Jeff I respectfully spit in your general direction. Just because the monkeys in the food chain don't know anything about technical specs and reference signals doesn't mean they don't get used by some. When I'm testing gamma shifts on QT imports and exports and checking my delivery files I use the bars to feed my scope to make sure I'm sending out what I think I should be. Clearly according to your perspective nobody uses them on the distribution side but then I still see them noted as complying with spec on QC reports. Also some of the Videotek and other legalizers offer the ability to make proc amp like adjustments to the signal. How you gonna do that properly without bars. How do I know what is hitting my record machine is proper level and what is coming back from that machine is also proper level. Without bars no chance of confirming unity in an infrastructure. Or worse case scenario somebody somebody records to a tape deck with the proc amp levels on the legalizer off a bit. If I have bars I stand a chance of correcting playback at the tape machine and not just running unity with the wrong levels. When everything is right it's a moot point but when things are off Bars can make the fix a lot easier. If the bay signal is off it will be reflected in the bars laid to tape but if I use internal bars on the deck they are no longer a "reference" to the important signal path of edit bay to record machine.
--- In Avid-L2@yahoogroups.com, "jeffsengpiehl" <jeffsengpiehl@...> wrote:
>
> but did anyone take a look at bars that had been through standards conversion? 1080-720, and vice versa? There's even MORE fun to be had there.
>
> Makes you realize that color bars are irrelevant in today's day and age, and they only exist on deliverables due to inertia. Punch in bars from your Deck's signal generator to your tape masters, and you've got a lot less to worry about then a QC flag for something no-one will ever use.
>
> JDS
>
> --- In Avid-L2@yahoogroups.com, "johnrobmoore" <bigfish@> wrote:
> >
> > Thanks for sharing with the L-2 group. If anyone would like to see the screen grabs and the actual files contact me off list and I'll send the same email to you that I did to Mr. Hullfish and others. I haven't had a chance to break out the calculator but it was obvious to me just using PhotoShop eye dropper that the I and Q RGB values were different in the various color bar files. Fun stuff for the nerd in me. Also I believe there is a debate over Union vs. NonUnion color bars in some circles. ;-) When I was in the unions I ended up in the bar every night. Now that I'm sober no bars at night and no union. Coincidence, Hmmmmmmmmmm. ;-(
> >
> > --- In Avid-L2@yahoogroups.com, Steve Hullfish <steve4lists@> wrote:
> > >
> > > A week or so ago, John Moore posted some links to some color bar examples from the web and from Avid and asked:
> > >
> > >
> > > I finally took a look at the Avid HD bars you mentioned and I do see the nice curly Q style of I and Q. I've included screen grabs off the Tek scope of 3 Bars The ARIB bars someone put up on the Avid L2, Smpte Bars of the 219 standard and Avid HD 1080 Bars. Interesting. I included the image files for each. So the challenge is which Steve H? Hullfish or Holmes will diagnose the image files and tell us how Avid got the math wrong for HD bars?. I'm betting it has something to do with the bigger stretch there is in HD bars as per one of the Tek seminairs. Here is what I've been told by Steven Holmes regarding 601 vs 709 bars. Could this be the reason Avid bars are off? Have they just repurposed SD bars?:
> > >
> > > The response is only coming THROUGH me, but it is indeed from the other Steve... Holmes, from Tektronix, with the answer:
> > >
> > >
> > > One thing we have to remember is that SMPTE Bars are a Composite color bar test pattern. I & Q do not exist in the Y Pb Pr or RGB domain.
> > > You cannot go as far negative as you need to reproduce proper I & Q with full White. I & Q fall outside of the Color space we have to work with they are a reference Composite phase vector not a color.
> > >
> > > I had one of my colleges look at the files also, and here is his response.
> > >
> > > I had a quick look at the TIF files it appears as if they allow support of color range that BMP does not as white is set a 235 and black at 16.
> > > Instead of white 255 and black at 0.
> > >
> > > Values of I and Q are from SDP2000
> > >
> > > G 40.68
> > > B 166.75
> > > R -144.56
> > > Y -0.3 C 307.9 and 303 degrees
> > >
> > > G -97.61
> > > B 258.27
> > > R 94.41
> > > Y -0.3 C 308.9 and 33 degrees
> > >
> > > So you cannot real produce I and Q values you have to change levels or degree in order to make it work.
> > >
> > > AVID values based on 16-235 for 0-700mv
> > > For 219
> > > G 62 = 146.4mv
> > > B 102 = 273.6mv
> > > R 4 = -38.2mv
> > > Gives Y 107 C 307.9 and 303.2 degrees
> > >
> > >
> > > G 4 = -38.2mv
> > > B 115 = 273.6mv
> > > R 64 = 152.7mv
> > >
> > > Gives Y 55.5 C 281.8 and 38.6 degrees
> > >
> > > For SMPTE Bars
> > >
> > > G 26 = 31.8mv
> > > B 64 = 152.2mv
> > > R 0 = -50.9mv
> > > Gives Y 21.2 C 184.1 and 315.8 degrees
> > >
> > > G 0 = -50.9mv
> > > B 80 = 203.6mv
> > > R 58 = 133.6mv
> > > Gives Y 33.9 C 248.0 and 46.4 degrees
> > >
> > > I choose for my pattern with RGB value 0-255
> > >
> > > G 0 = 0 mv
> > > B 122 = 334.9mv
> > > R 66 = 181.1mv
> > > Gives Y 94 C 290.1 and 33.1 degrees
> > >
> > > G 64 = 175.68mv
> > > B 107 = 293.7 mv
> > > R 0 = 0 mv
> > > Gives Y 134 C 291 and 302.8 degrees
> > >
> > > Therefore in AVID 235-16 would be about these value.
> > > G 16
> > > B 121
> > > R 73
> > >
> > > G 71
> > > B 108
> > > R 16
> > >
> > > This concludes my math for the day.
> > >
> > >
> > >
> > >
> > > Steve Hullfish
> > > contributor: www.provideocoalition.com
> > > author: "The Art and Technique of Digital Color Correction"
> > >
> > > [Non-text portions of this message have been removed]
> > >
> >
>
No comments:
Post a Comment