Rendering for CRT , LCD , TV ... suggestions please

FuTz wrote on 5/1/2005, 10:01 AM
I've been doing this clip for a band so I ditched this old TV set that was so beat up and bought another Toshiba TV to correct my colors.
I also did a complete other correcting for CRT screens that's gonna be used on the band's site.
And I ommited watching all of this while correcting on some LCD screen.

Now what I've got is, well, total mess (according to my personnal standards).

What I balanced thru 1394 for TV sets, I renderd in MPEG2 and AVI so these guys can author some DVD if they want. BUT : the guy has a DVD player that can read these files and when we watdched it on his TV set, it look almost exactly like it looked on my *CRT* computer screens ! I mean, no mids; hightlights and deep blacks. All of which DID NOT appear this way on my TV set while doing the works.

Also, the guy had his laptop so we watched the clip that I corrected for the net and yuk!!! way too much brightness so we could see a lot of things I didn't expect people to see in the editing (some masks for example, concealed in blacks that simply "jumped out" the whole picture right in our face).

I didn't watch this "net version" on a CRT at this guy's place since he was on the go and didn't have time to start up his comp'.

Questions:
-Will simply applying an "S" curve on the "CRT balanced" clip resolve the equations to get an "LCD balanced" clip ?

-How come this "TV balanced" clip looked like the way it looked on my CRT monitors and not on my TV monitor? I don't think DVD authoring will "improve" hence correct the clip since I don't think it does anything but take what is fed to render to DVD, ain't it ? Are our TV sets so much differently balanced ? I calibrated mine before doing all the work but even the colors don't look the same on our respective TVs. Whoa... is life so hard in Monitorland? I really miss the time all I did was only for computer purposes now...
Well, almost : )

Comments

rmack350 wrote on 5/1/2005, 7:02 PM
Life is really hard in monitor land. For instance, I run two CRTs and a fairly old and cheap JVC studio monitor. My CRTs don't look much like each other until they've been running for about a half hour. I've gone to great trouble with a colorvision spider to get them right but the truth of the matter is that most people don't make any effort to adjust computer screens. No two will look alike.

The studio monitor looks fine and I check it frequently against bars so I know where it's at. But of course no one else's TV will look as good.

I'm going to guess that if you can see your mattes on his TV then your TV isn't doing you justice. If there's a difference within the blacks that you couldn't see then your monitor setup is tricking you. The blacks are too crushed. It doesn't matter if his are too pumped because if the blacks are supposed to have the same value then they should really have the same numerical value. If you can see the difference, the difference is real and your monitor setup is off.

Rob Mack
BillyBoy wrote on 5/1/2005, 7:35 PM
If you live in the states and a few other places you deal with the NTSC "standard". Long time ago engineers mocked it, labeling it "Never twice same colors":

See http://aroundcny.com/technofile/texts/bye2ntsc89.html

Its the nature of the beast. There's a long winded technical reason why that it so, but since we are nearing the light at the end of the tunnel with HD TV, I guess we can put up with it for awhile yet.

No two TV's will look EXACTLY like is closer to the truth. With a little effort you can make multiple sets "look alike" to the point the differences will be so slight hardly anyone will notice. If you have a couple sets you want to try to match, feed them both the same signal at the same time, then adjust both till they are as close as you can get them. Of course you would calibrate each first. Many old threads, where I give more specifics on how I approached this "problem".

Everytime the topic monitors comes up the same bunch get their shorts all bunched up. I just laugh. I also shut them up like they got hit with a lightening bold when I challenge them to explain WHY is it so-called professional getting ready to go live over broadcast or cable TV all covering the same live event, and supposedly first doing "color bars" end up broadcasting slightly different hues, brightness levels when they are all broadcasting the same event. Nobody yet, can explain how come.

We come full circle: NTSC = Never twice same colors.
musicvid10 wrote on 5/1/2005, 8:33 PM
My best advice for the present is, adjust your TV so it looks the best from a rental WB video tape, then tweak your projects to match. At least you have found the baseline and concensus. Enough said.
John_Cline wrote on 5/1/2005, 8:38 PM
BB: "Nobody yet, can explain how come."

