Home

Please support mflenses.com if you need any graphic related work order it from us, click on above banner to order!

SearchSearch MemberlistMemberlist RegisterRegister ProfileProfile Log in to check your private messagesLog in to check your private messages Log inLog in

Metabones speedbooster
View previous topic :: View next topic  


PostPosted: Sun Jan 20, 2013 6:11 pm    Post subject: Reply with quote

The "winning a stop" thing is only relevant when comparing the performance of a given lens used on an APS-C sensor, with and without the adapter. It doesn' "win" a stop, in real life performance, when compared to using that lens on FF.

The "F" (or "T" for that matter) depends on the density of the light flux. That's why the "F" of a lens doesn't change when you mount a given lens on a smaller sensor via a glassless adapter. No matter how small an area of the image circle you crop out, the density of the light flux remains the same.

What this adapter does is concentrate the light so that you end up with a smaller image circle; that's why you get a wider FoV, and that's why, due to the density of the light flux increasing, you "win" a stop of F.

However, when you compare a given lens used on FF, and the same lens used on APS-C with this adapter, you will see that the "total amount of light flux that hits the sensor" remains the same.

Here's an illustration:



So how can using the adapter with APS-C have an advantage over using the lens on FF without adapter? The answer is, it doesn't!

"Winning" a stop is just a by-product of increasing the light flux density so that the same "total" amount of light can hit a smaller sensor.

In real life, because, in order to print a given image at the same size (say, 8x10), you have to magnify a shot taken on an APS-C 1.5 times more than you would a shot taken on FF, you are going to increase the amount of noise (and aberrations for that matter).

So, shooting a 50/1.4 lens as a 33/1.0 lens on APS-C with the Speed Booster @ ISO100 and 1/125s, when enlarged to 8x10 print, is going to be just as noisy/clean as shooting that same 50/1.4 lens on FF (without adapter) @ ISO200 and 1/125s and enlarging that to 8x10 print.

So, you're really not "winning" a stop with this adapter when compared to shooting the same lens on FF.


With that said, the benefits of this adapter is clear (as long as it performs near the theoretical optimal range); You get to shoot all of your FF SLR lenses at their native FoV/DoF/and real world speed (i.e., with apparent 1 stop enhancement on APS-C), on a super compact body, that is an order of magnitude cheaper than the alternative (unreleased Leica M, Sony VG-900). On top of that, you still have the option of using the same lenses without the adapter and you get a 1.5x more FL from the same lenses, which is useful for e.g., wildlife photography, etc. Really, the price tag is not that steep, if you think about it.


PostPosted: Sun Jan 20, 2013 6:20 pm    Post subject: Reply with quote

From the site of a manufacturer of reducers:
Please consider that the TS Reducer does not correct image distortions like coma, field curvature or astigmatism. Newtonian coma will even be increased by the wider field.
So it's quite sensible to stay sceptical until actual results are there to study. (Er, what's Newtonian coma?, is there someone here with enough knowledge of optics and photography to explain this?
Wink


PostPosted: Sun Jan 20, 2013 7:04 pm    Post subject: Reply with quote

Some time ago I made some tests using achromats to reduce the focal lenght of some FF lenses on Nex-7 (achromat+shortened adapters). The focal length of the lens is reduced in accordance with this formula:
New focal length of (lens+achromat) combination = 1/ (1/focal length of the lens in m. + achromat’s diopter number)
Using HQ achromats (I have used Leica Elpro) I have noticed that you get:
- shorten focal length and wider angle of view(covered by the FF lens)
- increased luminosity (the new f-number of the lens+achromat combination can be counted)
- no notable increase in CA
- no notable loss in central resolution (probably a little increase)
- some loss in corner resolution
- some increase in spherical aberrations
- narrower DOF and creamier bokeh (due to the increased spherical aberrations)
However, my tests were neither scientific nor very rigorous.
My conclusion was that the combination is interesting for portrait and close up work on some old lenses and I am now gearing some shortened M42 to Nex adapter for a more proper testing.
This, I think, could be a kind of poor man’s Speed Booster.
I am confident that the Metabones's Speed Booster, by using HQ glass and coatings, can achieve what it promised.


