]

From looking at the C code of HBCLI:

https://github.com/HandBrake/HandBrake/blob/master/libhb/param.c

--hqdn3d y-spatial=1:cb-spatial=1:cr-spatial=1:y-temporal=3:cb-temporal=3:cr-temporal=3

"should" work...

but then again, if its not an argument that is passed as a pre-named preset it shouldnt take.. I am thinking
if I specify custom, then I need to give the values elsewhere, but I cant find it in the code.



Andrew

[RogerTango]

The log shows this. Perhaps that's enough of a clue

Not really, the JSON file is just a config file, any variable and value can be placed in any order, specifically if you look at the DOS screenshot,
you will see the actual values, and none of them jive with the values I am assigning the filter.

Thanks for your time brother! Ive already encoded a video with the GUI and saved both presets and CLI settings
to JSON and neither one of them help resolve the command line syntax problem I appear to be having.

Andrew

[davexnet]

Did you look at the reference ?
https://handbrake.fr/docs/en/1.2.0/cli/command-line-reference.html

Perhaps you should post your complete command lne

[RogerTango]

I didnt spend the last 8 hours trying to get it to work without looking at the help file first & frequently... Google is about to shut me down for excessive use of bandwidth.

Heres the cmd line from the batch file:

set "quality=24"
set "denoise=--hqdn3d 1:1:1:3:3:3"

handbrakecli --input "%%a" --output "encoded.mkv" --encoder nvenc_h264 --crop 0:0:0:0 --width 1280 --height 720 --deinterlace --non-anamorphic --encoder-preset medium --quality %quality% --encoder-profile main --encoder-level 4.0 --aencoder av_aac --mixdown stereo --ab 128k %denoise%

I can go back in the cmd window and see the arguments were passed correctly; HBCLI would crash out if they werent anyway...

[RogerTango]

This is the puke from my ver of HBCLI via --help 2> helpfile.txt:






Usage: HandBrakeCLI [options] -i <source> -o <destination>

General Options --------------------------------------------------------------

-h, --help Print help
--version Print version
--json Log title, progress, and version info in
JSON format
-v, --verbose[=number] Be verbose (optional argument: logging level)
-Z. --preset <string> Select preset by name (case-sensitive)
Enclose names containing spaces in double quotation
marks (e.g. "Preset Name")
-z, --preset-list List available presets
--preset-import-file <filespec>
Import presets from a json preset file.
'filespec' may be a list of files separated
by spaces, or it may use shell wildcards.
--preset-import-gui Import presets from GUI config preset file.
--preset-export <string>
Create a new preset from command line options and
write a json representation of the preset to the
console or a file if '--preset-export-file' is
specified. The required argument will be the name
of the new preset.
--preset-export-file <filename>
Write new preset generated by '--preset-export'
to file 'filename'.
--preset-export-description <string>
Add a description to the new preset created with
'--preset-export'
--queue-import-file <filename>
Import an encode queue file created by the GUI
--no-dvdnav Do not use dvdnav for reading DVDs


Source Options ---------------------------------------------------------------

-i, --input <string> Set input file or device ("source")
-t, --title <number> Select a title to encode (0 to scan all titles
only, default: 1)
--min-duration Set the minimum title duration (in seconds).
Shorter titles will be ignored (default: 10).
--scan Scan selected title only.
--main-feature Detect and select the main feature title.
-c, --chapters <string> Select chapters (e.g. "1-3" for chapters
1 to 3 or "3" for chapter 3 only,
default: all chapters)
--angle <number> Select the video angle (DVD or Blu-ray only)
--previews <number:boolean>
Select how many preview images are generated,
and whether to store to disk (0 or 1).
(default: 10:0)
--start-at-preview <number>
Start encoding at a given preview.
--start-at <string:number>
Start encoding at a given offset in seconds,
frames, or pts (on a 90kHz clock)
(e.g. seconds:10, frames:300, pts:900000).
Units must match --stop-at units, if specified.
--stop-at <string:number>
Stop encoding after a given duration in seconds,
frames, or pts (on a 90kHz clock) has passed
(e.g. seconds:10, frames:300, pts:900000).
Duration is relative to --start-at, if specified.
Units must match --start-at units, if specified.


