Busting Speedbooster Myths
I recently came across a discussion online about how a Speedbooster changes the field of view, depth of field and exposure. There was so much misinformation and conflicting statements about how they affect your lens or camera that the users who were giving correct advice were virtually drowned out. Therefore, I hope to dispel some of these mistruths by boiling down what a speedbooster actually does and how, before busting some common Speedbooster myths at the bottom.
Update 7/6/21: Accompanying video embedded; originally uploaded 13/2/21. It covers many of the same topics as below, along with visual demonstrations.
I’ve tried to explain things as clearly as possible, which has the effect of making this article a bit of a lengthy, wordy one. If you want the TL;DR, it’s as follows: “Speedboosters make a lens wider and faster, but don’t change the depth of field”. If you want to know how or why, read on.
First thing’s first: Speedbooster® is actually a registered trademark of Metabones, the company arguably responsible for popularising them almost ten years ago with the advent of the Sony E mount and the accompanying E mount Super 35 video cameras. A Speedbooster is just a brand name for a focal length reducer: that’s what it’s proper name is. Metabones didn’t invent the focal length reducer, but in the same way we say ‘Hoover’ to refer to any vacuum cleaner, the moniker has become the defacto term for such a device in colloquial terminology. I’ll continue to call it a Speedbooster – but bear in mind I’m referring to any focal length reducer from any brand (the original online discussion mentioned above was in reference to Canon’s EF to RF focal length reducer).
Let’s start by addressing one of the most commonly misunderstood aspects of using a Speedbooster: it fundamentally changes the lens’ optical design. It’s literally adding in extra elements. With this extra glass comes a change in the lens’ focal length, but it does not change the lens’ entrance pupil.*
(*Quick note: a lens’ entrance pupil is not the same as its physical aperture opening, ie how wide the diameter of the iris blades open. It’s a related but slightly different concept, though for the purposes of this explanation, assume they’re the same and don’t worry about it.)
The final note before we begin is reminding ourselves of what an f-number is. A lens’ f-number is a basic fraction: focal length over entrance pupil, for instance a 50mm f/2 lens would have a 25mm entrance pupil (50/25=2). With that in mind, let’s cover three aspects of a lens: field of view, aperture and depth of field.
Field Of View
Let’s say we add a Speedbooster to our 50mm f/2 lens. It has a magnification of 0.71x, therefore, we can multiply the focal length by this factor: 50*0.71=35.5. The lens is now a 35.5mm lens (I’ll just say 35mm from now on for the sake of simplicity), but the entrance pupil has not changed. This bit is important to understand and is so often overlooked or misunderstood. For the rest of this article I’ll stick to using this example lens – when I say 50mm, I mean the original lens without the Speedbooster and when I say 35mm, I mean the same lens but with the speedbooster attached. (Additionally, we’re assuming the 50mm is a full frame lens and we’re ‘speedboosting’ it onto a Super 35/APS-C format camera. Speedboosters for other formats are available.)
Because the crop factor of Super 35 versus full frame is ~1.5 (this varies slightly camera to camera), a 35mm lens on a S35 camera will have (approximately) the same field of view that the original 50mm lens does on a full frame camera. 35.5*1.5=53.25.
Aperture
I’ve already explained above that the entrance pupil does not change. However, given that we’ve now reduced the focal length (from 50 to 35), the formula for calculating f-number has changed. The entrance pupil is still 25mm in diameter, so it now gives us 35/25=1.4. So a 50mm f/2 lens, speedboosted, becomes (not effectively, but actually) a 35mm f/1.4. That’s how Speedboosters ‘magically’ create an extra stop of light. To clarify, our 35mm f/1.4 will give you an exposure one stop brighter than the original 50mm f/2, whether that 50mm is used on full frame or S35.
If it helps, a Speedbooster (a focal length reducer) is the exact opposite of a focal length extender (aka teleconverter), like the Canon 1.4x or 2x. All an extender does is increase the focal length, the byproduct of which is the f-number drops and you lose one or two stops of light. This is evident to anyone who’s used one, it isn’t a hypothetical or theoretical change, it’s an actual, visible change in exposure.
