"Lens hood" for Sony TRV900 with wideangle lens

10 replies [Last post]
Christian Lett
Offline
Joined: Apr 26 1999

Hello everyone.

I was wondering if anyone here had any success either building or buying a good lens hood for a TRV900 (or similar) with a wide-angle adapter on the front. Of course when I say "lens hood" I'm really trying to describe the french flag set-up, i.e. a large flat black canopy that sits over the top of the lens, with some smaller flaps either side.

My previous effort was based on the Cokin A-series filter holder and a single clip-on lens hood, to which I attached a black cardboard hood. I could go down this path again (with the P-series), but was wondering if

a) this might cause vignetting due to the wide-angle adaptor and
b) if anyone has any better solutions.

I tried a Jessops screw-on hood but it was immediately apparrent that it wasn't designed for wide-angle lenses - the resultant image was nice and round! Without some kind of protection I get some terrible flare if shooting at the wrong angle and I want to eliminate it wherever possible.

My adapter is the Raynox lens with a 72mm thread. But I might upgrade in future to an anamorphic adapter, so my hood will idealy work with either.

Many thanks,

Christian.

Christian Lett After Effects and Maya Artist www.quarterlightpictures.com

Alan Roberts
Alan Roberts's picture
Offline
Joined: May 3 1999

If you're getting serious flaring, I'd recommend building a system of flags rather than trying to get a custom lens hood. A hood can only ever be maximally effective at the widest angle, where it just doesn't quite clip the image, and narrower angles it's pretty well useless.

A flag system, if you've not met one before, is a contraption like barn doors, but each flap is separately hinged maybe in two planes, so that you can flip it about to obscure the stray light. They are hugely successful, and you only ever need two flags, usually only one, because the stray-causing light usually comes from only one source.

Get my test cards document, and cards for 625, 525, 720 and 1080. Thanks to Gavin Gration for hosting them.
Camera settings documents are held by Daniel Browning and at the EBU
My book, 'Circles of Confusion' is available here.
Also EBU Tech.3335 tells how to test cameras, and R.118 tells how to use the results.

Christian Lett
Offline
Joined: Apr 26 1999

Thanks for the reply Alan.

Am I right in thinking that the flag system you've described is not attached to the camera? I've seen on film sets these flags, which are just black boards of varying sizes, set up around lights to block light from certain angles (it all looks very elaborate!).

For my use I'm guessing that a few pieces of thick black card placed in strategic positions will perform the same function.

Regards,

Christian.

Christian Lett After Effects and Maya Artist www.quarterlightpictures.com

Alan Roberts
Alan Roberts's picture
Offline
Joined: May 3 1999

Yes, spot on. Flags can be any size, placed anywhere. As far as the camera's concerned though, I mean something like a matte-box (since you've seen film shooting in action, you should know what that means). I'm not sure how to go about making one, but my idea was for a ring that surrounds (but needn't touch) the lens, with one or more rectangles of something black on hinges that you can twiddle until you're happy. A simple ball/socket swivel on one corner of each flag allows maximum flexibility. I'm sure you've got the idea by now.

Get my test cards document, and cards for 625, 525, 720 and 1080. Thanks to Gavin Gration for hosting them.
Camera settings documents are held by Daniel Browning and at the EBU
My book, 'Circles of Confusion' is available here.
Also EBU Tech.3335 tells how to test cameras, and R.118 tells how to use the results.

tom hardwick
Offline
Joined: Apr 8 1999

If you go to www.cavision.com you can take a look at their hoods and flags. It sounds as if you have the Raynox 6600PRO lens Christian, and this lens does indeed need carefull hooding because of its minimilist coating.

Beware the 900's v/f and side-screen masking too - as the full frame could well be suffing cut-off yet you'd not know it while shooting.

tom.

Alan Roberts
Alan Roberts's picture
Offline
Joined: May 3 1999

Yes, Tom, good point. Pretty well all con- and pro-sumer camera viewfinders and flip-outs overscan. I've no idea why they do it, totally unnecessary because the scans are fixed, but they all do it. It means that there's at least 3.5% of the picture on each edge that you can't see. So when you're sdjusting flags, bring it in until yuou can see it about 5% into the picture, then observe how much you have to move it to get it just out, and move it by the same amount again. Tedious, but it should work.

Get my test cards document, and cards for 625, 525, 720 and 1080. Thanks to Gavin Gration for hosting them.
Camera settings documents are held by Daniel Browning and at the EBU
My book, 'Circles of Confusion' is available here.
Also EBU Tech.3335 tells how to test cameras, and R.118 tells how to use the results.

tom hardwick
Offline
Joined: Apr 8 1999

"I've no idea why they do it, totally unnecessary because the scans are fixed".
Wonderful - at last I've found someone to agree with me that at least one of our camera's finders should underscan.

tom.

Christian Lett
Offline
Joined: Apr 26 1999

Thanks for the info chaps. I had heard somewhere about the underscan issue, which was confirmed a couple of months ago when I watched a live feed from the camera into my laptop via firewire. I don't pretend to know a lot about the nuts and bolts of how a camera works, but this does seem a little odd. Maybe the LCD panels are NTSC resolution (I wonder if NTSC cameras have this issue...).

Tom, you're correct about my Raynox lens - in fact it was you who recommended it. It's a great lens, improving on the TRV900's less-than-great wide-angle without too much distortion at the edges (in fact I hardly notice any). I'll let you know how my adventures in hood & flag making go.

Cheers,

Christian.

Christian Lett After Effects and Maya Artist www.quarterlightpictures.com

Alan Roberts
Alan Roberts's picture
Offline
Joined: May 3 1999

Underscan's nothing to do with display resolution, it's designed in, it's deliberate. But God knows why, they all do it. If you want to know exactly how much of the picture's missing from you view, why not download one of my "Test Cards" and find out? They're on the CV Mag website, you want the one called Overscan.tif. Import it into your NLE, then either put it to tape and play it in the camera, or just use the camera for live viewing. You can actually count the missing pixels/lines.

Get my test cards document, and cards for 625, 525, 720 and 1080. Thanks to Gavin Gration for hosting them.
Camera settings documents are held by Daniel Browning and at the EBU
My book, 'Circles of Confusion' is available here.
Also EBU Tech.3335 tells how to test cameras, and R.118 tells how to use the results.

Alan McKeown
Offline
Joined: May 9 2001

Alan,

Quote:
“Underscan's nothing to do with display resolution, it's designed in, it's deliberate“

Shouldn’t that read “Overscan’s nothing to do with........” ?

Alan

Alan Roberts
Alan Roberts's picture
Offline
Joined: May 3 1999

Rats, of course. I'm not well today, concentration's hard.

Get my test cards document, and cards for 625, 525, 720 and 1080. Thanks to Gavin Gration for hosting them.
Camera settings documents are held by Daniel Browning and at the EBU
My book, 'Circles of Confusion' is available here.
Also EBU Tech.3335 tells how to test cameras, and R.118 tells how to use the results.