I’m a real computer novice so I don’t know how to find the code for your program, but I do know that I’m running Windows 10.
1)As the header tells you I want to know the maximum for input of names.
2)Is there a Swedish language included?
3)If so, does it need a computer pro to make it function?
Looking forward to start an old hobby once again.
As of this morning I have 220,131 people. Another user imported multiple GEDCOMS and was well beyond 1million.
Obviously running filters, etc can become slow and sometimes seems to ‘hang’ before completing its task. But for the most part I have no problem.
Of greater importance is how ‘deep’ you set the Max Generation for relationships calculator. It starts to hang (I find) when the generations is greater than 20 and trying to label the relationship when the shared relative is greater than 20 generations back.
No maximum, somebody over on reddit is using Gramps with over 1.4 Million entries with the associated slowness you’d expect from parts of Gramps.
The maximum is the database size of 140TB…
If each record (row) in the database is 800KB of data, that would be something similar to 187 904 819,20
So even if you wrote each persons full biography (as in an essay or a short book) in a Note , you still would be able to have a few 100k records in the database…
From personal experience, I advise using Linux as the operating system for more than 2000 people. Under Windows, Gramps is really tough, under Linux everything goes smoothly.
I have 11500 people using Windows 10, without problems.
There are things that you should change immediately if you intend to make a VERY large tree:
Change the ID Formats preferences.
The default is for IDs with a letter identifier for category and 4 digits for the indexing. Person used “I%04d” (That’s 'I’ndividual with up to 4 leading zeroes.) So for the Person category, that’s individuals I0001 through I9999. The ten thousandth individual does NOT cause a failure. It just messes up the alphabetical sorting.
You’re likely to have more Events, Persons, Notes & Citations than the others. But I set them all to 6 leading zeroes regardless. (Events example: ‘E%06d’) That’s clean to a million.
the ultimate limit is currently just above 7.8 billion. in reality, it depends on how much storage you have and how much information each person gets. that 4GB memory stick is unlikely to be a decent backup.
I just imported the test database with 124k people on a Windows PC, took less than 4 min for the import and the loading of the database took less than 15 second ON A WINDOWS COMPUTER !
No issue with memory, no response error, even though I stress tested with fast scrolling and opening multiple editor Windows.
I also have a database with 32k places in hierarchy, no problem, even though Gramps reload the data nearly every time I open an Editor.
I have loaded 349000 places once, but it was just a test, so I didn’t clock it, also on a Windows 10 WS.
But with mongodb as backend.
so it’s time to stop that “anti-windows” preach some users deliver here.
likewise there needs to be no anti-whatever preach against any other OS, or other tool like editor. there’s not even any anti- preach needed for other genealogical software. the topic here is specific for help.
That’s one of the biggest reason I ask for interchangeable export/import formats, because I use multiple software to analyze my data, including a few Linux only solutions.
I don’t use Mac but I know that it has always been a easy OS to use, and it has some real advantages regarding DTP, Graphic and Video editing. All OS has their advantages and disadvantages, just as software do… and the easiest way to overcome those disadvantages for software developers are to support interchangeable import/export formats, so that people can chose to store data one in one software, analyze it in another and publish it in the 3rd or 4th.
we need to get all of the right people to agree on what that one format is. it needs to be text in UTF-8. Python3 handles this well. line handling methods can deal with the end-of-line variations quite easily so a common format does not need to specify that. outputting any of the three should be readable in all three platforms. the big question will be the other software if not developed in Python3. hopefully, most other languages have similar capability. end-of-line should be plenty easy to just ignore if we use a layer that doesn’t use lines, such as XML. then it’s a matter of deciding what and how to represent.
this is a topic for a whole other thread.
When I talk about interchangeable formats I think about OWL/XML, RDS/XML, JSON-LD and Graph formats like Graphml, gefx or the older format gml.
This is all linked formats that can be used in other research tools, and that most programming languages has libraries for…
One format for interchangeable web vocabular like OWL/RDS, and one format for graph interchangeability, and a full database dump to a tabular format like CSV (for tabular analyzes like pivot tables, or import to software that read tabular data i.e. QGIS or SAGA… I know of a lot of research applications that would have great benefits for peoples research, but without interchangeability of data, there are no point of writing about them, because the workflow of converting the Gramps xml to any other of those formats are way more difficult than most users want to get into…
But with those interchangeable formats, people can chose both what web apps (presentation or research tool) they want to use, and they will have the possibility to do analyze in Network Graph tools like Cytoscape, Gephi, yEd, Constellation or Social Networks Analyzer, or even multiple python graph tools like networkx or dash, or relatively easy import data to numpy and other spreadsheet tools, without any knowledge of file formats, development or scripting.
I have created a feature request for this, https://gramps-project.org/bugs/view.php?id=12052
And the topic has been opened multiple posts before.
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.