Multi exposures on a Panasonic G1

<p>Akira or Yoshio<br>

Can you recommend a third-party adapter for Olympus OM lenses? There appear to be several from China but I have no idea how they compare in quality and precision. Do you have any personal experience of using the excellent Zuiko 75>150mm f4 OM zoom lens on the G1?<br>

David</p>

<p>David, I know that at least jinfinance (RJ camera) and rainbowimaging are selling OM to m4/3 adapters for 60-75 USD, but I have no experience with them. Rayqual (sold at Stephen Gandy's shop as Cameraquest) seems to be the most reliable one (I have Nikon F to m4/3 adapter by Rayqual which works great), but costs as much as the genuine one. Novoflex makes one, too, but it is more expensive than genuine one or Rayqual one at least here in Japan.</p>

<p>As for the image quality of your lens on G1, I'm afraid I cannot say anything. In general, you never know about the performances of lenses made for the film cameras until you actually test them. A well reputed lens may turn out to be poor performer and a mediocre considered lens can perform surprisingly good.</p>

<p>David</p>

<p>I have RJ Camera adaptor ... it is well made and well priced.</p>

<p>I found <a href="http://cjeastwd.blogspot.com/2009/11/micro-43-om-adaptors.html"><strong>this blog</strong> </a> worth reading, as well you may like to lurk on <a href="http://www.flickr.com/groups/1084614@N23/"><strong>this flickr group</strong> </a> where you may find others using the same (m4/3 and legacy adaptors)</p>

<p>Also, with respect to 'reference pixels' I was not meaning that you will need to place something, simply refer to the pixel. If the camera is mounted solidly (it does not need to be in an engineering clamp, just a good tripod or mount) then each subsiquent exposure will be placing the same spatial component of the scene on the same pixel. For instance when I do hdr on a tripod I do not find that I need to "adjust" the images for spatial shifts during compositing.</p>

<p>I noticed on the weekend that using my OM50mm lens that in the x10 magnification focusing aid that I could see 3 or 4 pixels of lateral movement created by 'slop' in the mechanism ... though I could not feel such. This indicates the accuracy which this sort of optical recording sytem has. I think you will find it suitable to your above stated application.</p>

<p>Thank you both for the feedback on OM lens adapters - I will buy one from RJ.<br>

Yoshio - I can see how aligning the images by pixel count should ensure accurate positioning from one image to the next. However, most of the image area of my individual exposures will be of blank sky photographed through a 4.0 ND filter and should remain below the threshold of receptor activation, so there will be no pixel image to use for alignment without some form of static reference point(s), unless there some way of mapping the position of pixels that I don't know about?<br>

David</p>

<p>David</p>

<blockquote>

<p><em>However, most of the image area of my individual exposures will be of blank sky photographed through a 4.0 ND filter and should remain below the threshold of receptor activation, so there will be no pixel image to use for alignment without some form of static reference point(s), unless there some way of mapping the position of pixels that I don't know about?</em></p>

</blockquote>

<p>I believe (unless I mis understand) that you are thinking in terms of film. There is an absolute reference even if you photograph a totally dark thing. As long as your camera does not move, your sensor will therefore also not move and therefore you will always be looking at the exact point in space in each subsequent exposure.</p>

<p>Assuming you had a fully dark hall and illuminated a moving object with a strobe using a long shutter exposure time there will be zero difference between that and 500 exposures composited because each pixel always refers to the exact location in space of the sensor (which is fixed in space). Unlike film you do not have to concern youself with the precision of film advance because the sensor does not advance between exposures.</p>

<p>If you have a 8mm sensor which has 4000 pixels across the x axis you will have accuracy to around 0.002 mm of your projected image (lens projects image onto sensor).</p>

<p>this will likely be enough for all applications I can think of (expecially with the applicaion of the appropriate focal length lens you should be able to measure thousandths of a mm movement)</p>