Last edited by dan_ on Sun Jan 20, 2013 7:53 pm; edited 2 times in total


PostPosted: Sun Jan 20, 2013 7:06 pm    Post subject: Reply with quote

harry tibi wrote:
From the site of a manufacturer of reducers:
Please consider that the TS Reducer does not correct image distortions like coma, field curvature or astigmatism. Newtonian coma will even be increased by the wider field.
So it's quite sensible to stay sceptical until actual results are there to study. (Er, what's Newtonian coma?, is there someone here with enough knowledge of optics and photography to explain this?
Wink


Can you give us your source ? A telescope manufacturer ? Apparently coma can be a particular problem with Newtonian (reflecting) telescopes - but I'm not an expert.

Stay sceptical by all means - but there are available already some results to study - MTF curves and actual pictures taken with the Speed Booster.


PostPosted: Sun Jan 20, 2013 8:31 pm    Post subject: Reply with quote

Oops...
http://forums.dpreview.com/forums/post/50688709


PostPosted: Sun Jan 20, 2013 9:53 pm    Post subject: Reply with quote

Sorry for not mentioning my source:
https://www.teleskop-express.de/shop/product_info.php/language/en/info/p365_Brennweitenreduzierung-0-5x---2--Filtergewinde.html


PostPosted: Sun Jan 20, 2013 10:13 pm    Post subject: Reply with quote

The speed booster is a very good news indeed ;
A kind of crop sensor breaker ;
The process is well known for astronomy or photo hunting ;
Have a look here on a Rubinar 10/1000 which becomes a 7/700 with a x0.7 converter :
http://discount-photo-import.com/1_objectif_photo.html
and some images :
http://discount-photo-import.com/Photos_Rubinar.htm
I'm very optimistic.
(Just have a look on the Nikkor 50/1.2ais on the white paper Razz)


PostPosted: Sun Jan 20, 2013 10:22 pm    Post subject: Reply with quote

for now my most urgent question:
which already available, cheap reducer is best suited for a 'DIY trial NEX adapter'? Wink


PostPosted: Sun Jan 20, 2013 10:50 pm    Post subject: Reply with quote

I tried with my Celestron reducer/corrector (which can be bought on Ebay for ~170US), but it only displays modest expansion in FoV as well an increase in F-stop. I'd say 20% for the former, and 1/3~1/2 stop for the latter. Even though the reducer should be capable of reducing to a much greater degree, I guess it's just not designed to work with mirror less cameras and 135 lenses Razz


PostPosted: Mon Jan 21, 2013 7:42 pm    Post subject: Reply with quote

Some tests and another approach on that one IL gain :

"There’s also been some confusion about how an adaptor could possibly change a lens’s f-stop.

It’s confusing because Metabones has created a more-or-less new concept in the realm of DSLRs (the research is not new, but it’s new to mainstream users), and they need to describe it in normal photographer lingo. The adapter does increase the light hitting the sensor. Think of it like a magnifying glass, focusing light from a broader area (full-frame) onto a smaller area (M4/3). The problem is that the normal way to talk about an increase in light is to talk in f-stops. An f-stop is technically a measurement of the size of the aperture, and the aperture affects light input as well as other factors.

Now, the Speedbooster does not physically open the aperture any wider than it normally goes, but it does increase the light input. There’s confusion because photographers are normally used to thinking that “light input” and “f-stop” are two words for the same thing. They are directly related (they mutually affect each other), but they are not equivalent.

If you are familiar with cinematographer’s terminology, you might say that the Speedboster increases the t-stop , while the f-stop remains constant. t-stop is a pure measurement of light input, rather than aperture size. Hopefully that makes sense!!"

