WOW! Speed Booster adapter makes your lens faster, wider and sharper!

Philip Bloom announced and tested the new Metabones “Speed Booster” adapter. And what it does sounds as it would be some kind of magic trick! It makes any lens faster, wider and sharper! Philipp explains how it works:
“In essence, it’s an optical adapter that attaches to the E mount of a Sony NEX camera (Super 35 sensor) and then to a full frame SLR lens. The Metabones ‘Speed Booster’ is a 0.71 x focal reducer, that will effectively turn your full frame 50mm f/1.8 lens into a 35mm f/1.2 lens. Note, doing so (as a guide) will increase the aperture of that lens by one stop. Your Sony NEX Super 35 E-Mount will effectively have near full frame coverage on a full-frame lens. It also serves ‘double-duty’ as a lens mount adapter, from Canon EF lens (but not EF-S) to Sony NEX, with auto-aperture and image stabilisation.“
Amazing or not? It will be available in January 2013 from Metabones’ web site http://www.metabones.com and its worldwide dealer network for US$599 / £372 plus shipping and applicable taxes and duties.
Here is an image that shows you how it works:

The image circle gets reduced and this actually doesn’t decrease the image quality. The only real drawbacks are:
1) the lens has no fast autofocus like the recently announced EF to NEX Smart Adapter II (they are here on eBay).
2) You have to use a Full Frame lens and you can use it on an APS-C camera only (not on FF cameras).
And here is a video sample shot by Philip:
194 grams from James Miller on Vimeo.
Question now….





Oldmanv
5 months ago |Autofocus needed.
Jon R
5 months ago |There’s AF but it is on the slow side. It’s mentioned in the product description on the Metabones website.
http://www.metabones.com/sony/ef-e-speed-booster
Beachrider
5 months ago |Everyone be careful here. There is NO Sony Alpha-to-NEX focal reducer offering yet. The ALPA thing has nothing to do with Sony Alpha FF.
Vivek
5 months ago |I will buy it as soon as I can.
If one overlooks the AF part, the possibilities are endless.
Think about an Olmypus OM 50/1.2 with this Speed booste- An f/0.9 normal lens for the NEX-7/6. Absolutely fabulous!
Roke
5 months ago |Video credit to James Miller…details of the adapter by Phillip
Matthew Durr
5 months ago |After thinking for a while figuring out how this concept works, this seems brilliant, but only if the element(s) in the adapter do indeed at least keep image quality the same (if not necessarily better). Will probably work like teleconverters, in the sense that it may perform well with some lenses, and not so well with others.
http://matthewdurrphotography.com
xxxx
5 months ago |Reverse tele-converter, or focal reducer, as called by astrophotographists.
LifeStoryImages.com
5 months ago |Well said Matthew. The limitations are indeed going to show when you try to compress that f/1.2 down to 0.9. Any add-on back end vs a lens designed as such – you see where that’s going.
To take the explanation a bit farther, by shortening the focal length you “demagnify” the image and compress the details – making what was a soft image on a given pixel focal plane possibly sharper.
Sean Lancaster
5 months ago |I am excited about this . . . sort of a poor man’s FF camera (poorer in noise and poorer on the wallet), but I won’t be an early adopter. I don’t mind MF only, but I really like AF and the speeds being predicted are far less than stellar in that regard so I’ll wait to see reviews. But I am still intrigued by this concept.
DtEW
5 months ago |Even though Caldwell Photographic leads to a placeholder page, it seems like Dr. Brian Caldwell is indeed an optics designer mostly known for his work in large-format photography optics. This bodes very well.
Pipo
5 months ago |Will there be adapters for other lenses too? I have a couple of really nice Canon-FD-lenses and I always thought that it’s too bad, that the image-circle is cropped – especially with the wide angles.
But one thing I don’t get: Is the f-stop really “increased by” converting the large image-circle into a smaller one? All the other adapters, that use glass, absorb (or reflect? I’m not that good in physics) light/reduce the light-transmission.
Or does it refer to the DOF (so you just get the impression, that a 85mm 1.2 becomes a 50mm 0.95 concerning the shallow DOF, but the DOF actually stays the same)?
Vivek
5 months ago |The actual speed is increased. An 85/1.2 becomes a 59/0.90 (not f/0.95- there is a big difference!).
Separate the DOF changes from the focal length changes (0.71X) and the aperture changes (1.42X) to make things easier to understand.