I did, search for it.
pelladon wrote on 5/1/2005, 9:03 PM
Before you folks badmouth NTSC, you might want to read this interview from Yves Faroudja:

Interview - AVguide.com
Spot|DSE wrote on 5/1/2005, 9:12 PM
I also shut them up like they got hit with a lightening bold when I challenge them to explain WHY is it so-called professional getting ready to go live over broadcast or cable TV all covering the same live event, and supposedly first doing "color bars" end up broadcasting slightly different hues, brightness levels when they are all broadcasting the same event.

It's a union thing. AFTRA knows that additional time needs to be wasted at every shoot, live or EFP, and so they came up with this rule that says that outputting color bars to provide a field reference point is required. This means that overtime is almost always going to be a factor in a 2 hour broadcast due to setup times prior to the event. They lobbied to make white balancing an mandatory requirement, but the broadcasters wouldn't go for it. So, at the Geneva Convention held in the Hague in 1966, it was determined that ALL NTSC broadcasts begin with color bars and white balance. This has to date, resulted in more than 13.6 billion dollars in addtional union revenue. For clarification, color bars exist ONLY to allow AFTRA and SMPTE engineers to look professional. Amateurs know better, and never use color bars. This is why so many people are unaware that Vegas can generate all sorts of color bars or other reference tools. (BTW, they are only to be referred to, never actually used)
Further, SAG and Equity are both very unhappy at the level of diversity presented in color bars, and there is a movement afoot to add flesh tones from every major culture to the SMPTE standard. It's been proven that skin tones have changed significantly since the Convention in 1966.
SMPTE is expected to make a statement any day now, that adding flesh tones to generated color bars for calibration is a waste of time, because color bars are a waste of time. Finally, D.J. Lenoman is expected to step down as a SMPTE officer next week, as he's been outed as not being "professional." He was caught using color bars as a reference point, but they weren't genuine, non-union SMPTE-approved color bars. He was in dire straights and made them with crayons and butcher paper, and was caught when his blacks were really only 5.5 IRE.
It's a shame, because everyone who watches TV pays a subsidy to the union for the time they take to send those calibrated color bars.
My question is, who watches the guy who makes the color bars? Where did those silly standards come from?

BTW, PAL=Picture At Last
SECAM=System Essentially Contrary to the American Method
Cheno wrote on 5/1/2005, 9:42 PM
"Where did those silly standards come from? "

Actually in 1954, Crayola and NBC execs got together and were working on tying the peacock logo and a box of crayons together. Crayola also at that time had only been producing black, white and grey crayons and low and behold that magical day, the peacock came to life in living Crayola color.

It isn't coincidence that SMPTE color bars match the basic box of Crayola crayons you first got ahold of as a young child. It isn't ironic that D.J. Lenoman was caught using crayons, in fact it was a marketing ploy to tie in with the 61st anniversary of the venture.

I'm in fact teaching my 4-year old to color a blue-only chart. She's just a couple of shades from nailing it. :)

Man, I love history.
John_Cline wrote on 5/1/2005, 9:58 PM
"you might want to read this interview from Yves Faroudja"

Great link!!! Now... let's see... who you going to believe Yves Faroudja or BillyBoy?

As an aside, I used to make a point of stopping by the Faroudja booth every year at NAB. Yves was very approachable and just loved to talk about television. He and Philo Farnsworth are personal heroes of mine. (Along with Nikola Tesla and Frank Zappa.)

John
albatross1 wrote on 5/1/2005, 10:51 PM
You problem here is very simple. You are using a consumer TV and not a broadcast monitor.

Consumer TV's are designed to make bad pictures look good. Broadcast monitors are designed to show the flaws, so that you can fix them.

Your problem is caused by DC restoration. Your TV adjusts it's picture to make the contrast and levels look better than they do.

