Makten wrote:
F/0.9 on APS-C equals f/1.2 on FF, both for DOF and collecting of light per time. I really don't get why this is so hard to understand and thus comes up every time different formats are compared and discussed.
A sensor with the area X will have to get twice the exposure compared to a sensor with the area 2X, to get the same noise per image height. Forget about ISO. What you want is photons.
You did read exactly what I wrote, didn't you? I suggested that one needed to be careful about the nature of the comparison that one was making. I hope that you do not disagree with me about the image brightness statement, or about anything else I wrote. If so, exactly what do you disagree with? (I must admit that I have no idea why I wrote the 'only applies' bit because I do understand the total flux argument)
douglasf13 wrote:
I think it's pretty safe to assume that one is talking about like sensor technology when considering lens equivalency.
That is just it -- this "light scooping advantage" is so often cited as a FF coup de grace, regardless of the sensor techs involved, so that somehow f1.2 on APS-C isn't "really" f1.2.
In photons-per-unit-time world, that argument is true, but I say this is non-sensical point to argue in the first place.
F1.2 and the resultant shutter speed is f1.2, regardless of the sensor involved, which is ultimately what we are concerned with in this "light" debate, as is ultimately allowing lower ISO, etc.
DOF equivalancy of course another topic but not really contended.
Mescalamba wrote:
Ive sent this suggestion to Metabones about 10 minutes after I saw their Speedbooster. They said they will look into it. For 645 lens they could probably just increase size of whole thing as x0.71 would work for 645 to FF perfectly. Which would make from lets say Mamiya 80mm f1.9 - 55mm f1.3 (or 1.2). And that would be quite interesting..
Tho it depends if they will really make it or not..
+1
I hope they will lunch this adapter soon, then people can save money from buying digital-back.
cputeq wrote:
It's not hard to understand at all. The reason this stupid argument comes up is because people are arguing two different things when they bring this up, and unfortunately one of those things is completely irrelevant to the discussion when practically applied to photography, however technically correct it may be (with certain assumptions). Because of the difference in "vocabulary", this argument perpetuates.
The first "thing" one could be arguing is referring to resultant shutter speeds from various apertures. This is a relevant argument. Shutter speed is a known parameter as displayed by the camera and can be altered by the photographer through a variety of means, one of which is to alter the aperture of a lens.
The second "thing" is this completely irrelevant "photos per unit time" and its association with differing sensor sizes and performances. I say irrelevant because there are a hell of a lot assumptions made when talking this point (and the assumed conclusion, then, that the FF sensor is always better) ...Show more →
To be honest, the argument perpetuates not because people here don't understand there are two "things", but because people insist one of the two is irrelevant. I understand the difference between the two concepts, and they both come with their own assumptions in order to be meaningful in a given usage.
HelenB wrote:
You did read exactly what I wrote, didn't you? I suggested that one needed to be careful about the nature of the comparison that one was making. I hope that you do not disagree with me about the image brightness statement, or about anything else I wrote. If so, exactly what do you disagree with? (I must admit that I have no idea why I wrote the 'only applies' bit because I do understand the total flux argument)
With half the sensor size, the signal must be amplified twice as much at a certain ISO speed, if the pixel count is the same. With the same pixel size and half the amount of pixels, you also only get half the "signal" of course. Image brightness doesn't tell anything about what happened to the photons after the image was shot.
cputeq wrote:
F1.2 and the resultant shutter speed is f1.2, regardless of the sensor involved, which is ultimately what we are concerned with in this "light" debate, as is ultimately allowing lower ISO, etc.
Why are you trying to ignore physics? What we should be concerned about is how much light was collected, since it builds our image, and that depends just as much on sensor size as on aperture. Twice the sensor size allows twice the ISO to get the same noise, because the same amount of photons were collected per time.
I really don't understand why this isn't common knowledge. It should be perfectly understandable for anyone, even without education. Put a bucket out in the rain and get X liters of water per hour. Now put TWO buckets there, and get 2X liters per hour. At THE SAME bucket size (aperture).
Makten wrote:
I really don't understand why this isn't common knowledge. It should be perfectly understandable for anyone, even without education. Put a bucket out in the rain and get X liters of water per hour. Now put TWO buckets there, and get 2X liters per hour. At THE SAME bucket size (aperture).
On the theorethical level, I think many people have not yet understood that the reason you want to collect much water, is that the noise is in the water (in the random arrival of drops) before it even hits the buckets. I believe many people think you need much water because the methods to measure the amount of water is imperfect.
Photon shot noise.
