Laowa 25mm F/2.8 2.5x-5x ultra macro lens

Have questions about the equipment used for macro- or micro- photography? Post those questions in this forum.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

enricosavazzi
Posts: 1474
Joined: Sat Nov 21, 2009 2:41 pm
Location: Västerås, Sweden
Contact:

Post by enricosavazzi »

Judging/guessing from the specifications, the 25 mm FL (assuming it remains constant at all magnifications, which is not necessarily true) and 40 mm working distance at 5x seem to indicate that the optical scheme is strongly asymmetric, probably comparable to a reversed retrofocus wideangle.

One of the characteristics that often (albeit not always) accompany strongly asymmetric optical schemes is that the pupil ratio (a.k.a. pupil magnification) is significantly different from 1. A larger entrance pupil allows the entrance pupil (and, more usefully, the front lens element) to be placed farther away from the subject and still produce the same effective aperture as a smaller entrance pupil placed closer to the subject. The lens designer can use a non-unity pupil ratio to increase the working distance, or conversely to decrease the diameter (and cost) of the front optical group. The first of the two choices seems to have been done in this case.

The pupil ratio also affects the effective aperture. There is, however, nothing to be gained or lost from a non-unity pupil ratio for what concerns the loss of resolution caused by diffraction: at the same effective aperture and image magnification, the loss is the same regardless of FL and pupil ratio.

PS - if memory serves right, the Macro Nikkor 35mm and 19 mm also have a non-unity pupil ratio, albeit used in the opposite way, i.e. smaller front element placed closer to the subject.
--ES

Macro_Cosmos
Posts: 1511
Joined: Mon Jan 15, 2018 9:23 pm
Contact:

Post by Macro_Cosmos »

It’s certainly an interesting lens. The fact they're putting 8 elements into it indicates that they're serious.
But if someone has an MPE-65, I don't see many selling it to buy this one. The automatic diaphragm makes such a difference.
I do own their 12mm and 105mm lenses, really good stuff. The build quality is great and the output is pleasing too. That being said, I agree. I don't see why anyone would be selling their MPE, unless the IQ of this new Laowa puts the MPE to shame.
my 200/4 macro lens is actually f/5.6 at 1:1
I wonder if that's a Nikon macro lens - if so, some report an effective aperture to the body, which can be misleading if you didn't know.
An f/4 lens, used at 1:1, has an effective aperture two stops smaller, ie f/8. The only way that might not be true is if the lens mechanically alters its aperture as you focus, which I've never seen.
Effective Aperture = Marked Aperture multiplied by M+1, where M= magnification.
So eg f/2.8 at 5:1 becomes Eff/16.8. There is a factor which has to go in for the PMR, which usually makes you worse off than you thought. Rik's post about Pupillary Magnification Ratio, & discussion, is HERE .
You're right, my mistake. I was meant to say 1:2. At 1:1, it's f/8. I no longer own the 200mm AFD micro-nikkor, still miss it. If Nikon comes out with a new one, I really doubt it's gonna be as good. The 105 VR in my opinion is a wannabe portrait lens that ends up being a master at nothing. I hope whatever replaces the 200mm micro doesn't go down that route...

For reasonable Effective Aperture, at higher M than 1:1, good lenses DO have wide apertures. The sharpest ones live on microscopes. Their FLs are 25-60mm as seems normal, but their apertures, which you have to work back from "NA", get towards f/1.0 at 5x (NA 0.4), or wider at higher M.
Yeah, I should have been a bit more concise. Most "consumer grade" macro lenses. I know someone with a Repro-nikkor 85mm f/1 lens which is optimised at 1:1. The images are just razor sharp even wide open and lacks any CA whatsoever. I have a 95mm Rayfact which isn't fast at f/4, but it produces very sharp results free of CA. When there's softness, it's me not using it correctly. https://flic.kr/p/23ivYFh
The lens is a special edition made by Nikon for line scanners. It was salvaged from some sort of industrial camera. As far as I know, it's not optimised for any particular magnification, hence it's fun to use.