<p>Yoshio<br>

I do understand that each camera image will be in absolute reference to each other. That is not what I see as the potential problem. My concern is how to retain this registration when superimposing 18 individual images in Photoshop. I can only presume that even if the individual peripheral pixels cannot be distinguished, registration can still be achieved by reference to the outermost edges of the individual picture frames. Or am I missing something obvious?<br>

David</p>

<p>David</p>

<blockquote>

<p>My concern is how to retain this registration when superimposing 18 individual images in Photoshop. I can only presume that even if the individual peripheral pixels cannot be distinguished, registration can still be achieved by reference to the outermost edges of the individual picture frames. Or am I missing something obvious?</p>

</blockquote>

<p>I must say I am at a loss on how to explain this to you pixels can always be distinguished ... even if they are all black. They are simply an array.</p>

<p>have you ever used any photo viewer or editor? I can only guess that your thinking is so firmly fixed in film that you are not thinking of what this really all is.</p>

<p> </p>

<p>try loading this image in a proper editor (not the microsoft one) and you can see coordinates</p>

<p>it matters not if you composite 2 or 200,000</p><div></div>

<p>Now just hold on a moment Yoshio.<br>

No I am NOT "so firmly fixed in film that I am not thinking of what this really all is". I asked on my posting of the 7th " <strong>unless there some way of mapping the position of pixels that I don't know about?</strong> "<br>

I DID NOT KNOW there is a way of doing this. <strong>That's why I am posting on this site</strong> .<br>

I looked at digital images in MS Picture and Fax Viewer, MS Office Picture Manager and MS Paint. I also looked at images in a copy of Photoshop but <strong>NONE</strong> of these showed the coordinates of the pixels, so I was NOT aware that it was possible to do so. I even did a search in Photoshop but could not find anything that referred to pixel coordinates. I have now viewed a digital image in another editor and see that it is.<br>

David</p>

<p>David</p>

<p>I am sorry if I seemed offensive, it was quite flabbergasting for me to think of this and not expect that would be obvious. I think only microsoft photo software (can't think of the name, comes free with windows, I have never used it but I have seen it, I therefore can not confirm if it does or does not show pixels) which does not give you the pixels</p>

<p>Photoshop shows this (in every version I have ever worked with back to version 2 and you can choose to turn on rulers and change you desired scale), even lowly MS paint shows this information. <strong>I can not think of <em>any</em> application which does not show this information, none.<br /> </strong></p>

<p>If you consider these things perhaps you can understand my reaction.</p>

<p>I thought it was <em>so intuitive</em> that I could not imagine someone had not seen it. So much so that I was starting to be considering if you were deliberately baiting me.</p>

<p>I am quite surprised that after all the trouble I have gone to to patiently show you all these things that you so suddenly yell at me like that. <strong>I even prepare presentation images to demonstrate this to you. Your appreciation of my efforts seems to be easily over taken by something which looks like pride. </strong></p>

<p>Because I thought if you were so throughly steeped in film that you may not have been scanning (even these drivers show pixel coordinates) and thus may not have been exposed to this. It does not seem very appreciative of you in my way of seeing things</p>

<p>May I ask what is wrong with being entrenched in film? As a person who uregularly uses 4x5 sheet I see nothing wrong with it, but it is a distinct mindset, different entirely to film.</p>

<p>good day to you sir</p>

<p>PS: it might be helpful if you preface your posts with some introduction of what you consider your depth of knowledge on the subject as there are far far to many assumptions to make on things for my part.</p>

<p>David</p>

<p>Screen shots for helping you find the pixel coordinates</p>

<p><img src="http://farm3.static.flickr.com/2803/4169911040_205bbe2844_o.jpg" alt="" width="603" height="393" /></p>

<p><img src="http://farm5.static.flickr.com/4002/4169911114_148aa144d5_o.jpg" alt="" width="450" height="399" /></p>

