digitalFAQ.com Forum

digitalFAQ.com Forum (https://www.digitalfaq.com/forum/)
-   Capture, Record, Transfer (https://www.digitalfaq.com/forum/video-capture/)
-   -   Which of these clips is a better capture? (https://www.digitalfaq.com/forum/video-capture/12341-clips-better-capture.html)

RandomMixTape 12-14-2021 03:03 PM

Which of these clips is a better capture?
 
2 Attachment(s)
Hello

I have captured an extract from the same Hi8 tape using two different capture cards. Both extracts were captured using VirtualDub with HuffYUV lossless compression and with luminance levels set within the 16-235 range using the histogram.

As a beginner to the world of analogue video capturing, I would be very grateful if any more experienced members would be able to critique the two clips attached below (Lagarith avi) and let me know which one represents an objectively better capture and for what reason(s).

In order to remove any bias and conduct a fair comparison, I will not reveal which capture card was used for each clip until I have received a number of responses.

Many thanks

R

Hushpower 12-14-2021 07:37 PM

To my amateur untrained eye, clip 1 has better colours, blacker blacks and whiter whites, although Clip2 could be captured with proc amp adjustment to achieve the same?

The Clip 1 offset to the left is cropping off a few pixels on the left edge.

I like Clip 1 better.

lollo2 12-15-2021 06:00 AM

9 Attachment(s)
After a quick analysys, I prefere the second (Hauppauge USB-Live 2), because once quickly equalized to the first for the overall look, shows (very) slightly more details (the text on the bags on the chairs), better flesh tones contrast, a little bit less noise on uniform small surfaces, (very very) slightly more details in blacks and whites areas, and does not cut a portion of the frame on the left.

At frame 137 in clip1 the bags on the chairs show thin horizontal lines. I suspect these are not details, it may be an artifact in clip1 (or a bad nnedi3 interpolation, but it does not happen on clip2).

clip1 is sharper probably because card defaults, I did not compensate for that.

Do not shrink to Y=235 while capturing with the Haupauge (it can capture 16-254).

A small restoration flow can improve the overall aspect of both.

Your taste may differ!

Histograms
Attachment 14403

Attachment 14404

Attachment 14405

Histograms after equalization
Attachment 14406

Attachment 14407

Attachment 14408

Histograms after equalization and fake deinterlace
Attachment 14409

Attachment 14410

Attachment 14411

Image comparison after equalization and fake deinterlace (use fullscreen and zoom)

https://imgsli.com/ODYzNjI

https://imgsli.com/ODYzNjM

https://imgsli.com/ODYzNjQ

script used
Code:

# plugins directory
plugins_dir="C:\Users\giuse\Documents\VideoSoft\MPEG\AviSynth\extFilters\"

        # FFmpegSource2
loadPlugin(plugins_dir + "ffms2_87bae19\x86\ffms2.dll")

        # Nnedi3
loadPlugin(plugins_dir + "NNEDI3_v0_9_4_55\x86\Release_W7\nnedi3.dll")

video_dir_1="C:\Users\giuse\Desktop\X\"
video_1="clip1.avi"

video_dir_2="C:\Users\giuse\Desktop\X\"
video_2="clip2.avi"

v1=FFmpegSource2(video_dir_1+video_1)\
.crop(0,0,-8,0).addborders(8,0,0,0)\
.crop(8,6,-12,-10)\
.assumeTFF().nnedi3(field=-2)\
#.histogram("levels")\

v2=FFmpegSource2(video_dir_2+video_2)\
.crop(8,6,-12,-10)\
.ColorYUV(cont_y=20, off_y=-10).Tweak(sat=0.8)\
.assumeTFF().nnedi3(field=-2)\
#.histogram("levels")

#subtitle(v1,video_1,size=20,align=2)
#subtitle(v2,video_2,size=20,align=2)

f_stackhorizontal(v1,v2,video_1,video_2)
#f_interleave(v1,v2,video_1,video_2)
#f_Subtract(v1,v2,video_1,video_2)

# analyze frame5 (11 after deint), frame132 (265 after deint) and frame137 (275 after deint)


function f_stackhorizontal(clip c1, clip c2, string t1, string t2)
{
stackhorizontal(\
subtitle(c1,t1,size=20,align=2),\
subtitle(c2,t2,size=20,align=2)\
)
}

function f_interleave(clip c1, clip c2, string t1, string t2)
{
interleave(\
subtitle(c1,t1,size=20,align=2),\
subtitle(c2,t2,size=20,align=2)\
)
}



All times are GMT -5. The time now is 04:47 AM

Site design, images and content © 2002-2024 The Digital FAQ, www.digitalFAQ.com
Forum Software by vBulletin · Copyright © 2024 Jelsoft Enterprises Ltd.