Author Topic: Minimum bitrate requirement seems really high (Read 7238 times)

dragonblade · « **on:** May 30, 2018, 19:50 »

It's been a while since Ive read Videoblock's technical requirements. I read them when I first signed up but have since forgotten some of the finer details. I read them again just now and was surprised to read that the minimum bitrate they accept is 50,000 kbps. That seems unusually high if you're submitting HD content. None of my HD cameras capture bitrates anywhere near that figure and I wouldn't exactly call them cheap cameras. They're definitely not point and shoots. Among them is a Panasonic G6 which has a fairly good reputation for HD video and produces a bitrate on average about 23,629 kbps or thereabouts. I have an older GoPro model that boasts a lower bitrate than that.

Thinking about this some more, I'm curious why they accepted a 1280 x 720 video from my Panasonic G2 since the bitrate for that clip would have been around 5,812 kbps.

increasingdifficulty · « **Reply #1 on:** May 31, 2018, 03:05 »

In a professional situation, 50 mbps is on the low side. I know lots of cameras encode harder than that, and it looks fine, but just use ProRes or encode a new 50+ file and the problem is solved.

The thing with a 28 mbit file is that, while it may look perfectly fine, it leaves little room for grading since the invisible information has been thrown away.

SuperPhoto · « **Reply #2 on:** May 31, 2018, 06:01 »

Speaking of which - can someone please explain to me the advantage of a 'higher' bit rate?

I have a large monitor (25") as well as large TV (60") - and if I view HD/4K video on them at say 20MPS as opposed to 150MPS - they look pretty much identical to me.

So I don't get/see the advantage of a higher bit rate.

increasingdifficulty · « **Reply #3 on:** May 31, 2018, 06:26 »

Higher bitrate means less compression.

The goal of compression is to minimize file size while keeping an acceptable level of image quality. What is acceptable will differ from application to application, and from person to person. An untrained eye will not notice heavy compression, while a professional editor can spot it right away.

Modern compression algorithms are really good, and for simply VIEWING a video, you can get away with quite heavy compression.

But when you're selling stock footage, you're not selling it to be viewed directly. You're selling it to be EDITED, or at least to have the option to be graded and changed.

Heavy compression removes the ability to grade and change colors, as what you see is the only information still present. In an uncompressed file, there is A LOT of information that you simply don't see, but that is necessary when you want to change something.

Compare a RAW file to a JPG. They can look identical if you have JPG compression set at 100%, but as soon as you want to make changes, like raise shadows or change the colors, you will quickly notice that the RAW file has A TON more information and the JPG will quickly fall apart and look terrible.

SuperPhoto · « **Reply #4 on:** May 31, 2018, 06:41 »

Hmm, okay - 'kind' of making a bit more sense - but can you give me an example of a specific color change (i.e., say premiere pro) - where it would look bad in one but good in the other? I do editing myself (on both say 50kpbs, and 10kpbs) - and for the most part - looks identical.

The 'only' time I would say notice a difference is if you had gradients (i.e., a timelapse with a blue sky). *Then* yes, I admit I definitely notice the 'banding' on a lower kpbs/more compressed item...

increasingdifficulty · « **Reply #5 on:** May 31, 2018, 06:59 »

Well, as usual it comes down to each specific clip and situation, what is in the clip, how much movement there is, and how heavy your changes are.

Also, with more experience, you will start to notice more and more things that are unwanted. "For the most part" identical might not cut it in a high-end project. Just like when making music, where it can take years for your ears to develop enough to hear what's good and bad, it takes a long time for your eyes to recognize certain things when it comes to grading. That is why grading is a job in itself, and a beginner will almost always produce quite bad results.

There are also different kinds of compression - inter-frame and intra-frame for example. The most effective compression algorithms and codecs, like h264, use inter-frame encoding - which means it calculates only the changes between a set number of frames.

A common algorithm can look like this: 1 keyframe (completely new frame) and 7 frames following that with only the changes updated. If there is a lot of motion, you will see ugly blocking.

An intra-frame codec, like ProRes, encodes a completely new image on each frame. This will increase the bitrate, but is much better and easier to handle when editing. And all the information is there for each frame, should you need to cut three frames of a clip, or something like that.

Just raise the shadows on a RAW file to the max, and do the same with a JPG version of that image at level 5 quality, and you will see a big difference.

Brightontl · « **Reply #6 on:** May 31, 2018, 07:19 »

Quote from: SuperPhoto on May 31, 2018, 06:01

Speaking of which - can someone please explain to me the advantage of a 'higher' bit rate?

I have a large monitor (25") as well as large TV (60") - and if I view HD/4K video on them at say 20MPS as opposed to 150MPS - they look pretty much identical to me.

So I don't get/see the advantage of a higher bit rate.

I own two drones: A Phantom 4 and a Phantom 4 Pro.
The p4 compress to 50MPS.
The P4 Pro compress to 100MPS and to be honest also has a big sensor.
The P4 Pro blows the P4 completely out of the water, partly because of the bigger sensor and better lens (especially in bad light situations), but even more because of less compression (because of the compression fine details like trees and houses look extremely mushy in the P4).
And also, as said by Incr Difficulty, when you color grade, you have more latitude with the P4 Pro, although you still don't have much latitude because 100MBS is still a huge compression.
Just in case, this is my comparison:
https://youtu.be/9nOCKJQ1eb4

increasingdifficulty · « **Reply #7 on:** May 31, 2018, 08:10 »

Quote from: Brightontl on May 31, 2018, 07:19

Quote from: SuperPhoto on May 31, 2018, 06:01
Speaking of which - can someone please explain to me the advantage of a 'higher' bit rate?

I have a large monitor (25") as well as large TV (60") - and if I view HD/4K video on them at say 20MPS as opposed to 150MPS - they look pretty much identical to me.

So I don't get/see the advantage of a higher bit rate.
I own two drones: A Phantom 4 and a Phantom 4 Pro.
The p4 compress to 50MPS.
The P4 Pro compress to 100MPS and to be honest also has a big sensor.

Yes, big difference between the two, but worth mentioning is that the 100 mbit/sec is for 4k. That would be like HD compression at 25 mbit/sec.

Maximum HD bitrate on the P4P is 60 mbit/sec (for standard framerates).

That being said, GH4 4k compression is also 100 mbit/sec, and it can look absolutely fantastic.

SuperPhoto · « **Reply #8 on:** May 31, 2018, 17:36 »

Hmm, okay, thanks for the feedback. I think I am 'starting' to 'get' it.

dragonblade · « **Reply #9 on:** May 31, 2018, 22:00 »

Quote from: increasingdifficulty on May 31, 2018, 06:26

What is acceptable will differ from application to application, and from person to person. An untrained eye will not notice heavy compression, while a professional editor can spot it right away.

When it comes to movement / motion in a video with a low bitrate, I would say that even an untrained eye will see that there's something amiss. A few years ago, an Australian TV channel was broadcasting sports at very low bitrates. Every time a player on the field moved at a quick pace, their limbs would become a pixelated mess.