Topics

Has anyone tried TEAM brand Compact Flash for ML Raw? They're selling 1000x 64GB and 128GB for the cheap.

Like this: http://petapixel.com/2014/09/05/steadxp-hopes-add-accelerometer-dslrs-gopros-better-image-stabilization/#more-144279

This picture: http://i.imgur.com/cLsodHx.jpg

Was taken at almost the same time someone's flash fires, resulting half of the sensor read to be illuminated by the flash.

The feature requested is to have the ability to time the flash firing in relation to sensor readout so you can set how much of the frame is flash-lit and how much isn't.

This relates to the pop-up flash, ext. flash, and off-camera flash.


Source:

This is my wife's photo, taken with an iphone 4s, while the flash on my actual camera went off. So, no photoshop or any post editing, she was just taking pictures while I was and caught this! We were underneath an I-35 overpass in South Austin. The spinners name is Adranos and you can find more pictures taken of him at www.adranosflows.com

http://www.reddit.com/r/PerfectTiming/comments/27ftts/picture_taken_right_when_another_cameras_flash/ci0lr4u

DxO Optics Pro can take certain DNG files for processing, but only from certain cameras. The request is for a converter that mimics output from such cameras so that you can batch process the DNGs in DxO Optics Pro.

Especially for Prime Noise Reduction

Hey guize
Wondering what your render time is for raw videos?

I'm converting the DNGs into DNxHD and 10,525 frames has been rendering for more than 12 hours now, and it's only gone through about half of that.

I'm converting this using AfterFX (with adobe camera raw) and rendering with Adobe Media Encoder.

Wondering if this is normal or is there a better converter?

Thanks

The Multishot NR on the EOS-M (and I suspect several other cameras) will be disabled when you turn off and turn on the camera. Please add an option to keep this setting enabled (or revert it back to on) when restarting the camera.

Thank you.

As in, in photo mode, you can switch the camera to MF mode, and have the movie record button function to trigger autofocus (as in half-shutter).

Option to keep AF assist light on during photo/video. Very useful when you're in the night and had no other light source, except the one embedded in your camera. It's quite a bright light, even though it's totally red, a bit of round red pic/vid is better than no pic/vid at all.

In video hacks, add an option to disable the auto disable canon audio recording when turning on video hacks, and an option to save video hack settings.

As in, when starting a recording, switch image resolution setting to whatever the maximum memory is for that specific camera, then switch the image resolution back to its original settings when recording is stopped.

For example, someone who would usually put the camera in RAW+L would have the camera set back to RAW+L when they're done recording.

Given: the way Highlight Tone Priority works is that it pushes the ISO higher by one stop. This practically increases the dynamic range by one stop with the side effect of increasing the minimum ISO that can be set by one stop.

You can shoot underexposed in RAW and push it by several stops to "increase" or push the dynamic range in the highlight, effectively extending or simulating an extended form of highlight tone priority. (with a bit of extra noise on the shadow, but you can denoise that)

But for video, doing this will only raise the noise-laden compression in the shadow area and having the 8bit compression wreak havoc if you push it just a little bit.

The idea is to have a form of extended highlight tone priority. Whereas the default increases by one stop, I wonder how much DR we can push out of any given sensor. The default HTP increases the minimum ISO from 100 to 200, I wonder if it is possible to increase this to perhaps 800 or 1600 and maybe gain 3-4 stops of DR?

We can denoise the shadow area to some extent, but this will, to some extent achieve a total of perhaps 14-15 stops of DR, which threads into the level of DR of the Arri alexa (14 stops IIRC) and maybe Red's new Dragon sensor (16 stops) and give great highlight rolloff.

So, in my Zoom H4N portable audio recorder, there is an auto volume level setting that lowers the volume if the recorder detects clipping. This sometimes happen when you record someone talking, and then the room cheers and clapping insanely loudly at the end of the speech and the audio would clip like crazy. With the auto volume level setting on, the recorder would immediately reduce the volume to a level where it wouldn't clip.

Now, AGC does this, but the downside is that the AGC clamps down hard when the room becomes quiet again, raising the floor noise level and everything. The auto volume level thing does not raise the volume afterwards like AGC. It just keeps it at the lowest level so far that prevents further clipping.

Now, I know this has very limited use, but it's very useful when you just don't have the time to properly adjust the in-camera mic volume. (except on the 5D3 with the touch the back wheel to set the volume thing on the fly)

It also works great for 'lazy recording'. That is, you just want to make sure that the audio doesn't clip when you record casually. i.e. just hit record without checking the volume.

This way, we could just leave the recording level at reasonably high level all the time, and it will lower the record volume to the level that it wouldn't clip every time.


I'm sorry if this has been posted before, but I searched the forum and couldn't find anything like this.

Dead/hot pixel: pixel stuck in a certain condition that it would leave a spot in your JPEG/video. To clear this out, one must put on the body cap to the body, turn on manual sensor cleaning, wait for 30 seconds, and hope it triggers the hot pixel mapping function. The way canon hides the hot pixel mapping function this way is beyond me, but if there's a way for ML to trigger this function, that would be awesomely useful.

Hot pixel mapping tutorial: http://www.youtube.com/watch?feature=player_embedded&v=WJBuGhMnvFo

So this has been requested, but there's no response from anyone. I hope a dev could chime in on what ML can do regarding this topic.