720p vs 1080i
- Thread starter GrahamH
- Start date
BobDiaz
The Professor
I recall that in the 1080p mode, Barry tested the camera and found it to be around 800+ to 900+ lines, depending on how one wants to read the results.
I didn't do the test correctly, but calculated that if I had, I should have seen 650-->700 or so lines of resolution in the 720p mode.
Resolution testing is not exact and the results vary depending on how one wants to read the high end.
Keep in mind that it's impossible to resolve beyond the number of pixels and whatever result you get, it's going to be less than the number of pixels. Thus, even with the top of the line 2/3" camera and the best lens out there, one can't resolve 720 lines or more in the 720p mode. I believe that even with the best equipment, around 650 to 700 lines is the top end for 720p mode.
On the other hand, because the 1080 mode has more pixels, if the camera can resolve higher, it will show the extra detail. In the case of the hMC40, yes it does show more detail.
The real debate will be over is the extra detail that big a difference. :tongue:
Bob Diaz
I didn't do the test correctly, but calculated that if I had, I should have seen 650-->700 or so lines of resolution in the 720p mode.
Resolution testing is not exact and the results vary depending on how one wants to read the high end.
Keep in mind that it's impossible to resolve beyond the number of pixels and whatever result you get, it's going to be less than the number of pixels. Thus, even with the top of the line 2/3" camera and the best lens out there, one can't resolve 720 lines or more in the 720p mode. I believe that even with the best equipment, around 650 to 700 lines is the top end for 720p mode.
On the other hand, because the 1080 mode has more pixels, if the camera can resolve higher, it will show the extra detail. In the case of the hMC40, yes it does show more detail.
The real debate will be over is the extra detail that big a difference. :tongue:
Bob Diaz
Bic Parker
Active member
I did the math on this and basically came up with the notion that the fine resolution of the HMC40 really isn't going to be limited by its sensors (and pixel resolution), but rather, the optics. The calculations below show that the image scale of the sensors is well below the resolution requirements of HD, so that really only leaves the lens and atmospheric conditions as the limiting factors. So it is a bit of testament to the lens and matching it against the sensors for its finer resolution.
The resolution against the pixels, technically speaking, will be limited by the aperture and focal ratio (which can be reduced to focal length as a practical matter, for a camera). The correct way to get to that is to measure the image scale, which requires knowing the pixel size, which we could estimate from Panasonic's specs at the end of the book o' rules (the operator's manual).
Each sensor has ~3.05 million pixels on a ~1/4.1" square (I love converting metric to english and back!). So 1,747 (the square root of 3.05 million rounded up for conservative measure) divided into 6.1722 mm gives you ~0.0035mm (3.5 microns), which sounds about right for a chip that size with that many pixels (interestingly enough, this also gives a ballpark explanation of the well depth of the pixels, which goes to in part, not including dark current and quantum efficiency, its light sensitivity).
We use the image scale formula (this is a composite formula, btw, which uses a fudge constant in the numerator to incorporate various apertures and focal ratios to reduce it down to focal length dependency): (205x 3.5 microns)/4mm = 179.375 arc seconds resolution per pixel at the lowest/widest zoom (4mm). At 48mm, the most extended zoom, it will be 14.9, say 15 arc seconds of resolution per pixel.
Now you can convert those angles to the line resolution chart and the distance it is from the lens to calculate the line resolution, but the image scales will be the angular resolution at any distance.
But, I will tell you that 179 arc seconds is a fairly small number, i.e., 3 arc minutes or approximately 1/3,600 the field of view HMC40 with a standard lens). That is certainly well below the lines and resolution definitions for HD, so any additional detail perceived will mean that the camera is resolving closer to the limits of the lens.
The resolution against the pixels, technically speaking, will be limited by the aperture and focal ratio (which can be reduced to focal length as a practical matter, for a camera). The correct way to get to that is to measure the image scale, which requires knowing the pixel size, which we could estimate from Panasonic's specs at the end of the book o' rules (the operator's manual).
Each sensor has ~3.05 million pixels on a ~1/4.1" square (I love converting metric to english and back!). So 1,747 (the square root of 3.05 million rounded up for conservative measure) divided into 6.1722 mm gives you ~0.0035mm (3.5 microns), which sounds about right for a chip that size with that many pixels (interestingly enough, this also gives a ballpark explanation of the well depth of the pixels, which goes to in part, not including dark current and quantum efficiency, its light sensitivity).
We use the image scale formula (this is a composite formula, btw, which uses a fudge constant in the numerator to incorporate various apertures and focal ratios to reduce it down to focal length dependency): (205x 3.5 microns)/4mm = 179.375 arc seconds resolution per pixel at the lowest/widest zoom (4mm). At 48mm, the most extended zoom, it will be 14.9, say 15 arc seconds of resolution per pixel.
Now you can convert those angles to the line resolution chart and the distance it is from the lens to calculate the line resolution, but the image scales will be the angular resolution at any distance.
But, I will tell you that 179 arc seconds is a fairly small number, i.e., 3 arc minutes or approximately 1/3,600 the field of view HMC40 with a standard lens). That is certainly well below the lines and resolution definitions for HD, so any additional detail perceived will mean that the camera is resolving closer to the limits of the lens.
GrahamH
Well-known member
Whew! Thanks, I think.
