H.265 VS H.264 - D-Cinelike comparison

I haven't attempted H.265 yet, but I've given up on D-log, not because I can't get it to work, but because the settings I do have to use to get it to work is not even close to what I'm after when using D-log. I really wish DJI would give us the option to choose our own ISO in D-log, I think it would go a long way in making it more accessible to the masses. Also, I agree, a Prores option in a future Phantom iteration would be the sheet.
 
The ISO in any Log mode has to be locked at the native ISO of the camera in order to use the full dynamic range of the sensor.

The 35mm cinema cameras that I use in my production studio are locked at 2000 ISO when using S-Log. It's just the nature of shooting log. If it were any other way, the results would be less than optimal. We control the ISO with a LUT in post production. A handy dandy tool I like to use is LUTcalc.
 
The ISO in any Log mode has to be locked at the native ISO of the camera in order to use the full dynamic range of the sensor.

The 35mm cinema cameras that I use in my production studio are locked at 2000 ISO when using S-Log. It's just the nature of shooting log. If it were any other way, the results would be less than optimal. We control the ISO with a LUT in post production. A handy dandy tool I like to use is LUTcalc.
Thank you for the explanation, I had no idea. In order to get a noiseless result in D-log, I have to change the camera settings to a style that I'm not fond of on the P4P, however, I use D-log with an ISO of 100 on my Mavic Pro Platinum, and really like the results, but I guess that's apples and oranges.
 
There's no such thing.
I use them every day with my 35mm cinema camera footage. I have LUTs that change the effective ISO to 500, which is the sweet spot for noise-free footage. The LUTs used with Sony CineAlta series are rated in ISO equivalents. We also make our own LUTs with LUTcalc. In that program, you can choose the ISO of the effective output of the LUT when applied to footage. Check out LUTcalc and see what you've been missing.
 
I use them every day with my 35mm cinema camera footage. I have LUTs that change the effective ISO to 500, which is the sweet spot for noise-free footage. The LUTs used with Sony CineAlta series are rated in ISO equivalents. We also make our own LUTs with LUTcalc. In that program, you can choose the ISO of the effective output of the LUT when applied to footage. Check out LUTcalc and see what you've been missing.
But really, this misunderstands the reality of ISO numbers being apples to oranges: there is no benchmark of "an effective ISO" because 500 on one camera could look like 1000 on another. If you understand the nature of log curves, then you understand how it's critical to lock ISO at capture for consistent application of the log curve for maximum (e.g., accurate) dynamic range.

That doesn't stop companies like Sony, though, from being utterly arbitrary when it comes to over-conservative mandating of high ISOs for their log curves, resulting in a high noise floor. Occasionally they relent to pressure, like when they changed the minimum ISO of the PXW-FS5 from 3200 to 2000.
 
Sony probably did that to bring it more into line with the FS7. The "native" ISO is one in which the video (analog) amplifier after the imager has a unity gain.

Optimum ISO depends on the scene and the observer's sensitivity to noise, as well as the quality of noise from the sensor. Under bright conditions, I've got LUTs with -2EV built into the curve, which gives an effective ISO of 500. The LUT pushes noise to invisibility.
 
Again: "effective ISO of 500" relative to what camera sensor? Apples to oranges unless you're referring to an actual, specific camera model/generation.
 
It doesn't matter what sensor. What matters is that the ISO can be adjusted down after the fact through the use of an appropriately designed LUT.
 
H.265 was created to provide a smaller (more compressed) data stream for distribution of video. It is a last step before distributing a video after post. At that point it matters little how long this compression requires as it is a one time last step in the process.

For video streaming it will be years before H.265 becomes the norm due to the amount of processing power needed to decompress the data packets as they arrive at the viewing device. In that sense it is similar to Blu-Ray that was going to dominate data storage and distribution. As of 2018 Blu-Ray is a dying market for content distribution.
 