Vlad
5 months ago |The correct answer is “the 85mm 1.2″ stays an 85mm (in terms of 35mm) and it gains one stop of light.
The more manageable answer is the one Vivek gave. You`re going to have the equivalent angle of view of a 60mm lens on a DX camera, the speed of a noctilux and the DOF of an 85mm lens on FF.
It does sound like the holy grail. Almost too good to be true. I`m thinking of that 21/1.8 Voigtlander just announced for 1300$ … price-wise it feels like a no-brainer compared to the 21 summilux. Optic-wise … it`s questionable.
Nik Page
5 months ago |yeahaaaa sure the holy grail
makes image quality better – yeahaaaa for sure! look at quality of the product shot
anony
5 months ago |Have you even tried to understand the concept?
GreenLens
5 months ago |Available samples illustrate that it works for the center of the image. Corners are carefully avoided. This is not convincing for still image photographers.
markq
5 months ago |Leica M Speed Booster to NEX?… then I don’t have to wait for NEX FF?…
CW
5 months ago |Not enough room for Leica M and Contax G optics. This will be for SLR lenses only.
markq
5 months ago |metabones just sent me an email confirming your statement. would have been nice though
Pb
5 months ago |back to waiting for FF NEX then…
Brüno
5 months ago |This is ridiculos marketing chatter. One should know that every introduced glass element (= lens) makes the image a bit worse. Moreover one of the benefit of working with M42 lenses on a NEX-7 or a GH2 was that the sharp middle part of the lens was used and not the worse corner parts. Know one should all of the sudden get a better image by using the bad part also. Sure you will use more of the lens area but I doubt if the image really will be better…
James
5 months ago |Introducing a new glass lens makes the image worse? I guess that means Zeiss lenses only every use a couple of bits of glass then, right?
CW
5 months ago |Read the whitepaper Bruno
Pipo
5 months ago |Ok, but besides that: Isn’t the focal length factor and the aperture factor the same, it just depends on the “direction” of the calculation? (50mm*0.7~35mm and 1.8*0.7~1.2 vs 35mm*1.4~50mm and 1.2*1.4~1.8)
But really: Maybe the f-stop (which in my opinion is a result of a geometrical calculation) changes, but does the light-transmission really increase (t-stop)? Once again, I’m not that good in physics, but I thought that glass always reflects some light and if so, you would loose some light, that hits the sensor, right?
Pipo
5 months ago |hmm, this should be a reply to Vivek’s post above…
Vivek
5 months ago |Yes, actual speed is increased (ie., the amount of light hitting the sensor is more) with the maximum achievable being f/0.9.
Here is the white paper on it: http://www.metabones.com/images/metabones/Speed%20Booster%20White%20Paper.pdf
Read it carefully. There is a lot of information!
Enjoy!
Pipo
5 months ago |Thanks! It really helped and sounds promising. Of course I don’t expect the holy grail, but honestly: I’m not a professional, but an enthusiast, and if there is a possibility that wide-angle lenses stay wide-angle with an acceptable image quality, I’m in.
Carl
5 months ago |Normal wide converters don’t increase the effective aperture of a lens as doing so reduces the image circle rendering them unusable. With larger format->smaller format converters that isn’t an issue, though, so it has surprised me no one has made something like this until now.
As with all converters, the quality of the optics is paramount, with cheap converters with low grade optics being barely usable for serious photography. It’ll be interesting to see how well Metabones has done on that front; the sides of the road are littered with third party camera accessory companies who thought they’d try their hands at wide and tele converters.
Varn
5 months ago |Hope I remember correctly, I think they copied the idea from Contax adaptor for medium format lenses to use on 35mm camera body, but that adaptor cost a few k and it’s zeiss glass. The ‘condensed’ images are brighter and of much higher image quality.
Miroslav
5 months ago |Who needs FF NEX now
? No wonder they changed the roadmap so much of late
.
albhui
5 months ago |Going to metabones product site, they also have speed boaster for the following products coming:
Alpha mount to NEX mount
Alpha mount to X mount
Leica R to NEX mount
Leica R to X mount
http://www.metabones.com/product
Vivek
5 months ago |That is not Sony Alpha but ALPA (an old defunct manual focus system with fabulous Switar glass)!
Spuds
5 months ago |A shame they don’t have a Nikon to Nex adapter, would love to see the possibilities of a Nikon 1.2… though I still am not convinced it wont impact the sharpness of the lens used.