One of the side effects of using a Speedbooster known to most users, is that it changes the projected image circle of the lens. A full frame lens speedboosted will reduce the image circle, enough to only cover S35. Again, compare to an extender. They’re literally magnifiers: they ‘zoom in’ on the image by literally stretching the projected image over a larger area, so that using the same capture window (ie the sensor), you’re sampling a tighter part of the image = narrower field of view. Therefore a speedbooster is just a demagnifier. It makes the projected image smaller.
With the above in mind, let’s consider that a lens lets through a certain amount of light, limited by the aperture. Our 35mm f/1.4 lets through the same number of photons as the original 50mm f/2, albeit in a smaller image circle. All this is akin to is concentrating that same amount of light over a smaller area, increasing the brightness. Extenders project the same amount of light, but over a larger area, hence, a darker image. (I like to think of an analogy from Bilbo Baggins: “I feel thin, sort of stretched, like butter scraped over too much bread.”)
Depth of Field
Depth of field seems to be another one of those topics that’s often confused. The only things that affect depth of field are focal length, aperture and distance to subject. Format size does not affect depth of field. Put it like this: the image has already been formed in the lens before it reaches the sensor. You can hold a lens close to a wall and it will project a discernible image. No matter how big the wall is, it doesn’t change the depth of field!
All of the following assumes the same distance to subject, leaving only focal length and aperture affecting depth of field. We’ll also assume the lenses are used at maximum aperture. Larger formats are associated with thinner depth of field simply because of the need to use longer focal length lenses for any desired field of view (compared to smaller formats). A speedbooster gives you the ‘full frame look’ or ‘full frame field of view’ simply because you’re using a wider lens on a smaller format as demonstrated above: a 35mm on Super 35 is about the same field of view as 50mm on full frame.
Our speedboosted lens has now widened its field of view by decreasing the focal length. With shorter focal lengths comes deeper depth of field – but, we’ve also decreased the f-number, which makes the depth of field thinner. Both of these combined cancel each other out: a 35mm f/1.4 will have the same depth of field as a 50mm f/2.
Using larger formats does not automatically make depth of field thinner. The depth of field is only thinner when comparing the same field of view and the same aperture. Whilst our 50mm f/2 (on full frame) and 35mm f/1.4 (on Super 35) share the same field of view and the same depth of field, using a 50mm f/1.4 on full frame will yield a thinner depth of field.
Our original 50mm f/2, used on any format, will give you identical depth of field. By using smaller formats, all you do is sample a smaller image and therefore a narrower field of view. You can take a photo on a full frame camera, then crop it to the same perspective as a photo taken with a smaller format and the depth of field will be the same.
Myths
Myth 1: When using a full frame lens on a Super 35 camera, the image is darker. Using a Speedbooster gives you the same exposure as if you were using the lens on full frame.
False, a lens used on any format will yield the same exposure. Using a Speedbooster will give you an image brighter than when used on full frame.
Myth 2: When using a full frame lens on a Super 35 camera, the focal length is longer due to the crop factor. Using a Speedbooster changes the lens back to its original full frame focal length.
False, a lens’ focal length does not change no matter what camera it is used on. A crop factor is simply a tool to compare the relative sizes between formats. A full frame 50mm lens does not become a 75 or 80mm lens when used on Super 35, it is still a 50mm. The crop factor simply illustrates that it would be necessary to use a 75 or 80mm lens on full frame to match the same field of view as the 50mm when used on Super 35. Using a Speedbooster reduces the lens’ focal length, thereby widening the field of view on Super 35 to match what would be expected if the lens was used on full frame.
Myth 3: The depth of field gets thinner when using a Speedbooster.
False. Whilst the aperture does get faster which, in itself would reduce the depth of field, the lens also becomes wider in focal length, which increases the depth of field by the same factor. Assuming distance to subject remains the same, depth of field remains the same when using a Speedbooster. However, if you were to move closer to your subject to offset the wider field of view, depth of field would get thinner, but this is a result of reduced distance to subject and not the Speedbooster.