(...)
http://philipbloom.net/2013/01/13/speedbooster/


PostPosted: Mon Jan 21, 2013 10:34 pm    Post subject: Reply with quote

Phenix jc wrote:

Now, the Speedbooster does not physically open the aperture any wider than it normally goes, but it does increase the light input. There’s confusion because photographers are normally used to thinking that “light input” and “f-stop” are two words for the same thing. They are directly related (they mutually affect each other), but they are not equivalent.

If you are familiar with cinematographer’s terminology, you might say that the Speedboster increases the t-stop , while the f-stop remains constant. t-stop is a pure measurement of light input, rather than aperture size. Hopefully that makes sense!!"[/i]
(...)
http://philipbloom.net/2013/01/13/speedbooster/


But this is still not 100% accurate. From what I understand, T is different from F in that F is measured by the mechanics (aperture, lens size, etc), while T is determined by actual transmitted light. Because any optics is going to reflect at least some of the light that hits it, actual transmistted amount of light is going to be slightly less than the ideal "F" stop.

So a lens could, theoretically and mechanically be F1.2… but actual measurement of the transmitted light might make it equivalent to F1.25.. which would be written T1.25.

However, things like DoF is decided by the mechanics, the "F", not the "T". For example, you could put an ND2 filter on an F1.4 lens, and the transmitted light is going to be halved, so you will have a T2.0 lens (if you don't like the filter, think of a tinted lens element). But the DoF is still going to behave like an F1.4 lens. That's why you can put on ND filters on a fast lens when shooting them wide open in daylight. So, in this sense, T-stop is not directly correlated with DoF (circle of confusion, etc.).

Now, with this adapter, like I described above, the light flux density is increased by 2x, i.e., for every square inch, you are going to receive 2x more photons; if you are measuring the T based on this, then yes, the T does increase 1 stop. A 50mm F1.4 lens is going to become a 33mm T1.0 lens. However, since T, like mentioned above, is not directly correlated with DoF, if the T-stop increases but the F-stop remains the same, then you would expect the DoF etc., to remain the same, too.

But, in fact, the DoF etc. are also affected, and the lens truly does behave like a 33mm F1.0 lens.

So, to say that this adapter "increases the T-stop" but the "F-stop" remains the same, isn't very accurate either.

At least, that's what I think.


PostPosted: Mon Jan 21, 2013 11:39 pm    Post subject: Reply with quote

rawhead wrote:
Phenix jc wrote:

Now, the Speedbooster does not physically open the aperture any wider than it normally goes, but it does increase the light input. There’s confusion because photographers are normally used to thinking that “light input” and “f-stop” are two words for the same thing. They are directly related (they mutually affect each other), but they are not equivalent.

If you are familiar with cinematographer’s terminology, you might say that the Speedboster increases the t-stop , while the f-stop remains constant. t-stop is a pure measurement of light input, rather than aperture size. Hopefully that makes sense!!"[/i]
(...)
http://philipbloom.net/2013/01/13/speedbooster/


But this is still not 100% accurate.


I agree completely. Either the writer doesn't understand the difference between "F" and "T" or his attempted simplification has gone so far as to be wrong - as you indicate.

The writer tells us that ....."light input" and "f-stop" .... are directly related (they mutually affect each other) .... It's nonsense. By "light input" I assume that he means light passing through the lens and hitting the sensor. If so, he's correct. It is affected by the "f-stop". However the "f-stop" depends upon the optical & mechanical construction of the lens and not upon the light passing through the lens. If you put the lens-cap on no light passes through the lens. The "f-stop" is still defined.


PostPosted: Tue Jan 22, 2013 1:15 am    Post subject: Reply with quote

kuuan wrote:
for now my most urgent question:
which already available, cheap reducer is best suited for a 'DIY trial NEX adapter'? Wink

Any 2" focal reducer should work
Something like that for example: http://www.ebay.de/itm/Focal-Reducer-2-0-5fach-f-Teleskop-Brennweitenreducer-TSRed052-/390433863742?pt=LH_DefaultDomain_16&hash=item5ae7abf83e
1,25" are cheaper but won't fit the NEX without vignetting.
How to adapt them is another good question ^^
Maybe they would fit inside a P6-NEX adapter etc.???

