[falconeye]

Those are not crops - they are full frame resized. They are actually 10:1.

I wondered if you would understand what I mean.

A pixel with a K20D/K-7 displays 5µm at 1:1 and 0.5µm at 10:1.

However, if you're downscaling to 1024 for the web, this becomes 23µm and 2.3µm resp.

You may still call it 10:1 but it is meaningless. A 3x2px photo at 10:1 would have 0.8mm wide pixels and stll you would call it 10:1 ...

[Blue]

Lets keep this discussion to the optics and sensor regarding 1:1 etc. The resizing topic should be another thread.

[falconeye]

This in no way changes the basic physics of the matter

which is M=image distance(>=focal length)/subject distance.

With my setup, you can reverse mount two 300mm lenses and have a whopping 1 feet working distance at 1:1

May be worth a second thought when capturing wild insects...

EDIT:
As newarts has pointed out, my above statement is false. The working distance still is the registration distance (about 45mm).

[falconeye]

The resizing topic should be another thread.

You can't. Not as soon as people start to post miniaturized sample photos of their results. That's a valid thing in general, but not when it comes to comparing the various macro methods.

E.g., using a Raynox lens may look like an acceptable approach only until you see a 100% crop of it.

[Blue]

You can't. Not as soon as people start to post miniaturized sample photos of their results. That's a valid thing in general, but not when it comes to comparing the various macro methods.

E.g., using a Raynox lens may look like an acceptable approach only until you see a 100% crop of it.

That's why for situations that require measurement a scale or method of calibration is included in scientific publications. It is still independent of optics.

Edit: This wasp is ~ 1.75mm in length (excluding antennae) regardless of whether the images is printed or displayed larger or smaller.

[yeatzee]

Fascinating thread!

To the OP: The Canon MP-E65 is basically a reversed mounted 35mm. No magic; you can get that cheaper yourself.

I think a short lens reverse mounted to a long zoom is your best bet. But please use a male-to-male reverse adapter and no tape. If you do use tape don't tell us and please don't post images of the setup. Images of lenses taped together where you invariably see a slight kink in the connection make me sick. I mean it.

Sorry for striking a nerve

(edit: Just so we are clear, Class A, I CAN'T use a male to male adapter because my sigma 105mm's filter ring is DENTED IN. I have no clue as to why taping a $30 lens to another would make you "sick." Kind of odd if you ask me...)

[pentaxman]

The full frame picture shows approximately 2.5mm. Since the sensor is 24mm across, this is 10:1 magnification. This photo is not cropped in any way. This is not meaningless - if you can make a subject 2.5mm in size take up the entire 24mm frame - this is 10:1 magnification.

I wondered if you would understand what I mean.

A pixel with a K20D/K-7 displays 5µm at 1:1 and 0.5µm at 10:1.

However, if you're downscaling to 1024 for the web, this becomes 23µm and 2.3µm resp.

You may still call it 10:1 but it is meaningless. A 3x2px photo at 10:1 would have 0.8mm wide pixels and stll you would call it 10:1 ...

[Marc Sabatella]

E.g., using a Raynox lens may look like an acceptable approach only until you see a 100% crop of it.

Not sure what you mean by that. Mounted to a lens that itself looks good at 100%, the images still look pretty darned good to me at 100%. Maybe not *quite* as good as they would without the Raynox, but of course, it's impossible to run a direct comparison since there is no subject distance that you can achieve with or without it. Hmm, except maybe if you used it on lens that already focused that close. I guess that would be an interesting comparison - macro lens with Raynox focused at infinity (meaning actual working distance of 6 inches or whatever) versus macro lens without Raynox focused at the same actual distance.

[Marc Sabatella]

Sorry, not true, when using a retrofocus wideangle lens reverse mounted, you always have a relative generous working distance.
It is the same distance that is required for the reflex mirror box.