Dan Achatz
Steve Mann wrote on 5/1/2005, 11:34 PM
Yves is correct that NTSC is more efficient in bandwidth useage than either PAL or SECAM, and he is correct that an NTSC decoder (actually, it's a de-multiplexer) is simpler and cheaper than either PAL or NTSC, but that's where the advantages end. I have no doubt that one can get a perfect picture from an NTSC signal, as long as the lighting doesn't change.

It is really looking like Congress is going to hold fast to the DTV date of December 31, 2006. Then, NTSC will start dying slowly since it won't be broadcast over the air.

But the DTV switch is another topic, and my money is on yet another delay.

Steve Mann
pelladon wrote on 5/2/2005, 12:08 AM
Of course Faroudja likes NTSC, that's where most of his patents are!

NTSC won't die, it'll just be another format in a digital world. Along with 720p, 1080i, HDV, WMV-HD, MPEG-4, etc.

I've noticed, at least in Los Angeles, almost all of the stations are already transmitting in digital. Biggest problem is the *reception*. When it's good, it's good. But when it's bad, you get garbage or zilch. At least with analog RF, you get something.
FuTz wrote on 5/2/2005, 4:28 AM

Ok, I'll try the S curve for my "LCD balanced" clip taken from the "CRT balanced" master and see what happens when viewed on LCD screens.
I'll apply different curves to my "TV balanced" clip and watch it on different TV sets and go with the best of both worlds.

Questions remains: having "calibrated" my "toy-TV" monitor with blue gels, adjusting contrast and everything, how come I can see Poncharello's face on both my friend's TV and mine and not see the faces of performers from my clip only on his set...
Aaaahh right, I'll boost the mids and see what happens.
rs170a wrote on 5/2/2005, 6:10 AM
how come I can see Poncharello's face on both my friend's TV and mine and not see the faces of performers from my clip only on his set...

Futz, It sounds like you've set your TV similar to the setup described in the Video University article and that's fine. Have you done the same thing to your friend's TV? As others have pointed out here, no two TV sets will be exactly alike and I suspect this is where the problem lies.
As far as setting up your CRT goes, there's an article on the Paint Shop Pro site that might be useful.

Mike
FuTz wrote on 5/2/2005, 5:52 PM

Comes to mind: next time I'll correct colors, I won't close my cam's viewscreen so I'll get both LCD on the cam and TV thru 1394 so I'll try to find a way to make the two look ok at the same time.
MH_Stevens wrote on 5/2/2005, 7:10 PM
Now you pros will think I am a heathen, but once when I had no video monitor available and only an old poor computer screen, I color corrected some footage blind, by using the scopes only. Like flying at night on instruments. I'm pretty good at reading the scopes as I never edit without them all on. I burnt the finished footage to DVD and went watched it in the home theater and I found I got really good results. As good as an amateur like me would have got with my usual HDTV monitor. Try it.

Mike S

Spot|DSE wrote on 5/2/2005, 7:17 PM
Mike, you're no heathen, you're a guy who knows fine champagne!
Was good seeing you in Vegas, as brief as t'was.
All the broadcasters use scopes along with the monitor, it's just sorta standard. The Calypso and similar systems are all more scope oriented than image oriented. Using scopes is more advanced than using a broadcast monitor! But not everyone knows how to read them. Which makes for a good plug for Steve Hullfish' book on Color Correction, it's got an excellent chapter on scopes.
BillyBoy wrote on 5/2/2005, 7:18 PM
Oh nuts Mike, now you went and spoiled it. Stop giving away secrets. <wink>

BTW, I've done that many times. Works as good or better. But Mike, wait till someone in the peanut gallery pipes up and claims you need to use "professional" scopes. I'll just sit back in my "professional" video editing chair, I have on my "professional" video editing shirt and pants, and I just took a sip of my favorite beverage out of my "professional" coffee mug. Ain't coffee either.

