alfn
Forum Enthusiast
As if! Peter Green's Fleetwood Mac were a great blues band but the Rumours line up were touched with genius.
-
Welcome to the new forums! Please read this first. For known issues we are working to resolve, click here.
Why no ultrafast MFT lenses?
- Thread starter YellowBullet2001
- Start date
bobn2
Forum Pro
As if Peter Green's original band wasn't touched with genius.
LoneTree1
Senior Member
- Messages
- 2,467
- Solutions
- 2
- Reaction score
- 1,318
You have to be pretty low if you choose a drug-addled contralto over a supergroup from the 1970's.
bobn2
Forum Pro
I'm not sure that 'drug-addled' is a great distinguishing feature there.
In any case, Peter Green's Mac isn't really a supergroup, and was from the 60's. At their peak, they outsold the Beatles and the Stones combined.
Someone else may have already posted this, but optical limits has just finished testing the Olympus 45mm F1.2, and they declared it one of the best lens tested (by them) ever. It may not be ultra fast, but it's plenty fast enough, sharp from wide open, and benefits from m43's excellent IBIS.
--
Dostoy, from Oz
(Oz = Australia)
--
Dostoy, from Oz
(Oz = Australia)
Here's the link BTW:Someone else may have already posted this, but optical limits has just finished testing the Olympus 45mm F1.2, and they declared it one of the best lens tested (by them) ever. It may not be ultra fast, but it's plenty fast enough, sharp from wide open, and benefits from m43's excellent IBIS.
https://opticallimits.com/m43/1071-olympus45f12
I noticed they are still using the E-M5 II for testing, so that holds back the results a bit.
Mark Ransom
Veteran Member
I saw Fleetwood Mac in concert in the late '70's. The most memorable part was when Stevie Nicks stopped mid-song and apologized for messing up her lines. I'd say drug-addled is a great description.
I had no idea Peter Green's Mac was so popular. There's no question that the later band was equally if not more popular - wasn't Rumours the best selling album of all time for some years?
spider-mario
Senior Member
lenstip was enthusiastic too (I am linking to the summary but don’t hesitate to read the rest of the review, it has nice information):Someone else may have already posted this, but optical limits has just finished testing the Olympus 45mm F1.2, and they declared it one of the best lens tested (by them) ever. It may not be ultra fast, but it's plenty fast enough, sharp from wide open, and benefits from m43's excellent IBIS.
Olympus M.Zuiko Digital ED 45 mm f/1.2 PRO review - Summary - LensTip.com
In the list of pros, I would also mention the superb bokeh (qualitatively speaking). It has spoiled me.
RCicala
Leading Member
Samuel Dilworth
Senior Member
It was extensively tested and described throughout the 20th century by the likes of the American National Bureau of Standards (now NIST).
It’s not controversial to say smaller objects have tighter tolerances for a given effort. That’s why precision gauge blocks, for example, have lower expanded uncertainty values in smaller sizes. The same goes for lathes and CNC milling centres and all sorts of other manufacturing equipment.
If this wasn’t so, tiny $20 phone lenses would be unusable rather than amazingly good and microprocessors would be impossible to manufacture.
Testing complete camera lenses for variation and finding that lenses for smaller sensors have greater variation (if they do) doesn’t disprove the general claim though it’s probably more relevant to camera customers.
I thought the matchstick versus roof rafter example would make this understandable in an intuitive way but maybe it didn’t help.
It’s not controversial to say smaller objects have tighter tolerances for a given effort. That’s why precision gauge blocks, for example, have lower expanded uncertainty values in smaller sizes. The same goes for lathes and CNC milling centres and all sorts of other manufacturing equipment.
If this wasn’t so, tiny $20 phone lenses would be unusable rather than amazingly good and microprocessors would be impossible to manufacture.
Testing complete camera lenses for variation and finding that lenses for smaller sensors have greater variation (if they do) doesn’t disprove the general claim though it’s probably more relevant to camera customers.
I thought the matchstick versus roof rafter example would make this understandable in an intuitive way but maybe it didn’t help.
Mark Ransom
Veteran Member
Citation needed. There are very few characteristics that would automatically scale with the object size; the only one that comes to mind offhand is thermal expansion.
Unless you have a few billion dollars to spare, microprocessors really are impossible to manufacture. The tolerances required aren't even comprehensible unless you're in the business.
I have found that bad analogies outnumber good analogies by a huge margin. I don't usually pay much attention.