<p> </p>

<p>Now THAT was very helpful.<br>

I thought I had made my lack of knowledge in this area quite clear when I wrote:<br>

"unless there is some way of mapping the position of pixels that I don't know about?"<br>

I can see how obvious it may have been to you, and I confess the bottom of the Paint window was off the screen when I looked so did not see the pixal coordinates. I have only just installed Photoshop and could not find where the information was displayed - still can't, but at least I now know it is there.<br>

So thank you Yoshio.<br>

I am just here to obtain help for my first steps into digital imaging, and have been very grateful for the assistance received so far, so can we bring this subject to a close now.<br>

David<br>

P.S. - I never yell at anyone, it never achieves anything worthwhile.</p>

<p><strong><br /> </strong></p>

<p>David<br>

I am glad it helped, but even without seeing the pixle numerics consider again the things I said:</p>

<blockquote>

<p>I believe (unless I mis understand) that you are thinking in terms of film. There is an absolute reference even if you photograph a totally dark thing. As long as your camera does not move, your sensor will therefore also not move and therefore you will always be looking at the exact point in space in each subsequent exposure.</p>

</blockquote>

<p>and</p>

<blockquote>

<p>Unlike film you do not have to concern youself with the precision of film advance because the sensor does not advance between exposures.<br>

If you have a 8mm sensor which has 4000 pixels across the x axis you will have accuracy to around 0.002 mm of your projected image (lens projects image onto sensor).</p>

</blockquote>

<p>and so when you wrote:</p>

<blockquote>

<p>I do understand that each camera image will be in <strong>absolute reference to each other</strong> .</p>

</blockquote>

<p>and then</p>

<blockquote>That is not what I see as the potential problem. <strong>My concern is how to retain this registration when superimposing 18 individual images in Photoshop.</strong>

</blockquote>

<p>it was clear that you <em>did not understand what it was you were saying</em> , and that you did not follow my explaining of the image being an array and that pixels (which come from fixed spatial refrerences in the sensor) no matter where (not simply edges) would <strong>always represent absolute coordinates</strong> .</p>

<p>Thus I said if you composite 2 or 200,000 images they will register perfectly every time in every part of the image. I appreciate that because you <em>work in mono medium format exclusively </em> and have done for 60 years, that you may not easily grasp these points. Please try to re read these points again carefully. If there are ideas you are not familiar with please ask specific questions. Now that I understand you knowledge levels I can understand there may be many misplaced premises in my descriptions.</p>

<p>I recommend you download software called <a href="http://www.irfanview.com/"><strong>irfanview</strong> </a> I see you mention photoshop, I use that too, but irfanview is free and for many applications (just viewing and resizing images in batch) often faster to use.</p>

<p>David</p>

<p>you say:</p>

<blockquote>

<p>I am just here to obtain help for my first steps into digital imaging,</p>

</blockquote>

<p>I think it is worth remembering that this is not a classroom, it is a community. I am not a teacher for you we are two people having a conversation.</p>

<p>I am not by profession a teacher, I am just a community member who happens to believe that it is good to help others as others have helped me in my journey (it just so happens I have been on the digital journey for some time)</p>

<p>Yoshio<br>

Thank you for the link to Irfanview - It loads very quickly, which I always appreciate and looks as if it will be a very useful tool if my G1 ever arrives (delivery has been put back 2 weeks - so it will not be with me until the New Year now - assuming Panasonic have not sold out).<br>

In our discussion what was obvious to you, was not to me, and conversely what was obvious to me, was not to you. Even between two native English speakers such misunderstandings can occur so I do not think either of us was to blame. I did find the absence of coordinates puzzling in a digital image that relies on an array of receptors and had I posed that as an individual question it might have saved a great deal of frustration for both of us, but at the time I believed this was self-evident. I have already expressed my appreciation for the help received so far from both Akira and yourself, so can we please bring this subject to a close now.<br>

David</p>