The "Wide Angle" adjective applied to this lens, is really a mistake.
A lens has to be (effectively) (FL + (FL x M)) away from the sensor.
Consider a "simple" lens:
(If this seems wrong - draw it, remembering that a focused ray from the edge of the sensor goes straight, through the centre of the lens.)
Even at 1:1, a 25mm lens (ie when 50mm from the sensor) has the view angle of a 50mm lens at infinity. Hardly a wide angle, especially on APS.
At 2.5:1, it's 87.5mm. At 5:1, it's the same as a 150mm at infinity.

Granted a 50mm lens would have less angle of view, but you have so little in focus, you just don't see the difference; the perspective is so little changed. Some of the best lenses for high M macro are telecentric, which means they have the perspective of a lens with infinite FL. Again, you don't notice.
If you're stacking, then the program which does that can/will realign images so details line up, changing everything anyway.

The MPE shortens its FL as it focuses closer. The Laowa apparently does something too; it must, to give the working distance. Eventually someone will find out exactly what it does!


So far on these (small) points then, nothing's significantly advantageous. Also the claimed DOF difference is just not there.
But if the working distance is much the same as the MPE-65 at 5:1 that is a significant advantage, because of the smaller front end which makes lighting easier.
Yeah, it's not a wide angle lens as you have stated. The Laowa 15mm is a wide angle macro lens that goes down to 1:1, seen a lot of fun photos taken with it. I wonder how the 15mm achieves 1:1? A wide angle macro lens sounds like an oxymoron to most people.
As illustrated by Keith, the Laowa is also smaller which could be an advantage to some people. The lens collar is optional, I hope they make it supplementary, but then would a lens this tiny need a lens collar? Maybe for smaller bodies it serves a purpose by keeping the centre of gravity in the middle when the lens is configured to shoot 1:5?

https://flic.kr/p/23UVZs8
Thing I like about this collar is its arca swiss compatible foot. It triggers me that Canisonikon refuses to make collars like this. :roll: [/quote]

Lou Jost
Posts: 5944
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

"As far as I know, it's not optimised for any particular magnification, hence it's fun to use. "
I am afraid those Rayfact lenses are all highly optimized for particular magnifications. See their website for the details of each one.

ChrisR
Site Admin
Posts: 8668
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Post by ChrisR »

Agreed, the Laowa 15mm is very tempting. It's great to see a manufacturer making whacky lenses which do nice things, without bothering about VR and AF.
I wonder what aperture you need to use in order to see enough to focus, with that LED ring :smt104 .
Chris R

Keithcooper
Posts: 8
Joined: Sun Jan 28, 2018 12:04 pm
Location: Leicester
Contact:

Post by Keithcooper »

After lots of shots of the same thing a different magnifications/aperture (tedious) I decided to have a quick try out at 5x and f/2.8...

This is from a series of stacked shots of a small part of an old memory module from my MacBook pro.

Image

a 100% crop

Image

The working distance at 5x

Image


BTW, using Helicon for my stacking also lets me make this (based on the depth map from stacking)

https://youtu.be/FQq8v-L4uNg
bye for now
Keith Cooper

Lou Jost
Posts: 5944
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

Not bad. I like your Helicon animation which rotates through two axes. Rik, is there a way to make Zerene do that? I have only managed to create 1-axis rocking images. The two-axis approach would be better for scientific imaging.

ChrisR
Site Admin
Posts: 8668
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Post by ChrisR »

There is - In the Rocking window you can enter tilts in both X and Y.
Chris R

Lou Jost
Posts: 5944
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

Thanks Chris, I hadn't noticed that. This is going to revolutionize scientific illustration in botany.

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

ChrisR wrote:There is - In the Rocking window you can enter tilts in both X and Y.
But that won't do what Lou wants. It will just end up rocking across a tilted axis.

What Lou wants can be done with a batch script, one task per frame, that specifies the shifts using cos(t) and sin(t), t in 0 to 2*pi. (For circular rotation, the shift % should be 1.5 times larger for Y, if the image is 1.5 times wider than high.)

With no other parameters specified, and operating on %CurrentProject%, the same script can be used with any project.