Samuel Dilworth
Senior Member
Come on. Look up anything written on dimensional metrology. This document would do for a start: https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=820983
Of course if you take the same milling centre and simply make something smaller with it, you won’t enjoy the naturally tighter tolerances of working with smaller dimensions. And maybe that’s all that’s done in some cases, with camera lenses. Plastics manufacturing has its own problems that I know little about.
I also don’t deny that some dimensions won’t neatly take care of themselves when making smaller lenses. The starting point for all of this was scaling lenses as a theoretical exercise. I think that question was answered before the needless snark arrived.
I think the reason you can't just scale arbitrarily is because lenses are not made the same way. I believe smartphone lens elements are largely made from molded plastic, while most ILC lens elements are made from glass (with only a few elements of plastic). So the tolerances between the two are completely different.
And it's not always viable to switch between methods because there maybe other considerations: for example a process viable for a large volume product like smart phones, may not necessarily be viable for the much smaller volume of ILC lenses; the machines that can make small elements may not necessarily be able to scale to larger ones (and vice versa).
Last edited:
Mark Ransom
Veteran Member
I don't know what business you're in, so maybe these things are self evident for you. I assure you they are not universally known. Until you said "dimensional metrology" I wouldn't have even known what to search for.
If you hand me a pair of scissors and two sheets of paper, and ask me to cut a large square from one and a small square from the other, I have no reason to believe the small square will have better tolerances than the large one.
Last edited:
RCicala
Leading Member
I think the problem is taking the scaling of a part and considering the scaling to a complex item with hundreds of parts is the same. Yes, you can make a small screw more accurately than a large bolt (at least assuming cost is the same pre weight).
Since modern lens elements are CNC machined, the best possible curvature is the same for small and (reasonably) large lenses, so the analogy fails. It fails worse when you consider a smaller lens may require a higher curvature.
Moving from part to complex machine, though, you're analogy may actually get reversed. The adjusting mechanisms for a micro 4/3 lens are the same thread and screw size as a full-frame lens (other than supertelephoto size). Yet in the smaller lens it's required to make a more accurate adjustment; which it can't.
The phone lens analogy also fails for different reasons. Phone lenses are entirely machine made, as similar to photo lenses as microprocessors are to circuits made with transistors and capacitors. The assembly line for phone lenses rejects anywhere from 30% to 50% at end of line testing and throws them away, which it can afford to do since manufacturing costs are tiny compared to a photo lens.
If smaller camera lenses were made using the highly automated methodology of phone lenses, then I'd totally agree with you, but they aren't. They're made like all the other camera lenses.
Last edited:
bobn2
Forum Pro
I think it's something of a generalisation to say that 'modern lens elements are CNC machined'. The majority of lens elements in modern lenses are roughed, ground and polished in the traditional way .The videos of lens plants at work show that quite clearly. It's one of the areas where there is a big difference between what happens in volume production, even low volume, and what happens in the specialist optical industry.
RCicala
Leading Member
Fair point and you are correct on spherical elements, which are the majority. I was thinking about aspheres and overgeneralized.
camerosity
Veteran Member
Actually fast lenses were first introduced just after WWII, in the form of the 5cm f1.5 Zeiss Sonnar and, since copyrights in Germany were made public after WWII, Nikon and Canon followed suit with their own derivatives.
Here is a good short history of fast lens design if you are interested:
https://www.rangefinderforum.com/forums/showthread.php?t=156630
I think it was a very good explanation... what could be added?
I think the point a lot of people miss at first (and perhaps the OP is one) is f stop is not aperture, nor is it a ratio to the image circle, it is the ratio of aperture to focal length.
The "a 25mm f1.0 lens has a 25mm aperture, and a 25mm f2.0 lens would have a 12mm aperture.
Which is the why when you halve the focal length you can halve the aperture, and keep the same f stop.
The point i consider not actually 'wrong' but at least debatable is
"It would also defeat the purpose of Micro Four Thirds that is to offer a cheaper and smaller alternative to camera systems of that size and weight."
I think this is one purpose....but not necessarily the only purpose. I am a proponent of a m43 system that includes smaller and lighter, but should not only consist of smaller and lighter. That is: one system for both when wanting smaller and lighter and allowing skipping the size advantage on the rare occasions when you do need more light.
Photo Pete
Veteran Member
Similar threads
- Locked
Keyboard shortcuts
- f
- Forum