Myth 4: Using a Speedbooster improves the quality of the lens.
Sort of. By widening the field of view, you’re making use of the whole image circle of a lens, thereby ‘shrinking’ any inherent optical flaws in the original lens. A lens may appear sharper or have less aberrations, but this is only because we’re not looking through a smaller portion of the lens so closely. Equally, consider that a lens does not perform as well towards the edges of its image circle compared to the centre – we may start to see vignetting that we didn’t before. A Speedbooster however has optics of its own which, no matter how well it’s made, will inevitably introduce some flaws of its own. This probably negates any optical improvements it may offer; a speedboosted lens on Super 35 won’t necessarily be better or worse, but any imperfections may be slightly different.
Myth 5: Using a Speedbooster is better for low light shooting.
Maybe. Making any lens one stop faster (with an actual, measurable and visible effect on exposure) will definitely help. Compared to using the same camera without a Speedbooster, it is definitely better for low light, but compared to a full frame camera without speedbooster, well that depends on the cameras being used! Full frame cameras generally perform a little better in low light than smaller format cameras. This essentially boils down to two reasons: with a larger sensor it is possible to either use larger pixels or fit in more pixels. Larger pixels means more photons being captured, therefore lower noise. More pixels means that, when presented at the same size (on a phone, computer or cinema screen) any grain or noise is comparatively smaller and therefore less noticeable. Even when the original higher resolution is downsampled to a lower resolution, any noise is also shrunk and is effectively suppressed. So, whether a speedboosted Super 35 camera is better for low light than a full frame camera, that’s probably a close call and further testing would be necessary to determine the most effective system for low light.
Myth 6: I can Speedboost Super 35 lenses.
Technically, yes, but not if you want to use them on a Super 35 camera. The most popular use for Speedboosters is to use full frame lenses on Super 35 cameras. This compresses the larger image circle onto the smaller format. Speedboosting a Super 35 lens is totally possible, however trying to do so on a Super 35 camera would result in hard vignetting (called portholing) as the image circle is now too small to cover the sensor. You can make use of speedboosted Super 35 lenses on even smaller formats such as Micro Four Thirds however.
Myth 7: Using a Speedbooster turns my camera into full frame.
No – you can’t change the size of the sensor. A Speedbooster changes the lens’ focal length to make use of its full image circle by compressing it onto your smaller format sensor. It gives you a wider field of view when using the same lens and sensor.
Bonus misunderstanding: “I want more depth of field.”
I’ve worked with a number of directors who, after reviewing a composition, request more depth of field. What does that really mean? A greater depth of what’s in focus? I stop down, and the director corrects themselves – “No, no – I mean I want less depth of field.” What they’re really asking for is a more pronounced background blur. Describing depth of field in terms of ‘more’ or ‘less’ is often confusing. Depth of field should be described in unambiguous terms like shallow, narrow or deep!
20 Comments
-
Jay
Thorough and exceptionally helpful.
-
Domingo Riesco
Thanks for your full explanation Alex! Just landing to your blog. It is hard for me to fully grasp how distance to subject affects DOF, but I have a question. What would happen in terms of depth of field, when attaching a speedbooster to a 24mm f2 FF lens, onto a FF camera? Lots of vignetting, but after cropping... I guess the result would be the equivalent to a ~35mm f2? That means one extra stop depth of field shallowness than before? Or just the same DOF as before but simply brighter? I already own a 24mm f2 and I'm evaluating the difference between adding a focal length reducer to it, or having (another) 35mm f2 lens.
-
Alex Stone
The depth of field would be the same: a 24/2 with a Speedbooster would become a 17/1.4. Cropping in to avoid the vignetting would result in the same FoV as the native 24mm, at the expense of resolution and at the gain of a stop of light. It's still a 17/1.4 but the crop is as if you were using a smaller format - you'd now be using a S35 lens on a S35 format, compared to the original set up which is a FF lens on a FF format. Adding a focal length reducer to your 24mm and then cropping in won't give you a longer lens so you'd need a separate 35mm lens if you wanted something longer. The closer the subject, the thinner the DoF - this applies to any focal length, though the longer the lens the thinner this is for any given distance.