There are also a few which have build-in coma correctors and field flatteners - try to avoid them for normal lenses, they might over-correct and decrease the IQ.


PostPosted: Tue Jan 22, 2013 10:24 am    Post subject: Reply with quote

It is sometimes simpler than that

Shown is a Helios 44-2 with and without a +8 dioptr balanced on the back, giving 2x FOV and 0.5x magnification





ForenSeil wrote:
kuuan wrote:
for now my most urgent question:
which already available, cheap reducer is best suited for a 'DIY trial NEX adapter'? Wink

Any 2" focal reducer should work
Something like that for example: http://www.ebay.de/itm/Focal-Reducer-2-0-5fach-f-Teleskop-Brennweitenreducer-TSRed052-/390433863742?pt=LH_DefaultDomain_16&hash=item5ae7abf83e
1,25" are cheaper but won't fit the NEX without vignetting.
How to adapt them is another good question ^^
Maybe they would fit inside a P6-NEX adapter etc.???

There are also a few which have build-in coma correctors and field flatteners - try to avoid them for normal lenses, they might over-correct and decrease the IQ.
[/img]


PostPosted: Tue Jan 22, 2013 10:41 am    Post subject: Speed booster Reply with quote

No. I, the system is not new, it has been used on microscopes, and movie lenses, long times ago!

No. II if the "front lens" is not of excellent wide open quality, the speed booster cannot improve actual the wide open optical quality, even on smaller digital sensors.

No. III it is shows only with an Nikon 1,4/35mm > 1.0/24mm, and not with WA glasses below 35mm!

I'm not sure, willing to spend $ 600 for this toy aventure!


PostPosted: Tue Jan 22, 2013 1:37 pm    Post subject: Reply with quote

Looks like a nice idea but way too expensive though. I think many further real-world tests will reveal the quality loss when using both wide open and stopped down. Actually would be nice to own one with minimum lens IQ affect and half the price tag Smile


PostPosted: Tue Jan 22, 2013 6:14 pm    Post subject: Reply with quote

A detailed test from a reliable source: http://www.lensrentals.com/blog/2013/01/metabones-magic

In short: surprisingly, it works as announced Wink


PostPosted: Tue Jan 22, 2013 7:27 pm    Post subject: Reply with quote

I was going to post the same blog post as the above.

As I had suspected (and hoped), it really does sound like the adapter is going to do exactly what it says it's going to do, and if that is the case, $600 is not that expensive.

Can't wait to get my copy Smile


PostPosted: Tue Jan 22, 2013 7:32 pm    Post subject: Reply with quote

Well, in short, my eyes tell me that the lens with an adapter is worse than the naked lens on NEX-7. Maybe useably worse, but worse nevertheless. Second, there is no mention of vignetting that was pointed out on dpreview and perfectly visible on Philip Bloom test. In fact, vignetting is a very easy way to tell the shot on crop with speedbooster v a shot with the same lens on FF. Third, the price ATM is too high. $500-600 is too much much even for Canon/Nikon 4 element lens, not to mention for the optics coming from Metabones.

If someone got a ton of wide FF glass, it's a pretty compelling way to get a true wide angle on crop. My wides end at 24mm. At the proposed prices, I would rather get Zuiko 9-18 for my m4/3 outfit, not that fast, but much wider, and still cheaper. The other problem is that it's a per-mount expense, which will get pretty irrelevant once there is FF NEX.


PostPosted: Tue Jan 22, 2013 7:42 pm    Post subject: Reply with quote

I don't see why having wideangles is the only way you'd want this.

