Ad
Ad
Ad
Pages: « 1 [2]   Bottom of Page
Print
Author Topic: How good is focus-stacking in Focus?  (Read 4407 times)
Doug Peterson
Sr. Member
****
Offline Offline

Posts: 2838


WWW
« Reply #20 on: April 27, 2013, 10:46:11 AM »
ReplyReply

This is why I'm shooting with the Hasselblad. Focusing a DSLR compared to a large format camera is much more exact in this regard.

That depends on what large format camera you are comparing to.

Old film cameras have very low focusing resolution when used for high-mp digital backs.

But for instance an Arca Swiss RM3Di or Factum as the front standard on an Arca view camera has FAR finer focusing resolution than any dSLR, and has the benefit of distinct focus numbers and ticks for stacking. Though of course it has no remote focusing capability.
Logged

DOUG PETERSON (dep@digitaltransitions.com), Digital Transitions
Dealer for Phase One, Mamiya Leaf, Arca-Swiss, Cambo, Profoto
Office: 877.367.8537
Cell: 740.707.2183
Phase One IQ250 FAQ
Esben
Full Member
***
Offline Offline

Posts: 182


WWW
« Reply #21 on: April 27, 2013, 11:32:44 AM »
ReplyReply


I would think that the focusing mechanism on this camera would be very workable, but you have to cock the shutter manually 80 times with the risk of moving the camera just a fraction each time.
Helicon Focus is somehow tolerant to very small deviations but I would think this camera would introduce just another factor in the equation.
The RM3Di would have tilt and shift, correct?
Even if using a Press shutter, how exact is it? If it can't repeat itself 100%, it will introduce banding in smooth surfaces. I see this with my ProFoto flashes.
Logged

Digital Tech NYC // LA
www.captureforce.com

phone. 212 203 6427
Esben
Full Member
***
Offline Offline

Posts: 182


WWW
« Reply #22 on: April 27, 2013, 11:37:18 AM »
ReplyReply



When will we see an electronic global shutter built in to the DB?  Then we would be able to eliminate nearly all moving parts in a camera and who knows, maybe be able to freeze motion 1/20,000 sec.   Smiley
Logged

Digital Tech NYC // LA
www.captureforce.com

phone. 212 203 6427
henrikfoto
Sr. Member
****
Offline Offline

Posts: 708


« Reply #23 on: April 27, 2013, 11:47:54 AM »
ReplyReply

Do you need that many shots if you use f.8or f.11?
Is your back 80mp?
Logged
Esben
Full Member
***
Offline Offline

Posts: 182


WWW
« Reply #24 on: April 27, 2013, 11:57:25 AM »
ReplyReply



I'm shooting with the Aptus 75S, 33Mpix.
Most of the images are composed  into a final image so the 33Mpix works fine most of the time.

I most often shooting with an aperture between F.10 to F.13

I would say each their own but this is what works for me.

E
Logged

Digital Tech NYC // LA
www.captureforce.com

phone. 212 203 6427
FredBGG
Sr. Member
****
Offline Offline

Posts: 1651


« Reply #25 on: April 27, 2013, 12:17:13 PM »
ReplyReply

That depends on what large format camera you are comparing to.

Old film cameras have very low focusing resolution when used for high-mp digital backs.

But for instance an Arca Swiss RM3Di or Factum as the front standard on an Arca view camera has FAR finer focusing resolution than any dSLR, and has the benefit of distinct focus numbers and ticks for stacking. Though of course it has no remote focusing capability.

While the Arca Swiss does have a very low pitch helicoid focusing mechanism requiring rotating the focus ring quite a bit for even fine adjustments
you will never get the precision, speed and accurate repatability that you will get using computer controlled focus stacking.
Stating that the Arca camera has finer focusing :resolution" than any dSLR is misleading. There is no problem doing a 100 step focus stack with the
focus motor of recent recent better DSLR lenses. Doing a repeatable and accurate stack with many steps manually with a tech camera has many limitations. Just cocking the shutter shifting the focus every time is a problem. While the focus ring has many markings there are not enough
for a stack with many steps. Depending on the subject many steps may be needed.
Also repeatability is important to several macro techniques such as multiple lighting passes.