-
-
Alaaddin
Dear Alex: Thank you very much for this article. My question is simple: does it make sense to use a Speedbooster on a full frame? I am considering a Villtrox speedbooster ot use Nikon Lenses on my EF sony mount camera. Thanx.
-
Alex Stone
Unless you're using medium format (larger than full frame) lenses, it doesn't make sense to use a Speedbooster on a full frame camera. The Nikon lenses will porthole and you would have to crop in on your image by ~1.5x in the edit to get a usable image.
-
-
Gabriel Goncalves
Hi! I'd like to pose a question. Imagine I have an APS-C camera and I choose an FF lens 35mm 1.4 lens. I then could couple - or not - that lens to the APS-C camera with a speedbooster 0.71x. (Of course the number 0.71 is not by accident: it is the inverse of the squared root of 2, a fundamental number in f stops.) Please correct if i'm wrong, because i'm completely new to this thing of speedboosters. (But not to crop factor and all the nonsense written about it and "equivalences" and all that stuff around aperture and DoF in different cameras - as if the sensor size could modify the lens!...) Is it true that without the speedbooster, I would have, mounted on the APS-C camera, a lens (the FF 35 mm 1.4) that would give me an angle of view roughly equal to that of a 50mm 1.4 lens mounted on a FF camera? The angle of view would be the same of a 50mm 1.4 mounted on a FF... However, not the depth of field, since a 35mm 1.4, the lens I really mounted (regardless of what camera is mounted in, of course, although many people have wrong ideas about this) has a not so shallow depth of view as a 50mm 1.4. Correct me if I'm wrong, but this is because of the focal length, which is shorter in the 35mm lens (all other things being equal: distance to subjet, and both having been set to the same 1.4 aperture), and therefore would be the sole responsible for the difference in depth of field. (By the way: conversely, if I were to use a 35mm lens on both cameras, the depth of view would be the same (again, and sadly, many people are wrong about this, as you obviously know), but not the "shot", since we now would have a different, narrower, angle of view when we put the 35mm on the APS-C camera.) In order to have the same characteristics (both angle of view and depth of field, thus giving us the famous "same shot") I would either have to put a 24mm 1.0 lens on the APS-C camera so I could have the same angle of view AND the same depth of field that I would have with the FF 35mm 1.4 lens mounted on a FF camera (I checked this on a DoF calculator and it agrees) OR I could mount the FF 35mm 1.4 on the APS-C camera with a speedbooster - necessarily 0.71x, as to reach the same angle of view of a 24mm on the same APS-C camera without the speedbooster. And the DoF would be the same as of a 24mm 1.0 lens in the APS-C camera. Is there a flaw in my reasoning in the last sentence? If there is not a flaw in my reasoning (and only if), aren't somehow right those who say that a speedbooster makes depth of field shallower? What I think they mean - using my example - is that using a lens designed for FF cameras, a 35mm 1.4, for instance, and using a speedbooster with this lens, this combo will give a shallower DoF than a 24mm 1.4 would. I really think this is what they want to say - but I might be wrong! If I'm not wrong (and I ask your favour to assess this) I think this might be the source for so many misunderstandings about this subject of speedboosters. Any comment from you would be really appreciated! One thing I don't understand: you wrote «Let’s say we add a Speedbooster to our 50mm f/2 lens. It has a magnification of 0.71x, therefore, we can multiply the focal length by this factor: 50*0.71=35.5. The lens is now a 35.5mm lens» I don't get it. Does the speedbooster actually change the focal legth of a lens, shortening it, or were you refering to "35.5mm lens" considering only angle of view, and not focal length, which it a physical property of the lens? Thanks a lot for any comment of yours! Happy New Year 2023 and Thank You so much for the video, explanations, and comments!
