<div style="font-family: 'verdana'; font-size: 12px; color: #000;"><span style="background-color: rgb(255, 255, 255);">Hi Remco,</span><br><br></div>
<div style="font-family: 'verdana'; font-size: 12px; color: #000;">
<div>I don't really want to write metadata at the moment. </div>
<div style="font-family: verdana; font-size: 12px; color: rgb(0, 0, 0);"> </div>
<div>In the 'Behaviour settings', I unchecked all the boxes except 'Use lazy synchronisation'. I also initially checked 'Update file modification timestamp when files are modified', but as I didn't know if this would trigger write operations, I unchecked this as well. Am I missing something? </div>
<div style="font-family: verdana; font-size: 12px; color: rgb(0, 0, 0);"> </div>
<div>Nothing is ticked in the 'Sidecars Settings' for the same reason. Can I enable writing to sidecars here without triggering a write process if nothing is selected on the Behaviour page?</div>
<div style="font-family: verdana; font-size: 12px; color: rgb(0, 0, 0);"> </div>
<div style="font-family: verdana; font-size: 12px; color: rgb(0, 0, 0);">Best regards,</div>
<div style="font-family: verdana; font-size: 12px; color: rgb(0, 0, 0);">Thomas</div>
</div>
<div style="font-family: 'verdana'; font-size: 12px; color: #000;"> </div>
<div id="sub-body-container" style="margin: 10px 5px 5px 10px; padding: 10px 0px 10px 10px; border-left: 2px solid rgb(195, 217, 229);">
<div style="margin: 0px 0px 10px;">
<div><strong>Gesendet: </strong>Donnerstag, 23. Januar 2025 um 15:49</div>
<div><strong>Von: </strong>"Remco Viëtor" <remco.vietor@wanadoo.fr></div>
<div><strong>An: </strong>digikam-users@kde.org</div>
<div><strong>Betreff: </strong>Re: challenge of network share</div>
</div>
On jeudi 23 janvier 2025 15:28:13 heure normale d’Europe centrale <br>thomas.weg@abwesend.de wrote:<br>> Hello everyone,<br>> <br>> Thank you for this fine piece of software (8.6). It gets better and better<br>> with each release. <br>> However, I am currently somewhat dissatisfied with my workflow and would<br>> like to benefit from your experience. These points should not be taken as<br>> criticism, I'm just describing challenges in my workflow that I don't know<br>> how to improve as a beginner. <br>> I have some images on a network drive that is connected very slowly. The<br>> databases are all stored locally on a SSD. The fact that the initial import<br>> process takes a little longer is not a problem for me. Unfortunately, the<br>> subsequent keywording process is also very slow and it sometimes takes<br>> several seconds to switch to the next image. I currently open the images in<br>> Preview to add faces or other markers. I assume that Preview creates<br>> previews from the real images and does not use the thumbnails. Is there a<br>> way to change this? Or what is your workflow here? I would also like to be<br>> able to do keywording without network drives connected. The low resolution<br>> thumbnails would be sufficient for me. However, this is not part of the<br>> Digikam concept, is it? <br>> Face recognition has improved significantly since it was first introduced.<br>> This is a great evolution of the tool. Thanks. I set it to search for faces<br>> in new images. If I understand it correctly, jpg and raw files are<br>> considered separately. I would like to see an option to group newly added<br>> images directly by name and then run the face recognition on only one of<br>> the images to save some time. I think this is currently only possible if<br>> you search for faces directly on a group. <br>> When assigning names to the images or ignoring faces, I noticed that this<br>> also took several seconds (->network drive). I would have expected that<br>> with a lazy synchronisation of the metadata, there would be no need to<br>> retrieve the data from the network drive along with the generated<br>> thumbnails, and that the write process could be done later, so that there<br>> would be no delays here. I also didn't check any boxes, so I assumed that<br>> no metadata would be written back at all.<br>(...)<br><br>Are you writing keywords etc. to the images and/or sidecar files? Those two <br>both require access to the network storage, so will be slow. It also implies <br>that you cannot work without the network drives in that case (not without <br>risking inconsistencies between image/sidecar data and the database...).<br><br>As face names are basically keywords, same reasoning holds.<br><br>Remco<br><br><br><br><br></div>