Sample Clips: Using an 85B filter to reduce noise?

Sample Clips: Using an 85B filter to reduce noise?

  • CLIP A has less noise

    Votes: 5 71.4%

  • CLIP B has less noise

    Votes: 2 28.6%
  • Total voters
    7
Luis Caffesse

Luis Caffesse

Admin
So its been mentioned more than once that using an 85B filter in daylight situations can reduce noise in the image (when using the 3.2K preset instead of the 5.6K preset).

I could never find any samples showing that - so I finally bit the bullet and just bought a decent 85B filter to see if it really made any difference in the footage.

I just shot a quickie test and thought you guys might be interested.
(this was a pretty rough and quick setup just to see if there was any merit to taking the time to set up a more controlled test).

I shot with the camera in 720/24PN
All settings were at the default "0"
Gamma was set to CinelikeD on purpose because its said to be the noisiest.

I proped a Macbeth chart up in my backyard and rolled a few seconds both with and without the filter.
White balance on both clips were camera presents (3.2K & 5.6K).
I also had to adjust the exposure to compensate for the filter.

I exported one 5 second clip of each, seeing as there was no point in postsing compressed versions when you're looking for noise.
I didn't touch the clips at all, only imported to FCP and exported those 5 second segments.
The best way to view the noise is going into AE, or some other program where you can view the R G B channels seperately.

Originally I was going to tell you guys which was which.....
But it just occurred to me that it might be better to do this as a blind poll.
So check out both clips if you're interested, and vote for which clip you felt had the least amount of noise.

You can download the clips here.
The zip file is 50.3MB
The clips are labeled A.mov and B.mov

I'm curious to know how this turns out.



EDITED TO ADD:
Okay, clips are now uploaded and ready to go.
:thumbsup:
 
Last edited:
Just for the record - this is NOT the same thing Chris (Cardmaverick) was talking about recently with his Magenta filter noise reduction technique.
 
I guess I'm the only guy who voted so far. I don't want to skew the tests by announcing my vote, but Luis don't keep me in suspense too long.
I've wanted to test this myself, but never bothered.
How about a private email to :
nsll@pacbell.net


Lenny Levy
 
Well, I'm glad someone found this interesting
:)

I figured instead of just posting side by sides we should start trying to do more blind polling.... seems like a great way to really get some objective answers to these subtle tech issues. But maybe its not the most practical idea.
 
cordvision said:
both look the same to me...
Click to expand...

Thanks for taking the time to look at them.
That's exactly why I wanted to post this as a blind poll ... I was wondering if people would see any difference without knowing which clip was which.

It ocurred to me after the fact that I probably should have added a 3rd option to the poll which says "There is no difference" or something like that, but unfortunately you can't edit a poll once its posted (or at least I couldn't figure out how).

Hopefully some more people will chime in here.... but its looking like I may have reached the limits of techy geekness with this poll.
:)
 
watching these two in quicktime, i saw no difference. but in AE i could see a considerable difference on the blue channel. what was the light loss like when using the filter?
 
Luis... good poll and downloading it now. I know you mentioned a while back where you got your icon from.. and it still makes me laugh.. So friggin funny looking
 
Justyn said:
I know you mentioned a while back where you got your icon from.. and it still makes me laugh.. So friggin funny looking
Click to expand...

It's 'Lumpy' (Chewbacca's son) from the "Star Wars Holiday Special"
Of course, in the special he wasn't wearing a tuxedo.
I added that in Photoshop.
:)
 
Well I wasn't going to post anything at first so as not to influence people but seeing as its even right now, I gotta say i'm surprised. It seems clear to me that A is cleaner. Not a tremendous difference, but definitely noticable to me.

I'm wondering - if i notice a difference on this download would the original be more obvious also on a real video monitor?
 
Lenilenapi said:
would the original be more obvious also on a real video monitor?
Click to expand...

The clips you downloaded are DVCProHD clips exported out of FCP from the original footage. So feel free to output it to a monitor and see if you can see a difference.

I don't have an HD CRT monitor... though I do have a decent SD monitor sitting here - I suppose I could hook it up and see if I can tell a difference in a downconverted image. Though honestly, I think that's overkill seeing as the reactions seems to be so close right now.

I'll give it a while and if no one else responds then I'll go ahead and post which clip is which. But its starting to look like it might not make a huge difference - at least not to our small little focus group here.
:)
 
Re: Sample Clips: Using an 85B filter to reduce noise?

Re: Sample Clips: Using an 85B filter to reduce noise?

The one I voted for as having less noise seemed to have more or about the same as the other one in one particular color chip?
Steve
 
Well, it looks like everyone that is going to has viewed/voted on these clips.
I suppose ClipA seems to have less noise according to most of you (5 to 2) - though the comments lead me to believe most aren't seeing a major difference.

I'll post which is which shortly.
 
Okay - looks like no one else is going to check this out - so here it is.


CLIP A used the 85B filter and the 3.2K white balance preset

CLIP B used no filter and the 5.6K white balance preset


Personally - I don't see a whole lot of difference.
But it looks like the majority of you did, though its a very small sample.

I guess its worth doing some more test shots and see if it makes a difference.
But its definitely not a night/day difference.

Anyhow - I hope someone found this helpful other than just me.
:thumbsup:
 
Luis,
I found it helpful as I heard about this from day 1 from Barry and bought an 85B because a client requested it. I now see that it does make a bit of a difference. Its not night and day, but if I have the time i would go with the 85. You have to factor the question of putting extra glass on the front though. And then, do you want to stack an 85 + a Pola, do you buy a straight Pola and an 85 Pola, an 85N6 and a straight ND6 etc. it can get expensive.

Thanks for the testing though.

- Lenny

Lenny
 
Lenilenapi said:
Thanks for the testing though.
Click to expand...

No problem.
When I get around to it, I'm going to do some similar tests (albeit more controlled) with the other Gamma curves (its really only Bpress that I'm interested in).

If I find anything interesting, I'll let you know.
Also, when I can justify spending money for fun I plan on picking up a magenta filter and putting Chris' techinque to the test as well... we'll see if I can make that happen sometime in the near future.
 
You must log in or register to reply here.
Back
Top