Where's the 35mm F0.9 lens on APS-C? Oh yeah, SLR Magic's 35/0.95 for $1300. Or, I get the Metabones for $600, slap on my Zuiko 50/1.2, and I've already paid for my adapter and $700 left in my pocket.


PostPosted: Tue Jan 22, 2013 9:06 pm    Post subject: Reply with quote

Well, it's not so simple. First, you don't get F0.95 lens in terms of exposure, probably more like f1.1-f1.2 lens because of vignetting. Sure, if you were thinking about getting SLR Magic lens for $1300 (which IMHO is ridiculous), then this adapter is a good deal. Otherwise, if you buy any high quality lens from a known manufacturer, say your Zuiko 50/1.2, or CV 35/1.2 ASPH, it's almost impossible to lose money on it. What will be the resale value on this thing is unknown.


PostPosted: Tue Jan 22, 2013 10:27 pm    Post subject: Reply with quote

My best guess is that the universal usefulness of this adapter will not diminish; it is just too attractive for APS-C shooters (including videographers). So the resale value will remain around 10-15% less than what you'd pay for one of these new. Kind of like the cream-of-the-crop teleconverters from Canon and Nikon.

Now, with volume sales, there's a chance that the retail price of the Metabones will come down in a few years, in which case the resale value will drop accordingly. While some people may laugh at the perceived loss of value of my adapter at that point, I will have, up until that point, enjoyed the hell out of it, making up for whatever that perceived loss of value ($100? $200?) might be a long time ago Smile

I agree that the value of this adapter may diminish somewhat with the introduction of a FF NEX. Indeed, if/when that does happen, I'd probably purchase it. There's still utility in it though.

(1) I can continue to use it on the NEX-6
(2) I can use it on an APS-C E-mount video cam, which will always be MUCH cheaper than FF video cam, and considering I'd really only dabble in the videography world (unlike photography in which I'm pretty hardcore), being able to go cheap on the camera and getting nearly the same images is a huge benefit.
(3) Even with an FF NEX, if it comes equipped with an "APS-C" mode like the VG-900, you might still want to use this adapter to get basically the same image but with a smaller footprint (the Metabones adapter is shorter than your normal EF-NEX adapter)

Here's one more thing that just occurred to me that I'd like to find out about that's potentially exciting.

What if we mount a tiltshift lens that has a much larger image circle than normal lenses, on a full frame NEX camera? Isn't it possible that we'll get a wider view lens that still covers the FF sensor (with the caveat that the amount of movement will be limited?).

E.g., Canon TS-E 17/4L, which can shift 8mm and tilt 8˚. Perhaps this will become a 12mm equivalent T/S lens on the FF NEX, but with the limitation that shift will be limited to 5mm and tilt to 5˚. THAT would be pretty friggin' sweet.


PostPosted: Tue Jan 22, 2013 10:29 pm    Post subject: Reply with quote

In terms of vignetting; I can't see how that would be a problem. Since the reducer doesn't quite reduce the image circle by 0.67x but more like 0.71, you'll probably get slightly LESS vignetting than using the same lens on FF (with the caveat that you're losing a few mm in FoV).


PostPosted: Tue Jan 22, 2013 11:02 pm    Post subject: Reply with quote

rawhead wrote:
E.g., Canon TS-E 17/4L, which can shift 8mm and tilt 8˚. Perhaps this will become a 12mm equivalent T/S lens on the FF NEX, but with the limitation that shift will be limited to 5mm and tilt to 5˚.

Caldwell posted on the nikongear.com link (earlier post) that he is about to test just this lens. He's already tested the Nikon 28mm f/3.5 PC and it's OK - although it's shift only and the amont of shift is reduced as you expected.


PostPosted: Tue Jan 22, 2013 11:06 pm    Post subject: Reply with quote

Samyang 24mm f/1.4 :

http://www.eoshd.com/content/9485/metabones-speed-booster-adapter-full-review?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+EOSHD+%28EOSHD.com%29