Monday, April 02, 2007

Mastering to HDCAM-SR vs CineForm 444

HDCAM-SR has found its way into to very important parts of the market, on-set capturing and final mastering. The quality is very high so it is clearly suitable for both. In our previous tests we compared CineForm 444 and HDCAM-SR as an on-set recording format. In this test we compared the performance in mastering.

Our source material was 580 frames of the Magic Hour sequence from the Standard test Evaluation Material (StEM) which is used for DCI compliance testing. The footage is very demanding on compression, with a lot of film grain, complex motion and significant detail within each frame. We played an uncompressed 16x9 version out to the Sony SRW1 deck in 440 and 880Mb/s modes, and compared those to range of quality modes offered by the CineForm 444 codec. As the 880Mb/s mode is designed for two camera shooting or operating at 60p, it is really overkill for a single stream of 24p content, so we added CineForm Filmscan2 to the testing mix (our overkill mode.)

The graph says a lot (click to see a larger image):

The large swing in the SR PSNR quality measurements is true to the constant bit-rate codec design, all tape based or fixed rate codecs have this nature. Whereas the CineForm 444 variable bit-rate maintains a mostly constant quality. The 880 Mb/s mode didn't do as well as I expected; doubling the SR bit-rate didn't buy very much. The graphs below compares the 440 and 880 mode for their color channel performance:


The odd element is the red channel's performance in 880Mb/s mode, it should not should have been so much better than green or blue (first 200 frames.) It does in fact appear that the RGB channels are swapped, as if the dual link cables were interchanged for this test, yet they are fine for the 440Mb tests. I'm curious if anyone knows whether the SRW1 will even operate with the dual link cables reversed, or whether this a byproduct of using the 880 (HQ) mode on a single camera stream. We did perform the 880Mb/s test after all other camera and deck testing at 440Mb/s, so an error may have been made when configuring for this test.

The graphs comparing Filmscan1 and Filmscan2 (in keying modes) shows a predictable increase in PSNR quality over all channels equally.

At some point I would like to investigate the 880 mode again, mainly for curiosity, as it is not commonly used (only the SRW1 deck currently supports it.) The SRW1 is a mobile recorder, it is not used for mastering, all the studio decks only support 440Mb/s. At the standard rate of HDCAM-SR, CineForm 444 maintained a higher quality throughout the entire sequence, no mater which CineForm quality setting was selected during the test.

The implications are simple, the digital file based delivery of content is not compromising quality versus tape based mastering and delivery. An entire two hour feature can be mastered to a signal 350GB harddrive, for long term archive (long shelf life than tape), lossless duplication for storage redundancy and data migration, and simple delivery for film-out or mastering to other delivery formats (the drive will cost less than the tape(s).) Tape is still the dominate solution, but as the production world moves towards IT solutions, it is nice to know disk based compression can deliver the goods. For those uncompressed fans, the same feature would consume 1.7TBs, still do-able just less convenient.

17 comments:

Anonymous said...

Hi David

Really enjoying the blog at the moment as it has answered a lot of questions I had regarding compressed workflows. I do have a question however about Cineform and its 4K workflow. Apart from a press release about exeriments with Dalsa I've not heard much more. Can you fill us in in a future blog about the current state of progress. I'm assuming its the same RAW workflow as using the SI camera but I'd like to hear comments about capture/data rates for this solution, especially once the footage is "developed".

cheers

Dean

David said...

Dean,

We intend to be showing our 4K workflow within our booth at NAB. 4K is just a bigger version of 2K, but otherwise the same workflow -- which is pretty straight forward. Unfortunately we won't be capturing 4K CineForm RAW until one the camera vendors as an inexpensive output (Red One with the RAW port isn't availahble yet.) Data rates for high quality RAW 4K (at 24p) ranges between 35-50MBytes/s, you can go lower, but I figure if you are shooting 4K you want quality. Developed footage is 4:4:4 or 4:2:2, and data rate go up 2 to 2.5 times for these formats. We don't yet see much market for 4k 4:4:4 RGB at 100-120MB/s, as we don't need to "develop" 4K to edit it, we think it is more practical to finalize to 2K 4:4:4 for film out or export, leaving 4K for your timeline and composites.

