http://vimeo.com/40911166 its just audio (04:10)
If true, that is good news. So far most of the testers have been silent on the issue of DR.
If true, that is good news. So far most of the testers have been silent on the issue of DR.
Not official, but claiming 13 stops of DR on the FS 700?
- Thread starter Wendt
- Start date
Chris Johnston
Veteran
Well, since it's not official let me say that my FS100, I haven't put it on a chart, but some other guys have, and we, they determined that it's probably at about 18 stops of dynamic range on here. now that is UN believable.
There. see how easy that was? I can talk S&%T out my a$$ too.
There. see how easy that was? I can talk S&%T out my a$$ too.
The sensor probably is over 13 stops, maybe even 14. Remember that the Nikon DSLR sensors are built by Sony and the D800 is over 14 stops of DR. But the problem is 8-bit encoding. It basically limits everything to 12-stops max. You can get more to 8-bits but it will start deteriorating the middle tones.
Postmaster
Dark Side of the Camera
The preproduction model I had was maybe 11.5ish, max. 12, depending on the gamma curve- which isn´t too shabby ether.
Frank
Frank
So, once the 4k recorder comes, the F700 may be able to realize a higher DR than in its current state at 8 Bit 4:2:0? So that is why the DR of the BMC Camera loses a little bit of DR when going to 10 bit 4:2:2 compared to 12 bit RAW? Does color space also have an effect on DR (4:2:0 vs 4:2:2 vs 4:4:4) or just Bit depth?
Nirv
Carbonite Member
Land that's why camera companies hand out cameras to these numpties as they're guaranteed to get exaggerated statements. It is rewarded by getting the next camera before everyone else. A positive feedback loop.
alaskacameradude
Veteran
Really? I always thought that increasing the bit depth, would enables the capture of more gray shades or color tones.
Dynamic Range refers to the ability to express the full range of tonal variations from lightest light to darkest dark?
Wouldn't that be affected by bit depth? Or is it just that there would be more 'banding' with lower bit rate but
not more dynamic range?
And with 8-bit that ratio will always be 0-255.
With audio, a 16-signal can be about 96 db. With 24 bits you can get up to 144db.
Nirv
Carbonite Member
That's signal to noise ratio not dynamic range.
Speaking about DR, people often mix two things - the ratio of the brightness and the ratio of the coded values (digital numbers). These are not interchangeable terms.
Imagine 1-bit codec. The value can only be 1 or 0. Now, this doesn't mean much, unless we know which brightness levels that "1" corresponds to.
In general case, the camera has to squeeze its DR (or part of it) somehow into the available bits. There are various ways to achieve that and none is perfect - one of the reasons we have so many gamma curves. Assume 8-bit codec - even linear gamma curve can squeeze more than 8 f-stops into them - all it needs to do is to 'compress' the brightness range, so that, f.ex. digital 255 represents the brightness of 12 f-stops above what's represented by digital 0. It's just the mapping.
The problem is that the more we compress, the worse the tonal resolution becomes in the mids (and shadows), which is of great importance, f.ex. for skin tones - there's risk of severe banding, when there are not enough distinct (coded) levels to smoothly represent skin gradation. That's why non-linear gamma curves are used - they assign more coding space (distinct values representing various brightness levels) to where it makes more sense, taking into account things like logarithmic tonal perception of human vision, anticipated viewing conditions, noise, gradability, being able to use standard displays for monitoring etc. F.ex, LOG-based gammas sacrifices easy monitoring in order to improve on other areas.
So, while 8-bit codec doesn't necessarily limit the DR it can represent, there's no point in squeezing too much DR into it, as severe banding will occur in post, if we're trying to stretch it to get back to the original scene DR. Of course, some compression is fine - we don't need too much tonal resolution in highlights so we can squeeze them - this is similar to what film does, known as "shoulder".
Imagine 1-bit codec. The value can only be 1 or 0. Now, this doesn't mean much, unless we know which brightness levels that "1" corresponds to.
