Pages: [1] 2 3   Go Down

Author Topic: Best techniques for multi-row panoramas using tilt/shift lens  (Read 22486 times)

the_ether

  • Newbie
  • *
  • Offline Offline
  • Posts: 25

I'm planning a shoot of a landscape similar to the following picture whereby there is a long field going off into the distance. The printed result will be very large, say, 10 feet square. To create it I will need to stitch together several rows and several columns of shots.

A big issue I'm concerned about is depth of field, as it would have been for the photographer in the above example. I could use focus stacking but am concerned about the quality given that the final, printed picture will be so large. I suppose the best thing would be to hire a tilt / shift lens, but how do I use that when photographing the different rows I need? Do I change the lens tilt for each row? Do I simply set the focusing for the centre of the entire field of view and then keep it static when taking all the shots I need?

I assume I need to get special guides to attach to my tripod head so that I can move the camera appropriately through the shots I need for the rows and columns but I wasn't sure how such movements would work with a tilt / shift lens. Any advice would be much appreciated.
Logged

David Campbell

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 63
    • David Campbell Photography
Re: Best techniques for multi-row panoramas using tilt/shift lens
« Reply #1 on: March 19, 2012, 06:35:59 AM »

I am sure Bernard (the photographer that took the image you have posted as a sample) can give us some tips.
He is active on LuLa and may see this post.

For my panoramic stitches, I use the Really Right Stuff Panoramic head (old and latest one). If you are using a large telephoto for example a 300/2.8 you will need the more sturdy latest version. These heads make panoramic images so easy.

I have not tried multi-row using a tilt shift but have done single row panoramic images with a 24TS and camera body in portrait orientation with excellent results.

In terms of focus staking, I have not had good results, mainly due to the the extra time and care it takes to get it right. Often the lighting condition has changed etc. Instead I use the hyperfocal.
Depending on the difference between you near and far subjects, you maybe able to use different focus for each row so that enough of the over lapping area in the stitch is acceptably sharp so you don't notice it in the final stitch.

If you can get up high (and far) and shoot down on the scene that will help with your depth of field/long lens concern.

Another option is to use the highest resolution camera you can get your hands on (buy/rent) so you can use a wider lens, you will still get detail and will not have to take as many shots to get the resolution for your desired print size. So maybe you will only need 1 row?

I will be keen to see what others suggest. Hopefully, Bernard shares some of his techniques.

Cheers,

David



Tony Jay

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2856
Re: Best techniques for multi-row panoramas using tilt/shift lens
« Reply #2 on: March 19, 2012, 07:32:24 AM »

Actually there are a couple of options with regard to focus stacking.

Helicon Focus markets a utility piece of software that computes and controls the focus points required once it knows the near and far focus points of the scene in question.
In contrast to manually fiddling around this system will take shots as rapidly as the camera settings, buffer, and write speed of the memory card will allow.
The downside is that one is forced to shoot tethered to a computer, usually a laptop in the field, and this introduces its own layer of practical challenges. These can be overcome if one allows for the fact that shooting multishot panoramas is not an impulse exercise in the first place.

The second option is a remote release, the brand name escapes me for the moment despite the fact that I own it, that is also programmable and allows an essentially automated sequence of shots at appropriate focus points.

Both these utilities also do HDR shots where again the user can input the exposure intervals required.

I am not at home at the moment so if anyone one wants more info just post on the thread and I will get back to you in the next day or so with the details.

Regards

Tony Jay
Logged

the_ether

  • Newbie
  • *
  • Offline Offline
  • Posts: 25
Re: Best techniques for multi-row panoramas using tilt/shift lens
« Reply #3 on: March 19, 2012, 01:48:50 PM »

Quote
Helicon Focus markets a utility piece of software that computes and controls the focus points required once it knows the near and far focus points of the scene in question.

Thanks, but I don't see how that would help in my case whereby I need to stitch together several rows.

To expand on my original post, these are some issues I am pondering over:

- use a 50mm or 85mm lens rather than a wide angle so as not to have distortions at edges which may cause problems when stitching the columns + rows together;

- I probably don't need an L-frame on my head for this landscape scene as there aren't any verticals in the scene; I'm assuming I won't have parallax issues if I simply rotate and tilt the whole camera (?);

- thinking more about the tilt/shift lens, I think what I should do is to set the tilt such that when focusing the bottommost part of the bottom row and the topmost part of the top row will be in focus. I then do not change the focus or lens tilt and take all the shots I need for each row for stitching together in Photoshop. By initially setting the correct tilt that covers all the rows, the wedge of in-focus area will have been set and tilting the camera (rather than the lens) to move from one row to the next should not cause any adverse issues (?).

