steve phillipps
Veteran
So if you put a b4 2/3" lens onto the HDx35 and then mounted the HDx35 on a 2/3" camera there would be no light loss and no magnification?
Steve
Steve
F3: HD35 MK2 adaptor review by SPARKY FILM
- Thread starter Sparky874
- Start date
I assume you mean "....mounted the HDx35 on a s35 camera", in which case the answer is "yes".
If both cameras were (say) single chip and (say) a 1920x1080 sensor, you shouldn't be able to tell the results apart, assuming the lens settings stay the same. (Iris, zoom, focus)
That applies to sensitivity, angle of view, and depth of field. It's not intuitive but it's down to what Joshua said about "aperture equivalence".
As said before, practically there may be a sensitivity difference of about one stop - but that's down to single versus three chip and light loss in the single sensor filtration - *NOT* any light loss in the adaptor.
steve phillipps
Veteran
No, I meant what I said - take the PL mount off the back of the HDx35, bolt on a b4 and stick the whole combo onto an F900 - if all it does is spread the light wider and not lose any light you would get no light loss and no magnification? So the lens at 7.5mm and f1.8 without the adapter would still be 7.5mm and f1.8 with it?
Steve
Steve
steve phillipps
Veteran
I don't think anyone is suggesting that the light is lost in a particle physics way, as in disappears into thin air or is transformed into dark matter! Just that there is less intensity of light reaching the sensor.
Steve
Last edited:
steve phillipps
Veteran
...and HardDrive don't think I'm doubting you - you're obviously more technically expert than me, I just don't get what you're driving at :happy:
Steve
Steve
steve phillipps
Veteran
...and I think it's probably just gonna confuse people (including me!) with the over-complication.
As far as I understand, correct me if I'm wrong, the upshot is if I put a Nikon 300mm f2.8 lens on an F55 at f2.8 and set correct expsoure, then put a b4 zoom, say an HJ22x7.5 that was f2.8 (hypothetically) onto the HDx35 and then onto the F55 wide open I would need to add 2.6 stops more exposure, either by gain, shutter or whatever to get correct exposure. Correct? Or not? I thought I knew what the answer should be, but at this precise moment I genuinely don't know!
Steve
As far as I understand, correct me if I'm wrong, the upshot is if I put a Nikon 300mm f2.8 lens on an F55 at f2.8 and set correct expsoure, then put a b4 zoom, say an HJ22x7.5 that was f2.8 (hypothetically) onto the HDx35 and then onto the F55 wide open I would need to add 2.6 stops more exposure, either by gain, shutter or whatever to get correct exposure. Correct? Or not? I thought I knew what the answer should be, but at this precise moment I genuinely don't know!
Steve
No - because what you are describing there is a teleconvertor, what happens when you put the doubler in for a normal lens, if you like.
So you do get magnification, a narrower angle of view - and yes, real light loss. (At least as far as the sensor is concerned.)
Look again at Joshua's post #38. In principle, the difference between using a device like this as an adaptor and a teleconvertor is solely down to sensor size. All either device really does is spread a given image out by about 2.5x here, or about 6x the area. If it's classed as an adaptor, the new sensor size is also 6x the area - so same angle of view, no lost light. If it's classed as a teleconvertor, the "new" sensor size is the same as the lens is designed for - so only 1/6 of the area of the new enlarged image. Result is magnification, a change in angle of view, and only 1/6 of the original light now impacting on the sensor.
By "lost", there's no need to think of dark matter
- just absorbed in ND, Bayer filtration or simply falling outside the sensor area is what we're talking about. Anything that means it's not usable for the purpose of creating the image. In the case of an adaptor, all the photons that form the original image in the 2/3" camera also form a (bigger) *and fully useable* image in the s35 camera. In the case of a teleconvertor, only 1/6 of those photons will go towards image formation, the rest will be wasted ("lost") around the edges of the sensor.Look at it this way. For any given photosite with the lens on a 2/3" camera, let's say X photons fall on it, and it's area is Y. With the "all else equal" caveat, in the adaptor/s35 case, the image will have 1/6 the brightness - but each photosite will be 6x the area. Hence each photosite will still have the same X photons falling on it so should give similar performance.
steve phillipps
Veteran
But no matter what you put it on it's the same bit of glass so it must be having the same effect in terms of light intensity - it doesn't magickly change from a teleconverter to a format converter when it detects what camera it's put onto!
steve phillipps
Veteran
I think that's where the confusion lies - playing off the larger photosites against the light loss / drop or whatever caused by the adapter. And each photosite will only have 6x the area if the sensors have the same resolution of course - it's misleading and unneccesarily confusing.
Forgetting about photosite size then, putting the adapter between lens and camera does actually reduce the intensity of light passing through yes? So if I put a Nikon 300mm f2.8 with a Universal b4 mount onto the HDx35 and then onto an F55 or an F900 I'd be 2.6 stops darker than if I put the naked lens on it? Please say yes or I'm going to go have to go on an optics course!
