Talk:VHS/Archive 6
![]() | This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page. |
Archive 1 | ← | Archive 4 | Archive 5 | Archive 6 |
About DVD resolution
I just reverted an IP who changed the DVD resolution figure from 500 to 720 lines, with edit comment referring to pixel count .The IP is correct in that the DVD format encodes 720 pixels per line. However that number is not directly comparable to the "lines of horizontal resolution" specs for VHS or S-VHS. Kell factor is not the primary issue here.
When in TV work we speak of "lines of horizontal resolution" ("TV lines" or TVL for short) that number is in lines per picture height. If we convert the DVD pixel count to that (multiply it by 3/4) we get 540. It's usually quoted as 500 because darn few DVDs are ever as good as that.
If we converted the VHS and S-VHS numbers to lines per picture width, we'd get larger numbers... but they're never quoted that way, and anyway analog formats don't have pixels at all... so we stick to what appears in reliable sources, which is the "TV lines" figure.
This is the same metric btw by which NTSC analog broadcast is quoted at about "330 lines" (you get about 80 "lines" per MHz; you are limited to 4 MHz bandwidth by the presence of the sound subcarrier in the broadcast signal). Jeh (talk) 06:07, 26 April 2016 (UTC)
- Isn't the reduction due to either a horizontal Kell factor (!), or accounting for overscan? If you work it backwards in the latter case, you get about 444 visible scanlines for NTSC, and 533 for PAL (7.5 to 8.0% overscan), which is close enough to double the figures I've seen quoted for "typical effective maximum" vertical resolutions of old TV-connected computers and games consoles (around 224 and 264 respectively, hence the 320x224 mode of the Sega Genesis, and the 320x256 default mode of the PAL-region Commodore Amiga and BBC Micro) that operated in 240p or 288p modes that I'm very tempted to just straight up call it as that. With the standard being to imagine a square that encompassed the visible height of the screen and a horizontal width equal to that because a lot of early TV sets had visible images rather closer to circular or at least a fairly narrow oblong/ovoid shape rather than a 4:3 rectangle (or in the early days, an officially 5:4 rectangle) owing to the difficulty of making CRTs that deviated much from the shape of a chemist's conical flask... Thus if you quoted the definition that even someone with the cheapest possible receiver was likely to be able to enjoy, they wouldn't be able to have you for overselling a particular product, and the ultimate resolution would be equal to that number multiplied by the individual picture tube's aspect ratio... ie, pay for a more rectangular set, get more visible lines.
- I didn't know (ie, "I'm not convinced that") NTSC was limited to 4mhz because of the audio, btw. I thought it was essentially the amount of waveband that a channel was allowed via different transmission methods (OTA, cable, satellite) in different areas. I've heard as high as 6mhz and as low as 2.5 (!) - or by your estimate that would be 206 to 495 lines - with the fairly narrowband audio carrier (maybe modulated/heterodyned up to several megahertz centre frequency but still only 50khz wide at the absolute maximum) simply moved as far from the whole-signal centre frequency as any neighbouring channel's vestigial lower sideband would safely allow, with the luma filtered to below typical audio background noise level just outside of that band with as steep a cutoff as possible without causing ringing effects, and the chroma similarly placed within the overall band such that it was as far from centre as possible (higher frequencies = less obvious mosquito noise on monochrome receivers) whilst still having a filter roll-off on the less important upper band (Q, for NTSC, I think?) that also avoided the audio. Given that NTSC's normal colour frequency is 3.58mhz, I'm not sure quite how the 2.5 and 3.0mhz bandwidths I saw listed would have worked, and can only assume they were monochrome-only VHF transmissions. It's been a while since I read that actual list; 4mhz would be the minimum practical whole-channel bandwidth without adjusting the colour encoding frequency, however (I know some experiments used 2.2 or 2.6mhz, but that was abandoned fairly early on), so even where higher transmission bandwidths are the actual norm, it probably gets quoted as the baseline minimum. Of course, PAL and SECAM colour demands a slightly higher minimum bandwidth (at LEAST 4.5, but preferably 5 to 6mhz) given the 4.43mhz colour carrier used in most versions of those standards...
