50mbps vs 100mbps vs bit depth at DVinfo.net

Tony Reidsma · July 20th, 2019, 11:21 AM

I've never really dug deep into video tech to understand the differences and benefits of:

MBPS
- 8 bit vs 10 bit
- But I think it's time for me to ask those of you who are experienced with the tech side of these two factors.

I have the Sony A7R II and the Fuji XH1

NOTE: I am focused on 1080p, not 4K

The Sony will shoot 50mbps, and the Fuji will shoot 100mbps

Both shoot 8bit 4.2.0 (I believe)

QUESTION:

Will the Fuji at 100mbps give me more room for grading? Or will it merely give me more details in the image?

Thank you in advance for any insight!

Seth Bloombaum · July 20th, 2019, 05:24 PM

Mbps (small b, for bits, megabits per second), or Mb/s, sometimes called bitrate or bandwidth, is a measure of the rate at which information is being recorded. Large B is for bytes, which is how we measure storage.

With a couple caveats below...

Although 100 sounds twice as good as 50, that doesn’t necessarily lead to better gradeability. On a single camera’s settings, all other things being equal, a 100Mbps stream should show fewer compression artifacts, like:
Macroblocking, in which several pixels are grouped into one big one of 16x16 or 32x32px. This is sometimes incorrectly called “grainy”. It happens when an encoder doesn’t have enough available bandwidth to accommodate all the pixels that are changing between frames.
Banding, in which there seem to be lines of pixels of different colors where there should be smooth gradations, typically the sky would show this.
Mosquito noise, in which pixels of the wrong color occur near transitions between objects of high-contrast differences.

But, 50Mbps is a lot of bandwidth for FHD.

Unless, there’s high picture complexity in your content. Moving water. Wind in the leaves or grass. Noise from higher gain settings. Lots of subject movement. Handheld/jerky camerawork. Then you may find that 100Mbps makes a difference.

Normally, we’d think of higher color resolution as being a more important factor for heavy grading.

4:2:2 color subsampling, or 10-bit color can both help more directly.

But, if you’re seeing any of the above compression artifacts in your footage, or it shows with heavy grading, you may be looking at some bandwidth starvation of your original footage.

I’d think of this as uncommon, but it really depends on what you’re shooting, the lighting, camera movement, and other contributors to picture complexity.

It would be very cool to test with your cameras! Find some highly complex images to shoot...

Christopher Young · July 21st, 2019, 05:23 AM

Quote:

Originally Posted by Tony Reidsma

The Sony will shoot 50mbps, and the Fuji will shoot 100mbps

Both shoot 8bit 4.2.0 (I believe)

QUESTION:

Will the Fuji at 100mbps give me more room for grading?

Not necessarily so! Oh I wish it were all as simple as that. It's very hard to make direct comparisons bit rate for bit rate when quite different and often proprietary encoding techniques are used using proprietary algorithms, DSPs and FPGA's as in the case of Sony with its XAVC codec. This applies to all their XAVC codecs right through to the F55 camera for example.

In Sony's case they specifically developed their XAVC format ecompassing H.264 like many other manufacturers but in Sony's case they are the only ones to date who in real time multi-pass and "Select appropriate encoding parameters by 'pre-encoding' for accurate rate control which enables them to achieve stable high picture quality constantly at much lower bit rates." as they put it. Why are Sony the only ones multi-pass pre-encoding H.264 in real time? Sony have been consistently ahead of the curve when it comes to developing very high speed DSP processing which is used very heavily in real-time encoding. Most other camera manufacturers are constantly playing catch up with Sony in this particular field of technology.

This why people are often confused as to how often Sony seem be able to deliver a quality level that belies the bit rate the vision was encoded at. Sony's VBR 50Mbps XAVC very often, scene dependent to a degree, will be as good if not better than some competitors 100Mbps and sometimes higher bit rates. Especially when comparing 4:2:0 with 4:2:0. The biggest grading improvements tends to come with greater color depth as mentioned above by Seth, 4:2:2 vs 4:2:0 not necessarily bit rate is the big helper here.

So don't be surprised if you get very similar results between Sony's 50 versus Fuji's 100. Both need handling with care as neither are high bit rate compared to some of the 600Mbps and up XAVC flavors.

A bit more of an explanation can be found here:

https://bento.cdn.pbs.org/hostedbent...2013-v05_1.pdf

and at 06.12 and beyond on this video:

Chris Young
Sydney

Tony Reidsma · July 22nd, 2019, 06:01 AM

Thank you for your replies. Very helpful!

July 20th, 2019, 11:21 AM	#1
Tony Reidsma Regular Crew Join Date: Feb 2010 Location: Michigan Posts: 69	50mbps vs 100mbps vs bit depth I've never really dug deep into video tech to understand the differences and benefits of: MBPS - 8 bit vs 10 bit - But I think it's time for me to ask those of you who are experienced with the tech side of these two factors. I have the Sony A7R II and the Fuji XH1 NOTE: I am focused on 1080p, not 4K The Sony will shoot 50mbps, and the Fuji will shoot 100mbps Both shoot 8bit 4.2.0 (I believe) QUESTION: Will the Fuji at 100mbps give me more room for grading? Or will it merely give me more details in the image? Thank you in advance for any insight!

July 20th, 2019, 05:24 PM	#2
Seth Bloombaum Inner Circle Join Date: Sep 2003 Location: Portland, Oregon Posts: 3,420	Re: 50mbps vs 100mbps vs bit depth Mbps (small b, for bits, megabits per second), or Mb/s, sometimes called bitrate or bandwidth, is a measure of the rate at which information is being recorded. Large B is for bytes, which is how we measure storage. With a couple caveats below... Although 100 sounds twice as good as 50, that doesn’t necessarily lead to better gradeability. On a single camera’s settings, all other things being equal, a 100Mbps stream should show fewer compression artifacts, like: Macroblocking, in which several pixels are grouped into one big one of 16x16 or 32x32px. This is sometimes incorrectly called “grainy”. It happens when an encoder doesn’t have enough available bandwidth to accommodate all the pixels that are changing between frames. Banding, in which there seem to be lines of pixels of different colors where there should be smooth gradations, typically the sky would show this. Mosquito noise, in which pixels of the wrong color occur near transitions between objects of high-contrast differences. But, 50Mbps is a lot of bandwidth for FHD. Unless, there’s high picture complexity in your content. Moving water. Wind in the leaves or grass. Noise from higher gain settings. Lots of subject movement. Handheld/jerky camerawork. Then you may find that 100Mbps makes a difference. Normally, we’d think of higher color resolution as being a more important factor for heavy grading. 4:2:2 color subsampling, or 10-bit color can both help more directly. But, if you’re seeing any of the above compression artifacts in your footage, or it shows with heavy grading, you may be looking at some bandwidth starvation of your original footage. I’d think of this as uncommon, but it really depends on what you’re shooting, the lighting, camera movement, and other contributors to picture complexity. It would be very cool to test with your cameras! Find some highly complex images to shoot... __________________ 30 years of pro media production. Vegas user since 1.0. Webcaster since 1997. Freelancer since 2000. College instructor since 2001.

July 22nd, 2019, 06:01 AM	#4
Tony Reidsma Regular Crew Join Date: Feb 2010 Location: Michigan Posts: 69	Re: 50mbps vs 100mbps vs bit depth Thank you for your replies. Very helpful!