Destination Options ----------------------------------------------------------

-o, --output <filename> Set destination file name
-f, --format <string> Select container format:
av_mp4
av_mkv
av_webm
default: auto-detected from destination file name)
-m, --markers Add chapter markers
--no-markers Disable preset chapter markers
-O, --optimize Optimize MP4 files for HTTP streaming (fast start,
s.s. rewrite file to place MOOV atom at beginning)
--no-optimize Disable preset 'optimize'
-I, --ipod-atom Add iPod 5G compatibility atom to MP4 container
--no-ipod-atom Disable iPod 5G atom
--align-av Add audio silence or black video frames to start
of streams so that all streams start at exactly
the same time
--inline-parameter-sets Create adaptive streaming compatible output.
Inserts parameter sets (SPS and PPS) inline
in the video stream before each IDR.


Video Options ----------------------------------------------------------------

-e, --encoder <string> Select video encoder:
x264
x264_10bit
nvenc_h264
x265
x265_10bit
x265_12bit
nvenc_h265
mpeg4
mpeg2
VP8
VP9
theora
--encoder-preset <string>
Adjust video encoding settings for a particular
speed/efficiency tradeoff (encoder-specific)
--encoder-preset-list <string>
List supported --encoder-preset values for the
specified video encoder
--encoder-tune <string>
Adjust video encoding settings for a particular
type of source or situation (encoder-specific)
--encoder-tune-list <string>
List supported --encoder-tune values for the
specified video encoder
-x, --encopts <string> Specify advanced encoding options in the same
style as mencoder (all encoders except theora):
option1=value1ption2=value2
--encoder-profile <string>
Ensure compliance with the requested codec
profile (encoder-specific)
--encoder-profile-list <string>
List supported --encoder-profile values for the
specified video encoder
--encoder-level <string>
Ensures compliance with the requested codec
level (encoder-specific)
--encoder-level-list <string>
List supported --encoder-level values for the
specified video encoder
-q, --quality <float> Set video quality (e.g. 22.0)
-b, --vb <number> Set video bitrate in kbit/s (default: 1000)
-2, --two-pass Use two-pass mode
--no-two-pass Disable two-pass mode
-T, --turbo When using 2-pass use "turbo" options on the
first pass to improve speed
(works with x264 and x265)
--no-turbo Disable 2-pass mode's "turbo" first pass
-r, --rate <float> Set video framerate
(5/10/12/15/20/23.976/24/25/29.97/
30/48/50/59.94/60/72/75/90/100/120
or a number between 1 and 1000).
Be aware that not specifying a framerate lets
HandBrake preserve a source's time stamps,
potentially creating variable framerate video
--vfr, --cfr, --pfr Select variable, constant or peak-limited
frame rate control. VFR preserves the source
timing. CFR makes the output constant rate at
the rate given by the -r flag (or the source's
average rate if no -r is given). PFR doesn't
allow the rate to go over the rate specified
with the -r flag but won't change the source
timing if it's below that rate.
If none of these flags are given, the default
is --pfr when -r is given and --vfr otherwise