Steve
That is correct, and the reasoning is that one effect of the HDx35 usage is to make the f stop of the lens effectively be 2.6 approx. stops smaller. But you shouldn't ONLY think in terms of f stop, and that example may not be very helpful in terms of understanding the principle.
Let's assume your 300mm f2.8 lens is purpose designed to cover the F55 sensor for a given angle of view. It follows by definition that the diameter of the front element is going to be 300/2.8, or about 107mm, yes?
To make an equivalent lens - same angle of view - for a 2/3" camera, we'll need one whose focal length is 300/2.6 or about 115mm. Now if we want that also to be f2.8, that must then have a front element of about 115/2.8 or about 41mm. It goes without saying that such will be far smaller and lighter than the first version.
So, we now have two cameras, with lenses giving the same angle of view, and both of them the same aperture - f2.8. But one being far smaller and lighter than the other. So which do you think will behave best in low light - they're both f2.8, after all, aren't they? Or do you think their performance will be the same - if we assume basic technology, resolution etc is otherwise the same in each case?
I'll let you think about it until I finish work tomorrow!
Yes! But it's how the physics work I'm afraid. What IS misleading is when reference is made to the change in f stop number without thinking about the implications of the bigger sensor, bigger photosites etc. That and talk of "light loss" leads many to believe the performance of these convertors in sensitivity terms is worse than they actually are.
Ah-ha! But it DOES do the change. Not by magic, but by the Laws of Physics and *the size of the sensor*. That's the determinant.
It all comes down to the difference between luminous flux and luminance, in technical terms.
Yes, it's the same bit of glass, yes, it must be having the same effect in terms of light intensity (or luminance) but the total amount of light that gets used for image formation will be the intensity multiplied by the sensor area, and that's what important.
Mitch Gross
Veteran
I believe that when I wrote the original literature for the HDx35, I used the phrase "effective light loss". I still stand by that.
All else is counting how many angels can dance on the head of a pin.
All else is counting how many angels can dance on the head of a pin.
steve phillipps
Veteran
I think that's exactly the point - this is a film-making website and not a physics one, so for 99% of people here the science, semantics and pendantry of it is utterly irrelevant.
What is important it to know that if you're shooting an F55 in a low light and your light meter tells you that you need f1.8 for correct exposure. If you have a Nikon 50mm f1.8 in PL mount and a Canon HJ22 also f1.8 but needing the HDx35 to mount and cover - you know that the Nikon will be OK, but the Canon WON'T, it will be 2.6 stops underexposed. That's all that matters, whether you call it light loss or whatever, you have lost 2.6 stops worth of your exposure. To keep saying that there is no actual "light loss" gives the impression (to those who don't know or care about the science of it) that the HJ22 f1.8 will not lose light and therefore by f1.8 when used with the HDx35.
Sorted - we got to the bottom of it eventually!
Steve
For a day to day operational point of view, what you say is fair enough, and yes, all the underlying physics may not be necessary.
But people come to websites such as this to help understand basic theory, and to learn about new kit and technology to better help with future buying, renting etc decisions.
And with relevance to that I've now heard too many times the general use of such adaptors dismissed through completely fallacious arguments. Typical might be someone saying they want to be able to use existing B4 lenses on a new camera purchase. The suggestion of going s35 and an adaptor just gets dismissed out of hand due to "light loss" - the implication is that it's a B4 lens, it'll perform far worse (in low light) on a s35 camera, so for that reason alone they really need to get 2/3" cameras.
Which isn't true. What Nate said many posts ago about the use on a F55 may be accurate - but the performance hit he talks about comes from factors like 4k versus HD, and single versus three chip - not the adaptor itself, but the latter is what it generally gets attributed to. And it's not being pedantic to say that is just plain wrong.
If a 4k 2/3" camera was to be available, I'm sure purchases would be made on the (incorrect) basis of low light capability, of using B4 lenses without the f stop reduction of the adaptor. In which case, they'd be in for a nasty shock. Yes, the f stop would be an unchanged f1.8 - but the user may find that the camera can then indeed only be rated at something like the ISO 200 Nate speaks about for equivalent performance to his F55 at ISO 1250 - that would be the consequence of the much smaller photosites of a 4k 2/3" chip. And it doesn't matter whether you keep f1.8 and drop to ISO200, or keep ISO1250 and drop the f stop rating. In which case, with no disadvantage to the adaptor, such a purchaser may wish they had gone the s35 route and got the flexibility of using fast primes or short zooms when appropriate.
And that's why it's important to get this properly understood, at least the full implications if not all the underlying physics. (I'd have been happy to have left it solely at what I put in post #28, which really sums up all most people here really need to know.)
But hey, if anybody thinks all this is just being pedantic, if anybody is hesitant about going the s35/adaptor route "because adaptors lose light", that's up to them. I can only say I've tried to explain.