I realize it's been some time since I last posted about my model of the Clopton chantry chapel. Unfortunately, I've run into a bit of a problem with the second portion of the walkthrough. I wanted to be able to step you through the process of creating the model, but I've realized that my current hardware cannot actually create the model at a better quality than what already exists without the software crashing. I have started to do so -- a process that takes several days to a week of rendering time depending on the particular machine -- several times with no luck. When I started the recent attempt I only had one machine hypothetically capable of doing the work, but at the time I created the original model I briefly had access to multiple licenses of Photoscan (which has been subsequently renamed Metashape) and so I could split the work out amongst multiple machines and avoid too much of the workload being placed on any one of them. That option isn't available to me now, and my workarounds -- a beefier graphics card and a dedicated classic Mac Pro with dual cards that I could leave running -- have not been able to solve the problem. In the case of running it on my laptop I ran into a memory issue and in the case of the aforementioned Mac Pro I kept running into an indexing error. The workaround appears to be to build the model at a lesser quality, at which point there's not really a point in rebuilding it at all. I can still post the process I use to clean up pictures, but without the model-building portion I'm not sure it's worthwhile as a second section. Therefore, until I can solve this problem (likely with a dedicated machine built from scratch after I find stable, non-adjunct employment) I'm going to have to put that article series on hiatus. I'll update this with a new title and further information as soon as that happens, but I don't have a time frame in which it will.
This does, however, speak to one problem with my do-it-yourself model. Since you are basically doing everything yourself there may be times when you know precisely what needs to happen in order to solve a problem and still not be able to move forward because the solution relies on resources that you don't have. The concept is still valid because should the situation change you know precisely what to do to solve the problem rather than being reliant on someone else's expertise, but it doesn't fix things in that moment. It also speaks to a larger question of access and scholarship in academia today, but that's probably something to mention in another post. Since I hadn't posted any updates while I tried out these various solutions I thought it best to go ahead and post this, admit defeat, cut bait by selling the hardware I accumulated to try out these solutions, and move on.
Add new comment