What is a good way to edit and upscale 3D gameplay videos?

[Guernsey]

I usually record and edit videos from Sega Genesis and Super Nintendo games via emulation but I never had to do 3D games before. Not every game on the PS1/Saturn/Nintendo 64 games are 3D but 3D is different from 2D and even 2D sprites are better animated than the SNES or Genesis. I am mostly using emulation to dump the videos and it already has an FFMPEG resizer built in but I want to know if I should a Lanzcos resize via Virtualdub2 or some other method to make it presentable?

Edit: To add to my question, what should do about the settings of the recording when I am editing the videos? I usually like to make videos of individual levels / Should I use the same settings as the video below? Or different settings?

Complete name               : ******.mp4
Format                      : MPEG-4
Format profile              : Base Media
Codec ID                    : isom (isom/iso2/avc1/mp41)
File size                   : 1.51 MiB
Duration                    : 50 s 604 ms
Overall bit rate mode       : Variable
Overall bit rate            : 251 kb/s
Writing application         : Lavf58.22.100

Video
ID                          : 1
Format                      : AVC
Format/Info                 : Advanced Video Codec
Format profile              : High 4:4:4 Predictive@L2.1
Format settings             : CABAC / 4 Ref Frames
Format settings, CABAC      : Yes
Format settings, ReFrames   : 4 frames
Codec ID                    : avc1
Codec ID/Info               : Advanced Video Coding
Duration                    : 50 s 566 ms
Source duration             : 3 s 540 ms
Bit rate                    : 173 kb/s
Width                       : 352 pixels
Height                      : 240 pixels
Display aspect ratio        : 3:2
Frame rate mode             : Constant
Frame rate                  : 59.940 (59940/1000) FPS
Chroma subsampling          : 4:4:4
Bit depth                   : 8 bits
Scan type                   : Progressive
Bits/(Pixel*Frame)          : 0.034
Stream size                 : 1.04 MiB (69%)
Source stream size          : 1.04 MiB (69%)
Writing library             : x264 core 157 r2935 545de2f
Encoding settings           : cabac=1 / ref=3 / deblock=1:0:0 / analyse=0x1:0x111 / me=hex / subme=7 / psy=1 / psy_rd=1.00:0.00 / mixed_ref=1 / me_range=16 / chroma_me=1 / trellis=1 / 8x8dct=0 / cqm=0 / deadzone=21,11 / fast_pskip=1 / chroma_qp_offset=4 / threads=7 / lookahead_threads=1 / sliced_threads=0 / nr=0 / decimate=1 / interlaced=0 / bluray_compat=0 / constrained_intra=0 / bframes=3 / b_pyramid=2 / b_adapt=1 / b_bias=0 / direct=1 / weightb=1 / open_gop=0 / weightp=2 / keyint=250 / keyint_min=25 / scenecut=40 / intra_refresh=0 / rc_lookahead=40 / rc=crf / mbtree=1 / crf=23.0 / qcomp=0.60 / qpmin=0 / qpmax=69 / qpstep=4 / ip_ratio=1.40 / aq=1:1.00
mdhd_Duration               : 50565
Codec configuration box     : avcC

Audio
ID                          : 2
Format                      : AAC LC
Format/Info                 : Advanced Audio Codec Low Complexity
Codec ID                    : mp4a-40-2
Duration                    : 50 s 604 ms
Duration_LastFrame          : -16 ms
Bit rate mode               : Variable
Bit rate                    : 66.2 kb/s
Maximum bit rate            : 128 kb/s
Channel(s)                  : 2 channels
Channel layout              : L R
Sampling rate               : 44.1 kHz
Frame rate                  : 43.066 FPS (1024 SPF)
Compression mode            : Lossy
Stream size                 : 404 KiB (26%)
Default                     : Yes
Alternate group             : 1

[jagabo]

That bitrate looks rather low for a capture master. Use a high a bitrate as you can stand if you're going to use x264, crf 12 or less, crf 0 (lossless) if you can. At least you used 4:4:4 chroma. I would avoid resizing while capturing. It's best to avoid multiple resizings.

Can you post a sample?

[Guernsey]

Here is the sample.

[jagabo]

As I suspected that video has lots of compression and resizing artifacts. If you can, avoid resizing and use better quality compression. A better source would give smoother edges and an overall cleaner picture.

[Guernsey]

What do you suggest using then?

[jagabo]

Since you're already using x264 -- set it to crf=0 -- that will give you lossless encoding. And disable the resize filters -- capture the emulator frame pixel-for-pixel.

[Cauptain]

Hi Guernsey

All current emulators have a video capture engine (I use Retroarch for 95% my captures).

As Jagabo said, use the X264. My configuration is above:


[Attachment 57945 - Click to enlarge]

For resize you have 2 options:

1 - Use some editor like Virtualdub + Avisynth filters

2 - Use upscale from emulator Retroarch using shaders like xbr, xbrz (recommended) or nnedi3 and already record in upscaled mode.



Claudio

[Guernsey]

Hi Guernsey

All current emulators have a video capture engine (I use Retroarch for 95% my captures).

As Jagabo said, use the X264. My configuration is above:


[Attachment 57945 - Click to enlarge]

For resize you have 2 options:

1 - Use some editor like Virtualdub + Avisynth filters

2 - Use upscale from emulator Retroarch using shaders like xbr, xbrz (recommended) or nnedi3 and already record in upscaled mode.



Claudio

Can this configuration apply to most x264 codecs? Or just the new ones?

[jagabo]

Can this configuration apply to most x264 codecs? Or just the new ones?

That dialog is specific to x264vfw, but all x264 encoders have those settings.

[Guernsey]

I see. So are certain settings in x264 encoders such as DAR (Display Aspect Ratio) similar to SAR (Sample Aspect Ratio)? Or they different from each other? Other than that there isn't that much difference between most of those encoders right?

[jagabo]

I just noticed that Guernsey's x264vfw dialog shows DAR instead of SAR (the version I have has SAR there). DAR and SAR are different settings. DAR is the display aspect ratio, the final shape of the picture. SAR is the sampling aspect ratio, technically the relative distance between samples horizontally and vertically -- people think of this as the shape of individual pixels. I'm pretty sure x264 doesn't support DAR directly. So the program that Guernsey is using must convert DAR to a SAR for x264:

DAR = FAR * SAR

DAR = display aspect ratio, the final shape of the picture
FAR = frame aspect ratio, frame width:height
SAR = sample aspect ratio, the relative distance between samples horizontally and verticcally

I believe the hardware decoder settings don't have anything to do with x264 encoding.

[Guernsey]

Okay I think I should be good. I will give those a shot.

[Guernsey]

What do you suggest I use as far AVISynth + Virtualdub filters?

[jagabo]

nnedi3_rpow2(), aWarpSharp2(), Sharpen(), LSFMod(). For cartoons you may want to use line darkeners like Hysteria() or FastLineDarkenMod(). You also want to reduce noise and other crap before upscaling -- otherwise you just get bigger crap. RemoveDirt(), RemoveGrain(), TemporalDegrain(), etc.