Sky_walker
5 months ago |Show me the corners with UWA.
lol
don’t get fooled people, there is no way to magically improve everything with no price behind it. If so – we’d all be running with MF lenses mounted on 35mm cameras for years.
GreenLens
5 months ago |should make wonders with portrait/tele lenses though..
Rob
5 months ago |Why would You do that? Do you like if its heavier and more expensive?
Chad
5 months ago |The reason we couldn’t before was due to the longer flange distance of SLRs.
The manufacturer claims that on the NEX you will have improved MTF in the center, and native MTF in the corners.
Sky_walker
5 months ago |That’s why I mentioned MF lenses – these tend to have much longer flange distance then SLRs and, FYI similar adapters were made for SLRs.
Thing is that – they are not worth it, so they quickly vanished from the market.
CW
5 months ago |We’ll see soon enough. No one is going to “get fooled” if this adapter is properly reviewed and proves itself to live up to the claims.
Reading through the white paper I’m inclined to believe the claims and they have a respected team with a history of delivering high end optics to industry so this is no cheap solution.
Nelson
5 months ago |If you don’t want, you dont’ belive.
This in nothing new, Every single lens for 35mm SLR’s that has more than 40mm uses the very same technology in the lens itself.
The lens focal distance is the distance of the entrance pupil of the lens to the film/sensor, because SLR’s have mirrors, then how can a SLR have lens as wide as 14mm, there’s where the mirror is! Look at wide angle lens for Leica, they are tiny in comparison!
A 24-70 f/2.8 lens made is in fact a lens like 48-140 f/5.6 with an 0.5X. A variable aperture zoom, is a low-zoom-factor tele lens, with a variable magnification teledecompressor.
Thrud
5 months ago |This sounds really interesting and I’m really hoping it will turn out to be an awesome adaptor, but I’ve been trying to wrap my brain around it without much luck. (Maybe it was because of the huge dinner I just had) but can someone tell me the 35mm equiv would be on an NEX with these 2 lens?
Sigma 20mm f/1.8 = 21mm f/1.2? (This would be awesome to shoot the milky way with an NEX if the corners were ok)
SLR Magic 50mm f/0.95 = 53mm f/??? (Would this solve the softness issue wide open?)
Thanks in advance to the brainier people here.
Hans
5 months ago |the adapter is only compatible for input lens up to speed of F1.2 I read on their site, output 0.9. So no 0.95 lens can be adapted (maybe stopped down or with lower quality, I don’t know)
Carl
5 months ago |The adaptor itself acts as a [large] virtual aperture, so there’ll be a point where lenses will stop getting ‘faster’ with it, even if you open their aperture more. It’s not like they suddenly stop working wide open, there’ll just be no point from a light or depth of field perspective.
More importantly for fast lenses, the converter will induce some spherical aberration that will require stopping down to correct for (the amount will depend on the quality of the optics), so there’ll be something to be said for having a lens that’s faster than one needs.
The question will be whether the optics are sufficiently good that you end up with a faster lens than just using a normal glassless adaptor once you’ve stopped down to correct the converter’s aberrations.
Nelson
5 months ago |This is for SLR lens only!
MC
5 months ago |I have checked the whole piece of review and there are 100% crop image samples to proof that the images look really sharper and brighter than direct use of lens. I think is is just similar to converge the sun ray by a magnifier which can increase the intensity of the light. This explain why it can increase 1 stop of light. I should buy one, then all my FF lens collection can be utilized
Varn
5 months ago |This is precisely my view, as long as the glasses in the adaptor will not deteriorate as much as the image quality gain by condensing the larger images, we should have an overall improvement in image quality. I think how much is improved also depends on the focal length of the lens.
Good news to us.
Navigo
5 months ago |Thank you Andrea for this very useful post! Again you are very quick and busy. Great! Best regards
DtEW
5 months ago |I was under the impression that the focal-reducer was not made because Kodak was sitting on the patents. Can anybody confirm/deny this, and of confirm… what has changed aside from Kodak’s bankruptcy?
(I’m assuming most/all their patents got hawked off to pay their debtors.)
If this is doable again, I am hoping to see even dumb adaptors for other legacy lenses. That would be very sweet.
Vivek
5 months ago |Read the “white paper” linked above.
This Speed booster is newly patented by its creator.
DtEW
5 months ago |There’s a world of difference between “pat. pending” and “patented”.
Patent filing fees are only ~$500, and that is actually the bare minimum (i.e. non-researched, non patent-attorney involved no-hope-of-being-granted waste of paperwork) for filing a patent application to allow somebody to use “pat. pending” for that period of time without being being fined for false representation.