Logged

Tony Jay

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2856
Re: Best techniques for multi-row panoramas using tilt/shift lens
« Reply #4 on: March 19, 2012, 07:57:40 PM »

Your fixation on TS lenses as being the magic solution here will not work.
Your premise about a fixed tilt as described (particularly at longer focal lengths) may well result in areas of each shot being out of focus.

It is possibly true as well that focus stacking may not give the technical results you seek either.
However, my suggestion to use those utilities I mentioned is valid because not only will they compute what is required to have everything in focus the utilities will also control the control the camera and take the shots much much faster than you could trying to achieve the same thing manually.

If you really want to use a TS lens to attempt your goal you would be better off using the shift facility (that is not a magic bullet either) to position for each row rather than try to tilt to achieve infinite depth of field.

If I were in you situation (still experimental) I would have a much more open approach.
You may have to shoot, and reshoot many times to fine-tune the technique required for the sort of panoramic image you trying to attempt.

Regards

Tony Jay
Logged

Ellis Vener

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2138
    • http://www.ellisvener.com
Re: Best techniques for multi-row panoramas using tilt/shift lens
« Reply #5 on: March 19, 2012, 08:57:18 PM »

"- use a 50mm or 85mm lens rather than a wide angle so as not to have distortions at edges which may cause problems when stitching the columns + rows together."

I regularly use both 50mm and 85mm lens when shooting stitched pans but not for that reason, instead I use these longer lenses to increase final resolution in pixel dimensions. I do not hesitate to use shorter focal length lenses, down to 14mm lens on a "FF" 35mm based DSLR  if that is what the situation calls for.  The edge distortion problem you cite as a possible problem is solved by having sufficient overlap between frames.

"- I probably don't need an L-frame on my head for this landscape scene as there aren't any verticals in the scene; I'm assuming I won't have parallax issues if I simply rotate and tilt the whole camera (?);"

L brackets are very worthwhile! You end up with greater vertical resolution withthe camera in "portrait" orientation but at the price of needing shooting more frames to capture your desired horizontal sweep. You also need a "nodal slide" to position the lens' entrance pupil (AKA "nodal point") in the rotation axis.

"- thinking more about the tilt/shift lens..."

Tony jay  is basically right. Once you shift or tilt a lens you have moved the entrance pupil and this will create all sorts of parallax and spatial rendering problems when it is time to line up frames.*  A mount that lens you move the camera body instead of the lens solves the problem but there are only a couple of very expensive alternate lenses that allow you do this directly with a DSLR.  if you are photographing a relative narrow angle of view (like 100˚) than there is also "the strap your camera to the back of a larger format camera with rear movements" solution but that has limitations as well.

*Keith Cooper of http://www.northlight-images.co.uk/  has worked out a stitching method for using shifted (not tilted) lenses using Autopano giga software. Look towards the bottom of http://www.northlight-images.co.uk/reviews/software/autopano-giga_2-6.html presumably the same approach will work in PTGui pro but I have not tried it.
Logged
Ellis Vener
http://www.ellisvener.com
Creating photographs for advertising, corporate and industrial clients since 1984.

Tony Jay

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2856
Re: Best techniques for multi-row panoramas using tilt/shift lens
« Reply #6 on: March 20, 2012, 05:16:46 AM »

GigaPan Epic Pro, if you can afford it, will combine the ability to do multirow panoramic shooting with multiple shots at each location for either HDR, focus stacking or other reasons.

This is an expensive option but can work - again this option is automated.

Tony Jay
Logged

Paul2660

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3672
    • Photos of Arkansas
Re: Best techniques for multi-row panoramas using tilt/shift lens
« Reply #7 on: March 20, 2012, 11:29:58 AM »

A few other thoughts, this is what works for me, there is really no "right answer" you have to test out various solutions and see which works best for you.