Audio Options ----------------------------------------------------------------

--audio-lang-list <string>
Specify a comma separated list of audio
languages you would like to select from the
source title. By default, the first audio
matching each language will be added to your
output. Provide the language's ISO 639-2 code
(e.g. fre, eng, spa, dut, et cetera)
Use code 'und' (Unknown) to match all languages.
--all-audio Select all audio tracks matching languages in
the specified language list (--audio-lang-list).
Any language if list is not specified.
--first-audio Select first audio track matching languages in
the specified language list (--audio-lang-list).
Any language if list is not specified.
-a, --audio <string> Select audio track(s), separated by commas
("none" for no audio, "1,2,3" for multiple
tracks, default: first one).
Multiple output tracks can be used for one input.
-E, --aencoder <string> Select audio encoder(s):
none
av_aac
copy:aac
ac3
copy:ac3
eac3
copy:eac3
copy:truehd
copy:dts
copy:dtshd
copy:mp2
mp3
copy:mp3
vorbis
flac16
flac24
copy:flac
opus
copy
"copy:<type>" will pass through the corresponding
audio track without modification, if pass through
is supported for the audio type.
Separate tracks by commas.
Defaults:
av_mp4 av_aac
av_mkv mp3
av_webm vorbis
--audio-copy-mask <string>
Set audio codecs that are permitted when the
"copy" audio encoder option is specified
(aac/ac3/eac3/truehd/dts/dtshd/mp2/mp3/flac)
Separated by commas for multiple allowed options.
--audio-fallback <string>
Set audio codec to use when it is not possible
to copy an audio track without re-encoding.
-B, --ab <number> Set audio track bitrate(s) in kbit/s.
(default: determined by the selected codec, mixdown,
and samplerate combination).
Separate tracks by commas.
-Q, --aq <float> Set audio quality metric.
Separate tracks by commas.
-C, --ac <float> Set audio compression metric.
(available depending on selected codec)
Separate tracks by commas.
-6, --mixdown <string> Format(s) for audio downmixing/upmixing:
mono
left_only
right_only
stereo
dpl1
dpl2
5point1
6point1
7point1
5_2_lfe
Separate tracks by commas.
Defaults:
none up to dpl2
av_aac up to 7point1
ac3 up to 5point1
eac3 up to 5point1
mp3 up to dpl2
vorbis up to dpl2
flac16 up to 7point1
flac24 up to 7point1
opus up to 7point1
--normalize-mix Normalize audio mix levels to prevent clipping.
<string> Separate tracks by commas.
0 = Disable Normalization (default)
1 = Enable Normalization
-R, --arate Set audio samplerate(s)
(8/11.025/12/16/22.05/24/32/44.1/48 kHz)
or "auto". Separate tracks by commas.
-D, --drc <float> Apply extra dynamic range compression to the
audio, making soft sounds louder. Range is 1.0
to 4.0 (too loud), with 1.5 - 2.5 being a useful
range.
Separate tracks by commas.
--gain <float> Amplify or attenuate audio before encoding. Does
NOT work with audio passthru (copy). Values are
in dB. Negative values attenuate, positive
values amplify. A 1 dB difference is barely
audible.
--adither <string> Select dithering to apply before encoding audio:
auto (default)
none
rectangular
triangular
triangular_hp
lipshitz_ns
Separate tracks by commas.
Supported by encoder(s):
flac16
-A, --aname <string> Set audio track name(s).
Separate tracks by commas.


Picture Options --------------------------------------------------------------

-w, --width <number> Set storage width in pixels
-l, --height <number> Set storage height in pixels
--crop <top:bottom:left:right>
Set picture cropping in pixels
(default: automatically remove black bars)
--loose-crop Always crop to a multiple of the modulus
--no-loose-crop Disable preset 'loose-crop'
-Y, --maxHeight <number>
Set maximum height in pixels
-X, --maxWidth <number>
Set maximum width in pixels
--non-anamorphic Set pixel aspect ratio to 1:1
--auto-anamorphic Store pixel aspect ratio that maximizes storage
resolution
--loose-anamorphic Store pixel aspect ratio that is as close as
possible to the source video pixel aspect ratio
--custom-anamorphic Store pixel aspect ratio in video stream and
directly control all parameters.
--display-width <number>
Set display width in pixels, for custom anamorphic.
This determines the display aspect during playback,
which may differ from the storage aspect.
--keep-display-aspect Preserve the source's display aspect ratio
when using custom anamorphic
--no-keep-display-aspect
Disable preset 'keep-display-aspect'
--pixel-aspect <par_xar_y>
Set pixel aspect for custom anamorphic
(--display-width and --pixel-aspect are mutually
exclusive.
--itu-par Use wider ITU pixel aspect values for loose and
custom anamorphic, useful with underscanned sources
--no-itu-par Disable preset 'itu-par'
--modulus <number> Set storage width and height modulus
Dimensions will be made divisible by this number.
(default: set by preset, typically 2)
-M, --color-matrix <string>
Set the color space signaled by the output:
Overrides color signalling with no conversion.
2020
709
601
ntsc (same as 601)
pal
(default: auto-detected from source)