And it generally takes 3+ years for the USPTO to grant/deny a patent, for which period the filer can continue to use “pat. pending”.
They would be allowed to make such a product and be protected from competitors that might want to make a similar product, unless somebody sues them for infringement of an existing patent/patent filing. Perhaps the calculation is that they would sell enough product to recoup whatever outlays they made before they get the cease-and-desist or lose the “pat. pending” claim.
DtEW
5 months ago |(Not that I really care if they really do have a patent or not, though. I just want my adapters and I could care less if they’re doing shenanigans to put it on the market, if only temporarily.
)
Anonymous
5 months ago |Wait, why won’t they make a Minolta/Sony A-mount booster to E-Mount adapter? Why Canon? I have a Minolta 35/2, 50/1.7, 85/1.4 and 135/2.8 waiting to be used!!!
Varn
5 months ago |Because of $$$, Canon has the biggest market share.
Varn
5 months ago |But I still prefer to use my FF lenses on the coming FF Nex as the image quality will still be better than APS-C Nex with this type of adaptor.
This adaptor is bound to attract many Canon users to Nex system, so will benefit us eventually.
john
5 months ago |there will be an alpha mount -> nex adapter .
http://www.metabones.com/product/sony-nex/alpa-lens-to-sony-nex-speed-booster-detail
it will be cheaper too !!!
Anonymous
5 months ago |That is for the Alpa mount, not Alpha.
Vlad3D
5 months ago |US$599 ? I hope China is already working on low cost version
Alpha Mark
5 months ago |I hope this doesn’t burn your sensor instantly when you point the camera directly to sun light…lol.
J
5 months ago |Sample photos?
Vlad
5 months ago |Provided link?
jason
5 months ago |At $600, no thanks! These things costs as much as a new body!!! A sucker and his money will soon part.
Carl
5 months ago |The product will sink or swim on the quality of its optics, and good optics aren’t cheap, so it’s not really surprising.
I’m sure there will be copycats who copy the design, but if they use cheap optics that turn the image into mush they won’t see a lot of use.
Rob
5 months ago |This product isn’t aimed at NEX users trying to expand and “push” their current lens lineup, it’s targeted at pro’s and invested amateurs with tens of thousands of dollars in high quality Canon EF glass. Some of those folks I’m sure have harbored a curiosity about the NEX system or have wanted to give Sony a try without having to abandon their collection of glass. Now with a cheap relatively cheap used NEX 5N body and a $600 adapter, you could open up your whole library of EF lenses on a system shooting great 16mp stills 1080/60p video, with a much smaller form factor, etc…
If I was heavily invested in Canon EF glass then $600 isn’t too big of a cost to open up my library of lenses to a completely new system, and improve them along the way.
Hellven
5 months ago |does this adapter works with Canon’s TILT/SHIFT lens?
DD
5 months ago |That’s what I want to know too.
CW
5 months ago |They claim specifically that it does – with all movements!
Mike
5 months ago |That guy who invented this adapter must be a genius.
But, not a technical genius, but a business genius!
CTPhotographX.com
5 months ago |Fabulous spot Philip Bloom!
Vlad
5 months ago |Hasselblad to Canon adapter and a little bit of stacking. Who`s got the bigger balls now? All them nipples are going to get hard at the sight of the Frankencamera!
Roni
5 months ago |Very expensive.
Fants
5 months ago |This is a cool concept but WAY expensive. They’d sell a whole lot more of them, I think, if they made a model with no electronic contacts. In fact all of the lenses they use in their “white paper” are old MF glass, which seems to defeat the purpose of having that $250-300 of electronics in the adapter.
Fants
5 months ago |Aaaah ok, just looked at the site. The Leica R version is only $399. Still overpriced, but a hell of a lot better than $599.
ItsaChris
5 months ago |This has also been used in very old Nikon DSLRs. The E2 and E3 had Reduction optics in between the lens and CCD.
I am surprised we are getting this and not a tilt/shift adapter like the Hasselblad HTS 1.5. But this just seems to be another stop gap measure because the nex line is lacking lens. Hopefully in a few years everyone who buys will no longer need it…
gojiguy
5 months ago |so… can i pop in my old canon FD lenses or Jupiter/Helios M42 lenses? I’m still quite confused.
Rob
5 months ago |It’s stated multiple times that this adapter is for Canon EF lenses (EOS) specifically.
