Newbie questions

57 views
Skip to first unread message

PeterB

unread,
Oct 10, 2012, 2:40:22 AM10/10/12
to pt...@googlegroups.com
I just purchased PTGui and I'm waiting for my registration code so that I can download a full version. I mention this because the use of the trial version conceivably might be related to my issues.

I just put together a HDR panorama. The stitched final product looks pretty good except for two important issues.

Some sections of the resulting panoramic image look grayish, with low saturation, and the wrong tonality. The originals were taken in Ap mode, so all have the same aperture, with only the exposure time varied. The nominal exposure times are all mostly identical. The sections at issue don't obviously seem to correlate with varying exposure times.

The images were preprocessed with DXO which should have removed image distortion. Yet, I see in the panoramic image that the shapes of the sub-images are 'barrel' shaped. Are the pair, DXO and PTGui double correcting? Should I tell DXO to not to correct for distortion and let PTGui do all of that? I don't want to throw away any more image than necessary. This was a rare opportunity panoramic image and I may never get another chance. Or should I have over-sized vertically the originals? (The originals are four in a single row.)

Thanks, Peter

PTGui Support

unread,
Oct 10, 2012, 4:29:23 AM10/10/12
to pt...@googlegroups.com
Hi Peter,

On 10-10-2012 8:40, PeterB wrote:
> Some sections of the resulting panoramic image look grayish, with low
> saturation, and the wrong tonality. The originals were taken in Ap mode,
> so all have the same aperture, with only the exposure time varied. The
> nominal exposure times are all mostly identical. The sections at issue
> don't obviously seem to correlate with varying exposure times.

HDR panoramas in PTGui should always be taken in M mode, not Ap. PTGui
expects a full panorama for every exposure time.

For details see section 6 in the FAQ:
http://www.ptgui.com/support.html#6_1

And the HDR tutorial:
http://www.ptgui.com/hdrtutorial.html

As a last resort to stitching your photos you could try converting them
to HDR in another application (Photomatix, Photoshop) and stitch the
result in PTGui.

> The images were preprocessed with DXO which should have removed image
> distortion. Yet, I see in the panoramic image that the shapes of the
> sub-images are 'barrel' shaped. Are the pair, DXO and PTGui double
> correcting? Should I tell DXO to not to correct for distortion and let
> PTGui do all of that? I don't want to throw away any more image than
> necessary. This was a rare opportunity panoramic image and I may never
> get another chance. Or should I have over-sized vertically the
> originals? (The originals are four in a single row.)

It's not necessary and usually not a good idea to correct your images
for barrel distortion prior to stitching. PTGui can perfectly correct
for this.

I have not seen your photos, but what you are probably seeing is the
normal distortion which happens with wide panoramic projections. For
details see:
http://www.ptgui.com/support.html#5_1
and
http://www.ptgui.com/man/projections.html

If you need more help feel free to post the images here, see 3.17:
http://www.ptgui.com/support.html#3_17

Joost

tracy....@comcast.net

unread,
Oct 10, 2012, 2:01:12 PM10/10/12
to pt...@googlegroups.com

This may be a silly question.  But if I'm shooting in RAW, is there any reason to bracket?  Can't the exposures be changed in Photoshop, Lightroom, etc?

 

Thanks!

Tracy R. Willis

Erik Krause

unread,
Oct 10, 2012, 2:26:39 PM10/10/12
to pt...@googlegroups.com
Am 10.10.2012 20:01, schrieb tracy....@comcast.net:
> This may be a silly question. But if I'm shooting in RAW, is there any
> reason to bracket? Can't the exposures be changed in Photoshop,
> Lightroom, etc?

There is probably less reason to bracket. However, depending from your
camera you can capture between 8 and 12 EV in a raw image. If you need
more dynamic range (a bright sunny day with deep shadows might need 21
EV) you still need to bracket. Another benefit from physical bracketing
is less noise. If you do virtual bracketing in your raw converter the
shadows are boosted which also amplifies shadow noise.

--
Erik Krause
http://www.erik-krause.de

Geoff - Spherical Visions

unread,
Oct 10, 2012, 2:28:25 PM10/10/12
to pt...@googlegroups.com
Depends if the camera can handle the dynamic range of the light, if it can't then you need to bracket and work in HDR mode.
I've gone over to using UniWB to get the absolute max. out of the camera, but if you try to get stained glass windows in side dark churches then I still need to bracket!

Ken Warner

unread,
Oct 10, 2012, 2:47:11 PM10/10/12
to pt...@googlegroups.com
I did a quick google for UniWB and it sounds interesting. From a brief glance it looks like you take a jpeg + raw and then combine the two. The jpeg is somehow adjusted. That's the part I don't understand. Also, the examples are all for Nikon cameras.