So this bit - "I did the math on this and basically came up with the notion that the fine resolution of the HMC40 really isn't going to be limited by its sensors (and pixel resolution), but rather, the optics" - was the thought that prompted my question. Whether the HMC40 can really use all that extra pixel goodness that 1080 provides?
Obviously I should just shoot footage and decide for myself, but it gets a bit complicated since the 1080 will be interlaced, and since when watching on a 720 or 1080 HDTV one size or the other will be getting rescaled .
So this bit - "I did the math on this and basically came up with the notion that the fine resolution of the HMC40 really isn't going to be limited by its sensors (and pixel resolution), but rather, the optics" - was the thought that prompted my question. Whether the HMC40 can really use all that extra pixel goodness that 1080 provides?
Obviously I should just shoot footage and decide for myself, but it gets a bit complicated since the 1080 will be interlaced, and since when watching on a 720 or 1080 HDTV one size or the other will be getting rescaled .
Bic Parker
Active member
So, the answer is, quantitatively, no, the HMC40 can't really use all that extra pixel goodness to the pixel limit, but this is actually a good thing. Since the you have a image scale smaller than your useable resolution, but you are shooting at the larger image scale, you are effectively binning (3x pixels for each point), or at least oversampling the pixels in use. This has a few benefits:
1) It increases the light sensitivity a bit, and probably makes the higher gain settings less objectionable, in other words, it gives a higher signal to noise ratio with an effectively higher well depth (something Robert W. has noted before, and something I have observed with mine, as well).
2) The image will be less affected by the quality of the sensor. All CCD and CMOS sensors have some imperfections. They are, in fact, rated by their quality. These imperfections include hot pixels, dead pixels, and column errors. Only chips like those used in the Hubble or Keck instruments will have class 0 errors (a very very few, if any). These imperfections are effectively "cleaned" by the redundancy of the system.
3) You get a better dynamic range. Again, this goes back to the binning and oversampling.
So, in the end, the HMC40 does use the abundance of its pixels to maximize the quality of its images against the optics and the required resolutions of HD.
1) It increases the light sensitivity a bit, and probably makes the higher gain settings less objectionable, in other words, it gives a higher signal to noise ratio with an effectively higher well depth (something Robert W. has noted before, and something I have observed with mine, as well).
2) The image will be less affected by the quality of the sensor. All CCD and CMOS sensors have some imperfections. They are, in fact, rated by their quality. These imperfections include hot pixels, dead pixels, and column errors. Only chips like those used in the Hubble or Keck instruments will have class 0 errors (a very very few, if any). These imperfections are effectively "cleaned" by the redundancy of the system.
3) You get a better dynamic range. Again, this goes back to the binning and oversampling.
So, in the end, the HMC40 does use the abundance of its pixels to maximize the quality of its images against the optics and the required resolutions of HD.
Bic Parker
Active member
Yes, they would for the reason you just said.
This is of food for thought when adjusting settings and setting up the camera (and a bit of argument against using automatic settings where possible).
There are some things you should be able to really push with this camera, such as zoom, or if you get a tele-extender. You should also have a pretty good histogram to work with in post if you have to make adjustments. This also means that you can push some settings in post without degrading the image significantly (which can be good if you are adding an effect).
Consider the mounting. Stable mounting and using the OIS to its advantage (I found there are tricks to that) are especially critical with this camera. With smaller pixels (and smaller image scale), the overall signal and image quality will degrade more significantly than with a comparable larger chip/larger pixels camera with unstable or shaky images. For handheld shooting, this makes a strong argument for using steadying mounts (steadicam, glidecam, etc.) when possible, or at the least, working hand held shots from stable positions. For static mounting, it is worth the investment to get the most stable, fluid head you can.
This is of food for thought when adjusting settings and setting up the camera (and a bit of argument against using automatic settings where possible).
There are some things you should be able to really push with this camera, such as zoom, or if you get a tele-extender. You should also have a pretty good histogram to work with in post if you have to make adjustments. This also means that you can push some settings in post without degrading the image significantly (which can be good if you are adding an effect).
Consider the mounting. Stable mounting and using the OIS to its advantage (I found there are tricks to that) are especially critical with this camera. With smaller pixels (and smaller image scale), the overall signal and image quality will degrade more significantly than with a comparable larger chip/larger pixels camera with unstable or shaky images. For handheld shooting, this makes a strong argument for using steadying mounts (steadicam, glidecam, etc.) when possible, or at the least, working hand held shots from stable positions. For static mounting, it is worth the investment to get the most stable, fluid head you can.
Robert M Wright
Well-known member
The short of all this is, an HMC40 can resolve more image detail than can possibly be recorded using 720 lines - and you don't need to shoot a res chart to see the difference - just shoot a scene with some real detail, in both modes, and look at it on a big 1080 line HDTV and you'll be able to see a difference.
Last edited:
Bic Parker
Active member
Something else that can be taken away from this is that you shouldn't get low quality accessories with this camera. It will be more adversely affected by bad glass (wide field adapters, tele-extenders, filters, etc.) and bad mounting... imperfections will be more easily seen.
Additionally, this also means that shooting techniques become paramount. Mistakes can become more amplified.
With great resolution comes great responsibilities
Additionally, this also means that shooting techniques become paramount. Mistakes can become more amplified.
With great resolution comes great responsibilities