If you are looking for a true pano, I would define a pano at least a 3:1  ratio print, the the use of TS-E Canon or Nikon lens will not really get there as even at the max of 12mm of shift (newer 17mm and 24mm) won't get there at least from my work with Panos.  You can get a very good high res 3:2 ratio print using the camera in the vertical and taking 3 images, but a true pano you will run out of image, even if you shift in landscape mode.  You can mount the camera on an L Bracket with a Pano head like the one from Really Right Stuff, and take three sets of shifts, bottom, middle and top, but you are still at 12mm max shift.  This can create a very high res image BTW, but not a true Pano unless you crop into the final image.  Shifting only with no tilt applied.  Note also with the new Canon the light fall off at Max shift is pretty harsh most times for me.  I have recently started to take a LCC on each shift as I would with a tech camera, then then use Capture One's LCC tool to correct for light fall off, which is a very nice solution, just takes a bit more time.   Also I have found that with at least the Canon 24mm TS-E II, at 12mm max shift you start to see a good bit of detail smearing from 10mm on, even at F11.   From 2003 to 2008 90% of my work was stitched with either a Canon/TS-E or Canon Zork solution as I wanted max resolution for large prints.  Now with cameras starting at 24MP that need is not as great, but I still stitch.

I feel you need at least 20mm of shift in either direction to get to where the image is going to be close to 3:1, and I have yet to find any 35mm TS lens combination that will get there.  You can use a Zork Adapter, (allows for a medium format lens with larger image circle to be used on a 35mm camera) and then get to 18mm of shift and depending on the medium format lensyou pick you can get a very good 3:1 image.  With the use an L bracket and a pano head mounted to the L bracket you can get a least 3 row with the Zork and stitch all that into 1 image, total 9 images 3 across top, 3 across middle, 3 across the bottom.  And with the Zork it's possible to setup the rig so you shift the camera not the lens.  Shifting the lens as with a TS lens on a Canon can still bring parallax into play.
One great solution is the Pentax 645 35mm or 45mm FA lenses (unfortunately hard to find them now) and a zork, with a Canon.  I easily could take a full 18mm of shift to each side since I had the larger image circle of the 645 lens to work with.  

Tilt can help, but I have found only on a 3 stitched solution Often I will use tilt on a 3 stitch image to help pull in the foreground as most times the top of the image will be sky and if I lose a bit offocus on the top due to tilt, no loss.  However I am not doing a multi-row solution as you are looking for.  As has already been mentioned adding tilt to a multi row will cause problems.  

Taking the time to get your specific lens nodal point and then using a setup where you can pan around the nodal point will get you some great pano's but not multi row as the camera needs to be levelor parallax will really cause problems in the creation of the final image.  With CS5 or PTgui, many times slight parallax can be resolved, but it's very dependent on the subject.

I also agree that the Gigapan solution can produce some really amazing images.  I happened across a photographer last fall using one with a 5D MKII and he sent me a final file of a shot and the resolution was really amazing.  However in many situations here in Arkansas, the light is going to change or wind will blow before you can complete a full series.   But if you can a complete series the results are very impressive.  

Worse case scenario, hand hold your camera and try to bracket visually, you might be able to get 9 frames hand held.  And with practice even more.  

www.reallyrightstuff.com has written a ton of material on working out nodal points, and stitching and they make some very nice equipment.  Most of the literature in on their website and it's a great place to start.   I can't remember zork USA website, but I looked them up the other day for another photographer and they are still around.  Zork is made in Germany so it take a bit of time to get unless it's in stock, but all their equipment I used was very well made and still works to this day.

Paul
« Last Edit: March 20, 2012, 06:52:19 PM by Paul2660 »
Logged
Paul Caldwell
Little Rock, Arkansas U.S.
Photography > http://photosofarkansas.com
Blog> http://paulcaldwellphotography.com

fike

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1390
  • Hiker Photographer
    • trailpixie.net
Re: Best techniques for multi-row panoramas using tilt/shift lens
« Reply #8 on: March 20, 2012, 02:18:01 PM »

I spent a bit of time working with stitching, tilting and shifting in the same images. It is difficult to make work well. For most applications, it isn't worth the trouble.  If this is your first panoramic project, I would recommend against being so aggressive. Tilt shift lenses are amazing, but they take some learning to get good at using them in extreme ways like this.  Here is one thread discussing my journey to tilt, shift, and pan for a panoramic image. http://www.luminous-landscape.com/forum/index.php?topic=42605.0