How would one do UniWB on a Sony NEX camera I wonder. What are the general principles?

Erik Krause

unread,
Oct 10, 2012, 3:35:55 PM10/10/12
to pt...@googlegroups.com
Am 10.10.2012 20:47, schrieb Ken Warner:
> I did a quick google for UniWB and it sounds interesting. From a
> brief glance it looks like you take a jpeg + raw and then combine the
> two. The jpeg is somehow adjusted. That's the part I don't
> understand. Also, the examples are all for Nikon cameras.
>
> How would one do UniWB on a Sony NEX camera I wonder. What are the
> general principles?

I guess you refer to http://www.malch.com/nikon/UniWB.html
As far as I understand the two are not combined. The whole procedure is
only to get a histogram from the jpeg that is more or less identical to
the one you would get from raw if you shoot at the same exposure.

Most cameras compute the histogram after conversion to jpeg. At this
step the extra dynamic range is already lost and hence you can't use the
histogram properly to adjust exposure. If you expose "to the right" you
always would expose a bit too dark because the histogram doesn't extend
as far "to the right" as the raw data permits.

Apparently it is possible to modify whitebalance such that those extra
data is recorded into jpeg and hence the histogram displays the extra
data. The jpeg, however, is not useable otherwise.

Erik Krause

unread,
Oct 10, 2012, 3:46:08 PM10/10/12
to pt...@googlegroups.com
Am 10.10.2012 20:47, schrieb Ken Warner:
> How would one do UniWB on a Sony NEX camera I wonder.

There is a step by step description on how to create your own UniWB:
http://forums.dpreview.com/forums/post/41245342

Ken Warner

unread,
Oct 10, 2012, 4:04:40 PM10/10/12
to pt...@googlegroups.com
Thanks...

Erik Krause

unread,
Oct 10, 2012, 4:28:18 PM10/10/12
to pt...@googlegroups.com
Am 10.10.2012 22:04, schrieb Ken Warner:
> Thanks...

Another longish article with raw files for different cameras that can be
used to establish UniWB:
http://www.guillermoluijk.com/tutorial/uniwb/index_en.htm

Magic Lantern BTW seems to have a very easy method to set UniWB: simply
dial in 1.0 multipliers in the ML WB submenu...

Ken Warner

unread,
Oct 10, 2012, 6:47:42 PM10/10/12
to pt...@googlegroups.com
Thanks Eric, I had a wrong idea of what UniWB was. I thought it was some trick way to get a little bit more dynamic range. But I see it's just for JPG and I've kind of lost interest for JPG since I got Lightroom.

Geoff - Spherical Visions

unread,
Oct 10, 2012, 6:48:45 PM10/10/12
to pt...@googlegroups.com
Yup thats right the Jpg is merely present to allow the correct parameters to be loaded as a pre-set white balance into the camera.
Once that is done the Histogram displayed shows much more accurate information at the right hand end and so allows you to maximise the exposure, at the cost of a very greenish tinge to the preview image.  However once you get use to this it presents no problem.  Once the RAW file is downloaded then you just set a normal White balance value up and the image appears as expected.
What you cannot then do is manually increase the exposure in any pre-procesing - you really are at the limits of the sensor itself so there is no point in doing so!  Previously I could increase the exposure by at least 1.5 and sometimes 2ev in post processing, which means of course that I had already lost that same amount of info at the dark end of the range.

Its simple enough on a Nikon to load the image and set up camera so that I have the UniWB available on one set of presets and other more normal types on the other ones, so its simple to switch over.  It might be worth while asking around on the DPReview forums for your camera to see if any one has come up with the relevant white balance and curve settings.

Erik Krause

unread,
Oct 11, 2012, 2:12:01 PM10/11/12
to pt...@googlegroups.com
Am 11.10.2012 00:47, schrieb Ken Warner:
> Thanks Eric, I had a wrong idea of what UniWB was. I thought it was
> some trick way to get a little bit more dynamic range. But I see
> it's just for JPG


No, it's for raw. Just a way to determine how to best expose raw for
maximum dynamic range. It is necessary if you want to choose exposure
for ETTR (Expose To The Right), since normal histogram wouldn't show the
true "Right".

--
Erik Krause
Herchersgarten 1
79249 Merzhausen

Ken Warner

unread,
Oct 11, 2012, 2:25:30 PM10/11/12
to pt...@googlegroups.com
I need to study it more to understand it then.
Reply all
Reply to author
Forward
0 new messages