Comparison of Native 4k, Native 1080p, and Bicubic Downsampled images

Comments

Musicvid wrote on 2/5/2018, 9:37 AM

Well grace us with your commentary in a thread of your making. This one is obviously not suited. If you think that bicubic scaling in Vegas is not a pattern, which contradicts your previous statement, I'm not interested in the obfuscation.

Sorry you are feeling frustrated, but for the purposes of my challenge thread (which is about learning), it's strike three.

Musicvid wrote on 2/5/2018, 9:55 AM

At least wwjd is trying...

But if I set up the testing stage, it would be sabotaged. Not going there.

wwjd wrote on 2/5/2018, 12:12 PM

Aight, you guys, pop a cold one and chillax. :) Aint no prizes for internet squabbles.

SO which Cameras?

Canon: "soft", orange colored

Sony: yellowish output, low light king

Panny: great bang for buck, reds are more orange than red, not good at low light (except the newest gh5)

Black Magic: get it now!, $6k min with parts

Nikon, Samsung, others... good luck

bitman wrote on 2/10/2018, 6:56 AM

You want examples:

Just shot this today on an Sony RX10iii

So which is which? Half of it is shot in 25P HD, the other half in 25P 4K, both are added together in Vegas15 and rendered to 25P. Normally I film in 50P HD, but for this exercise I use 25P as I cannot film in 4K 50P.

I used the highest bitrate settings in both cases.

So which one comes from 4K? The first or second part?

Does it make sense for the RX10iii to film in 4K if you need HD (and do not need cropping)?

I noticed that filming in 4K with this camera you loose one F-stop of light with the same shutter speed. One F-stop is half the light lost folks! Not to mention you cannot shoot beyond 25P in 4K, in HD you can go up to 100 frames.

http://www.dropbox.com/s/ec9fr4ikrp5isdg/TwoVideoSources.mp4?dl=0

APPS: VIDEO: VP 365 suite (VP 22 build 194) VP 21 build 315, VP 365 20, VP 19 post (latest build -651), (uninstalled VP 12,13,14,15,16 Suite,17, VP18 post), Vegasaur, a lot of NEWBLUE plugins, Mercalli 6.0, Respeedr, Vasco Da Gamma 17 HDpro XXL, Boris Continuum 2025, Davinci Resolve Studio 18, SOUND: RX 10 advanced Audio Editor, Sound Forge Pro 18, Spectral Layers Pro 10, Audacity, FOTO: Zoner studio X, DXO photolab (8), Luminar, Topaz...

  • OS: Windows 11 Pro 64, version 24H2 (since October 2024)
  • CPU: i9-13900K (upgraded my former CPU i9-12900K),
  • Air Cooler: Noctua NH-D15 G2 HBC (September 2024 upgrade from Noctua NH-D15s)
  • RAM: DDR5 Corsair 64GB (5600-40 Vengeance)
  • Graphics card: ASUS GeForce RTX 3090 TUF OC GAMING (24GB) 
  • Monitor: LG 38 inch ultra-wide (21x9) - Resolution: 3840x1600
  • C-drive: Corsair MP600 PRO XT NVMe SSD 4TB (PCIe Gen. 4)
  • Video drives: Samsung NVMe SSD 2TB (980 pro and 970 EVO plus) each 2TB
  • Mass Data storage & Backup: WD gold 6TB + WD Yellow 4TB
  • MOBO: Gigabyte Z690 AORUS MASTER
  • PSU: Corsair HX1500i, Case: Fractal Design Define 7 (PCGH edition)
  • Misc.: Logitech G915, Evoluent Vertical Mouse, shuttlePROv2

 

 

Musicvid wrote on 2/12/2018, 12:44 PM

Yes I saw your tests and they are appreciated. Although framing and levels are a little different, the left image reads sharper on my 50"

Zooming/cropping are the RIGHT reasons to shoot 4k. Not even the Blusterer-in-Chief would disagree with that.

I wish someone with sdi could capture and record simultaneously at two resolutions. That way we might get a clean SSIM to consider.

wwjd wrote on 2/13/2018, 7:43 PM

Unless you REALLY botched the render, on my 4K monitor, from 3 different video players, the 2nd half is 4k downscaled. Unless you played with it, and defocused the first half on purpose, and sharpened the 2nd half. :)

I upscaled some HD to 4K today, and it took really well!

Musicvid wrote on 2/14/2018, 9:58 AM

How was the second image downscaled? Bicubic?

wwjd wrote on 2/14/2018, 10:10 AM

Me? I upscaled, whatever defaults. Left is the HD frame, right is upped to 4K, 32bit, and graded, sharpened. There's not much different, but it looks useable on a 4K monitor. That from a crappy Canon T6i HD video.