Show posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Show posts Menu

Messages - Seanoneil72

Quote from: nanomad on May 20, 2013, 11:24:42 PM
I have no clue about altering the downsampling algorithm (it could be hardcoded or programmable, who knows). What's sure is that using the CPU to do so is impossible

About focus peaking, you can raise the treshold as you wish (the higher, the more "false" in focus points you have)

Good to hear that focus peaking treshold is adjustable, thanks. Not sure what would be in control of downsampling, cpu was just a catch all term, intended to be anyway, not sure if there was more in the camera for processing than that digic 5 chip because have no clue about the hardware in the camera.  :-) 

I do have a feeling that Canon would make a point to make such a thing impossible because if these cameras simply had an average downsampling matrix, say like the default windows picture viewer, then the video quality would smoke movie cameras costing over 20 times more.
A few random question for the devs,

When shooting video, I have read that the camera uses a pixel skipping process to scale down the 18mp image captured by the sensor to a ~2mp 1080 video frame, or a ~1mp 720 frame. So then the cpu is basically just remapping select pixels from the sensor into the video frame of whatever size if I am understanding correctly?

Is there any possibility that the sensor data could be looked at on the whole and downsampled so that averaged values were used for the remapped pixels or is there not enough horsepower in the cpu to process that amound of data quickly enough to fill the frame buffer in real time?

If downsampling (mathematically averaging and redrawing not simply remapping values) is not possible maybe its possible to fine tune focus peaking so that the peaking would register as ideal when ever so slightly defocused, which actually seems to produce an equally sharp video image with less aliasing and less moire? 
Quote from: 1ricca on March 23, 2013, 03:35:40 AM
That totally burns! My main concern is getting a proper build with Kelvin WB controls on the 650d before its abandoned (if it gets abandoned). Also, the 700d will likely not take a whole lot of work for the port from wherever the current build of 650d ML is at the time of that project start, considering it's basically a T2i-T3i type of jump. The 700d is basicallty a slightly tuned 650d.

Why the hell did Canon have to make a t5i so damn fast?!

Looks like the same camera but they put back the crop zoom, that is probably the most irritating thing of all unless they offer a firmware update to enable it in the t4i.
Quote from: msrobi on January 29, 2013, 02:20:26 AM
I'm willing to be patient, because the longer we have to wait, the more stable the alpha will be.

As for Magic Zoom, it allows you to see a magnified view of the image while recording video.  This helps with manual focusing using only the back of the camera.

Ahh magic zoom sounds like a lifesaver. Any hope for the return of the t3i video zoom or maybe a better downsampling script to help with the moire and softness? Or is that stuff beyond the scope of ml?

In any case magic zoom will be great to have, especially for those of us without stm lenses.


I just wanted to pop in and say that this is very exciting for me as a new t4i / 650d owner (upgrading from my trusty 300d I got close to 10 years ago.)

I did some research into the t4i before I bought it, but in all of my research into the t2i, t3i, and t4i, I had never come across any information on magic lantern. If I had I may have bought a different camera but it seems that the t4i has potential for magic lantern to make it pretty special.

I still know little about magic lantern, so I wanted to ask if some of my hopes could be considered realistic. I have read a few posts asking for the ability of the 10x digital zoom for video to be unlocked like the t3i and would like to add my voice to that request, if such a thing is possible.

Some maybe more complicated features to implement that would be great (again I don't know how realistic some of my hopes are so take that into consideration) would be to unlock the ability to get 1080p60 video and maybe a better downsampling script to eliminate the heavy moire patterns that show up in video. If we are really lucky maybe there will be a way to make the continuous auto focus actually useful.

Maybe opening up some more control over HDR would be possible too. Anyways, I plan to follow this thread for a while and see where things go. Thank you to everyone working on magic lantern for this camera, it certainly gives us t4i owners something to look forward to.

Thank you!