50D Raw video

[jcdenton]

No it won't do that.

Shooting in a lower resolution will just give you a softer image when blown up to 1080P, if you only plan to put it on a DVD then this won't matter. (although, in my experience, shooting a higher res and downscaling does tend to give better results than shooting SD from the outset)

The angle of view, crop factor, DOF, f-stop etc will stay exactly the same regardless of whether you shoot at 1584x1058 or less. All these things will  only change if you enter 5x shooting mode, then the crop factor is around 4x I believe.

Ok, But when I change the resolution on my 50d in raw_rec module (for example to 1440x890) the shooting frame becomes smaller and don't cover the whole screen. And records only the image that stays within the frame. And it always been this way in every build I tried. Your experience is different?

[Wlad81]

I think shooting at 900-something by 500-something  is enough to get a good picture for DVD, and it gives the time for the computer to copy the raw files from the other card, am I right? I've been shooting on Beta SP (PAL), at the picture was quite good, though it has worse resolution than 50D at that res might have.

[rommex]

Interesting comparison of BMPC against 5dM3 and 7d ML (why not 50d?):

http://www.eoshd.com/content/11350/depth-test-5d-mark-iii-7d-raw-vs-blackmagic-pocket-vs-gh3

[Andy600]

My 1000th post  and to think, I only started using Magic Lantern for zebras and focus peaking 

I've learned so much. Thankyou guys!

[Andy600]

Interesting comparison of BMPC against 5dM3 and 7d ML (why not 50d?):

http://www.eoshd.com/content/11350/depth-test-5d-mark-iii-7d-raw-vs-blackmagic-pocket-vs-gh3

Yea, I saw that and thought the same thing. Is it just me or does the 50D produce a better raw video image than the 7D? I think the smaller sensor helps a bit with aliasing and moire compared to the 7D. It's still there of course but there is something very nice and, dare I say, unique about the 50D's image quality (if you treat it right ).

[rockfallfilms]

Ok, But when I change the resolution on my 50d in raw_rec module (for example to 1440x890) the shooting frame becomes smaller and don't cover the whole screen. And records only the image that stays within the frame. And it always been this way in every build I tried. Your experience is different?

Yes that is how it is going to be.

You are shooting a smaller frame resolution so you would need to expand it to fill the frame. This will result in a softer image because you have less resolution to begin with.

In the same way that if people used to  shoot in SD they would have to expand the image to HD.

This chart might help.

[rockfallfilms]

Interesting comparison of BMPC against 5dM3 and 7d ML (why not 50d?):

http://www.eoshd.com/content/11350/depth-test-5d-mark-iii-7d-raw-vs-blackmagic-pocket-vs-gh3

I think the 50D would have faired pretty well and been quite close to the pocket. Not quite sure why he chose the 7D unless he has sold his 50D. It's like a warzone on those forums so I don't tend to comment there!

Also people going on about the GH3, saying it's better than Raw. To me they are just trying to justify their purchases. Now I've got a taste for Raw I could never go back to 8bit h264.

There is something very nice about the 50D image, I think it's pretty close to the Blackmagic in terms of motion cadence.

I'm still hoping a VAF might materialise at some point.

[LEVISDAVIS]

I was hitting an 81.5 MB/s with the 1050x when shooting 1584x1058. The card was formatted in camera too.

[SleeperNinja]

Ok, But when I change the resolution on my 50d in raw_rec module (for example to 1440x890) the shooting frame becomes smaller and don't cover the whole screen. And records only the image that stays within the frame. And it always been this way in every build I tried. Your experience is different?

Probably the same and different, at the same time.

You should spend more time reading in the forums, rather than posting. Everything you need to know is here, you just have to digest it.

The point is that it's capturing raw live feed as sequential images. If you capture smaller than the whole frame, where would you take it from? It's not resizing it, it's just capturing as many pixels as you tell it to.

[jcdenton]

Probably the same and different, at the same time.

You should spend more time reading in the forums, rather than posting. Everything you need to know is here, you just have to digest it.

The point is that it's capturing raw live feed as sequential images. If you capture smaller than the whole frame, where would you take it from? It's not resizing it, it's just capturing as many pixels as you tell it to.

I know that and that is what I'm trying to tell. You capturing a part of image you suppose to take from sensor. So if you have 30mm lens in FF it will be 50mm on 50d and it will 60mm or even more when you choose to shoot in lower resolution with RAW on 50d. I't will just crop your angle. And in result you will have more noticeable noise. It will just be bigger and softer just like you zoom Full HD on computer. You just began to notice more then with max resolution.

[CITY-U1001]

Hi All,
Dual ISO not work on 50D in video mode ? I see Dual ISO update everyday

[rockfallfilms]

Hi All,
Dual ISO not work on 50D in video mode ? I see Dual ISO update everyday

No it doesn't

[CITY-U1001]

No it doesn't

it isn't planned?