- I expect that "4mhz" is a "line" or "wave" frequency (able to show a black and a white pulse within full cycle as there's a positive-going and a negative-going peak within that), btw, given that a typical pixel clock for the default TV-compatible resolution in the aforementioned machines was in the range of 6 to 8mhz and produced about 256 to 360 pixels across the visible scan width (and 352-360 across the entire line would of course be equal to 6.75mhz, half the DVD frequency), so those "line" counts would actually be 412 to 990 luma pixels (or 660 at 4mhz) across the central square and maybe 550-1320 (880 @ 4mhz) visible... not that bad really, and I guess why hi-rez modes with about 512-640-720 pixels across were considered fair game, at least in reduced colour depth and sometimes using a monochrome-only composite output where RGB wasn't (easily) available.
- ((yeah, guess what subject I've been researching recently that's led me to this page in a roundabout way.))
- ((also I guess that means minimum NTSC colour bandwidth is approx 400khz for the lower resolution component? (4mhz - 3.58 = 0.42, minus 20khz for the lower sideband of the audio signal) So, like ... 33 lines?! Or 88 visible "pixels" of that component, though they would be smoothed out and not on clearly defined centres... Don't know what the other one would be, but ISTR it was either a 2:1 or 3:1 typical ratio between them, thus 0.8-1.2mhz, 66-99 lines, 132-198 equivalent pixels. Which is... OK, I suppose? So long as your equipment was reasonable quality and could preserve that as a reliable minimum throughout, and you were sympathetic in choice of material and framing? ... and given that it probably wasn't possible to cut the vestigial lower sideband down to less than maybe 400-450khz itself, that would add up to an overall minimum channel separation of 4.5mhz, for 4mhz of useful signal... possibly the "2.5mhz" VHF channel might actually be 2mhz of video, then? 440 pixel-equivalent across the visible screen in monochrome (=440 x 444 overall, with fat-ish pixels, ignoring Kell factors). So long as you ignore the loss of colour, that might still be an acceptable picture on a smaller screen.))
- (((and, therefore, ultimately, if each of the colour components gets that same 400khz when recorded onto VHS at the lower carrier frequency - which seems unlikely, but let's run with it - it's not AS big a reduction as first thought, because the one that was already blurry isn't reduced any further, and the higher definition one is "only" cut down by 2, MAYBE 3x, to match it. With the luma also reducing by a good 25% or so, therefore the loss of clarity vs broadcast might seem relatively uniform, and actually be better than most people would expect - actually around 660 visible pixel equivalent for luma, despite a rather rough 44 pixel equivalent for colour...))
- So yeah, interesting things to think about there
- Also I guess that means visible DVD pixels across *should* be 1485, or at least 742, on a typical screen, if it's "500 line"? Something doesn't quite add up here... 702~714 pixels of usable image within a scanline (usual quoted figure for PAL/NTSC, regardless of the 704~720 digitally encoded size) with 7.5-8.0% overscan comes out to 649~660 pixels ... which is ... uh...
- ...487 to 495 TV lines :/
- OK, I give up, I have obviously lost the ability to do maths. That or something's got twisted somewhere.
- ...or, wait, it's that Kell factor stuff, isn't it. 330 lines / approx 440 actually scanned within the visible height = 0.75 ... is that the usual one counted for interlaced NTSC? :)
- anyhow, time to sign off 193.63.174.254 (talk) 10:52, 4 May 2017 (UTC)
- Briefly, yes, that's how NTSC gets to 330 lines of vertical resolution for a camera aimed at a real-world scene. They do count the entire raster and Kell factor was measured at about 0.7. Hence 330. You would only get 480 lines of v.res if your scene's details happened to fall on the scan lines. The converging lines in the standard test pattern will clearly show this.
- For h.res in analog (like NTSC) with a pickup that does not have pixels, Kell factor does not apply, because the scene is not sampled across the width of a scan line. The max bandwidth of broadcast NTSC is 4.2 MHz. Not counting the horizontal retrace interval, etc., you have 52.4 uS per scan line for visible picture info. In which time a 4.2 MHz signal has time for 220 cycles - 220 light spots alternating with 220 dark spots, which TV counts as "440 lines". Allowing for the 4:3 screen size gets us to the NTSC standard of 330 lines of horizontal resolution per picture height, or 78.6 per MHz.