Nobody has mentioned your target image size, so I will.  10 feet square is very large.  (Do you have a plan for printing it?) Depending on the intended viewing distance, you will need somewhere between 150 DPI and 300 DPI on the output print.  My quick calculations show that you would need between 18,000 and 36,000 pixels on a side (over a gigapixel image).  I have made images this large.  They are difficult to work with in photoshop.  I have 5' x 6' image at well over 300 DPI and using PT Assembler it wouldn't stitch as a single process. I had to stitch the left half of the image and then the right half and then the two together.  It looks fabulous, but it's composition doesn't require great depth of field and as a result the sharpness is great throughout.  http://marcshaffer.net/fine-art-panoramics-for-sale/very-large-images/bindex.html#zion_wall

Tilt shift lenses won't have enough travel (shift movement) to capture enough images to get the resolutions you need with a flat stitch. Generally shift lenses can do about two flat stitched frames in the landscape or three in the portrait orientation  (give or take).  In the landscape orientation That would get you about 10,000 or 15,000 pixels in one row.  Then you are left with a second row that will need to be made somehow.

With an image that is as simple as the one you mentioned, you could consider shooting several high-depth-of-field rows of images with lots of overlap and slightly different focus points gradually moving out into the scene.  This is sort of a simplified focus stacking except that you manually blend the layers.  I've done this.  It is finicky, but it can work particularly if you don't have trees or branches in the intervening area.

I have also done some peculiar experiments with tilt shift lenses where the tilt actually extends the focus plane vertically instead of horizontally.  By doing this you end up with a narrow slice of focus through the center of your image that runs from the foreground to the background.  This then necessitates a very large amount of overlap between image frames. It was flaky, but cool.   

Using a tilt shift lens, don't mix tilting, shifting and panning into one panoramic operation. It is impossible to keep distortion and parallax from being a problem...that is if parallax is likely to enter the equation at all.  The image you showed wouldn't have any substantial parallax issues.  Where you run into problems is when you have trees or power lines that are part of the composition.  They are impossible to line up accurately without a well calibrated pano head and lens.

Don't try to rent a tilt-shift lens to get the shot.  It will be frustrating and it won't happen in a day.  Use a good prime lens and stitch in the vertical orientation. Don't worry too much about a full pano head. Remember that the TOP of your ballhead needs to be level and then you need to rotate around the top, not the bottom of the ballhead where most have the rotational control. If you can't do that, level the base of the tripod as well as the top of the ballhead.  Your stitching process will be easier if you level the rotational surface. 

Consider how much resolution you really need/want.  Don't make this harder than it needs to be.  The difference in complexity between an 18,000 x 18,000 pixel image and a 36,000 x 36,000 image is huge. 

Logged
Fike, Trailpixie, or Marc Shaffer
marcshaffer.net
TrailPixie.net

I carry an M43 ILC, a couple of good lenses, and a tripod.

Ellis Vener

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2138
    • http://www.ellisvener.com
Re: Best techniques for multi-row panoramas using tilt/shift lens
« Reply #9 on: March 20, 2012, 03:50:09 PM »

1O feet square (10 feet x 10 feet) or 10 square feet ( 2.5 feet x 4 feet or 2 feet x 5 feet)?
Logged
Ellis Vener
http://www.ellisvener.com
Creating photographs for advertising, corporate and industrial clients since 1984.

the_ether

  • Newbie
  • *
  • Offline Offline
  • Posts: 25
Re: Best techniques for multi-row panoramas using tilt/shift lens
« Reply #10 on: March 20, 2012, 04:01:41 PM »

Quote
I spent a bit of time working with stitching, tilting and shifting in the same images.

Thank you flke. There was some interesting titbits in your post and you obviously took care to read what I wrote and reply appropriately.

I was not planning on using the shift capability of the lens, only the tilt to get the depth of field I need.

I noticed that you experimented with having a tilt only on the bottom row with the top row having no tilt at all. I had been considering setting the tilt only once and keeping it the same for all rows - assuming that just one setting could encompass such a wide area. The shot I'm considering has no vertical subjects so I won't have parallax issues and therefore is simpler than your scene.

I'll be using a D800E for the shot but even so I intend to print at 300dpi and so I'll need quite a few image tiles.

Quote
I had to stitch the left half of the image and then the right half and then the two together

Just out of interest:

a) is there any reason you stitched left to right instead of top to bottom?;

b) do you know if any of the specialist applications have any benefits over CS5 for stitching such a large area?