One lighting pass for detail and one lighting pass for the style you want.

Non computer (or mobile device) controlled stacking is very limiting.
Logged
Dustbak
Sr. Member
****
Offline Offline

Posts: 2371


« Reply #26 on: April 27, 2013, 01:11:49 PM »
ReplyReply

Henrik,
I mark off with tape on the lens, start and stop points as when pulling focus for motion



Dustbak,
Do you know, or would you have the possibility to test if the focus adjuster in Phocus is fine enough to shoot around 80 frames of a watch, with focus starting from around 10cm from the lens, and ending around 12-13cm from the lens?
Does the focus tool work when using a space-adaptor?


E


It would be most certainly possible if you use the smallest amount of adjustment (control key + the adjustment key + or - ). It works with the extention rings too, not with the HTS where you loose AF.
« Last Edit: April 27, 2013, 01:15:50 PM by Dustbak » Logged
JoeKitchen
Sr. Member
****
Offline Offline

Posts: 719



« Reply #27 on: April 27, 2013, 01:16:03 PM »
ReplyReply

Why not use a bellows camera that has tilt and swing on both standards?  

I seem to think that the old school methods will in the end be much easier and faster, and probably cheaper if you factor in the amount of time you spend doing this.  

Disclaimer: I do not do macro photography; this post is a question, not a statement. 
« Last Edit: April 27, 2013, 01:26:52 PM by JoeKitchen » Logged

Joe Kitchen
www.josephmkitchen.com

"Photography is one percent inspiration and ninety-nine percent moving furniture."  Arnold Newman
"Try not to be just better than your rivals and contemporaries, try to be better than yourself."  William Faulkner
Dustbak
Sr. Member
****
Offline Offline

Posts: 2371


« Reply #28 on: April 27, 2013, 03:17:35 PM »
ReplyReply

Because sometimes the old ways of using a TC is not enough. I have never been able to do the same with my Rollei X-Act2 as I can with focus stacking (that is when you need lots of DoF).

Auomated a stack of many images is even as fast as using a TC, in my case it is anyway Smiley
Logged
Doug Peterson
Sr. Member
****
Offline Offline

Posts: 2838


WWW
« Reply #29 on: April 27, 2013, 06:04:04 PM »
ReplyReply

While the Arca Swiss does have a very low pitch helicoid focusing mechanism requiring rotating the focus ring quite a bit for even fine adjustments
you will never get the precision, speed and accurate repatability that you will get using computer controlled focus stacking.
Stating that the Arca camera has finer focusing :resolution" than any dSLR is misleading. There is no problem doing a 100 step focus stack with the
focus motor of recent recent better DSLR lenses. Doing a repeatable and accurate stack with many steps manually with a tech camera has many limitations. Just cocking the shutter shifting the focus every time is a problem. While the focus ring has many markings there are not enough
for a stack with many steps. Depending on the subject many steps may be needed.
Also repeatability is important to several macro techniques such as multiple lighting passes.

I'm assuming you've never actually used an RM3Di.

The resolution is in fact finer focusing and is both more repeatable and more accurate than any computer controlled dSLR lens focus. It is for sure not as fast - which you correctly note. But given the length of the entire process (prep/shooting/stacking/retouching) the increase in time would be very small, and the use of tilt may offset that increase in time.

I love that you feel comfortable using a phrase like "While the focus ring has many markings there are not enough for a stack with many steps." It just shows you have never used an RM3Di at all, let alone for the application for which you are providing such conclusive advice.

At a magnification of 1:1 (approximate watch face magnification for full-frame digital back) at a steep angle to the watch face, dividing the watch face into 80 focus-step shots there are around 10 times more markings (or 10 times the focus resolution) which is required.