Andrzej said...

Hi David,

Don't the graphs above show that HDCam SR works in CBR mode?

It semas to be that Cineform advantage comes mostly from its VBR nature.

I can be totally wrong :)

David said...

I would agree that HDCAM-SR CBR behavior doesn't help it have a smooth PSNR graph, but CBR codec should not be inherently limited in quality. If two codecs where equal yet one used CBR and the other VBR you would expect the to PSNR graphs to cross each other -- maybe with a similar average PSNR. Yet here the HDCAM-SR 440 graph never touch CineForm 444 for any of the encoding modes. So a 350Mb/s CineForm 444 is better than HDCAM-SR at 440Mb/s whether CineForm was VBR or CBR.

Andrzej said...

Yes, I agree that overall Cineform is better, but I would like to see graph with Cineform at 350Mb in CBR mode. It would look much closer to HDCam SR, but Cineform is VBR what is really good.

Can you coment Cineform vs. Canopus HQ codec?
My internal tests showed that Canopus is slightly better (and faster).
I would like to see your tests (as my workflow is limited).

David said...

If we operate in CBR we would have a graph 4dB higher than SR.

As for Canopus HQ that is an 8-bit 4:2:2 YUV DCT codec, it doesn't compare to a 12-bit 4:4:4 RGB wavelet codec. You may have been testing against our old 8-bit wavelet codec, as of NAB that will be discontinued, and new tools will replace Connect HD, our low end product. All our codec line is getting a big upgrade -- a sneak pre-annoucement here.

Andrzej said...

Yes, I've tested old Cineform not 12bit 4:4:4.

I have a question abput 10bit to 8bit conversion. If the source is 10bit and I'm encoding to Cineform 8bit where is the conversion done?
By Cineform, inside codec?
If yes what metod is used (Round, or advanced Noise Shaping)?

thx

David said...

We encode what we are given going in, so 10-bit is not converted to 8-bit on encode. If the codec supports 10-bit (like Prospect HD) it is encoded as 10-bit. Upon decode if 8-bit is requested from a 10-bit file, the 8-bit is dithered from the 10-bit to reduce banding.

Anonymous said...

Hi David

Thanks for the reply. Not going to NAB but look forward to the comments from people who do.

cheers

Dean

Rith said...

Can't wait to upgrade my 10 to a new 12 bit 444. Thanks Cineform and all of the staffs for all of the hard works to bring this technologies to the market.

I'm enjoy using Cineform products.

Cheer,

Rith.

dany schelby said...

Sounds all very great - can't wait to test the codec with our HDCAM SR workflow on Speedgrade. But we are mac users - why this delay for 12/444 codec ?

Dany

David said...

We did the PC 4:4:4 codec at the same time as the Mac 4:2:2 beta release. We will be adding 4:4:4 to the Mac before any of the NEO product line is release on the mac. It will not be too long.

Andrzej said...

Hi

I've just seen your updated site.
Is it good idea for us (as I HD authoring studio) to go for Cineform codec instead of capturing to uncompressed?
Can I just simply capture with Decklink card (and Premiere) into Cineform 10bit codec in realtime?

Can you (or your friends) contact me about possible workflow?

andrew.kolakowski.gmail.com

Thank you

Andrzej

David said...

Yes to both questions.

Ambarish Natu said...

Hi
I am new to this group, want to know where can i find some 4:4:4 12 or 10 bit /sample video sequences in YUV format . and what tool can i use to test these to calculate parameters such as PSNR , SSIM, MSE etc for such sequences. Moreover can i view such sequences on a standard desktop or laptop, or do i need some sort of calibration done to the DirectX SDK on my machine, answers would be of great help here.

PixMoveMaker said...

Hi,

Where can I read a comparison of ProRes422HQ-HD and yours?

Thanks,
Jay Bala.

criticpapa said...

whoa!!!, you had a passion in blogging, thumbs up for your work of love.. Hehe very inspiring ideas,


anyway I'm william
mind if I put a link back to you?


see my works here ------> Black Suit