XF Noise reduction

[Bill Weaver]

During my first low-light run-through of my XF, using standard settings as well as Alan Robert's recommended settings, I was appalled at the noise in the picture, even at 0 db, and found the noise levels at 6 db very unpleasant.

This prompted me to experiment a bit with the noise reduction.

First of all, I found very little difference between automatic NR and none at all. So I went into manual mode, and discovered that with a setting of 3, I lost only enough detail to give me a less harsh picture, and got a an impressive reduction in noise as a result. Looks really good at 6db.

Have I sacrificed anything else to achieve this? Looks good to me.

Anyone else doing any tweaking in this area?

[Doug Jensen]

I've never noticed any objectionable noise in my picture with the paint settings I use, so I haven't looked for any extra methods of reducing noise. If you ask me, the camera is very clean for 1/3" sensors.

[Doug Tessler]

I think this camera has the lowest noise levels of any camera i have owned very clean

Doug

Maybe you had the AGC switch on that would create massive amounts of noise

[Robert Turchick]

I have been reviewing footage from last weekends shoot which included a cinealta hdc1500 mounted in a cineflex system recording 100Mbps to a ki pro, an HMC150, HMC40, 7D, and my Xf300.
Out of all those cameras, guess which is the cleanest noise-wise....it's not the $100k cinealta!
The Canons were the best with the XF better than the 7D. Also had the truest color. Shoot was outdoors in the sun so all the cams had plenty of light.
Of course I'm sticking with the claim that it is all due to my amazing skills as a cinematographer! Ha ha!

Disclaimer... The cinealta normally records to an srw1 deck for full 4:4:4 video. They literally got the Ki pro the day before so it may not be fully dialed in yet. I've seen the full-bore 4:4:4 footage the system is capable of and it is simply stunning.

Canon's XF300 however, is a force to be reckoned with!! And I second Doug in statin this is about the cleanest camera I've ever used!

[Doug Jensen]

I have a PDW-F800 that has the exact same CCDs and a lot of the same guts as an HDC-1500, and I can assure you that it puts out a far better and cleaner picture than my XF305. The XF is good, but it's not THAT good. Of course, it doesn't cost nearly as much either. :-)

I'd say your HDC-1500 was not dialed in right if it looked better than the 300.

[Robert Turchick]

as I mentioned, I don't think they had the KiPro dialed in. I know the guys who own the system, I think I'm going to discuss this with them.

Here's a couple of screengrabs that pretty clearly show it. Wondering if some sort of noise is being introduced between the camera and Ki Pro?

And I know there's no way the XF is a better camera! Just calling this one like I see it!

[Doug Jensen]

Wow, are you sure #2 wasn't shot with a 1990 era VHS camcorder?
They've got more problems that just not "dialing in" the right settings. I really don't know how you could get a HDC-1500 to look that bad. I'd be embarrassed.

[Daniel Doherty]

If they were recording 100mb/s with the Ki Pro they were doing themselves a great disservice. The Ki Pro goes up to 220mb/s at 30p.

[Doug Jensen]

I can't see how the Ki Pro could be the problem. My F800 looks fantastic at 50Mbps 30P when recording to XDCAM optical. I could barely see any difference at higher bit rates when I tested the F800 with my NanoFlash. For green screen work, sure there's a little difference, but for outdoor daytime shooting you'd really have troulble telling 50 vs 100 or 220. Same thing goes for my XF305. Recording at higher bit rates than what you can get on the CF cards directly has very little benefit for routine shooting

The problems that freeze frame #2 show go way beyond any recording device problems. The camera itself could not have been setup properly. The bit rate is not the issue. Just to prove me point, here's some aircraft footage I shot last year at only 35Mps with an EX3. And what you're seeing on Vimeo doesn't look anywhere near as good as the original (no grading has been done).

2009 Rhode Island Air Show Highlights on Vimeo

[Robert Turchick]

Yeah...not sure what's going on. The sr1 footage looks perfect so it's something they did getting it to the KiPro. I'll be going over this with them and this was a test run for a few things in their rig...not a paying client.

[Robert Turchick]

This is starting to feel like an unintentional threadjacking so let me finish my dilemma by tying it back to the OP...
The XF is a very clean camera. If excessive noise is in the picture something is wrong. As you can see by my example above, even $100k camera systems can be messed up. It's usually something simple like a setting or switch. In the hdc1500/KiPro instance, there's a lot of pieces to the puzzle and hopefully it is a simple fix.
As I get into troubleshooting with the company, I will post to a different thread. I am bidding on a gig that they would be involved in so it's important to solve the issue.

[Bill Weaver]

Thanks for the confidence re: noise. I was testing the camera in extremely low light settings, during last vestiges of light in the evening, and indoors with little more than desk lamp illuminating the room. In those situations, lots of noise at 6 db.

I'm stil trying to see a difference between auto NR off and on. But, again with manual at a low setting seems to shore things up.....

[Peter Moretti]

Originally Posted by Robert Turchick

as I mentioned, I don't think they had the KiPro dialed in. I know the guys who own the system, I think I'm going to discuss this with them.

Here's a couple of screengrabs that pretty clearly show it. Wondering if some sort of noise is being introduced between the camera and Ki Pro?

And I know there's no way the XF is a better camera! Just calling this one like I see it!

My guess, which could be way off base, is that the aperture on the Sony was set way too small, causing diffraction and gain. Just a wild a$$ guess.