Given the number of shots I have to take I'll probably use a full panoramic head so that I can position the entrance pupil properly and to also guide me through all the shots.

I hadn't realised that I could do focus stacking whereby the shots aren't fully stacked but instead just overlap. I thought maybe I'd only end up with the narrow strip where they overlap rather than the sum of the two. This may therefore be the best way to go though I may still hire a Nikon PC-E lens just for the experience. I'm planning a weekend trip just for this one picture and the minimum renting period is a week anyway. I can therefore try both focus stacking and tilting. If the tilting works I may even throw in HDR if the lighting conditions mean the photo may benefit from it and I have any patience left. I'll begin with the focus stacking method though as it would seem to be the most straightforward.

As for printing, this photo is for a fine art project and will be made into a triptych. Printing will therefore be easier.

« Last Edit: March 20, 2012, 04:05:22 PM by the_ether »
Logged

the_ether

  • Newbie
  • *
  • Offline Offline
  • Posts: 25
Re: Best techniques for multi-row panoramas using tilt/shift lens
« Reply #11 on: March 20, 2012, 04:02:48 PM »

1O feet square (10 feet x 10 feet) or 10 square feet ( 2.5 feet x 4 feet or 2 feet x 5 feet)?

10x10
Logged

fike

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1390
  • Hiker Photographer
    • trailpixie.net
Re: Best techniques for multi-row panoramas using tilt/shift lens
« Reply #12 on: March 20, 2012, 05:00:52 PM »

Thank you flke. There was some interesting titbits in your post and you obviously took care to read what I wrote and reply appropriately.

I was not planning on using the shift capability of the lens, only the tilt to get the depth of field I need.
Good, do one or the other. Not both.
Quote
I noticed that you experimented with having a tilt only on the bottom row with the top row having no tilt at all. I had been considering setting the tilt only once and keeping it the same for all rows - assuming that just one setting could encompass such a wide area. The shot I'm considering has no vertical subjects so I won't have parallax issues and therefore is simpler than your scene.

I did make some successful panos with the lower row tilted and the middle and upper rows untilted.  This can work if you don't tilt too aggressively. Unfortunately, if you can't tilt aggressively, you don't really get much benefit from the enhanced focus in the horizontal plane.  When the tilt is different between the rows of the pano, the distortion at the edges of the frame is very different, so things don't line up and the stitchers out there can't easily handle that sort of atypical distortion. 
Quote
I'll be using a D800E for the shot but even so I intend to print at 300dpi and so I'll need quite a few image tiles.
Wow, so many things to discuss on this.  With an image like this, with a high-resolution camera like that one, you will be pushing and pulling every attribute against one another.  For example, you will want a small aperture to maximize depth of field, but with a camera of that high a resolution, you will be running into diffraction limits, so you may not see the benefit of all that resolution at say f/16.  (I am not sure exactly what the calculated diffraction limits of this camera are. You will need to look it up.) Some sort of focus stacking or manual focus blending may be the best you can do while staying at lower or mid f-stops needed to fully realize that cameras resolution. Nikon guys can probably comment about when you are starting to diminish your returns at higher apertures. 
Quote
Just out of interest:

a) is there any reason you stitched left to right instead of top to bottom?;
No reason. 
Quote
b) do you know if any of the specialist applications have any benefits over CS5 for stitching such a large area?
CS5 will not do a very good job with this sort of project. Multirow mosaics are notoriously difficult for CS to work with and CS will not allow you to control pitch and choose your projection. You will need a purpose-built tool. I  like PTGui. There are several on the market that are good.
Quote
...
I hadn't realised that I could do focus stacking whereby the shots aren't fully stacked but instead just overlap. I thought maybe I'd only end up with the narrow strip where they overlap rather than the sum of the two. This may therefore be the best way to go though I may still hire a Nikon PC-E lens just for the experience. I'm planning a weekend trip just for this one picture and the minimum renting period is a week anyway. I can therefore try both focus stacking and tilting. If the tilting works I may even throw in HDR if the lighting conditions mean the photo may benefit from it and I have any patience left. I'll begin with the focus stacking method though as it would seem to be the most straightforward.

As for printing, this photo is for a fine art project and will be made into a triptych. Printing will therefore be easier.