Maybe you could consider limiting your advice to systems you've actually used?
« Last Edit: April 27, 2013, 06:15:07 PM by Doug Peterson » Logged

DOUG PETERSON (dep@digitaltransitions.com), Digital Transitions
Dealer for Phase One, Mamiya Leaf, Arca-Swiss, Cambo, Profoto
Office: 877.367.8537
Cell: 740.707.2183
Phase One IQ250 FAQ
Doug Peterson
Sr. Member
****
Offline Offline

Posts: 2838


WWW
« Reply #30 on: April 27, 2013, 06:07:51 PM »
ReplyReply

I would think that the focusing mechanism on this camera would be very workable, but you have to cock the shutter manually 80 times with the risk of moving the camera just a fraction each time.
Helicon Focus is somehow tolerant to very small deviations but I would think this camera would introduce just another factor in the equation.
The RM3Di would have tilt and shift, correct?
Even if using a Press shutter, how exact is it? If it can't repeat itself 100%, it will introduce banding in smooth surfaces. I see this with my ProFoto flashes.

No need to cock the lens. For this application you'd use a Schneider Electronic Shutter which is made for the sort of ultra-repeatability this application calls for.

You would, for sure, want a sturdy camera stand to prevent translating helical movement into any lateral/vertical/rotational movement. But the built-in full size arca rail mount and overal construction quality will help that quite a bit.

The use of tilt sometimes will do nothing to help this application. But sometimes it can greatly reduce the number of shots required.
Logged

DOUG PETERSON (dep@digitaltransitions.com), Digital Transitions
Dealer for Phase One, Mamiya Leaf, Arca-Swiss, Cambo, Profoto
Office: 877.367.8537
Cell: 740.707.2183
Phase One IQ250 FAQ
Esben
Full Member
***
Offline Offline

Posts: 182


WWW
« Reply #31 on: April 27, 2013, 07:57:46 PM »
ReplyReply



hmm, this camera would/could be pretty cool. I like that I would be able to do tilt and shift. Would it work with the new Rodenstock eShutters as well?
The DSLR still has a trick up its sleeve. When working with anything as reflective as watches, I need to be able see through the lens in order to see what each reflector does. The live-view on the Aptus is not god enough for a pleasant workflow, and even using an IQ, having the live-view activated for hours each day doesn't sound like a good idea. I know the alternative would be a sliding adaptor but I've never really been too fond of the. The camera sounds pretty sweet (and expensive) though.


I have a Foba Alpha to help me standing still   Smiley

E
Logged

Digital Tech NYC // LA
www.captureforce.com

phone. 212 203 6427
BartvanderWolf
Sr. Member
****
Online Online

Posts: 3755


« Reply #32 on: April 27, 2013, 08:52:57 PM »
ReplyReply

When working with anything as reflective as watches, I need to be able see through the lens in order to see what each reflector does.

Hi,

Yes, that may be an issue with all suggested approaches. The only way to avoid that, is by moving the sensor plane (and keep the lens stationary), which would keep the entrance pupil at a fixed perspective point.

An alternative not mentioned so far (although it wouldn't really solve the reflection consistency, just transform it to a light tunnel), is moving the subject while keeping the lighting, lens, and camera fixed. A Stackshot rail with the subject mounted on a platform, would allow the required positioning accuracy (slowing down the acceleration of the Stackshot could reduce the risk of vibration having an impact on the subject's positioning).

Cheers,
Bart
Logged
FredBGG
Sr. Member
****
Offline Offline

Posts: 1651


« Reply #33 on: April 28, 2013, 03:19:45 PM »
ReplyReply

I'm assuming you've never actually used an RM3Di.

The resolution is in fact finer focusing and is both more repeatable and more accurate than any computer controlled dSLR lens focus. It is for sure not as fast - which you correctly note. But given the length of the entire process (prep/shooting/stacking/retouching) the increase in time would be very small, and the use of tilt may offset that increase in time.

I love that you feel comfortable using a phrase like "While the focus ring has many markings there are not enough for a stack with many steps." It just shows you have never used an RM3Di at all, let alone for the application for which you are providing such conclusive advice.

At a magnification of 1:1 (approximate watch face magnification for full-frame digital back) at a steep angle to the watch face, dividing the watch face into 80 focus-step shots there are around 10 times more markings (or 10 times the focus resolution) which is required.

Maybe you could consider limiting your advice to systems you've actually used?