- This btw was deliberately chosen so that H and V res (in lpph) would be about the same. In other words those converging lines on the old "Indian head" test pattern should "fuzz out" at about the same point in both the H and V dimension.
- But if you have a camera with a digital sensor, then the Kell factor does apply to h luma res. Say you had a camera pickup with, oh, say 720 pixels horizontally. Then Kell factor says you can count on about 500 pixels' worth of resolution in real-world scenes. Allow for 4:3 and this is 375 "lines of horizontal resolution per picture height". Hm, not that much better than NTSC!
- You can believe or not believe anything you want, but that 4.2 MHz limit of NTSC as broadcast (or emitted from an RF modulator) is very definitely due to the presence in the channel band of the sound carrier, which any old-school TV repair tech will tell you is at 4.5 MHz up from the luma carrier. Occasionally on a misaligned TV - or with a broadcaster with a badly adjusted modulator - you will hear a buzz in the sound when the white levels are up too high (often on self-keyed white graphics) and it slops over into the sound channel. If you're not putting NTSC on a broadcast-standard RF carrier then there is no sound carrier at 4.5, or anywhere else, and that is why S-VHS, Super Beta, and everything else that claims more "lines" than 330 can do it. (Unless you're using the playback machine's RF output; then you're limited to 330.)
- "the whole-signal centre frequency " - there is no single "center frequency". The channel is 6 MHz wide. The carrier for the "4.2 MHz wide" luma+chroma signal is 1.25 MHz up from the channel lower edge (it's AM, but the LSB is suppressed). The 4.2 bandwidth of the luma+chroma signal takes us up to 5.45 MHz. The FM sound carrier is at 5.75 up from the channel lower edge. See here for a diagram.
- Chroma resolution for NTSC, horizontally, is encoded at 125 lpph for the I signal and 25 lpph for the Q. [1] (Vertically of course it is the same as the luminance res.) Sorry but subtracting 3.58 MHz from 4 MHz to get the "bandwidth for the chroma" is not a useful thing to do. The carrier of the chroma signal is at 3.58 MHz, but after modulation it is actually splattered all through the 4.2 MHz bandwidth of the luma. But its energy is practically all in a series of narrow "spikes" that correspond to places where the luma has none, or very little. This is the purpose of the "comb filter" you may have heard of that's used to separate them - its frequency response looks like a comb. If you've ever seen areas in a color TV picture like a herringbone fabric show up with color bands, that's because the filtering isn't being done perfectly, and the luma is getting into the chroma.
- The chroma res. sounds terribly low but our eyes' chroma resolution is much worse than for luma (we have a lot more rod cells than cones) and it's really not much less than what we can discern anyway. The imbalance in resolution in the two components was picked to give the best color rendition in the range of colors that matter most to us - flesh tones. Unfortunately most color TVs' chroma decoders cheat and decode both to just 25 lpph. See YIQ.
- I'm not going to say much about speculations on how the width of the broadcast channel could have been reduced, except to say that the various guard bands, etc., were picked to give good selectivity - that is, lack of adjacent-channel interference - with commercially viable tuners of the day. Even so, two strong adjacent VHF channels won't play nice with each other, and we have to settle for decent alternate-channel performance. That's why channels in urban areas were assigned on an every-other-one basis.
- This should give you some food for thought re NTSC. I only skimmed the DVD aspects of your comment - will have to get back to those later. Jeh (talk) 01:01, 6 May 2017 (UTC)
Objectives
The seven bullets in VHS development are actually eleven of the twelve objectives, according to the indicated source http://www.rickmaybury.com/Altarcs/homent/he97/vhstoryhtm.htm. (5 "key features" + 6 "six key consumer and manufacturing requirements".) The twelfth, something about "social implications" is so vaguely worded in the source that it can't be used for the article. Do we have a better source? If not, what should we do about this? — Sebastian 06:50, 10 May 2017 (UTC)
External links modified
Hello fellow Wikipedians,
I have just modified 3 external links on VHS. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
- Corrected formatting/usage for http://www.ce.org/Press/CEA_Pubs/941.asp
- Added archive https://web.archive.org/web/20110111005825/http://www.nipponsei.jp/n-hajimete/n-hajimete009.html to http://www.nipponsei.jp/n-hajimete/n-hajimete009.html
- Corrected formatting/usage for http://tafkac.org/products/beta_vs_vhs.html
When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.
