Cheap 15 Monitor for FX1 Z1

[paultv]

Has anybody tested one of these??

http://www.247electrical.co.uk/epages/twentyfourseven.storefront/EN/Product/RELRLT1520BS?SOURCE=TD

Thought it might make a cheap viewing device for 1080i component users - according to the spec it should work...but

[Alan Roberts]

You're right, according to the spec it should accept the signals and make pictures. BUT, it's only 1024x768 pixels, and the FX1/Z1 image is 1440x1080. So, you're going to throw away the top third of the horizontal resolution; you'll probably not miss it visually though, but you could be fooled if you're doing things like adding titles, keying and so on.

The 768 lines aren't really a problem, since the FX1/Z1 is interlaced, so you can't get the full 1080-line resolution, you get a realistic maximum of about 3/4 of that (i.e. about 768).

For the price, it's a good start, but if I'd spent the price of a FX1/Z1, I think I'd be prepared to spend a bit more than £200 for a display that doesn't quite make it (and 500:1 contrast isn't anything like good enough for tv).

[paultv]

Yes, I agree, I was thinking of using it as a cheap field display, as I'm often shooting live music in studios, and the massive Sony 14 inch HD monitors are just too cumbersome to use, despite producing good pictures.

We wanted something for focus monitoring which is much more critical with HDV than SD, particularly in less than perfect lighting where the lens is wide open, giving poor depth of field.

I find the expand focus facility a good idea but useless, as it fails to work during record, so not much help at critical focus points when shooting.

These observations are after shooting a 4 Z1 recording session with 100 musicians, and cutting on a Canopus NXHDV system; any slight focal errors really show up big time.

paul

[paultv]

aaaahh!
this is more like it!

http://www.novatech.co.uk/novatech/specpage.html?SAM-MH17WS

thanks Richard

paul

[Alan Roberts]

Yep, that's much more like it.

[Gyr]

I may be missing something very basic here, but the max. resolution is 1280 x 1024 (not 1440 x 1080) and like the Relisys model mentioned at the start the contrast ratio is still only 500:1.

So doesn't the Samsung still miss the mark (only not by as much)?

[Alan Roberts]

Yep, s'right. 1280's closer to 1440 pixels than 1024 is, and 1024 lines is more than enough for 1080 lines.

[Gyr]

Thanks for confirming that Alan. I've come to video from a non-technical background, so please bear with me if I ask another question. If I find a monitor that is 1440 x 1024 and has component input will that be good enough for an FX1/Z1 or are other connectivity features needed?

And if a contrast ratio of 500:1 isn't really good enough, what figure is OK (without moving into the realms of the more expensive monitors).

[Alan Roberts]

You'll struggle to find 1440x1024 (or 1440x1080 which would, in theory, be the right numbers). 1280x768 should be fine for FX1/Z1 because tyhe camera doesn't make detaiul right up to the 1440 limit (actually, it can make some detail (because the lens is too good at short focal length) beyond that limit but it's spectrum-folded back and is best removed, using a slightly lower resolution display is a good enough way to do it). Since the FX1/Z1 is an interlaced camera, you don't get detail all the way up to the full 1080-line limit; interlace can't do that, so you get detail up to about 65% of the limit, about 720, and that means that a 768-line display is fine.

Contrast ratio claims are highly dubious in this game, particularly in flat panel displays. The reason's a bit complex, but I'll try to keep it simple. The "target" display for tv is still the crt, it has a power law relationship between the signal voltage (video level) and light output, typically L=V^2.35 for studio monitors. The value 2.35 is the result of my measurements of many such displays over the years. Consumer tv sets tend to have a lower value, my Pansonic 28" set has a value of 2.05.

Other displays have to mimic this law so that puctures look right. Plasma panels are inherently linear so they include a look-up table to convert the non-linear input signals (video) to produce the right output. They're not very good at it because they don't use enough digits, they would ideally need a 14-bit system or better, but the early panels were only 6-bit, and we're only seeing 8-bit panels in the high-end market now. They make up some of the short-fall by using processes very like the half-toning used in printing, so you get raging dancing patterns in low-lights. Nasty.

LCDs are a bit different because they are inherently non-linear, but don't exhibit a power law relationship. Some cheap displays don't bother to try correcting for it, and produce odd pictures. Others have look-up table corrections which work, to some degree. The lcd on my Vaio does a decent job, but still doesn't get it anything like right.

But, what has this to do with contrast?

Well, the contrast ratio that a display can make is the ratio of the brightest light level it can make over the darkest. So, contrast gets better as you make the blacks blacker. CRTs are very good at this, I've measured over 3,000:1 on a decent HDTV crt, and over 1,000:1 on a consumer tv set. The problem with lcds is that the light level changes with viewing direction. That's because the light modulation is done using polarising filters and they're notorious for giving odd polar responses. So the manufacturers tend to use the measurement that gives the best numbers, even if they don't relate to actual viewing conditions. And that's why the advertised contrast ratios often bear little relationship to what you actually get.

In and lcd or plasma with 500:1 contrast, the black level (signal voltage equals 0%) is 0.2% of peak white. The maximum light level they make in the consumer market is usually 250 candelas/square metre or higher, so the "black" level for 500:1 would be 250/500=0.5 candela/square metre. And that's clearly visible in subdued lighting. In a crt with 500:1 contrast, you actually get much darker blacks because the measured contrast in the ratio of white to black that you get withing a typical picture, and crts tend to spread some of the light within the glass screenb, thus raising the black level. But when you fade the video to black, there's no light to polllute the black so you get much darker blacks.

This effect is called "flare" in crts, and explains why a crt with "only" 500:1 contrast can look so good. Plasma and lcd displays don't exhibit this effect, the black level made is not affected by the whites in the rest of the picture.

So, 500:1 is probably ok for viewing in full lighting, but for subdued environments, you'll notice the black level as a veiling dark grey.

Hope that helps.

[Gyr]

Many thanks for that Alan, and thanks for keeping it simple - it is clearer in my mind now.