Baked in Look

[stoneinapond]

It seems a shame to start a new thread, but I don't want to step all over the firmware thread.

Would someone be so kind as to try and explain what the term means? Is this something left over from film technology? It meant very little to me when watching the Zacuto shoot out.

Thanks in advance.

 

[alignment1]

"Baked in" is when the in-camera settings such as contrast, sharpness, saturation, etc. are pre-set via internal camera adjustments. It's as if someone has already applied these things to your footage before you even get it into post.

An example would be: Shooting something 'raw', vs. using an internal B+W filter. Well now that your footage is B+W when you bring into post, it's B+W THAT'S IT!, -there's no color detail to make any adjustments...so in a sense your shot has been baked B+W--very little room for any post adjusting.

Make sense?

So if you use an in-camera 'setting' like "SUPER HIGH CONTRASTY-MEGA SATURATED" - well you're kind of cornering yourself into that 'look'.

I think what a lot of shooters like is a very 'flat' image, in which you the 'color god' can add your richness, contrast, saturation, or whatever look your going for....many more options this way.

However- if you don't have time for all that CC/grading and such, and you enjoy the cam's scene settings then that's there for you to mess with as well.

 

[stoneinapond]

Make sense?

Thanks and yes, makes perfect sense. However, "Baked In" (and that is how you began your kind reply) refers to a "process". But I hear the term used as an adjective ("Look") that appears to have certain properties.

Are people in this instance referring to a specific set of visual attributes? To me it sounds like someone burnt something in the oven, and it came out looking "burnt" (i.e. blackened).

Am I making sense?

 

[alignment1]

Thanks and yes, makes perfect sense. However, "Baked In" (and that is how you began your kind reply) refers to a "process". But I hear the term used as an adjective ("Look") that appears to have certain properties.

Are people in this instance referring to a specific set of visual attributes? To me it sounds like someone burnt something in the oven, and it came out looking "burnt" (i.e. blackened).

Am I making sense?

yep I understand what you're saying--as far as I know, 'baked-in' is not an aesthetic attribute that is to say the description of burning something in the oven- more that it's 'look' is a little more 'set in stone'. So the scene file has been 'pre-set', 'baked-in', 'pre-determined', but I could be completely wrong--that's just been my understanding--I'm sure someone else will chime in.

cheers

 

[Martti Ekstrand]

"Baked-in" implies that the encoding is so lossy that the look can't be altered later. Which is patently untrue in regards to GH1. Watch exhibit A, original camera image with cocked-up cold white-balance and exhibit B; graded sunny warm result.

 

[stoneinapond]

"Baked-in" implies that the encoding is so lossy that the look can't be altered later. Which is patently untrue in regards to GH1. Watch exhibit A, original camera image with cocked-up cold white-balance and exhibit B; graded sunny warm result.

Martti,

Thanks for your reply.

Yes, I remember watching your very enjoyable video a while back.

Don't want to beat what might appear to be a dead horse, but my question remains. How does one know that a look is baked in? Or are people just assuming when they see something they don't like, that it is baked in and permanent.

 

[Martti Ekstrand]

The assumption many people sling around is that the encoding in HDSLRs is so lossy that nothing can be done with it in post - ie: "a baked-in look" - but with the right image settings there's is some alteration room as I've shown.

 

[stoneinapond]

Got it.

Many thanks to you both!

 

[alignment1]

The assumption many people sling around is that the encoding in HDSLRs is so lossy that nothing can be done with it in post - ie: "a baked-in look" - but with the right image settings there's is some alteration room as I've shown.

The GH-1 has lovely latitude for sure- in fact, I was just looking over some old footage from the winter -- sunset scene- very little daylight left. One particular shot was of a guys face which was really silhouetted, no features to be made out on face- and with a little grading- PRESTO-- ther his face was- it's really quite amazing what info is actually there- even when at first it doesn't appear to be.:thumbsup:

I have a question for you Martti- in that example you showed - that is good example how to correct white balance- but (and I've never tried this but I'm curious), Are you telling me that if you shot w/ gh1 settings : Contrast +2, Saturation +2, etc. that in post you'd be able to dial those settings back to a more nuetral, flatter image?? I can't imagine that's possible without some type of loss. That's what my interpretation of being 'baked in' was. perhaps I'm off:engel017:

Z

 

[Essami]

Are you telling me that if you shot w/ gh1 settings : Contrast +2, Saturation +2, etc. that in post you'd be able to dial those settings back to a more nuetral, flatter image?? I can't imagine that's possible without some type of loss. That's what my interpretation of being 'baked in' was. perhaps I'm off:engel017:

Z

In my experience too as long as you dont over or underexpose (ie. totally clip) you can do quite a bit of tweaking.

 

[Cavemandude]

The GH-1 has lovely latitude for sure- in fact, I was just looking over some old footage from the winter -- sunset scene- very little daylight left. One particular shot was of a guys face which was really silhouetted, no features to be made out on face- and with a little grading- PRESTO-- ther his face was- it's really quite amazing what info is actually there- even when at first it doesn't appear to be.:thumbsup:

Which is what puzzles me about the Zacuto Shootout tests and stating that they color corrected all the videos from all the cameras to try and match them to film as much as possible. Why weren't their color correction guys in post able to do this with the GH1 videos to show how much latitude it could have? It looks like they didn't color correct the GH1 videos at all since the blacks were still crushed.