This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}}
(last update: 5 June 2024).
- If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
- If you found an error with any archives or the URLs themselves, you can fix them with this tool.
Cheers.—InternetArchiveBot (Report bug) 01:36, 27 July 2017 (UTC)
Regarding the "Industry Losses" statement
I noticed that an industry assertion about the hundreds of millions of dollars being lost to illicit copies was listed unchallenged. I don't feel that's NPOV, providing only an (unattributed) statement by one of the parties most concerned, and one which in addition has a strong incentive to present biased information. In my previous experience, I've seen an underlying assumption of very questionable accuracy: that if they did not have illicit copies, the people would have purchased all of the same products at full retail price. (for example, every teenage warez dood who uses a cracked copy of Photoshop to put captions on his Facebook pictures would have shelled out hundreds of dollars to Adobe, instead of just using MS Paint) Accepting this assertion as fact is probably a bad idea, because it's presenting a biased, one-side account. IMO, this should be either removed, expanded with other involved parties' opinions on the subject, or at least better explained as an industry statement rather than a known fact. Worldwalker (talk) 17:20, 21 August 2018 (UTC)
Unnecessary edit that was also poorly worded?
https://wiki.riteme.site/w/index.php?title=VHS&diff=869394971&oldid=869389811 The reverting user said the above to the last edit at the time of the revert. I don't agree, but maybe someone does?211.27.126.189 (talk) 12:37, 21 November 2018 (UTC)
- The edit I reverted had added this text:
which would add up to cacophony in audio and video, especially for the user
- Let's see:
- "cacophony" is defined as "a harsh, discordant mixture of sounds". (The -phon suffix should have been a clue.) So you can't have a "cacophony" in video, unless you're writing metaphorically - which is fine for poetry, but a poor style fit for an encyclopedia.
- "especially for the user". These words convey no information. They are fluff words, i.e. needless padding. Think: Who else would be affected but the users?
- Exactly what you get with incomplete replacement of old tracks with new is highly situation-dependent. For example, if the signal from the control track isn't "clean", many VCRs (particularly more recent ones) will mute the audio. This is by design, because manufacturers found that most users prefer that to noise.
- In short, what you had there was very carelessly worded. We are trying for a reasonably competent level of English composition here.
- As for "unnecessary" - I disagree with the notion that we need to explain to the reader that incomplete replacement of old signal with new would be a bad thing. Jeh (talk) 13:24, 21 November 2018 (UTC)
- I didn't apply an effort even close to 0% in the wording (I never do a O don't know why you said it was careless). I didn't know the audio can be muted. Some people care about their tapes not being erased, some don't.211.27.126.189 (talk) 20:12, 21 November 2018 (UTC)
- I don't know why the wording you said careless (there was something wrong when I said it 1st time). It makes perfect sense to me (why can't you do that for yourself?), which means I think it wasn't careless. — Preceding unsigned comment added by 211.27.126.189 (talk) 21:09, 21 November 2018 (UTC)
- I explained already. I could explain it again, but I can't understand it for you. Ask your English teacher. Jeh (talk) 08:26, 22 November 2018 (UTC)
- I didn't carelessly makes the wording up, whether I can be made to understand why its careless.211.27.126.189 (talk) 11:17, 22 November 2018 (UTC)
- A truly careful writer cares about the meanings of the words and phrases they use. And a good Wikipedia editor, or anyone else who cares about writing effectively, accepts valid criticism and corrections (such as "you can't use that word there, because it doesn't mean what it would have to be mean to be used there") and says "thank you, I learned something," rather than quibbling over specific words in the criticism.
- As I explained already: you used a word without knowing what it meant; you used a phrase that added no meaning whatsoever, only nonsensical puffery; you made a general claim without realizing that there are important exceptions. Seems pretty careless and poorly worded to me.
