just like to inform you that a new update is out.
Thanks for the update :) I hate softwares :th_wink: Each time you start any software now, it pop-ups asking to download latest version ! At least with hardware, once you bought it you are quiet for years of use :biggrin:
A small warning about PTGui 9.1.1.
When using viewpoint correction for source images with a FOV of >180 degree (f.e when using a full size sensor camera and a Tokina 10-17 mm or a Canon 8-15 < @13 mm zoom setting) you will get problems with an incomplete output so when using such images and vp correction it is better to keep and use the previous version 9.1.
In case PTGui 9.1.1 is already installed you can download 9.1 from the download section of PTGui where older versions are also available (you need to enter the license key to get access to the downloads).
The developer of PTGui knows about the issue and it will be fixed.
Thx for sharing,
I have send this to Joost,
However there is a blending issue when you have parts of the panohead in the image. They are now much more prone to be blended into the nadir if you do not mask them.
This may result in what you call incomplete output.
Unfortunately the editor does not show it always.
Do you have an example of how your incomplete output looks.
The error occurs in both the Panorama Editor and the output of PTGui 9.1.1.
When stitching a pano with only fisheye images with a FOV > 180 degree, VP correction, no zenith and/or nadir patch, so only roundshots, then you will see that the warped images will miss a big part of zenith or nadir resulting in a lot of grey/white transparant blocks in the editor and a lot of black parts in the output. After I reported the issue Joost has confirmed it so I am pretty sure that it will be solved soon.
BTW, when using fisheye images with a FOV < 180 degree and VP correction there is absolutely no issue so in this case you can use PTGui 9.1.1. without any problem.
Ok I managed to trigger it after trying about 6-7 projects. But only after I had done some completely weird things.
I mean you can of course not expect that you can use VP on a full fisheye with CP applied at the very edge of the images.
I had to set CP at around 170 degrees before I could see it. You can not expect that this has a usable effect on a full 180 degree fisheye.
BTW it is not visible in the detail viewer.
And using the VP did not have any good effect on the stitch either.
Are you kidding me ? I am sure you know that there are situations that you need a precise template and when using an optimum spread set of CP's (and that includes sometimes setting CP close to the edges) and using VP correction (to eliminate errors due to a lot of parallax) this can help a lot to get a better result, f.i. when (batch) processing hand held pole panos.
BTW, it should not matter what workflow is used, the only thing that matters is that an issue that is absent in PTGui 9.1 is present in version 9.1.1 but soon will be solved so soon everyone can use his/her own workflow that could be weird, or not, can have CP's close to the image edge, or not, and can have VP correction, or not, just as the user of PTGui seems to be fit for a given project.
I am sure you agree with that, or not ;-))
No, using VP for parallax on a full 180 degree image is of absolutely no positive effect. You can use it on flat areas if you can mask it out from the rest of the image.
You can sometimes use individual shift to correct for parallax though but even that effect is limited.
Hans, come next year to the PanoTools Meeting and we will discuss this topic off forum, it could be very well that despite all your knowledge I can still learn you a new trick about how to process a hand held shot pole pano in such a way that you can use VP correction to get the best result.
Hans means that when the NPP is correctly set, parallax at the zenith/ nadir region can not be corrected by vp.
Wim means that when the NPP is offset (whether mistakenly or deliberately/ unavoidably, eg in pole photography), VP correction will improve the overall stitching.
Am I right?
The use of VP correction is often a point of discussion on forums because there are only a few situations where it can help to improve the results, one of those situations is when shooting hand held pole panos with a lot of NPP offset (a combination of a deliberately forward shift of the lens + an unavoidable wobble of the lens caused by the hand held leveling of the pole).
Sometimes it is hard to solve a misunderstanding when having an online discussion and in such situations it is much easier to take away the misunderstanding when talking in person, I think the last postings of this thread are a good example for this ;-)
When talking about an improvement when using VP correction it is impossible to judge the improvement without numbers so I did some tests with a few handheld pole panos and depending on the scene (= spread of CP's) and how much the pole is shaken I get an overall improvement from 10-20% for both CP max and CP average.
For me these numbers are worth the use of VP correction but for other people perhaps not, at the end it is just about personal preferences and workflows.
Joost has now released PTGui V9.1.2 to fix the VP correction problem.