Brüno
5 months ago |M42 will work because you can easily adapt it with a very cheap additional adapter to Canon EOS mount. The flange distance of FD is smaller than EOS and this is why you can normally adapt it to an EOS mount without loosing the capability to focus for infinity (without an additional glass element in the adapter). As the Metabones adapter semms to make miracles become true, maybe this will work nevertheless *lol* but I would not be sure about the FD mount. M42 should definitively work and also Mamyja 645 lenses which you can adapt to Canon EOS mount.
Brüno
5 months ago |Correction for FD: can normally not adapt it
Pipo
5 months ago |I asked the guys at metabones for a Canon-FD-version: They told me to wait, it’s already on their roadmad. And I expect it to be cheaper than this version (no electronic parts)…but if the glass is good, then maybe ~400USD?
jan
5 months ago |I sent them an email about the FD adapter..it is on there product road map
Jim
5 months ago |My main question about this is how it actually alters the perceived depth of field. If for instance it turns your 50 1.8 into a 35 1.2, your image won’t have the same sized bokeh as a 35mm 1.2 on a fullframe camera?
CW
5 months ago |Jim – It will “almost” preserve the perceived DOF of the native lens. So while you get a full stop of extra light gathering the DOF perception will be just a hair over that of the native lens. Still awesome!
Jim
5 months ago |By ‘native’, do you mean the lens it’s attached to or the lens it would be mimicking? Sorry if this is all in the White Paper, downloaded it but not had time to trawl through yet.
CW
5 months ago |A 50/1.4 Zeiss Planar, becomes the equivalent field of view
and DOF of a 53mm f/1.5 Planar, with light gathering of an
f/1.0
Chris K.
5 months ago |I hope this will start a trend of adapters that include a lens to correct for the flange distance. (So you could use Leica M glass on an SLR, or m43 glass on NEX [without vignetting], etc.) Of course, quality glass/optics required but the one-time investment in the one correcting adapter opens up the whole world of lenses from a previously-unattainable line of glass.
Log
5 months ago |Now make one for A-mount SLT cameras please! I don’t care if I can’t focus all the way to infinity because of the longer flange distance. But, I’m sure if you tried you could figure out a way around it.
Jason
5 months ago |I am saving for the FF NEX in 2014, forget silly adapters which costs more than my 5N did.
NEXfive
5 months ago |Okay, I do believe that this sort of magnifying glass widens the lens and makes it faster, but I don’t believe the quality increase by additional lenses. Great option for those having a lot of Canon EF glass and like to get additional use for this. No need to think about for me, since I would need a Speed Booster for Canon FD Mount.
But if I had a choice to use my Canon Lenses with an adapter like that or with a Full Frame NEX, I’d go for the Full Frame NEX, of course.
CW
5 months ago |maybe you should read the white paper?
NEXfive
5 months ago |I’ve downloaded and read it, but under 6) the MTF increase is simply rubbish, because they used Full Frame Lenses and compared MTF with and without Speed Booster, instead of the lenses’ MTF on a Full Frame Sensor against the MTF on APS-C with Speed Booster (both sensors would need the same pixels per inch for serious comparison!). Of course the x0.71 MTF with good glass should be better than the cropped MTF without it, but it will never reach the Full Frame MTF, because of the extra elements.
CW
5 months ago |I’m not an optical engineer, are you?
What I am concerned about is whether the lens performs better than it would on a standard adapter, not better than it would on a FF camera. If, compared to the current adapter solution, you can get better MTF, maintain the perceived DOF and gain an extra stop, what is your complaint?
NEXfive
5 months ago |I’m not complaining about this additional option to use lenses with E-Mount. I simply don’t believe this Speed Booster can really perform better reducing Full Frame Lenses’ image circles to APS-C with extra lens elements than a Full Frame Nex will do without the extra glass.
CW
5 months ago |Why bring a currently non-existent FF NEX into the conversation? It has nothing to do with this product which supports m/43 and Fuji X in addition to NEX.
And why make claims that something is rubbish without any proof? I’m not claiming it is awesome, just saying that the science and implementation looks sound and I am willing to give them the benefit of the doubt, which will be judged soon enough by enthusiasts and reviewers.
NEXfive
5 months ago |My initial posting praised the new option to use Full Frame lenses, especially for those with a lot of Canon EF glass, but I would prefer a Full Frame NEX instead, because I don’t believe this crap of better MTF by using wrong relations. If you’re unable to deal with such a clear and simple opinion, it’s your problem, mate.
CW
5 months ago |Alright Mate!
