Cineform equivalent to ProRes LT?

[Michael Wisniewski]

Is there a Cineform equivalent to ProRes LT? I will be traveling with a Sony Vegas laptop and I would like to be able lower the storage and rendering requirements.

[David Newman]

ProRes has three quality settings: LT, SQ & HQ
CineForm has five quality settings: Low, Medium, High, Filmscan 1 & Filmscan 2

Pick the one you like best.

[Perrone Ford]

David,

ProRes now also have 5 offerings:

(based on 1080/24p)

ProRes Proxy ~40 Mbps
ProResLT ~115Mbps
ProRes ~145 Mbps
ProResHQ ~220Mbps
ProRes 444 ~440Mbps

[David Newman]

But we are getting silly if we include 444 (although Proxy is fair,) as we have 444 and 4444 and RAW modes, with 5 quality level is each.

[Michael Wisniewski]

Is this a fair comparison between Cineform and ProRes?

Cineform Low = ProRes Proxy
Cineform Medium = ProResLT
Cineform High = ProResSQ
Cineform Filmscan1&2 = ProResHQ
Cineform 444 = ProRes444
Cineform 4444 = N/A
Cineform RAW = N/A

[David Newman]

Cineform Low is a higher datarate than ProRes Proxy
Cineform Medium is a lower datarate than ProResLT
Cineform High = ProResSQ
Cineform Filmscan1 = ProResHQ
Cineform Filmscan2 = Doesn't have that high
Cineform 444 not equal ProRes444 -- CineForm 444 is RGB based, ProRES is always YUV
Cineform 4444 not equal ProRes4444 -- RGBA vs YUVA
Cineform RAW = N/A

[Jay Bloomfield]

David,

One thing that I've wondered about. I assume that the higher bit rate Cineform options put more load on the hard disk subsystem. Do the lower bit rate options (more compression) put more load on the CPU? If so, is there some break even point between the two? The reason that I'm asking is that for some reason, my RAID0 array (which is old and is made up of only two 150GB WD Raptors) has recently been dropping a frame here and there during Intensity Pro captures. I know I should upgrade the RAID array, but I'm wondering if I went to medium, rather than high, whether it might cure the problem. The entire RAID array is empty prior to capture and the BMD (and other) disk speed tests are showing that the array is fast enough to capture 1080i or 720p uncompressed, so I'm somewhat baffled.

J

[David Newman]

That is baffling, as one of those drives would be fast enough. Something could be causing the raid to block occasionally. Try capturing to your system drive to confirm.

As for the question, weirdly (it might seem) for many I-frame codecs (JPEG & Wavelet) more compressed is less CPU load.

[Tim Kolb]

Originally Posted by David Newman

Cineform Low is a higher datarate than ProRes Proxy
Cineform Medium is a lower datarate than ProResLT
Cineform High = ProResSQ
Cineform Filmscan1 = ProResHQ
Cineform Filmscan2 = Doesn't have that high
Cineform 444 not equal ProRes444 -- CineForm 444 is RGB based, ProRES is always YUV
Cineform 4444 not equal ProRes4444 -- RGBA vs YUVA
Cineform RAW = N/A

David,

Are you only comparing data rates? Or...are you implying that the comparisons you lined up are of similarly maintained pre-compression image quality?

[David Newman]

Data rate and approximate quality. We never bothered with an extensive ProRES comparison, as it will depend too greatly on the source. The only test I did 3 years ago showed PSNR numbers (error/distortion) for ProRES are good, and it is nice codec, yet practically you can push CineForm more in post. Ask Shane Hurlbut and guys over a Bandito Brothers, as they did an extensive ProRES to CineForm comparisons in post workflow that lend to choose only CineForm.

[William Boehm]

david..i used cineform neoscene 4 back in march to capture hdv footage. the only three quality settings that were from low to high. i copied in high...filmscan 1 and 2 were not necessary for a canon xha1 hdv qualtiy tape? bill

[David Newman]

Filmscan 1 & 2 only make sense for uncompressed sources, live captures and DPX files. HDV is already too heavily compressed to benefit form filmscan.