In general case, the camera has to squeeze its DR (or part of it) somehow into the available bits. There are various ways to achieve that and none is perfect - one of the reasons we have so many gamma curves. Assume 8-bit codec - even linear gamma curve can squeeze more than 8 f-stops into them - all it needs to do is to 'compress' the brightness range, so that, f.ex. digital 255 represents the brightness of 12 f-stops above what's represented by digital 0. It's just the mapping.
The problem is that the more we compress, the worse the tonal resolution becomes in the mids (and shadows), which is of great importance, f.ex. for skin tones - there's risk of severe banding, when there are not enough distinct (coded) levels to smoothly represent skin gradation. That's why non-linear gamma curves are used - they assign more coding space (distinct values representing various brightness levels) to where it makes more sense, taking into account things like logarithmic tonal perception of human vision, anticipated viewing conditions, noise, gradability, being able to use standard displays for monitoring etc. F.ex, LOG-based gammas sacrifices easy monitoring in order to improve on other areas.
So, while 8-bit codec doesn't necessarily limit the DR it can represent, there's no point in squeezing too much DR into it, as severe banding will occur in post, if we're trying to stretch it to get back to the original scene DR. Of course, some compression is fine - we don't need too much tonal resolution in highlights so we can squeeze them - this is similar to what film does, known as "shoulder".
Nirv
Carbonite Member
I agree with everything Trez said.
The banding in the image is due to the quantisation of the signal due to limited bit depth. More bit depth means less quantisation and therefore greater signal to noise ratio.
I like to think of SNR as the real world dynamic range and I think that's why many people think bit depth increase dynamic range.
I am an advocate of having camera profiles that match the specific lighting condition so that the very best image signal to noise is captured. As you rightly point out there are limitations as there can be compromises that have to be made but by understanding the effect of a particular profile (or gamma curve) on the scene these limitations can be minimised. It's really important that you give your footage the best chance by capturing the maximum available information by utilising the very best dynamic range or signal to noise ratio, however people want to express it..
Using cinemgammas and Slogs in reality means you're more likely to have to grade the footage or do secondary adjustments to skin etc.
Having said that, Sony have done a tremendous jobs of the AVCHD Codec as I barely notice that I'm grading 8 bit highly compressed material. I am absolutely loving this camera.
PS. Since purchasing the camera I have caused the UK weather to go to rats, apologies fellow UK citizens. Shooting was cancelled today but the Leveson enq. is proving to be great watching....
The banding in the image is due to the quantisation of the signal due to limited bit depth. More bit depth means less quantisation and therefore greater signal to noise ratio.
I like to think of SNR as the real world dynamic range and I think that's why many people think bit depth increase dynamic range.
I am an advocate of having camera profiles that match the specific lighting condition so that the very best image signal to noise is captured. As you rightly point out there are limitations as there can be compromises that have to be made but by understanding the effect of a particular profile (or gamma curve) on the scene these limitations can be minimised. It's really important that you give your footage the best chance by capturing the maximum available information by utilising the very best dynamic range or signal to noise ratio, however people want to express it..
Using cinemgammas and Slogs in reality means you're more likely to have to grade the footage or do secondary adjustments to skin etc.
Having said that, Sony have done a tremendous jobs of the AVCHD Codec as I barely notice that I'm grading 8 bit highly compressed material. I am absolutely loving this camera.
PS. Since purchasing the camera I have caused the UK weather to go to rats, apologies fellow UK citizens. Shooting was cancelled today but the Leveson enq. is proving to be great watching....
plasticam
DVXuser Sponsor
Trez,
Thanks for that enlightening post. I am not an engineer and have tried for many years to really understand gamma and Log E.
I've glommed on to about 85% of it and the rest is a constant struggle. I believe that our eye sees about 100 stops of scene luminance in an exterior day. Thats hard to measure with a spot meter but a blue sky is often over F64 and you have to place that F64 onto a sliding 10 stop scale and near an F16 if your skin tone is 5.6. Want more detail in the sky? slide your scale toward it and as soon as you do - start losing your precious mids. The contrast gangs up fast. So you end up working with a sliding scale you place over your target F stop- with blatant trade offs if you move it more than a stop either way.
