8K News Round-Up – Panasonic Embraces 8K, Codecs and More
This week saw a number of announcements that added to the momentum of 8K so we thought it worth detailing them in a round-up.
Panasonic Comes Round on 8K
Back at the end of 2023, Panasonic’s camera developers were quoted as being very negative about the adoption of 8K video in its cameras. That surprised a few people, as Panasonic had previously said that it had planned to develop professional 8K cameras as part of the build-up to the 2020 Olympics. However, the group has clearly changed its mind and last week announced the Lumix S1RII full frame mirrorless camera which has a new 44.3MP sensor and supports 8K video capture. There’s a detailed analysis of the new camera here.
Features of the new hybrid camera include open-gate (full sensor aspect ratio) in 8.1K (after a firmware update). CineD interviewed Matt Frazer of Panasonic in some depth (its 44 minutes if you listen in full). The firm said in the interview that the 6.4K open-gate video is downsampled from the full 8.1K sensor image. Frazer also explained that the camera can also support external recording of 8K video using the camera’s HDMI port. Although portable recorders to capture this are not yet widely available, computer capture cards can be used for example in stadiums and worship facilities to exploit the 8K feature. Panasonic is working with Blackmagic Design and Atomos to support 8.1K RAW capture.
The camera will also allow 8.1K open-gate video to be recorded on an external SSD if an additional battery grip is used on the camera. This allows more power draw for the camera to handle the extra processing. A 3 amp dummy battery/PSU can also be used for this.
Getting True 8K Video
There is a group of dedicated individuals that are really keen to ensure that what we see as ‘8K’ (or other high resolutions) is really that quality and not just a lower resolution image in a high resolution container. One of those is Dirk Hildebrandt of Wavelet Beam. We interviewed him after meeting him at IBC last year and you can see that conversation here.
Hildebrandt took to Linked In with three articles that discussed the topic we reported on several weeks ago, that you need to be really careful when trying to use simple numbers to evaluate different codecs. His first article includes a video image of the moon in 4K. As the phases of the moon are shown, the true video resolution in pixels (measured by his software) drops from 3.7K which is effectively ‘full resolution’ to under half resolution as the moon goes into darkness. His other two articles shows how it is important to understand the bit rate limits for different codecs. As he puts it “Each video encoder technology operates within a specific range of bitrates that are most effective for a given video resolution and content type”.
The 8K Association believes that the problem of low resolution video in a high resolution container is one of the reasons that some of the comments that ‘you can’t see the difference between 4K and 8K’ are made. Often, the content may be labelled as 8K but there is not enough detail to really show the difference. Researcher Hosub Lee and Bill Mandel of Samsung Research, along with camera maker, Pawel Achtel, will be presenting a paper on their research into the visibility of 8K at the SID Display Week in May. If you will be attending, make sure you note it. Once it is published, we will, of course be reporting on it.
Nvidia Boosts 8K GPU Processing in new SDK
In separate codec-related news, Nvidia released its Video Codec SDK 13.0 which exploits the latest Nvidia Blackwell GPU technology to improve video encoding and decoding in H.264 and HEVC as well as boost performance in AV1. There are also improvements in suport for stereoscopic content using MV-HEVC. Our attention was drawn to the H.264 8K support and the firm said that the GPU:
Now handles ultra-high-definition video decoding with ease, providing future-proof capabilities for 8K content.
There is a detailed description of the improvements on the Nvidia Developer blog but they include better chroma sub-sampling, more 10 bit support and higher encoding efficiency. The firm said that In support of the Video Codec SDK 13.0 launch, Blackmagic, Capcut, and Wondershare have already integrated features such as 4:2:2 encode, 4:2:2 decode, AV1 UHQ, and split encoding in their video pipelines. The company will be presenting more details in the annual GTC event.
Meta Provides Tool for Video Quality Assessment in AV1
Also in the area of codec development, David Ronca of Meta posted on Linked In about a new tool, VCAT, that can be used to benchmark hardware or software decoding of AV1 content. The tool fills a gap that the Meta team found in tools to understand how mobile devices could use AV1. The firm said that the tool would help SOC and handset vendors, Cellular Service Providers (CSPs), and Content Access Providers (CAPs).
