noise = no noise?

[Mennovandijk]

Today I plugged in my old crt 22" Lacie monitor as a second monitor next to my 30" lcd. I played back some clips which I thought the noise levels wouldn't be acceptable but on the crt there was no noise visible.

Anyone else using an old crt as reference?
Or would you advise not to keep a crt as reference because almost everybody uses lcd/plasma tv's to view movies?

 

[David Mullen ASC]

CRT's are great references for contrast / blacks, but not so much for noise since the raster lines tend to blur some of the noise. A lot of movie studios that mastered their library to HD using large studio-grade CRT HD monitors as a reference are now discovering a lot of inconsistent noise levels when viewed on a consumer-grade LCD HD monitor... and some remastering is going on.

 

[NateWeaver]

Well, its's nice to know for all the pontificating that goes on about monitors that even the sharpest sticks in the bunch have made some boo-boos.

Not schadenfreude quite, but maybe nice to have your own observations validated I guess.

 

[Charles Angus]

My rule of thumb is lowest common denominator. If it looks good on the screen it looks the worst on, (and on an accurate screen) then you're laughing.

Don't use the least noisy monitor to judge noise, is I guess what I'm saying.

 

[Mennovandijk]

Judging the images on your worst monitor is of course a good idea.
But with all noise we see on our monitors does it mean it is there, or do we only see it because of the 8-bit lcd limitations?

 

[Dylan Reeve]

The problem is compounded in modern TVs (LCD especially) where they often perform a lot of image manipulation. Lots of issues show up on consumer displays that are virtually undetectable on other monitors.

I've watched a satellite-broadcast MPEG-2 sports game on a PVM CRT monitor and a 40" Plasma - the compression artifacts were much much more noticeable on the Plasma.

 

[Gunleik Groven]

As to watching on ypur worst monitor... It's not always a good idea, as some monitors are so bad that noise and banding is a function of the monitor, not the signal.

 

[GlennChan]

I've watched a satellite-broadcast MPEG-2 sports game on a PVM CRT monitor and a 40" Plasma - the compression artifacts were much much more noticeable on the Plasma.

Was this a SD or HD signal?

If it's a SD signal, then the rectangular pixels are hard for non-CRTs to deal with since they have to scale the image, and that introduces scaling artifacts. For monitoring SD, I'd prefer a SD CRT as they won't introduce scaling artifacts (and handle interlacing like a CRT, which should be our reference for how interlacing should look; deinterlacing doesn't really work).

If it was a HD signal, then a PVM CRT will likely be putting out a soft image.

2- For monitoring HD, I think a flat panel of some sort would be the way to go. (And for SD, probably a CRT.) Ideally, we would make our images look good on a reference CRT so there is continuity between display technologies. (There are some SMPTE members working on this, and this is the direction they would like to go.) Ideally, switching over to different display technologies shouldn't make our images look different. That would go against the content creators' creative intent.

In practice, I don't think there's any way that consumer TV manufacturers might try to replicate the softness of the CRT (which is arguably nice because it hides artifacts, and doesn't have funky looking moire... especially bad on some apple cinema displays). It's theoretically possible for them to do this by putting blurry glass in front of the display... but it's highly doubtful they will do that.

Whether we like it or not, some aspects of our images will look different because of what the display manufacturers (and broadcasters) will do.

Accurate color in terms of the chromaticity of the primaries (exact shade of red,green,blue) will remain a problem... and we won't notice it. Right now, we don't really do the conversion between rec. 709, SMPTE C, and EBU primaries right. e.g. the popular BVM CRTs have SMPTE C phosphors... which isn't 100% correct for HD.

Black reproduction will be something to pay attention to, since this is something we can easily notice. I think we should grade material for a display that does black, and not try to fudge things to look good for LCDs with their raised black level.
The CRT does flare and this reduces black level (a lot) for non-dark scenes... I don't think it will be a big deal if we ignore it. (The flaring is likely why Sony can make their BVM LCDs and CRTs look alike, even though the LCD has only a 500:1 contrast ratio.)

3- I think compression right now is a big problem. I saw a lot of displays at NAB, and some manufacturers were showing material with banding and/or compression issues. We probably want displays that show these issues and I think these are the bigger problems we face now. Most LCDs will do this. Consumer ones will not if they have excessive noise reduction turned on (and some of them are so excessive they make people appear plastic). Some but not all plasmas seem to have noisy-ish blacks... but I haven't looked into them much.

(Compression is a little trickier since you'd have to monitor a signal post compression.)