I'm now placing the same sliding scale over my target stop (a credit to HD sensors) with one fatal flaw- the gray scale - my mids have taken a hit and I have to really work with "profiles" to stretch the middle part of the curve into submission.
That's why I read your post with such interest. I don't have a process yet to due this PP stretch. I sit and mash buttons until a test scene matches my eye the best- then I compare this to profiles I find online. I have been going with my own, but am trying to learn more about what is under the hood. I don't know what crispening is and to tell you the truth don't bother to read the thread. I care about my shots but dont want to become an engineer- I want to shoot not program an orbital trajectory. I have to sit down soon and really go thru Frank's G-log a chore I've been avoiding, like cleaning the garage.
I guess what started all this is DR and Db (signal to noise). These are linked in my mind because one, DR is quantitative, but the useability of it (noise) is a qualitative response. Larry Thorpe suggests that and ideal camera db and one very hard to achieve until recently was around 63. Manufacturers seldom publish this stat. I wish they would as it says so much about usable sensitivity.
Thanks for that enlightening post. I am not an engineer and have tried for many years to really understand gamma and Log E.
I've glommed on to about 85% of it and the rest is a constant struggle. I believe that our eye sees about 100 stops of scene luminance in an exterior day. Thats hard to measure with a spot meter but a blue sky is often over F64 and you have to place that F64 onto a sliding 10 stop scale and near an F16 if your skin tone is 5.6. Want more detail in the sky? slide your scale toward it and as soon as you do - start losing your precious mids. The contrast gangs up fast. So you end up working with a sliding scale you place over your target F stop- with blatant trade offs if you move it more than a stop either way.
I'm now placing the same sliding scale over my target stop (a credit to HD sensors) with one fatal flaw- the gray scale - my mids have taken a hit and I have to really work with "profiles" to stretch the middle part of the curve into submission.
That's why I read your post with such interest. I don't have a process yet to due this PP stretch. I sit and mash buttons until a test scene matches my eye the best- then I compare this to profiles I find online. I have been going with my own, but am trying to learn more about what is under the hood. I don't know what crispening is and to tell you the truth don't bother to read the thread. I care about my shots but dont want to become an engineer- I want to shoot not program an orbital trajectory. I have to sit down soon and really go thru Frank's G-log a chore I've been avoiding, like cleaning the garage.
I guess what started all this is DR and Db (signal to noise). These are linked in my mind because one, DR is quantitative, but the useability of it (noise) is a qualitative response. Larry Thorpe suggests that and ideal camera db and one very hard to achieve until recently was around 63. Manufacturers seldom publish this stat. I wish they would as it says so much about usable sensitivity.
Chris Johnston
Veteran
yea, and he keeps saying 2K alot too.
vincent thomas
Well-known member
Frank
I suppose you have seen the still HD frame from the Black Magic Camera. How do you compare it to the Fs700? DR lloks superior for the second one, what about now in term of sharpness and micro film detail resolution?
what is your feedback about this?
I suppose you have seen the still HD frame from the Black Magic Camera. How do you compare it to the Fs700? DR lloks superior for the second one, what about now in term of sharpness and micro film detail resolution?
what is your feedback about this?
Postmaster
Dark Side of the Camera
Both cameras where pre-production models though, so it´s hard to say anything waterproof.
Sharpness/details should be in the same ballpark. The 700 downsamples from 4k but records compressed, the BMCC has less photosites, but records raw.
Regarding DR - the ProRes recording of the BMCC shold be similar than recording the 700 with a Samurai.
When Sony releases the raw update, the 700 should be better, cause of the larger sensor and I trust Sony a bit more, regarding sensor and signal VooDoo.
We will see.
Frank
Sharpness/details should be in the same ballpark. The 700 downsamples from 4k but records compressed, the BMCC has less photosites, but records raw.
Regarding DR - the ProRes recording of the BMCC shold be similar than recording the 700 with a Samurai.
When Sony releases the raw update, the 700 should be better, cause of the larger sensor and I trust Sony a bit more, regarding sensor and signal VooDoo.
We will see.
Frank