H.265 is also in the process of being replaced with AV1 which is royalty free unlike HEVC where the costs increased a great deal over H.264. There are changes taking place and they are being pushed by the big dogs in AOMedia whose founders include Netflix, Facebook, Microsoft, Apple, Google, Cisco, Intel, nVidia, and IBM.
 
Some amazing replies in this thread. I was noticing leaves being more distinct in H.264 than H.265. I had to compare numerous times but I'm not sure if it's because I'm viewing 2.7k on a 1080p screen and the H.265 is indeed more detailed than H.264?
 
Some amazing replies in this thread. I was noticing leaves being more distinct in H.264 than H.265. I had to compare numerous times but I'm not sure if it's because I'm viewing 2.7k on a 1080p screen and the H.265 is indeed more detailed than H.264?

I wouldn't bother with H.265 at 2.7K particularly if you're only planning to view in 1080P. You need to use 4K to and view it in 4K to make that kind of assessment.


Brian
 
Comparison footage in 4K between H.265 and H.264 codecs with D-Cinelike in DJI Phantom 4 Pro.
ISO 100, Style: Sharpness -1, Contrast -3, Saturation -2

I think we must use h.265 in D-Cinelike because we get more accurate colors, less artifacts and less noise than in h.264


H.265 provides for more data compression of video files so less bandwidth is needed for streaming and nothing more. It also requires more than 10 times as much processing power do decrypt the data which is why it will take a whole new generation of monitors and video transcoders to process in real time. It is only of value in a limited number of commercial applications and of no value in a home theater environment where even Blu-ray has been a commercial failure.

Different applications are going produce different results with the processing of image data and that is what you are seeing and not a image quality gain despite a 40-50% increase in compression with H.265.
 
H.265 provides for more data compression of video files so less bandwidth is needed for streaming and nothing more. It also requires more than 10 times as much processing power do decrypt the data which is why it will take a whole new generation of monitors and video transcoders to process in real time. It is only of value in a limited number of commercial applications and of no value in a home theater environment where even Blu-ray has been a commercial failure.

Different applications are going produce different results with the processing of image data and that is what you are seeing and not a image quality gain despite a 40-50% increase in compression with H.265.

yeah, not an image quality gain at all but a loss if anything

Some youtube videos showed H.265 being better than H.264 but it could be a scam.
 
I have tried importing h265 in adobe cs6 and DaVinci with no luck. What are you guys editing with?
 
With Apple 4K ProRes 4444 XQ a production camera shooting 100 hours of footage can produce 76 TB of data and all that data needs to be stored somewhere. The Phantom Flex 4K RAW at 24fps produces 1500 GB/hour and numbers like this are what is driving the need for data compression that is effective and efficient.
Sounds like a good reason to employ DOPs trained to work with film! It's an interesting concept though - video producers needing to direct tighter sequences.
It's the most compatible format. It is one of the factors driving me toward getting an Inspire 3 when it hits the market.
I'd definitely skip the Inspire 2 - a load of trouble at the moment.
 
If you don't have lots of footage to work with I assume H.264 is more ideal in terms of quality? Of course, if you are dealing with data in the TB the H.265 is important? How does the computer handle uncompressing H.265? You can't even film in C4K60 at H.265, only H.264 if not it drops it down to 48 fps!
 
If you don't have lots of footage to work with I assume H.264 is more ideal in terms of quality? Of course, if you are dealing with data in the TB the H.265 is important? How does the computer handle uncompressing H.265? You can't even film in C4K60 at H.265, only H.264 if not it drops it down to 48 fps!

Video recorded using H.265 is not smaller than H.264 -- they both consume 100Mbps so the file sizes are pretty much the same for a minute of video. The theory is that with H.265 being more efficient it can record the same detail in a smaller file OR more detail in the same sized file. Again, the file size will be the same for a minute of video. You will not save any drive space using H.265 with the DJI drones -- not a byte!


Brian
 

Recent Posts

Members online

No members online now.

Forum statistics

Threads
143,087
Messages
1,467,528
Members
104,965
Latest member
cokersean20