Talk:Uncompressed video
This article is rated Start-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects: | |||||||||||
|
Article Title?
editI think that the title of this article should not be "Uncompressed Video." Why? Because HDMI/SDI/DVI always transmit uncompressed video; they cannot transmit compressed video. Whenever someone watches video, they are watching uncompressed video. The fairly new concept of "Clean HDMI," referring to a device which has never compressed the video, is similarly confusing, because there is nothing at all different about the HDMI signal itself.
I propose that a better title/definition is "Clean Video." This would refer to video which has never had lossy compression applied since its creation. Such video therefore has no motion or other compression artifacts. Other than chroma sub-sampling, the video is "Clean". In this way, "Clean Video" refers to the video content, whether it is streamed live from the image sensor, or captured to a storage card in RAW format, or subsequently transferred to a computer and edited without lossy compression ever being applied.
If we define "Clean Video" in this fashion, we could subsequently add designations to the "Clean Video." Why? At all stages of capture/storage/transmission, the chroma sub-sampling and/or bit depth of the video can easily be increased or decreased, which is also lossy compression. For example, Clean Video 10-bit 4:4:4 would refer to video which has never been decimated to less than 4:4:4 and/or 10-bits. The Clean Video designation should always reflect the minimum quality to which the video has been processed. If it has ever been reduced to 4:2:0, for example, its CV designation should remain 4:2:0, even if it is later up-sampled to 4:2:2 or 4:4:4. Such upsampling is meaningless, because chroma (color) data has been stripped out when the video was down-sampled to 4:2:0.
Whatever it is called, this product feature is important and costly to implement and use, and valuable. There is a very substantial visual difference between Clean Video and video which has been compressed, particularly when the video is paused. Therefore, having a clear understanding and definition of what this means is important to consumers.
- Is there any evidence (i.e. in secondary sources) that Clean video is an established term? I beleive Uncompressed video is an established term. ~Kvng (talk) 21:20, 23 December 2017 (UTC)
Bitrates
editThose bitrates should be in megabits (Mbps) and gigabits (Gbps). The uppercase "B" means byte, the values displayed there are represented in terms of bits, which is represented by a lowercase "b". Need to replace "MB" for "Mb" and "GB" for "Gb". I would do it, but I don't know how.
- I'm unable to immediately reproduce the calculations in Uncompressed video § 4:2:2 format which I beleive are the figures in question here. My issue may have something to do with incomplete understanding of Chroma subsampling. In any case, I don't think this sub-section is much help to readers and I propose to strike it. ~Kvng (talk) 21:43, 23 December 2017 (UTC)
Agreed that any chroma subsampling must be taken into account, i.e. 4:2:0, 4:2:2 or 4:4:4.
Not Enough Information
editIt appears that there are some unstated assumptions used here. In the example data rate calculations (24-bit, 1080i @ 60 fps: 24 × 1920×1080 × 60/2 = 1.49 Gbit/s; 24-bit, 1080p @ 60 fps: 24 × 1920×1080 × 60 = 2.98 Gbit/s), the discrepancies between the two examples are explained (Interlaced video formats transmit every other line, half the picture content, per field period. Two field are required for a full frame so the frame rate is halved in this calculation). Although the colour space is unstated in the example above, 24 bits are stated (RGB?) and the mathematics works.
The remaining examples appear to assume prior knowledge of an unstated colour space (RGB, YUV etc.) and the mathematics does not work without this assumed knowledge. This makes it difficult to understand what is being described here. As this is a topic that is frequently misunderstood (based on my research), this article does not help dispel the confusion.
It would be very helpful if the original author could state the missing assumptions, for the benefit of the rest of us.
Thanks for everyone's efforts to date.
Scj242 (talk) 21:19, 7 October 2018 (UTC)
This article seems to be lacking important information. Two problems in particular:
1. There is no explanation of what "8 bit" and "10 bit" refer to. I presume this is not colour depth, since that seems way too low.
2. Two rows with exactly the same values have different data rates. Specifically, "10 bit @ 1920 x 1080 @ 24fps" under 1080i and 1080p HDTV uncompressed and "10 bit @ 1920 x 1080 @ 24fps" under 1080i and 1080p HDTV RGB uncompressed have two different data rates. Why?
Please add enough information to solve both of these problems. I would do so myself, but I don't know enough about this.