- And in your defense you keep harping on the word "careless" - thereby showing that you're missing the point. Whatever effort you made, it was obviated because you didn't (and apparently still don't) understand the meaning (or lack thereof) of the words and phrases you were using.
- In any case, whether you think you were being careless or not, your edit reduced the quality of the article. So I reverted it, and I stand by that action whether you accept the judgment of "careless" and "poorly worded" or not. Jeh (talk) 13:14, 22 November 2018 (UTC)
- Final post: Careless means without effort. I am allowed to make edits that I feel improve the article but don't actually do so. I think something is poorly worded if it makes no sense to my or not as much as I want it to. I try to improve how text Ames sense on Wikioedia. That's mostly what I do on Wikipedia.211.27.126.189 (talk) 19:50, 22 November 2018 (UTC)
- I didn't carelessly makes the wording up, whether I can be made to understand why its careless.211.27.126.189 (talk) 11:17, 22 November 2018 (UTC)
- I explained already. I could explain it again, but I can't understand it for you. Ask your English teacher. Jeh (talk) 08:26, 22 November 2018 (UTC)
Is VHS truly obsolete?
IIRC, there's a handful of movies and TV shows which still haven't made the jump to DVD or other digital formats. Plus, some stores still carry blank VHSs. Sure new movies aren't put on VHS anymore, but the fact that there are some movies and TV shows which literally CANNOT be watched legally outside of VHS must mean something. MightyArms (talk) 23:20, 16 May 2019 (UTC)
Official Standards?
The article starts off by saying that VHS is a standard, but it seems like there are no ANSI/ISO or similar standards for it, nor does the article cite the official standard. Does that mean it's a proprietary standard? I'd like to know if such an official spec for VHS even exists, even the name of it would be helpful. 50.68.13.81 (talk) 06:06, 15 September 2019 (UTC)
"In popular culture" section
Some anon user added the "In popular culture" section in which reads:
In an episode of Regular Show, entitled "The Best VHS in the World", Rigby and Mordecai have a VHS named "The Best VHS in the World".
The section will likely to be removed if they are not backed up by sources. I've been looking for sources, and there isn't seem to be any at the moment. Using the Regular Show Wiki as a source (wikis are not acceptable as sources) doesn't do any good. – Hounder4 18:20, 25 September 2019 (UTC)
Time Lapse Security VCRs
Maybe I missed it, but this article doesn’t address time lapse security VCRs.
We still use these at my business in 24 hour and 960 hour speeds.
A few other businesses still use it.
2600:8801:E00:5300:6C89:DF6F:53C:8F7F (talk) 05:16, 9 January 2020 (UTC)
Digital horizontal resolution
Where do the numbers of 333 and 335 for NTSC and PAL respectively stem from? They seem like unnecessarily rough approximations of what could be theoretically specified exactly. If we consider a bandwidth of 3 MHz and an active line duration of 52 microseconds, we would essentially get a resolution of 312x576 under the Nyquist sampling theorem. 80.162.33.59 (talk) 23:16, 22 March 2021 (UTC)
- Since VHS is analog, you don't have clear definitions of where one pixel ends and the next begins. Instead you get reductions in high frequency data and possibly subjective opinions on how much noise and blur one can have before a line on a test chart becomes indistinct. — Preceding unsigned comment added by Algr (talk • contribs) 22:48, 20 November 2021 (UTC)
VHS HiFi
I remember the days of Beta-HiFi, and then VHS HiFi. Beta writes the combined video and HiFi audio signal with the same heads. It moves the FM video signal up in frequency enough to fit the signal in. Players can find the signal at this higher frequency. As well as I remember, VHS couldn't do that, so that is why the complication with extra heads and depth recording. If there was enough unused bandwidth, they could have done it directly. But yes, the audio signal is recorded first, and written over by the shallow shvideo signal. I presume, like the two video heads, they use different head azimuth to keep them apart, but I don't have the numbers for that. Much less convenient than Beta, and also I suspect sensitive to tape quality. It would be nice to get this right. Gah4 (talk) 21:23, 23 May 2022 (UTC)