[rockfallfilms]

it isn't planned?

As far as I'm aware, the hardware doesn't support it.

Developer 1% will be able to confirm whether this is correct or not.

[1%]

Changing the registers in LV has no effect on 7D/50D/5DII, basically that whole generation.

[Wlad81]

The common question: does ML plan to introduce a 12 bit or on-the-fly lossless compressed version of raw? Sorry for the offtopic

[1%]

That's not very possible.

[goldenchild9to5]

Yea, I saw that and thought the same thing. Is it just me or does the 50D produce a better raw video image than the 7D? I think the smaller sensor helps a bit with aliasing and moire compared to the 7D. It's still there of course but there is something very nice and, dare I say, unique about the 50D's image quality (if you treat it right ).

Yes it does Andy.. to me nothing right now beats the 50D as far as it's filmic qualities only with the exception of the Blackmagic Cinema Camera. 

[goldenchild9to5]

Guy's I'm back haven't been posting for a while now due to a couple music projects that I was working on.  I'm thankful for all the advances I'm seeing @a1ex @1% @andy600 props to you guy's for taking it this far, and let's not forget @gregoryofmanhattan for getting the 50D hacks properly running @ the beginning.  I'm currently working on a treatment for a music video shoot that I'm shooting and Directing, if anybody lives close to Connecticut, or lives close to Manhattan shoot me and email or reply this post looking for collaborators for this project.  All I can say right now it will all be shot on the 50D, maybe with a couple slowmotion shoots from my GH2.  Looking forward for your replies. 

GOD Bless

[1%]

I like the 7D in 1x and 50D in crop. 7D is better for dual purpose photo/video, 50D is better for video only.

[N/A]

I wonder how 40D raw would look, if even possible....

[50Deezil]

Yea, I saw that and thought the same thing. Is it just me or does the 50D produce a better raw video image than the 7D? I think the smaller sensor helps a bit with aliasing and moire compared to the 7D. It's still there of course but there is something very nice and, dare I say, unique about the 50D's image quality (if you treat it right ).

The 7D and 50D have the same size sensor of 14.9 x 22.3mm APS-C - but the Resolution of the sensor in the 7D is 5360 x 3515 and the 50D is 4770 x 3177.    They have different amount of pixels  7D 18.84 Mpix and 50D 15.15 Mpix.    The Pixel pitch of the 7D is 4.2 and the 50D is 4.7.  So it seems they have totally different sensors which could easily explain the difference some think they see.

The strange thing is that the Dynamic Range of the 7D and 50D are almost the same.  The most they can muster at ISO 100 is 11.7 for the 7D and 11.4 for the 50D.  The 2 cameras track almost identical thru the rest of their ISO range in terms of DR.  I don't know about Noise tho.     I'd love to see some data on the Noise going thru the ISO range.

Has anyone done a comparison video between the 7D and 50D to prove if the image in indeed different with the same settings? 

[50Deezil]

After doing some technical research on the 7D and 50D I found that the 50D has far more noise in RAW as you increase ISO than the 7D.  The 7D also retains more detail and I think this is probably one of the things people like about the 50D.  The slight softness is likely more pleasing than the 7D for some.  This might also be the same for the  5D3 in some peoples opinion.  I think some people like the slightly softer look of the 50D.

Nothing i've seen would suggest that the 50D would be better in low light than the 7D tho.  Another issue is how the cameras handle color.  From what i've seen they seem to have the same color profile.  So I think another issue would be highlight handling.  The only thing I see is that it seems the 50D is slightly shifted towards the highlight range and the 7D the shadow range.   Not a wide margin but it's there.   Throughout the usable ISO range the 50D holds a slight advantage in highlights.  The 7D does better in the shadow range.  According to DP Review the best they could get from the 7D in RAW was 9.8EV and the best they got from the 50D was 11.6 in RAW.  So this could be another aspect that is helping the look of the 50D.

[a1ex]

According to ML raw histogram, how much DR do you get?

(it measures DR on the fly, with the engineering definition, so it should be objective and comparable across all cameras).

DR shifted to highlights or shadows is nonsense because (1) all digital cameras have a hard clipping point and (2) you are free to meter wherever you want, relative to that clipping point. To compare DR between two cameras, you should make sure the clipping point is the same in both images (otherwise you are comparing apples with oranges).

[Andy600]

According to ML raw histogram, how much DR do you get?

(it measures DR on the fly, with the engineering definition, so it should be objective and comparable across all cameras).

DR shifted to highlights or shadows is nonsense because (1) all digital cameras have a hard clipping point and (2) you are free to meter wherever you want, relative to that clipping point. To compare DR between two cameras, you should make sure the clipping point is the same in both images (otherwise you are comparing apples with oranges).

I just did a quick, unscientific test and the max I can get it to show is 11.4EV at ISO 100 & 200 (which proves the Dx0 mark ISO charts) using center weighted average metering.