Best way to capture pc screen ?????????
WHAT IS THE BEST WAY TO CAPTURE PC SCREEN ? SHOOTING OR TAKE CAPTURE PC SOFTWARE ? AM SHOOTING WITH BETACAM SP ! I TRY CANTASIA BUT THE FINAL RESULT IS NO GOOD THE SMILE LINE FLICK ! WHAT IS THE BEST WAY ? NEED HELP ASAP THANX
Assuming your caps key is stuck and you're not shouting - I'll take a shot at this...
The central problem is that a computer screen and a TV screen are two completely different beasts.
A television screen is a fixed raster with a fixed refresh rate and NO scalability involved. Well - sorta. In fact the different world standards of TV make the first sentence of this paragraph arguably untrue. There are multiple WORLD STANDARDS for broadcast TV. But let's ignore that. Pick any one of them to produce your TV content for - and you're working in a fixed format.
COMPUTER displays are anything BUT fixed. They are progressively scanned rather than interlaced. They are variable in size, screen dot pitch, and can be electronically refreshed at any rate their software drives them to. There are "some standards" for screen raster sizes. But it's a choice for the user. If you buy a better card, or a better monitor, you're free to ALTER your screen dot pitch and refresh rate all you like.
So when you say you want to "CAPTURE PC SCREEN" it's like saying you want to "COOK CHICKEN". Your library has 10, 000 different recipes for how to do that. And doing it in a satisfying fashion depends on a lot of stuff - again like cooking chicken.
To do it really well, you have to know a lot about that original PC signal. Is it complex? Simple? Is the color pallet used TV safe? Is there type? Is that type large or small? What about patterns and detail? Are their patterns that look fine on TV, but will shimmy and crawl when converted to the raster of TV?
Those questions are the first of hundreds that can come up when you're transcoding between a good COMPUTER signal - and a good VIDEO signal.
Some general rules are: Better quality scalers, transcoders, and rasterizers are typically the BEST way to do direct transfer of computer output to video - but sometimes even the best work poorly if the original computer signal design is "video hostile" with broadcast unsafe color, tight pattern design, or too-small type choices
Shooting traditional cathode-ray TV picture tubes is possible, but is best done with higher end cameras that feature very adjustable shutter speeds (Clear-scan, etc) that allow you to precision sync the camera shutter with the monitor refresh rate in order to avoid the "rolling white bar" issue of sync incompatibilities.
Shooting LCD video screens is easier - and sometimes you can get decent results just videotaping a laptop screen - however for best results, you might need neutral density gels or color correction gels (laptop backlights tend to be more daylight like than tungsten-like) to get the laptop screen dimmed and color corrected enough for a proper exposure.
I know you just wanted a simple "set it up like this - and hit record" but while capturing video from a computer can sometimes be that simple when the gods are smiling on you - in practice it's usually annoyingly complex - and if the original computer designer made poor choices - it can be a solid nightmare.
Bill gave a very good answer. I'm just wanting to piggyback a little extra bit onto it.
He left out one other way to go, a method falling out of favor as old-fashioned, which is to use a scan converter, attached to the VGA monitor output (actually, it sits in between the VGA port and the monitor, so you can still look at what you are doing)
The scan converter is a black box that will turn the computer video into an analog or digital TV signal you can record and work with in the way you are used to.
You get what you pay for, more or less, with these (you can also rent them sometimes), They range from 99 bucks and just good enough for some powerpoint work, to five figures and up. We use a Scan-Do Pro Ultra, I think it was $13k new. But I also sometimes still use a 99-dollar one, depends on the job and other factors. My own opinion is they are all pretty much the same below about five grand, but above that, they have added features and many, many controls that let you really customize the view and quality. You can program multiple preset window views in some of these converters/scalers, and they will act like a multicam swticher, with tight shots and wide shots and whatever you need, of any part of the screen, as if each set was a separate camera. Set this up right at the start, and you can save having to re-shoot multiple times just to be able to cut from extreme close-ups to wide shots, for example.
These scan converters are still great for live work; for one thing, they do not use any processing or memory power of the machine they are looking at, so for critical apps and shakey machines that are already using the computer processor board and hard drive to the fullest, the scan converter will not slow you down in any way.
But I think most people are now using snapzpro or camtesia software-based screen grabbers instead. Camtesia makes huge video files, is one downside of using it, so you have to have enough hard drive space free in the machine you're recording to hold that, while you're doing whatever it is you wanted to record in the first place.
I'm surprised you did not get good results with camtasia. Everybody else does. Many of the COW demo or tutorial online vids are made with it, I understand. So it is not the grab program that's bad. You might want to go back and experiment on that again. Not just on the camtasia capture settings, but on the way you import the camtasia-generated video files; you may actually already have GREAT quality, that just looks bad because you imported it wrong, and the NLE system has re-compressed it and lost resolution. I have seen still graphics look awful when imported into FCP using the wrong settings, this can happen to motion video too.
Good luck and write back with more questions if you still have trouble; maybe we can figure out where in the chain things are breaking down for you.
I am having an issue similar to Patrick Deslauriers', and I was wondering what solution he was able to come to.
For my issue I am receiving video that was made using camstudio and the divx codec (720x480). The video looks crisp and clear and is around 236mb for 2hrs. It is an avi file, but I have tried converting it to quicktime, but that makes no difference in render times. When I bring it into final cut to work on it tells me it is going to take 30+ hrs to render in a dv/dvcpro ntsc time line (square pixels looks best). I have tried different compression settings like pro res, animation, dv/dvcpro ntsc, none, photo jpeg, and video and all are giving me ungodly render times. Is there a codec you would recommend using in camstudio or a compressor setting in FCP that would work good.
If not the scan converter method sounds promising.
Thanks for any help you can give