I've posted a sample script for 12 frames with +-1.5% in X and +-2.25% in Y, at https://zerenesystems.com/temp/ZereneBa ... ocking.xml

The body of the script looks like this:

Code: Select all

<?xml version="1.0" encoding="UTF-8"?>
<!-- Generate 12 frames, circular rocking with +- 1.5% in X, +- 2.25% in Y -->
<ZereneStackerBatchScript>
  <WrittenBy value="custom script generator" />
  <BatchQueue>
    <Batches length="1">
      <Batch>
        <Sources length="1">
          <Source value="%CurrentProject%" />
        </Sources>
        <ProjectDispositionCode value="101" />
        <Tasks length="12">
          <Task>
            <OutputImageDispositionCode value="1" />
            <Preferences>
              <SkewSequence.FirstImage.MaximumShiftXPct value="1.5" />
              <SkewSequence.FirstImage.MaximumShiftYPct value="0" />
             <SkewSequence.NumberOfOutputImages value="1" />
              <SkewSequence.Selected value="true" />
            </Preferences>
            <TaskIndicatorCode value="1" />
          </Task>
          <Task>
            <OutputImageDispositionCode value="1" />
            <Preferences>
              <SkewSequence.FirstImage.MaximumShiftXPct value="1.29903810567666" />
              <SkewSequence.FirstImage.MaximumShiftYPct value="1.125" />
             <SkewSequence.NumberOfOutputImages value="1" />
              <SkewSequence.Selected value="true" />
            </Preferences>
            <TaskIndicatorCode value="1" />
          </Task>
          ... another 10 tasks ...
        </Tasks>
      </Batch>
    </Batches>
  </BatchQueue>
</ZereneStackerBatchScript>
Animated, those 12 frames look like this:

Image

Give that script a shot and see if it does what you want.

I can put a switch in the GUI to do circular rocking if that's what people want to do.

--Rik

Lou Jost
Posts: 5944
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

Thanks very much for that, Rik! The 2-axis rocking stereo fully utilizes the information in the stack, so for scientific work, it is considerably more informative. I would love to see a switch in Zerene that makes this easily accessible for all users who might want it. I know that in my field this will be in high demand once I start publishing examples.

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

"2-axis rocking stereo", eh?

So are you thinking to make one circular rotation with say X in -2.5 to +0.5, and another with X in -0.5 to +2.5, thus giving a 3% rotation with 2% disparity between?

--Rik

Lou Jost
Posts: 5944
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

No, I was thinking of combining displacements in the x direction with displacements in the y direction, just as you did.

ChrisR
Site Admin
Posts: 8668
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Post by ChrisR »

ChrisR wrote:
There is - In the Rocking window you can enter tilts in both X and Y.

Rik wrote:
But that won't do what Lou wants. It will just end up rocking across a tilted axis.
I was assuming Lou would see how to use multiple combinations!

This was a crude attempt I posted in 2009 (gulp)

The awful dust trail gives a good idea what's going on:
Chris R

JH
Posts: 1307
Joined: Sat Mar 09, 2013 9:46 am
Location: Vallentuna, Stockholm, Sweden
Contact:

Post by JH »

rjlittlefield wrote: I can put a switch in the GUI to do circular rocking if that's what people want to do.
--Rik
Yes please :D

For ”free” movements with ”simple” shapes it is possible to use Zerenes depthmap to create and scale a 3D object in Adobe PS. I did this video to test what can be done with Zerene and Adobe PS

https://youtu.be/LLnxE-HUP-8

But for complex subjects like the flower in Riks example I prefer ”real” rocking because it reveals parts of the flower that is hidden in a 3D modell that is created from one depth map.

Best regards
Jörgen Hellberg
Jörgen Hellberg, my webbsite www.hellberg.photo

JH
Posts: 1307
Joined: Sat Mar 09, 2013 9:46 am
Location: Vallentuna, Stockholm, Sweden
Contact:

Post by JH »

Here is a video that shows how ”rocking” reveals hidden areas. The second clip in the video:

https://youtu.be/Ah9qZtaRQsE

Best regards
Jörgen Hellberg
Jörgen Hellberg, my webbsite www.hellberg.photo

Post Reply Previous topicNext topic