DVL-Digest 841 - Postings: Index color bars/NTSC monitor confusion Length of y/c cable - followup question Video dynamic range of VX2000 vs XL1 color bars/NTSC monitor confusion - "Perry" The problem is that using bars to calibrate a single monitor takes us only to the starting blocks. It makes no allowance of viewing color temperature, screen phosphors and small misadjustments of the monitor tracking. All it does is set the correct black level and channel gains. These are important but limited. In my edit suite I currently have 4 computer monitors, an industrial grade video monitor, and a Grade 1 broadcast monitor. I can set them all up with color bars and they don't match too badly. If I put a face on them, they all look very different from each other! I set the industrial monitor to give me the most attractive but neutral picture with a known well exposed and balanced picture of a girl's face. I use this monitor because it matches most of my clients' resources in the corporate sector, and is also close to most domestic tv sets. I then try to match my main computer image monitor, which I originally bought because it has video style phosphors. It never looks quite the same but it is good enough to use for video image creation in Photoshop etc. None of the other computer monitors come close to matching but they are not used for critical image purposes. The grade 1 is aligned professionally for me and kept as a reference for broadcast output. What an editor needs is an accurate (ie sensitive) indication of color balance so that he/she can be sure that the balance is correct. Unfortunately we don't have an accurate ABSOLUTE color memory, but we have a good relative color matching ability. What works for me is to put a known good balanced face into one input of my video monitor and then switch between a few suitable shots in my programme clips. This assesses whether the clips all have a color bias that needs correcting. Thereafter, I only worry about whether the edit cuts appear to keep the same balance, since this is where the viewer will notice any changes. If you start making changes to lots of shots it is important to go back to the original reference shot to make sure that you don't start drifting away. If you don't have twin inputs, then you can periodically (temporarily!) cut your reference shot into your programme and check that the cut looks OK. While we are at it, a few other hints. Don't edit in the complete dark or you'll easily lose any sort of color balance reference. Ideally you want a gently lit and neutral (grey) environment of a similar color temperature to your monitors, but in practice most folks get used to a much lower CT background. I have an opaque Venetian blind which I adjust to let in just a little (north) daylight. Set all the computer desktops to have a neutral background color, it's amazing how many folk try to color balance with a bright blue background! Lastly, ask third parties for advice as to how they see your pictures, remember that video is a conjuring trick and VERY dependent on the viewer! just my two penn'orth of course Perry Mitchell Video Consultant http://www.perrybits.co.uk Length of y/c cable - followup question - "Perry" >How do I tell a good cable from a cheap one?< Once you have them, I guess this answer is obvious, but in the shop it may be more tricky. Studio system installers will buy cable by the mile to an industry standard, but a shrink wrapped cable in your local store could be any old junk. I've noticed that the same obsession as afflicted the hi-fi business is being exploited but this is nonsense. >Good place to find them where quality is reliable?< Difficult for me to say being based in UK! >By the "twin BNC" type, I assume you mean running two BNC cables with a BNC to DIN adapter - am I assuming correctly?< Yes, but I would make up my own adaptor to avoid too many converters. >You refer to "properly terminated broadcast type video cable." I'm not clear what this means. If the cable is plugged into the source or record deck, isn't that fully terminated?< It depends! Lots of consumer based kit don't terminate with the correct 75ohm. On long runs this can promote reflections and other problems. >Finally, I understand the need for matching the length of cables, but what is the margin for error? Is within a half inch sufficiently similar?< I don't have a suitable reference book to hand that give the delay per foot. Since we are concerned here with Y/C differential delay I wouldn't think it was particularly critical. One word of warning: I tried to use the Y/C output of my Sony DHR-1000 deck for the first time when it was installed in my system, and it needed about 5m of cable to reach. I had a 10m Y/C cable to hand (typical video store type) and sure enough it gave horrible differential errors and some 'ringing' on the edges. I spent a good while making up a decent connection from some cannibalised BNC cables - result was identical! Turned out I had a faulty Y/C output on the deck! Moral - check your picture quality first on a short connection, then put the long one in and check you cannot see any difference. Perry Mitchell Video Consultant http://www.perrybits.co.uk Video dynamic range of VX2000 vs XL1 - "Perry" This is an enormous subject and could be discussed till the cows come home (as my granny used to say). In short: 1) CCD imagers do indeed have less dynamic range than negative stock, but effectively as much as a print shown in a theatre. 2) Consumer cameras are designed to look good without fill light, and the shadow detail is severely limited to lower apparent noise. 3) Consumer cameras also have restricted 'knee' circuitry to compress and preserve highlight detail. I am not familiar with the XL-1, but certainly these are two areas where say the DSR-300 would show clear superiority over the VX-2000. The ironic thing is that good quality production will use lighting, make-up and set design to lower required dynamic range so it's a double whammy! Perry Mitchell Video Consultant http://www.perrybits.co.uk (diese posts stammen von der DV-L Mailingliste - THX to Adam Wilt and Perry Mitchell :-) [up] |
 slashCAM  News HD Camcorder database One-on-One Cam comparison About our tests About us |
|
|