The noise is in the light before it even hits the sensor
On the practical level (The 2 stop ISO advantage the larger sensor has in practical use), I think it is a combination of:
1) Some people go from an old FF sensor to a modern u4/3 sensor (typically 5D or 1DsII to OM-D). That is a technically valid argument, as the photon shot noise after recording by a sensor is the product of (total amount of light into the sensor X sensor efficiency). Try comparing the 6D and the OM-D and the two stop difference in noise performance is there (save for some fraction of a stop).
2) Some people openly admit that they can tolerate more noise when they use a smaller camera, or that as long as the noise is below some level of tolerance, they don't care if there is a two stop difference. This is of course not a technically nor logically valid argument.
alundeb wrote:
On the theorethical level, I think many people have not yet understood that the reason you want to collect much water, is that the noise is in the water (in the random arrival of drops) before it even hits the buckets.
This is only part of the truth. There is also noise, lets call it dirt, in the buckets (pixels).
The more water you collect, the less the dirt matters, no matter where it came from.
Makten wrote:
This is only part of the truth. There is also noise, lets call it dirt, in the buckets (pixels).
The more water you collect, the less the dirt matters, no matter where it came from.
Yes, but with modern sensors the dirt is under control up to about ISO 25600..
Makten wrote:
Put a bucket out in the rain and get X liters of water per hour. Now put TWO buckets there, and get 2X liters per hour. At THE SAME bucket size (aperture).
thats bizarre
I was trying to explain to a friend how exposure works with the aperture and shutter speed and thats the example I came up with... bucket, tap and water
alwang wrote:
To be honest, the argument perpetuates not because people here don't understand there are two "things", but because people insist one of the two is irrelevant. I understand the difference between the two concepts, and they both come with their own assumptions in order to be meaningful in a given usage.
Aren't there more "things" we are discussing?
- Focal length equivalence: 50mm on MFT = 100mm on FF. This one is being forgotten a bit in this thread.
- Exposure equivalence: f/2 on MFT = f/2 on FF
- DoF equivalence: f/2 on MFT = f/4 on FF.
- Noise equivalence (a more theoretical comparison, due to sensor differences): f2 on MFT = f/4 on FF.
Personally, for my use the last one is irrelevant, the IBIS of the E-M5 more than takes care of getting the same performance for the smaller sensor. I am mostly concerned with #2, and this is the one which Metabones is talking about too, not #4, if I understand correctly.
Spyro P. wrote:
I was trying to explain to a friend how exposure works with the aperture and shutter speed and thats the example I came up with... bucket, tap and water
That is a very good example of how common exposure analogies don't include photon shot noise.
Put a bucket under a tap with a certain opening for a ceration amount of time. Do it again and you get the same amount of water in your buckets.
Put a bucket under a constant rainfall for a certain amount of time. It collects say 100 real droplets of water.
Do it again under exactly the same rain intensity for exactly the same amount of time, and It will collect maybe 90, maybe 105 droplets of water.
carstenw wrote:
Aren't there more "things" we are discussing?
- Focal length equivalence: 50mm on MFT = 100mm on FF. This one is being forgotten a bit in this thread.
- Exposure equivalence: f/2 on MFT = f/2 on FF
- DoF equivalence: f/2 on MFT = f/4 on FF.
- Noise equivalence (a more theoretical comparison, due to sensor differences): f2 on MFT = f/4 on FF.
Personally, for my use the last one is irrelevant, the IBIS of the E-M5 more than takes care of getting the same performance for the smaller sensor. I am mostly concerned with #2, and this is the one which Metabones is talking about too, not #4, if I understand correctly....Show more →
What about #3? Meta just says one stop gain in aperture, is the DoF smaller now too?
My gut feel is that the adapter adds aberrations that are comparable to the types of aberrations you get when using a faster, wider lens without the adapter, both mounted on the APS-C sensor.
I would say more. They don't have the ability to design a unified system, which lens designers do. They just design a universal device which doesn't know anything about the lens being attached.
carstenw wrote:
I would say more. They don't have the ability to design a unified system, which lens designers do. They just design a universal device which doesn't know anything about the lens being attached.
If so, we are down to avaliablity of lenses, and for NEX users with a suitable FF lens collection it will open up some possibilities right away.
Except for the gut feeling, we can know a lot already. Mind this is an ages old concept. Rollei already experimented with a twin lens reflex model for 120 and 135 film in the same body where the upper lens had "speedbooster" type of lens design. Remember optical engineering know-how tells us:
- compatibility of the primary system (the lens mounted) is an important factor. The converter / speed booster works with the light gathered and image rendered through the lens that is mounted.
Highly corrected slower lenses will be a better companion than fast lenses with residual aberrations. By the nature of design this type of condensor probably works best with (short) tele lenses.
- The edges and corners of full frame lenses are included. A lot of excitement about alt glass on APS-C and smaller sensors comes from the fact that the worst part of the image quality is left out. That great old manual focus glass may look less attractive on a "full frame" APS-C setup (is this a contradictio in terminis or does the speedbooster deserve that title ?