Filters Options --------------------------------------------------------------

--comb-detect[=string] Detect interlace artifacts in frames.
If not accompanied by the decomb or deinterlace
filters, this filter only logs the interlaced
frame count to the activity log.
If accompanied by the decomb or deinterlace
filters, it causes these filters to selectively
deinterlace only those frames where interlacing
is detected.
Presets:
permissive
fast
Custom Format:
mode=mpatial-metric=s:motion-thresh=m:
spatial-thresh=s:filter-mode=f:block-thresh=b:
block-width=b:block-height=b:disable=d
Default:
mode=3patial-metric=2:motion-thresh=1:
spatial-thresh=1:filter-mode=2:block-thresh=40:
block-width=16:block-height=16
--no-comb-detect Disable preset comb-detect filter
-d, --deinterlace[=string]
Deinterlace video using FFmpeg yadif.
Presets:
skip-spatial
bob
Custom Format:
mode=marity=p
Default:
mode=3
--no-deinterlace Disable preset deinterlace filter
-5, --decomb[=string] Deinterlace video using a combination of yadif,
blend, cubic, or EEDI2 interpolation.
Presets:
bob
eedi2
eedi2bob
Custom Format:
mode=m:magnitude-thresh=m:variance-thresh=v:
laplacian-thresh=l:dilation-thresh=d:
erosion-thresh=e:noise-thresh=n:
search-distance=sostproc=parity=p
Default:
mode=7
--no-decomb Disable preset decomb filter
-9, --detelecine[=string]
Detelecine (ivtc) video with pullup filter
Note: this filter drops duplicate frames to
restore the pre-telecine framerate, unless you
specify a constant framerate
(--rate 29.97 --cfr)
Custom Format:
skip-left=skip-right=skip-top=s:
skip-bottom=strict-breaks=slane=parity=p:
disable=d
Default:
skip-left=1kip-right=1kip-top=4:
skip-bottom=4lane=0
--no-detelecine Disable preset detelecine filter
-8, --hqdn3d[=string] Denoise video with hqdn3d filter
Presets:
ultralight
light
medium
strong
Custom Format:
y-spatial=y:cb-spatial=c:cr-spatial=c:
y-temporal=y:cb-temporal=c:cr-temporal=c
Default:
y-spatial=3:cb-spatial=2:cr-spatial=2:
y-temporal=2:cb-temporal=3:cr-temporal=3
--no-hqdn3d Disable preset hqdn3d filter
--denoise[=string] Legacy alias for '--hqdn3d'
--nlmeans[=string] Denoise video with NLMeans filter
Presets:
ultralight
light
medium
strong
Custom Format:
y-strength=y:y-origin-tune=y:y-patch-size=y:
y-range=y:y-frame-count=y:y-prefilter=y:
cb-strength=c:cb-origin-tune=c:cb-patch-size=c:
cb-range=c:cb-frame-count=c:cb-prefilter=c:
cr-strength=c:cr-origin-tune=c:cr-patch-size=c:
cr-range=c:cr-frame-count=c:cr-prefilter=c:
threads=t
Default:
y-strength=6:y-origin-tune=1:y-patch-size=7:
y-range=3:y-frame-count=2:y-prefilter=0:
cb-strength=6:cb-origin-tune=1:cb-patch-size=7:
cb-range=3:cb-frame-count=2:cb-prefilter=0
--no-nlmeans Disable preset NLMeans filter
--nlmeans-tune <string> Tune NLMeans filter to content type
Tunes:
none
film
grain
highmotion
animation
tape
sprite
Applies to NLMeans presets only (does not affect
custom settings)
--chroma-smooth[=string] Sharpen video with chroma smooth filter
Presets:
ultralight
light
medium
strong
stronger
verystrong
Custom Format:
cb-strength=c:cb-size=c:cr-strength=c:cr-size=c
Default:
cb-strength=1.2:cb-size=7
--no-chroma-smooth Disable preset chroma smooth filter
--chroma-smooth-tune <string> Tune chroma smooth filter
Tunes:
none
tiny
small
medium
wide
verywide
Applies to chroma smooth presets only (does
not affect custom settings)
--unsharp[=string] Sharpen video with unsharp filter
Presets:
ultralight
light
medium
strong
stronger
verystrong
Custom Format:
y-strength=y:y-size=y:cb-strength=c:cb-size=c:
cr-strength=c:cr-size=c
Default:
y-strength=0.25:y-size=7:cb-strength=0.25:
cb-size=7
--no-unsharp Disable preset unsharp filter
--unsharp-tune <string> Tune unsharp filter
Tunes:
none
ultrafine
fine
medium
coarse
verycoarse
Applies to unsharp presets only (does not affect
custom settings)
--lapsharp[=string] Sharpen video with lapsharp filter
Presets:
ultralight
light
medium
strong
stronger
verystrong
Custom Format:
y-strength=y:y-kernel=y:cb-strength=c:
cb-kernel=c:cr-strength=c:cr-kernel=c
Default:
y-strength=0.2:y-kernel=isolap:cb-strength=0.2:
cb-kernel=isolap
--no-lapsharp Disable preset lapsharp filter
--lapsharp-tune <string>
Tune lapsharp filter
Tunes:
none
film
grain
animation
sprite
Applies to lapsharp presets only (does not affect
custom settings)
-7, --deblock[=string] Deblock video with avfilter deblock
Presets:
ultralight
light
medium
strong
stronger
verystrong
Custom Format:
strength=s:thresh=t:blocksize=b:disable=d
Default:
strength=strong:thresh=20
--no-deblock Disable preset deblock filter
--deblock-tune <string>
Tune deblock filter
Tunes:
small
medium
large
Applies to deblock presets only (does not affect
custom settings)
--rotate[=string] Rotate image or flip its axes.
angle rotates clockwise, can be one of:
0, 90, 180, 270
hflip=1 flips the image on the x axis (horizontally).
Custom Format:
angle=a:hflip=h:disable=d
Default:
angle=180:hflip=0
--pad <string> Pad image with borders (e.g. letterbox).
The padding color may be set (default black).
Color may be an HTML color name or RGB value.
The position of image in pad may also be set.
Custom Format:
width=w:height=h:color=c=x:y=y:top=t:bottom=b:
left=l:right=r
--colorspace <string> Convert colorspace, transfer characteristics or color primaries.
Presets:
bt2020
bt709
bt601-6-525
bt601-6-625
Custom Format:
primaries=p:transfer=t:matrix=m:range=r:
tonemap=taram=p:desat=d
-g, --grayscale Grayscale encoding
--no-grayscale Disable preset 'grayscale'