Secarrie (talk) 20:06, 16 February 2012 (UTC)
- Yes, it refers to color depth which is reported in two different ways, the first being all three channels per pixel (for example, 24 bit, with 8 bits for each channel), or 8/10/12/14/16 bit per channel, for an overall value of 24/30/36/42/48 bits. Bumblebritches57 (talk) 18:43, 17 March 2015 (UTC)
- 1080i has half as much information as 1080p. Interlaced formats only transmit every other line per field period. The example in Uncompressed video § Data rates is correct and I have added an explanatory note. I haven't been able to reproduce the calcualtions in Uncompressed video § 4:2:2 format and have proposed deleting this subsection (see #Bitrates above). ~Kvng (talk) 21:43, 23 December 2017 (UTC)
Potentially wrong information
edit“ | 720 @ 59.94field | ” |
How come? 720p is only a format for progressive scan, not interlaced, so it should have a frame rate, not a field rate. --Mahmudmasri (talk) 18:54, 9 July 2012 (UTC)
Container formats
editWhich container formats can use uncomoressed video? There are loads of articles on wikipedia about h264 and other compression codec contsiners but the only 10 bit rgb format mentioned is quicktime what others can do this for Windows/Linux machines? — Preceding unsigned comment added by 86.27.131.165 (talk) 13:42, 31 December 2014 (UTC)
Raw vs Uncompressed
editThere is a lot of fast and loose wording in this article that seems to interchange the idea of RAW video and uncompressed video. These are two very different things. Doorzki (talk) 02:03, 2 January 2015 (UTC)
There is also a bizare claim that uncompressed video has better resolution than "raw" formats. I read through the linked article on (still image) raw formats. Apparently Adobe's "Digital Negative" format is essentially a newer version of TIFF; which is a container format that does support lossy formats such as jpg. 70.74.233.146 (talk) 18:34, 16 February 2015 (UTC)
- Summary, this article is bad, but not a candidate for deletion. How about renaming it, the lede suggests "clean HDMI"? –Be..anyone (talk) 01:22, 17 February 2015 (UTC)
- I agree, RAW should refer to RAW video codecs like DNG for example) while uncompressed should refer to straight PCM. Bumblebritches57 (talk) 18:40, 17 March 2015 (UTC)
- I agree that the wording is fast and loose, but disagree that the article should be renamed "clean HDMI" -- a term that lacks reliable, consistent definition. The article purports to be about "original" (vs. "decompressed lossy") video, but there is no common term for such video, and what could one really say about it anyway, except that it has no compression artifacts? After a brief introduction, the article basically turns into a detailed rehash of HDMI. Perhaps it would make sense to merge the short mention of "original" video into HDMI and then delete this article. Lambtron (talk) 16:09, 19 July 2015 (UTC)
Massive Nikon spam in lead section
editI've removed the long list of Nikon models from the lead section, because it's covered by the info template at the bottom, and off topic here. –Be..anyone (talk) 18:16, 7 March 2015 (UTC)
- The navigation template is not an article info. And its not spam. 77.12.4.177 (talk) 15:30, 8 March 2015 (UTC)
- This article is already too bad without a long off-topic list of Nikon models in its lead section. If you want to enumerate those cameras do it on an article about Nikons, not here. I've invited a third opinion. –Be..anyone (talk) 04:50, 9 March 2015 (UTC)
- As a third opinion was requested, in my opinion, the list in the navigation template is more than sufficient. That type of information is far too detailed for the lead, and would be overly detailed even for the article. Seraphimblade Talk to me 05:04, 9 March 2015 (UTC)
- This article is already too bad without a long off-topic list of Nikon models in its lead section. If you want to enumerate those cameras do it on an article about Nikons, not here. I've invited a third opinion. –Be..anyone (talk) 04:50, 9 March 2015 (UTC)
Move Lossless video compression to it's own article
editThere's tons on info available about lossless video compression for it to have it's own article, and calling losslessly compressed video uncompressed is 100% inaccurate, it's very much compressed, it just doesn't throw data away, while uncompressed video is just straight PCM. what does everyone else think? Bumblebritches57 (talk) 18:28, 17 March 2015 (UTC)
- Maybe you meant "like PCM for audio", with lots of possible containers (WAV/RIFF, CAF, MKV, OGG, NUT, "voc" seen on requested aticles, etc.) Otherwise ACK, the page as is doesn't cover Y4M, YUVMPEG1PIPE, or similar names/formats for "raw" video, where I'm not sure what "raw" actually is, but "losslessly compressed" sounds good. –Be..anyone (talk) 01:57, 19 March 2015 (UTC)
- I second moving lossless video to its own article, with uncompressed in its own section, because uncompressed is a subset of lossless. --Traal (talk) 19:46, 11 June 2023 (UTC)
- At present, I don't see where lossless compression is covered in this article. Lossy and lossless compression are mentioned in contrast to uncompressed. This article is fine as it is. I think it is fine to keep coverage lossless compression in Video coding format beaches some of the same formats are used just in lossless mode. But if someone wants to make an argument at Talk:Video coding format that lossless should be handled separately, have at it. ~Kvng (talk) 14:42, 14 June 2023 (UTC)
Reorganization
editI'm going to just completely reorganize this article and removing the inaccurate information, and split things such as Clean HDMI, Lossless Video Compression, and anything else that needs to be split out. Bumblebritches57 (talk) 18:50, 17 March 2015 (UTC)
Lousy vs. lossy log
edit129.12.102.116, 79.64.175.114, and 146.200.74.205 so far. –Be..anyone (talk) 10:41, 15 April 2015 (UTC)
What is the primary topic of this article?
editIs this article supposed to be about raw video or HDMI? In video engineering, "raw video" is uncompressed digital video that comes directly from a digital camera (e.g., with Camera Link/DVI/HDMI output), or from an analog video decoder (e.g., with BT.656/BT.1120 output), or from a decompressor (e.g. H.264 decoder). HDMI is just one of many formats of raw video. Right now the article is a jumbled mixture of raw video and HDMI -- two vastly different topics.
As an aside, the lede currently says "clean HDMI" is synonymous with "uncompressed video" -- a patently false claim. Other sources define "clean HDMI" in various other, completely different ways (e.g., as HDMI video that lacks text/graphic overlays). The article needs a concise definition of the primary topic, and then a complete overhaul to expand that topic in a coherent way.Lambtron (talk) 15:25, 26 May 2015 (UTC)
Spam
editSeeing the contents of this article and the references, this mostly seems a way for companies to promote their products... Instead of of giving actual insight, they just mention they have products available which support certain standards... 37.153.194.221 (talk) 10:55, 10 September 2015 (UTC)
- If I'm active at the time I'll support an AfD (hopelessly unclear topic). –Be..anyone (talk) 22:11, 11 April 2016 (UTC)