From the perspective of optical engineering it is interesting how issues will be resolved like correction of:
- field curvature, common in fast wide angles
- distortion, adding a strong condensing set of lenses to an exististing system will almost certainly introduce more distortion. If the primary system already suffers from some slight distortion aberrations the end result could be a complex and hard to correct moustache type of distortion
- chromatic aberrations, if the primary system is not APO corrected (which most fast lenses are not) this is going to be amplified by the condensor effect of the speed booster. Even if the primary lens is not suffering from chromatic aberrations the speedbooster will probably introduce some colour fringing.
- AF speed and precision, maybe not that important for all those manual focussing alt glass lovers here, but the way the speed booster works will have effects on AF. Nothing to worry about, we accept that with teleconverters as well, but it's realistic to set expectations right.
- Something that I look forward to see, is how this speedbooster effects bokeh quality. To my taste, in general tele converters have a destructive effect on bokeh quality. I am afraid this won't be different for the speedbooster and that the design leaves the bokeh quality of the original lens intact. But the fact that lens design becomes decentered for the position of the aperture (an important factor for bokeh quality) is not very promising. I hope my suspicion is wrong, since the smaller DOF which hopefully comes in a nice bokeh quality, should be one of the selling points of the speedbooster.
The enthusiasm shown for this metabones speedbooster so far is mostly from videographers. As long as they work in the 720P and 1080 resolution area, and not in the 4K class, this speed booster is a nice additional add on to their setup. Still photographers, especially those here at the boards of Fred Miranda and even more so in the alt section, have higher quality standards than average. I am afraid a few, now excited about the concept, will be desillusioned soon by the reality standards and live results from the speedbooster for their photographic applications. There aren't a lot enthusiasts about the optical results of teleconverters among the high standards photographers. I guess they won't be over enthusiast about a speedbooster either. It's definitely an interesting creative tool that will be able to get you shots that you couldn't without the speedbooster and your current lens setup, but by the nature of optial design it won't come without compromises.
So, not having any real knowledge in mathematics or optics, my simple understanding of the device is the following:
- by "bundeling" the light beams that the FF lens projects, the actual full image is projected onto the APS-c (roughly) surface:
- By the projection the image is actually scaled down. It still contains all the information, but smaller and (like sun through a loupe) brighter. Everything is by the way smaller, including abberations (if new ones are not introduced), unsharpness in the tiniest details. The "scaled down" image contains the whole angle of view that the FF lens originally sees, as opposed to the 1,5/1,6x crop in a normal APS- DSLR. Instead of the cropped part of the image the whole FF image is sent to the sensor, thus scaled (down).
- It is the opposite - as many have said - of a teleconverter which magnifies everything, also the weaknesses of the lens. This makes the opposite, it makes the weaknesses smaller.
- on the other hand: As opposed to the normal case where with FF lenses on Dx sensors the outer parts of the lenses are left out, and you'll only use the sweet spot, here you use (almost) the whole image circle that the FF lens draws, including the weaker outer parts.
- About DOF I am not so sure, but I assume you'll get nearly the same DOF as in a FF camera. Which is what the samples suggest....
Did I get something wrong here?
In the end speculation about how the adapter works in practice will not tell us how it actually will perform. I personally assume much better than many think, the concept is original and I am very surprised that noone has come up with such a device since APS DSLRs have become the mainstream and sell in millons.
There are obviously some creative thinkers at Metabones and I wish them a huge success with this interesting product.
What I wonder is: How come it is so expensive: There is one optical element, right?
eosfun wrote:
- chromatic aberrations, if the primary system is not APO corrected (which most fast lenses are not) this is going to be amplified by the condensor effect of the speed booster. Even if the primary lens is not suffering from chromatic aberrations the speedbooster will probably introduce some colour fringing.
Can you elaborat a bit on this? How will the condenser amplify chromatic aberrations at the same time as it compresses the resolved detail?
eosfun wrote:
There aren't a lot enthusiasts about the optical results of teleconverters among the high standards photographers. I guess they won't be over enthusiast about a speedbooster either. It's definitely an interesting creative tool that will be able to get you shots that you couldn't without the speedbooster and your current lens setup, but by the nature of optial design it won't come without compromises.
The latest generation of teleconverters from Canon leave little to be desired regarding image quality. But I haven't seen satisfactory results from teleconverters of any other brand.
Anyway, I can afford to be an early adopter on this, and I don't expect FF quality
They said they will look into it. For 645 lens they could probably just increase size of whole thing as x0.71 would work for 645 to FF perfectly. Which would make from lets say Mamiya 80mm f1.9 - 55mm f1.3 (or 1.2). And that would be quite interesting.. 
?