Subtitles Options ------------------------------------------------------------

--subtitle-lang-list <string>
Specify a comma separated list of subtitle
languages you would like to select from the
source title. By default, the first subtitle
matching each language will be added to your
output. Provide the language's ISO 639-2 code
(e.g. fre, eng, spa, dut, et cetera)
--all-subtitles Select all subtitle tracks matching languages in
the specified language list
(--subtitle-lang-list).
Any language if list is not specified.
--first-subtitle Select first subtitle track matching languages in
the specified language list
(--subtitle-lang-list).
Any language if list is not specified.
-s, --subtitle <string> Select subtitle track(s), separated by commas
More than one output track can be used for one
input. "none" for no subtitles.
Example: "1,2,3" for multiple tracks.
A special track name "scan" adds an extra first
pass. This extra pass scans subtitles matching
the language of the first audio or the language
selected by --native-language.
The one that's only used 10 percent of the time
or less is selected. This should locate subtitles
for short foreign language segments. Best used in
conjunction with --subtitle-forced.
-S, --subname <string> Set subtitle track name(s).
Separate tracks by commas.
-F, --subtitle-forced[=string]
Only display subtitles from the selected stream
if the subtitle has the forced flag set. The
values in 'string' are indexes into the
subtitle list specified with '--subtitle'.
Separate tracks by commas.
Example: "1,2,3" for multiple tracks.
If "string" is omitted, the first track is
forced.
--subtitle-burned[=number, "native", or "none"]
"Burn" the selected subtitle into the video
track. If "subtitle" is omitted, the first
track is burned. "subtitle" is an index into
the subtitle list specified with '--subtitle'
or "native" to burn the subtitle track that may
be added by the 'native-language' option.
--subtitle-default[=number or "none"]
Flag the selected subtitle as the default
subtitle to be displayed upon playback. Setting
no default means no subtitle will be displayed
automatically. 'number' is an index into the
subtitle list specified with '--subtitle'.
"none" may be used to override an automatically
selected default subtitle track.
-N, --native-language <string>
Specify your language preference. When the first
audio track does not match your native language
then select the first subtitle that does. When
used in conjunction with --native-dub the audio
track is changed in preference to subtitles.
Provide the language's ISO 639-2 code
(e.g. fre, eng, spa, dut, et cetera)
--native-dub Used in conjunction with --native-language
requests that if no audio tracks are selected the
default selected audio track will be the first
one that matches the --native-language. If there
are no matching audio tracks then the first
matching subtitle track is used instead.
--srt-file <string> SubRip SRT filename(s), separated by commas.
--srt-codeset <string>
Character codeset(s) that the SRT file(s) are
encoded as, separated by commas.
If not specified, 'latin1' is assumed.
Command 'iconv -l' provides a list of valid codesets.
--srt-offset <string> Offset (in milliseconds) to apply to the SRT
file(s), separated by commas. If not specified,
zero is assumed. Offsets may be negative.
--srt-lang <string> SRT track language as an ISO 639-2 code
(e.g. fre, eng, spa, dut, et cetera)
If not specified, then 'und' is used.
Separate by commas.
--srt-default[=number]
Flag the selected SRT as the default subtitle
to be displayed during playback.
Setting no default means no subtitle will be
automatically displayed. If 'number' is omitted,
the first SRT is the default.
'number' is a 1-based index into the 'srt-file' list
--srt-burn[=number] "Burn" the selected SRT subtitle into
the video track.
If 'number' is omitted, the first SRT is burned.
'number' is a 1-based index into the 'srt-file' list
--ssa-file <string> SubStationAlpha SSA filename(s), separated by
commas.
--ssa-offset <string> Offset (in milliseconds) to apply to the SSA
file(s), separated by commas. If not specified,
zero is assumed. Offsets may be negative.
--ssa-lang <string> SSA track language as an ISO 639-2 code
(e.g. fre, eng, spa, dut, et cetera)
If not specified, then 'und' is used.
Separate by commas.
--ssa-default[=number]
Flag the selected SSA as the default subtitle
to be displayed during playback.
Setting no default means no subtitle will be
automatically displayed. If 'number' is omitted,
the first SSA is the default.
'number' is a 1-based index into the 'ssa-file' list
--ssa-burn[=number] "Burn" the selected SSA subtitle into
the video track.
If 'number' is omitted, the first SSA is burned.
'number' is a 1-based index into the 'ssa-file' list