OK so in the case you mentioned you have enough markings. Not sure that will be the case in all situations, however
the real issue is sitting there and going through the 80 steps without getting it wrong, actually being able to see the markings depending on
shooting position of the camera and touching the camera 80 times and not moving the camera. In macro situations even the smallest movement is an issue.


Also regarding watches.... haven't nearly all watch makers moved away from macro photos to 3D rendering?

http://www.plus421.com/projects/breitling/3d-graphics



« Last Edit: April 28, 2013, 04:03:47 PM by FredBGG » Logged
FredBGG
Sr. Member
****
Offline Offline

Posts: 1651


« Reply #34 on: April 28, 2013, 03:23:35 PM »
ReplyReply

Why not use a bellows camera that has tilt and swing on both standards?  

I seem to think that the old school methods will in the end be much easier and faster, and probably cheaper if you factor in the amount of time you spend doing this.  

Disclaimer: I do not do macro photography; this post is a question, not a statement. 

Tilt does not increase depth of field, it only inclines the plane of focus.
If you have a relatively flat subject that is inclined this will work.
However there are limitations as to how much tilt cam be used with digital sensors before you get unmanigable
color shifts.
Logged
henrikfoto
Sr. Member
****
Offline Offline

Posts: 708


« Reply #35 on: April 28, 2013, 04:06:54 PM »
ReplyReply

OK so in the case you mentioned you have enough markings. Not sure that will be the case in all situations, however
the real issue is sitting there and going through the 80 steps without getting it wrong, actually being able to see the markings depending on
shooting position of the camera and touching the camera 80 times and not moving the camera. In macro situations even the smallest movement is an issue.




I would be more worried about having to touch the setup 80 times in a very fine-tuned macro-setup.
I am sure this must be much better to do from the software. As far as I know only Hasselblad and
Sinar M has this possibillities. I use the Sinar M. A great camera with great lenses, but the software
is not very good and crashes a lot..

Henrik
Logged
ErikKaffehr
Sr. Member
****
Online Online

Posts: 7642


WWW
« Reply #36 on: April 28, 2013, 04:30:34 PM »
ReplyReply

Hi,

I would consider using a focusing rail, like this one: http://www.cognisys-inc.com/stackshot/stackshot.php .

Best regards
Erik



I would be more worried about having to touch the setup 80 times in a very fine-tuned macro-setup.
I am sure this must be much better to do from the software. As far as I know only Hasselblad and
Sinar M has this possibillities. I use the Sinar M. A great camera with great lenses, but the software
is not very good and crashes a lot..

Henrik
Logged

Esben
Full Member
***
Offline Offline

Posts: 182


WWW
« Reply #37 on: April 28, 2013, 07:02:27 PM »
ReplyReply


Also regarding watches.... haven't nearly all watch makers moved away from macro photos to 3D rendering?



Good question. It seems like everything is moving toward CGI. I still think there're some of these luxury manufactures who would like to show images of their watches without having to include the "Photo may have been enlarged, enhanced and/or computer rendered". Buying a Rolex is very much about buying an old world time pice. This is of course just me thinking, but I do know that JWT refused to have the glass removed or the date magnifying area Photoshopped in in post. This was a couple of years ago and I don't know if this is still the case.

E

Logged

Digital Tech NYC // LA
www.captureforce.com

phone. 212 203 6427
8x10 user
Sr. Member
****
Offline Offline

Posts: 303


« Reply #38 on: May 11, 2013, 11:02:41 PM »
ReplyReply

You can do that. You have to be careful not to move the camera when you change the focus. Having automation does sound nice but view camera movements are also special and give you the ability to control perspective in a better way then photoshop. Its really about what your trying to do the best way for you.

I heard about people doing massive focus stacking projects with an 8x10 view camera!



Just an idea:
What about stacking using a LF camera with the rail just attached to the front standard and
the bach standard in fixed position. That should give a minimum of focus shift and using the
best lenses on the market this should be good?

Is it hard to make?
Logged
Pages: « 1 [2]   Top of Page
Print
Jump to:  

Ad
Ad
Ad