OK, then I stand corrected - except I'd observe it's no longer functioning as a wide angle lens, but as a lens of a different focal length. The point being, the Canon lens in question does not (to my knowledge) employ magic to obtain 5:1 magnification at longer working distances than would be possible via other means.

[Class A]

I CAN'T use a male to male adapter because my sigma 105mm's filter ring is DENTED IN.

1. Don't dent your filter rings.
2. You are excused
3. Still, don't post images of taped together lenses.

I think you missed the slight humorous slant in my message.
I didn't see an image of tapped together lenses in this thread and didn't mean to address anyone specifically. I've just seen this method before (printed in a magazine ) and it just makes me cringe.

EDIT: I just went back to see whether you posted an image of lenses taped together and found your video. I'm really sorry that my flippant remark must have come across as a comment about your setup. It wasn't addressing you at all. Sorry!

BTW, I'm all for cost effective DIY in general. However, I think a male-to-male reversing ring should be worth the expenditure. If one of the lenses has a dented filter ring, that's of course a different story.

I'm very impressed by your macro shots of spiders and praying mantises, BTW. I saw a number of them and typically don't comment since I'm left speechless pretty much every time.

This wasp is ~ 1.75mm in length (excluding antennae)

Fascinating image!

[yeatzee]

Its the strong wind and a cheap POS tripod I used for a 10 sec. video's fault

and yes, I have no images of them taped

[glasbak]

OK, then I stand corrected - except I'd observe it's no longer functioning as a wide angle lens, but as a lens of a different focal length. The point being, the Canon lens in question does not (to my knowledge) employ magic to obtain 5:1 magnification at longer working distances than would be possible via other means.

It is still a wide angle, with the same focal length, but normally it projects the huge world onto a small sensor, exchange world and sensor, then it projects a sensor sized part of the world onto a huge image circle.
Put in this huge image circle a small sensor, and you get some serious magnification.

And it is indeed magic, why would the other name of a projector, which does essential the same, be 'laterna magica'

[falconeye]

The full frame picture shows approximately 2.5mm. Since the sensor is 24mm across, this is 10:1 magnification. This photo is not cropped in any way.

Of course, I know what magnification is.

Unfortunately, you didn't try to understand what I was trying to say. And I won't repeat. Everybody had his fair chance.

[newarts]

As you can see from this link, the working distance is not too bad. This is a 24mm reversed with a 2X macroconverter.

https://www.pentaxforums.com/forums/pentax-slr-lens-discussion/43370-ultra-cl...tml#post411715

Sorry, not true, when using a retrofocus wideangle lens reverse mounted, you always have a relative generous working distance.
It is the same distance that is required for the reflex mirror box.

The Working Distance for a camera lens mounted in a normal fashion cannot usually be predicted a priori because the actual physical location of the "front of the lens" is usually unknown.

However a reversed lens has a predictable working distance; it is:

Reversed_Lens_Working_Distance=Registration_Distance + Focal_Length/Magnification

This is true because camera lens construction is such that the location of the image cast by the lens at infinite magnification is exactly the registration distance (the distance from the lens' flange to the image sensor plane.) Hence the effective lens location in space of the "rear of the lens"is known and when used in reversed mode the front and rear of the lens switch physical locations.

Note that this relationship is valid for both long and short focal length camera lenses. It shows that the minimum clear working distance for a reversed lens is the Registration Distance (adjusted for any protrusion of the lens beyond the mounting flange.)

When a reversed lens is used as a close-up lens (ie. reversed in front of a normal lens,) the effective working distance of the combination will be the registration distance of the reversed lens.

The combination of a 300mm lens reversed in front of a 300mm lens will have a magnification of 1:1 and a working distance equal to the registration distance.

Dave

I have not seen this relationship in print before but am confident it is true. Its derivation is simple and based simply on the optical meaning of the registration distance.

[Blue]

I don't know what brand of tape Yeatzee is using, but if it will improve my images, I'm willing to try it.