[davexnet]

Does using a software encoder (eg. X264) make any difference?

[RogerTango]

Changing it to use presets works fine even with NVENC, so I would say the software filter is getting applied to both HW & SW encoding.

I am using the default with a simple --hqdn3d call and no arguments right now, the outcome is pretty good... I was hoping to only apply temp filtering and leave spat to nil.

I like the way you think though!!!

Thanks,
Andrew

[davexnet]

I didn't bother with the batch variable substitution, just straight command line - this seems to work

--hqdn3d=y-spatial=33:cb-spatial=31:cr-spatial=31:y-temporal=7:cb-temporal=7:cr-temporal=7

I used high values so I could easy see the encode effect - the file got much smaller as the details were removed

[RogerTango]

WHAT?!?!

That is the *one* combination I have yet to try! You have got to be kidding, I overlooked that?!?!

I just plugged it in, slick as snot on a brass doorknob

Dude, you are the bomb!

Andrew

[RogerTango]

Im glad I didnt just throw the project in the trash! I am experimenting with 1 1 1 4 4 4 and 1 1 1 5 5 5 right now, I wanted some light/medium degrain and very light "regular denoise", the results are awesome! Both SD and HD are much improved... Now Ill plug this into my batch for bulk processing.

I own you a coffee and a hamburger, hit me up when your in town!

The still frame does it no justice, when playing (naturally) it looks stunning!


[Attachment 62292 - Click to enlarge]

[RogerTango]

Even stretched out on a 1080 monitor it looks good up close, where you should be seeing pronounced grain... A hint of it remains to keep it "natural". This should look awesome on a TV when you have some distance...


[Attachment 62294 - Click to enlarge]

All these years of programming, and I got tripped up by simple syntax. Thanks for your help!!

[RogerTango]

Asus Nvidia GTX 1050 Ti

This is with upscaling and denoising... direct VOB read and encode.


[Attachment 62295 - Click to enlarge]

I would *like* to have a Turing GPU, but heck.. this card is working great! (Plus Im poor..)

[davexnet]

Any chance of a video sample, something to show off the work. 30 seconds would be nice. I'd like to try it on my TV

[RogerTango]

Any chance of a video sample, something to show off the work. 30 seconds would be nice. I'd like to try it on my TV

Sure! Ill send you a PM link... give me till tomorrow at the latest..