Grainy Footage even in good light
 

i have the HVR-V1U and im still not aware if im capturing the footage right, another videographer told me to use these settings

Frame size of 1440x1080
Upper field interlacing (i wasnt sure where this was could someone tell me?)
The capture setting should HDV
Device control Sony HDV

i did that and captured my footage, however in some areas i was gettin really grainy stuff. I know the dark areas are goin to be grainy, but if you go 1:36 in the movie i uploaded and notice how bad the blacks are in that footage
Host Day Part 1 on Vimeo

thanks for any help

Those settings are totally irrelevant to your problem.

Until someone more experienced with the cam answers, check your gain settings. Do NOT set to auto. You should be on 0dB gain unless you are in a dark situation.

Sony cams have to be on manual control for iris, gain and shutter speed if you want true manual control. If two are on manual, and the other auto, the camera happily changes the third automatically to suit what it thinks is right.

So you can easily get grainy footage in good light if you are on manual iris and shutter, but auto gain (for example). If you stop down in bright light, the camera adds gain if it thinks you are underexposing, giving you grainy footage.

Could that be the case for you?

Yet more encouragement to be very careful with gain and autogain.

If you ever want to shoot automatic, you should go into the menues and set agc limit to +6db. You may want to change that for a particular shot, knowing that some grain/noise will be the result.

If you're shooting fully manual, you have more choices, and can, if needed, set your gain at +9db with little perceptible grain.

so i have my shutter at 1/24, 24 fps, the Fstop at F1.6 are you sayin i shud keep my gain between 0 and +9 to have the least amount of noise?

There is a setting in the menu to limit the gain level (called AGC). You should set this to +6 dB.

Another setting worth using is to limit the aperture to F5.6. This avoids any possible fringing due to diffraction with very small apertures.

Your advice is good, Alan, although I hear that f4 is the 'sweet spot' for the lens - where you get maximal sharpness. I would suggest going with 5.6 maximum in bright light, plus neutral density filters as required and in anything less than that (short of a true low-light situation), f4.

The tutorial over at HDR-FX7 / HDR-FX7E tutorial recommends use of manual white-balance and even gives you a series of settings that can be used to achieve the best balance of colour and sharpness for outdoor situations.

Using the right WB settings can help to improve perceived sharpness and reduce grain.

My personal opinion regarding the 'grain' issue is that it may have a fair bit to do with the fact that the FX-7 uses sensor chips that have a 4:3 aspect ratio, and pixel-shifting is done to allow users to shoot in 16:9.

The only way I could see Sony accomplishing this is by cropping the 4:3 image in-camera so that it results in a 16:9 image. In other words, not using the entire sensor area, which means your effective sensor size is not 1/4" but something less than that.

Rishi, you can turn on the "Cam Data" display in your menu. If yo do, when you play a tape, it will then display the settings that the camera used when taping. One of the settings it will display is the gain so you can tell if the camera automatically used increased gain without you realizing it.

are you sure of your facts here? i was under the impression that the sensor was 16:9.

Well it *is* a 1440x1080 sensor, which is 4x3 - but isn't it stretched horizontally (each pixel is 1.33 times as wide as it is high) to get 16x9? Not "true" 1920x1080, but I don't think it's cropped.

I don't know your application with this camera, but if you've got some render time to spare you might wanna look into the NeatVideo plugin. If you're after that smooth as glass look you can get out of an EX1, run this plugin on your V1/FX7 footage. For $100 and the way I work it was a no-brainer. Renders slow as heck but produces very nice results.

Ethan,
Do you have any real-world before/after samples, and have you used this product on Cineform HD and on M2T files?

Thanks,
Matt

I don't have anything handy to post that wouldn't eat up a nice chunk of my day but seriously, it's 100 bucks. If you have any grainy footage drop the cash and check it out. Actually I think you can download a demo from their site but it only covers a cropped middle portion of the image. That might be enough to give you a good idea of what it can do.

I can say if you use it, render overnight or you're in for a long day of doing nothing.

I can't say how it works with Cineform and M2T's since I'm a ProRes guy. I can't imagine the format would affect it much.

Download the demo, or do a little searching around, people have been talking about it for a while now.

Their Photoshop version of this plugin was so good that before they came out with an FCP compatible variation I was willing to export out TIFF sequences of trouble footage and run those through Photoshop to use it.

I think I will. I just happen to have just shot a dance recital, where the director was very enthusiastic about the audience experience (Yay! for the audience, and it did look pretty good), but good for audience usually is not the same as good for video...there were definitely 12dB parts...a LOT.

Reserving judgement for now, but if the results are close to the advertised results, it'll be pretty darn good.

Thanks,
Matt

Matt - if you're using it on an entire dance recital you might want to consider rendering over a whole weekend.

12db should clean up fairly well, 15 or 18 is pushing it. Try out the demo.