PTGui will allow you to stitch an image and then have them output as a layered PSD (or PSB in your case with such large files).  To do something like this you will want to output layers from PTGui and then in Photoshop merge the rows into single layers and then manually blend them with clipping masks using a feathered brush at something like 3-10 pixels. 

Your project is ambitious and looks like lots of fun.  Try to keep from adding too many variables.  HDR on top of all that...yikes! You can do it, but it won't be perfect on your first time out.
Logged
Fike, Trailpixie, or Marc Shaffer
marcshaffer.net
TrailPixie.net

I carry an M43 ILC, a couple of good lenses, and a tripod.

the_ether

  • Newbie
  • *
  • Offline Offline
  • Posts: 25
Re: Best techniques for multi-row panoramas using tilt/shift lens
« Reply #13 on: March 20, 2012, 06:43:06 PM »

For example, you will want a small aperture to maximize depth of field, but with a camera of that high a resolution, you will be running into diffraction limits, so you may not see the benefit of all that resolution at say f/16.  (I am not sure exactly what the calculated diffraction limits of this camera are. You will need to look it up.)

According to the camera's documentation Nikon recommends not exceeding f/11.

There is a further discussion on this here.

Quote
CS5 will not do a very good job with this sort of project.
Thanks for the advice. I will try a few different programs.

Seems I will be spending most of my time in front of the computer on this, what with diffraction and lens distortion compensation for all the tiles before I even begin to stitch.
Logged

Ellis Vener

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2138
    • http://www.ellisvener.com
Re: Best techniques for multi-row panoramas using tilt/shift lens
« Reply #14 on: March 20, 2012, 06:50:02 PM »

Beware the paralysis of pre-shoot analysis.

"one experiment is worth a hundred expert opinions."

Rent a tilt shift lens and try a small scale experiment first.

Edit:

I have been to Fike's  website and he ( not the OP) seems to know what he is about.

« Last Edit: March 20, 2012, 11:27:26 PM by Ellis Vener »
Logged
Ellis Vener
http://www.ellisvener.com
Creating photographs for advertising, corporate and industrial clients since 1984.

theguywitha645d

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 970
Re: Best techniques for multi-row panoramas using tilt/shift lens
« Reply #15 on: March 20, 2012, 11:20:03 PM »

This is really not such a tough problem. Given a viewing distance of about five feet, your lower limit for pixels would be 6000 on one side or your print or a 36MP image. There are single shoot camera that can do that. DoF is not that difficult to achieve with that viewing condition by just stopping down. Or you can use focus stacking--a three-shot stack should work nicely. I would not worry about the tilt/shift lens.
Logged

Ellis Vener

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2138
    • http://www.ellisvener.com
Re: Best techniques for multi-row panoramas using tilt/shift lens
« Reply #16 on: March 20, 2012, 11:24:40 PM »

If he is looking to make a print that is ten feet x ten feet square at 300dpi printing resolution, then 6000 / 300 = 20 inches. Could you expalain how you come up with 6,000 pixels on a side?
Logged
Ellis Vener
http://www.ellisvener.com
Creating photographs for advertising, corporate and industrial clients since 1984.

theguywitha645d

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 970
Re: Best techniques for multi-row panoramas using tilt/shift lens
« Reply #17 on: March 20, 2012, 11:31:02 PM »

If he is looking to make a print that is ten feet x ten feet square at 300dpi printing resolution, then 6000 / 300 = 20 inches. Could you expalain how you come up with 6,000 pixels on a side?


You don't need to make it 300dpi. 300dpi is the minimum for an 8x10 print viewed from 10". Viewing distance is a relative problem, not an absolute one. So for a five foot viewing distance, a 36MP image will be fine.
Logged

kevk

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 88
    • Kevin Kucks Photography
Re: Best techniques for multi-row panoramas using tilt/shift lens
« Reply #18 on: March 21, 2012, 04:43:48 AM »

Given a viewing distance of about five feet...
I never get it about this viewing distance stuff...whenever I see a big print the first thing I do is walk up close to it to look at the details. Is it only me?

Kevin
Logged

Walter_temp

  • Newbie
  • *
  • Offline Offline
  • Posts: 28
Re: Best techniques for multi-row panoramas using tilt/shift lens
« Reply #19 on: March 21, 2012, 04:53:56 AM »

How to spot a photographer in a photo exhibition? It's the person leaving nose marks on the framing glass.

Ciao, Walter
Logged
Pages: [1] 2 3   Go Up