fultro wrote on 5/2/2005, 7:21 PM
Mike - being relatively new to video editing I'd be interested in knowing what you rtechniques were in using the scopes - or at least how tou learned them - I am quite new to color correcting with scopes and BillyBoys color correction tutorials look to be pretty good - but in the next couple of weeks I have to bear down with some serious corrective actions and I only have a basic tv monitor - so any extra tips you could provide would be greatly appreciated....
Jeffrey Cline wrote on 5/2/2005, 7:22 PM
It takes a trained eye to set up a monitor (professional or consumer) to look the best it can. However, with the proper tools and knowledge a consumer tv can look decent. But, a consumer tv will still not allow you to monitor your signal adequately. The quality of the image is directly related to how much money is spent on that monitor. For editing, you use what you can afford. If your bank account will not allow you to spend $5000 on a monitor, you use what you've got and make the best of it, but your image quality will suffer.

There are so many variables involved in the broadcast chain that making sure the edited program is properly color corrected is critical and you hope the rest of the chain is maintained by other competent professionals, which it usually is. I will not edit a project without a broadcast monitor, that I calibrate myself, along with a waveform monitor and vector scope. No professional should work without any of the three. But, that's the world I work in. Yes, Billy, I am a professional editor and yes, I am related to John Cline. I am not here to defend John, Spot or anyone else for that matter. I am here to defend a point of view. As I said above, you work with what you can afford. Luckily, I have all of the edit gear I work with provided for me. I edit content that is used in network sports broadcasts and if I produced something sub-standard I wouldn't be called back again. That's the difference! Professionals use professional gear to produce the best product possible. It's not marketing hype. It's fact! If this weren't true, then why would networks and facilities spend millions of dollars on monitors and hire competent professionals to use it. There is a reason for it! To produce the most consistent, highest quality content for distribution possible.
BillyBoy wrote on 5/2/2005, 7:25 PM
I'll just whistle past the graveyard.
MH_Stevens wrote on 5/2/2005, 7:35 PM
Fultro:
You need do a lot of color correcting using your eye with the scopes on. Every time you make an adjustment watch what the scopes do. Soon you will be able to predict what the scopes will do and then you can reverse your technique. Adjust the scopes and watch what the picture does. It will come.

In general, at the very basic level, what you can do with scopes is set the black and white in and out points and set the middle, gamma to the feel you want with the histogram. Then use the vectors to get the color ranges and central images within the brightnes windows you want. Keep everyhting below 110% and get you major spread from 40% to 80% with your skin tones at about 70% and you will be close. That's my science and the rest is just art, and that a piece of cake, right?
fultro wrote on 5/2/2005, 8:01 PM
thanks for that Mike - I am going to check that out later this week


BillyBoy - it might be helpful for some of us to understand you if you reveal what it is you are sipping out of that "professional" coffee cup if it ain't coffee.....

MH_Stevens wrote on 5/2/2005, 8:06 PM
Fultro:

An after thought. What I said assumes you have correctly exposed and white balanced footage. If your white balance is out, ie what is really white does not look white, you need set the white point with the negative eye dropper on the color corrector.

AND don't take this too seriously. Have you ever been in BestBuy and seen the rows of $5,000 HDTV plasma screens? They ALL LOOK DIFFERENT!

Mike

Jeffrey Cline wrote on 5/2/2005, 8:15 PM
Sure there's a human element to this. I regularly edit material acquired with Digital Betacam of Panasonic VariCam on an Avid system via a serial digital interface with a broadcast monitor, export the timeline to a .mov file and apply plug-ins in After Effects such as an Ultimatte key, text effects or a film look on my notebook LCD screen. The reason I use a notebook is that I can continue editing while the render is completed or I can take it back to my hotel room and finish the project without having to stay at the venue while waiting for a render to complete. After importing the result back into my edit system and looking at it on that same broadcast monitor, I sometimes find artifacts on the key edges unseen on the LCD that require a re-adjustment of the plug-in and re-render to correct those issues. No consumer tv set or LCD monitor would allow me to see problems that affect the final renders. It's pretty simple. Only a professional monitor has the resolution necessary to show me what I am actually delivering.