-
Alex Stone
Hi Gabriel, thanks for your comment. Yes, you are correct in your understanding all the way down to "Is there a flaw in my reasoning in the last sentence?" (There is no flaw in your reasoning.) With regards to your last two questions, a 35mm f/1.4 lens with a speedbooster will give a shallower depth of field than a 24mm f/1.4, because it is now a 24mm f/1.0. It is a whole stop faster and the depth of field (when used wide open) is therefore one stop shallower - assuming all other things equal as you preconditioned. A 24mm f/1.0 (a 35mm f/1.4 with speedbooster) would have the same depth of field as 35mm f/1.4 on full frame however. This leads in to your final question too - yes, a speedbooster actually changes the focal length of the lens (and therefore the corresponding maximum aperture.) This is where I find most misunderstandings stem from - many people consider their 50mm with speedbooster to still be a 50mm. The text on the lens might still say that, but the optics have changed.
-
Gabriel Goncalves
Thank you so much for your explanations! Happy New Year 2023!
-
-
-
Niek Deventer
Hi Alex, good story! The depth of field part is even simpler when you look at the 'image size' on the sensor (that's how Google translates 'Afbeeldingsmaatstaf' from Dutch...) Example: A close-up of Pinocchio with a 50 mm lens at 2 meters distance. To get eyes and nose tip exactly in focus you need aperture F8. The same close-up can also be made with a 25 mm lens at 1 meter distance, a 500 mm lens at 20 meters distance or a 2000 mm lens at 80 meters distance. If you take those shots with F8 then in all cases Pinocchio's head is and the same size and in all those shots the sharpness is exactly from his eyes to his nose. If Pinocchio starts to lie in that close-up then the ONLY way to keep eyes and nose tip sharp is to stop down to F11 or even smaller. If that's not enough you have to use a wider lens or step back. But then your 'image size' gets smaller and your close-up becomes a medium shot. So if you take a knee-shot of an actor and you want another actor a half meter behind him also in focus then you need a certain aperture. And with that aperture it doesn't matter what kind of lens you use to make the knee-shot, close with a normal lens or from a distance with a telephoto lens, the depth of field, or how sharp the other actor becomes in your image stays the same.
-
Roman
Hello dear Alex thank you for this article it help me understand better the way lenses work because you used simple language not technical misleading terms, thank you for that. I must inform you that although your article is technically sound and mostly correct i find it lacks deeper explanation of how lens work and could potentially lead readers that have little understanding about lens mechanics to the wrong conclusions about effectiveness of Speed Boosters. I find important omission to not mention that FF 50mm f/2.0 lens mounted on APS-C camera would be actually reduced to f/3.0 aperture since the projection image it outputs is intended for 35mm sensor when actual sensor is 25mm (more details in links below). Indeed its marketing statement that speed booster makes 50mm f/2.0 lens into f/1.4 lens, it simply changes it from f/3.0 back to f/2.0 by focusing entire projection on the sensor. Readers could benefit greater from more physical explanation rather than pure mathematical one. Speed Booster physically takes place of 35mm sensor and receives 35mm projection FF lens generates then outputs smaller 25mm projection to into APS-C sensor. Please see second page for diagram. Two pages https://www.dpreview.com/articles/2666934640/what-is-equivalence-and-why-should-i-care Please see diagram under "Why do you say a smaller sensor sees less light?" https://www.dpreview.com/articles/5365920428/the-effect-of-pixel-and-sensor-sizes-on-noise
-
Alex Stone
A 50mm f/2 on an APS-C camera is still a 50mm f/2. The actual aperture doesn't change, only the equivalent aperture. The exposure and depth of field on both APS-C and full frame would be the same, only the field of view would be narrower. If you wanted to match both the field of view and depth of field, you would need to use a 35mm f/1.4 on the APS-C camera. This is where the confusion of equivalent aperture stems from: even though it's a stop faster (and the exposure one stop brighter), a 35mm f/1.4 would have the same depth of field as the 50mm f/2 does on full frame. The sooner one puts equivalent apertures out of their mind, the easier it is to accept and understand what any given focal length and aperture looks like on the format you're shooting on. Don't worry about how to emulate a different format's look or what that lens would look like on a different format - just enjoy the act of looking through the viewfinder! The article is correct in that the light projected by the lens onto a smaller sensor is the same intensity (photons per mm^2), but that there's less total light hitting the sensor. In the same way, a small wall doesn't require as much paint to cover it than a larger wall, but it isn't any less colourful. As they've demonstrated, an APS-C and a full frame camera with the same pixel size produces the same level of noise; only when the images are printed out or blown up to equal sizes does full frame have the advantage. This is purely down to the higher resolution: each pixel is printed smaller and any noise is less noticeable.