Pipo
5 months ago |as above: Canon-FD-Version will be released, the guys at metabones told me, that it is on their roadmap…
xxxx
5 months ago |Don’t compare it with the real FF. Performance increase is for the crop sensor, not against FF sensor. That’s what they claim. It’s a “Speed Booster vs. plain mechanical adapter” not “speed booster vs FF”.
Against FF sensor, the amount of light should be the same, minus very little light loss.
tom
5 months ago |So can someone help me understand; isnt this talking about only the nex video cams with the super35 sensor? How about normal nex?
John Maverick
5 months ago |It increases aperture and changes FL while reducing the resulting image-circle of the lens. Simply using a fullframe sensor would give the same result, and without the deteoration of image-quality induced by the extra optical components.
CW
5 months ago |um, why talk about using a FF sensor? This product is for people with APS-C and m/43 sensors.
FlashBFE
5 months ago |Please make such an adaptor for M42-Mount.
Nelson
5 months ago |Why?
Use a M42-to-EF adapter (doesn’t need chip) after the new adapter.
skeptical
5 months ago |This adapter would increase sharpness as claimed (compared to the same lens on APS-C), under two idealized assumptions:
(1) the extra glass is optically flawless
(2) your FF lens has uniform sharpness corner to corner
Both of these are likely to be violated, because as others noted no piece of glass is flawless, but also because the transformation of the image circle brings in the extreme corners where sharpness is typically poor. (As opposed to an FF lens directly used on APS-C which exploits the center area of the lens.) I’m curious to see the real-world results, but I would not have very high expectations.
Sam Waldron
5 months ago |Very cool…
NEX-6 ($800) plus metabones ($600) plus EF35 1.4 ($1000) = roughly effective 35mm F1.0 for about $2400
Imaginary full frame NEX ($2000 minimum) plus EF35 1.4 (Because no FF NEX lenses exist) ($1000) = 35mm F1.4 for about $3000.
The FF NEX set will be a little larger as well as a normal NEX to EF adapter is thicker.
Very interesting.
Sam Waldron
5 months ago |…Meaning with the adapter you get the same FOV and DOF, but gain a stop in light, lost a little length, add a little weight.
FF NEX is bandied around as the holy grail, but how long will it take for Sony to release a decent line up of top quality optics? Projecting a FF image form that kind of flange distance is going to be a crazy challenge.
ihur
5 months ago |Nah dude, you get 24/f1.0 out of that 35/f1.4
You dont get shallower DOF, it’s just wider and faster
mc
5 months ago |just to remind you, Sigma 35mm f/1.4 might be a better option. And I.don’t think you can get a Canon 35mm f1.4 dor $1000.
WTH
5 months ago |PL0X MAKE ONE FOR ALPHA FF LENSES TO ALPHA APS-C CAMERAS!!!!!
JOhn
5 months ago |This is false marketing.
You cannot take a 50/1.4 add extra glass and get a 35/1.2 in terms of light and speed. That extra glass is going to introduce both a resolution loss and light loss.
It might be possible to get a shallower depth of field, but you cannot make more light.
At best what this will do is present the angle of view (AoV) from a 50mm full frame camera on a crop camera so that when you look through the eye piece, you see the same image with both cameras.
In other words, this lens adapter negates the “magnification factor” of APS-C cameras when full frame lenses are used on them.
aardvark7
5 months ago |This is all to do with the image circle produced by a lens, not the total light gathering. Relative to the original lens used, you certainly won’t gather more light.
However, when you normally use a 50mm f2.8 full frame lens on an APS camera all that happens presently is that the outer light is ‘wasted’ because the image circle is much wider than the sensor.
In this case, it is simply focussing that larger image circle down to fit the smaller sensor, thereby increasing the light intensity; effectively providing a faster f stop.
Think of it the same way as a magnifying glass when you were a kid: if you reduce the size of the image of the sun the light intensity increases until it burns a hole! The smaller sensor can make use of the smaller, more intense image.
The only reason why it hasn’t been done as a matter of course before is that the flange distance needed in SLR cameras to allow the mirror meant it was not possible. With mirrorless cameras and smaller sensors it is entirely feasible. The big problem comes in controlling all the other optical effects, such as aberration, especially with different makes, designs and focal lengths.
Previously, such devices would be dedicated to a particular lens and it was often just as easy to make the appropriate lens. This seems to be suggesting the adapter will improve all cases, so will save a bucketload of money. We will have to wait and see if all reviews will agree.