Randy

 

[PDR]

Which is what puzzles me about the Zacuto Shootout tests and stating that they color corrected all the videos from all the cameras to try and match them to film as much as possible. Why weren't their color correction guys in post able to do this with the GH1 videos to show how much latitude it could have? It looks like they didn't color correct the GH1 videos at all since the blacks were still crushed.

I think that's what they were implying by the term "baked in"

I think the GH1's compression tends drops more shadow detail than the 5D or 7D for example (which also use poor h.264 implentations, but make up for it by using ~3x the bitrate). They weren't able to bring out any more detail without having the footage turn to mush. I think the compression artifacts limited what they were able to do in post.

 

[ojejones]

I have had the same experience the image is definitely flatter.

 

[Razz16mm]

The range of corrections possible with any form of encoded video are limited compared to the possibilities with RAW image processing where nothing about the look is baked in. "Baked in" means you should y try to get as close to the look you want as you can for video as it is recorded from the camera to minimize the range of corrections needed in post. This is especially true of optimizing white balance, and exposure settings before you shoot.
With encoded video, some issues can't be corrected in post. You don't for instance have post processing tools that literally let you pick the ISO speed, and color temperature you want in degrees K before you even begin color correction, or the ability to choose from multiple different gamma curves or even create a custom gamma curve.
Its not that you can't correct encoded video, just that the range and choice of tools to do so is more limited.

 

[dvbrother]

Are you telling me that if you shot w/ gh1 settings : Contrast +2, Saturation +2, etc. that in post you'd be able to dial those settings back to a more nuetral, flatter image??

It's difficult to remove contrast after it's been added...you will lose detail in the shadows.

Saturation is a different story. Reducing saturation in post is not detrimental to the image. The luminance values of the pixels remain, so the image just has less color but is still full resolution.

 

[Cassius]

There's more to it than color correcting. The Canon cameras also allow you to modify the pre-encoded image in huge ways, which is a very powerful tool. The simple settings on the GH1 don't allow the same tweaking.

 

[Cavemandude]

It's difficult to remove contrast after it's been added...you will lose detail in the shadows.

They weren't able to bring out any more detail without having the footage turn to mush. I think the compression artifacts limited what they were able to do in post.

All you have to do is look at still frames that Martti Ekstrand posted from his video on the first page of this thread to see how much detail can be brought out of the crushed black area from a GH1 video.

Randy

 

[PDR]

No one is saying you can't bring out more detail or produce excellent results; We've all seen excellent footage and grading. Martti provided some excellent examples, as have others.

It's a relative thing here, not an all or none phenomenon

You're going to be able to have more freedom and be less limited with 16-bit 444 SGI film scans. Similarly, you are going to have more freedom with 45Mb/s vs. 18 Mb/s footage, that uses a similar compression implementation. If the GH1 was able to record 45Mb/s, everything else the same, wouldn't you get better results? The quantization and DCT decimation (especially with AVC) is known to adversely affect shadow areas. Software encoders address this with adaptive quantization techniques, but the hardware encoder is very limited, the implementation is very rudimentary.

The fact is, there is going to be more noise and compression artifacts in the lower bitrate footage, and that will limit your options comparatively. The differences will be even larger in higher scene complexity scenarios, where the codec is taxed even harder.

 

[Isaac_Brody]

Baked in to me means commiting to a look in camera and not relying on post for that. I notice it when I see crushed black levels and highlights blown out because of high contrast. The good thing about GH1 is it's standard look is really filmic. I think the straight out of camera look for GH1 footage is more filmic looking then the Canons or Nikons. The bad thing is depending on your settings you could be throwing away some dynamic range. I brought up the Adobe RGB setting which to my eyes looks more natural and still filmic and retains highlight information.

I think it's worth experimenting because on the GH1 the mjpeg/photojpeg footage looks overly sharp and contrasty to my eyes. So if there's a way to compensate for that through Adobe RGB and lowering contrast and sharpness settings then that's a good thing. Lowering some contrast just gives you a little more to work with in post.

 

[Ian-T]

I think the GH1's compression tends drops more shadow detail than the 5D or 7D for example (which also use poor h.264 implentations, but make up for it by using ~3x the bitrate).

I see folks say this a lot along with "poor AVCHD implementation etc." but I disagree. I think Canon knew what they were doing and did it right. They say it's poor because of missing B Frames. But from what I understand B Frames are only for lower bit rates to help them become more efficient. If anything Panasonic's is the only poor implementation of AVCHD because at anything under 24Mbps is considered AVCHD (with included B Frames). I guarantee you if Panasonic did have the needed B Frames in the 17MBps codec it would have easily matched the Canons in IQ. Canon got around this by (like you mentioned) trippling the data rate which was the correct thing to do. Their codec does not need B Frames. I think the next step up would be AVC-Intra. I could be wrong about all of this but I believe this to be correct.