-
-
Ola
I’ve got a question about using a speed booster with a standard zoom lens. I’ve got a viltrox speed lens that I attached to a recently bought 28-80mm f2.8-4. My observation is that the aperture does not open wider than the f2.8. This is not the case with the nifty 50 f1.8 that I have as it opens up to f1.2 when used with the speed booster. I guess my question is, does zoom lens open any wider when used with the speed booster? N.B Camera is canon m50 and the lens is also a canon, L series
-
Alex Stone
Yes, a zoom behaves the same as a prime when used with a speedbooster. Remember that the 28-80L is a variable aperture lens. It's usually f/2.8 at 28mm and stops down to f/4 at 80mm. Does the lens open up to f/2.8 at 80mm with the Viltrox? If so, this is working as it should - the lens would open to f/2 at 28mm. If the camera is still reading the lens as f/2.8 @28mm and f/4 @ 80mm, I suspect that the reason it's not communicating an updated (faster) f-number is down to the adapter's firmware. Given the lens' age, the adapter may not 'recognise' it, but rest assured that the physics are still at work - the lens will become brighter and wider than if you were using it with a regular adapter.
-
-
Bob
Actually, very informative and helpful in understanding exactly what and how a speed booster works!
-
YX
as far as conclusion goes about depth of field: Vladimir is correct; Alex Stone is wrong.
-
Alex Stone
Can you elaborate? I demonstrate the effects - including depth of field - in this video here. I'll embed it in the at the top of the article for future readers. https://youtu.be/xlnciK2cyto
-
-
-
Vladimir
Wow, you just made everything even more confusing and complicated :) The main thing about the speedbooster is, it makes your 50mm on a crop body to behave like 50mm on a FF, everything will be exactly the same as if you have a FF sensor, and that means your FOV is the same, your DOF is the same, your bokeh is the same, the amount of light hitting the sensor is the same, that's all. It's like moving your sensor back and gathering all the light lens is projecting, that's why you get all the properties of a FF sensor. There is no need for confusing theory, you could just post a photo with and without a speedbooster and you will see both are identical, given the speedbooster is making the exact 1.0x crop. If not, everything still applies, but you will have 1.05 or 1.1x or 1.3x, but a speedbooster is working exactly like cropping backwards and all the image properties are affected.
-
Alex Stone
I think you've missed the point of the article. It isn't an opinion piece or a 'confusing theory' that I've come up with, it's optical physics. To describe using a Speedbooster akin to 'moving your sensor back' is a misleading statement - the sensor doesn't move, and if it did, would not result in the same effect but would have significant negative side effects, namely the inability to focus at infinity. Everything is not exactly the same as if you were using a full frame sensor. Whilst many factors are equivalent, the image is one stop brighter - a key difference - and the article explains how.
-
Mladen
You are confusing people (unnecessarily) with your approach to this issue. Nobody is measuring equivalence the way you have done it here. At least, not in a photographic world. I don't see why videographers would be any different. ''50*0.71=35.5. The lens is now a 35.5mm lens'' You reversed the order of how most people measure equivalence. I wouldn't go about this issue in that order. A 50mm lens on APS-C sensor will be equivalent to 75mm on full frame (35mm). That is the crop factor. That is the ''equivalence''! Period. Now, one of the functions of a focal reducer (Speed Booster, Lens Turbo, Pixco...) is to ''reduce'' the focal length according to a focal reducer specification - 0.7X or whichever, to make it closer back to 50mm (specified focal length)!