#7 Installing GeigerLog on Python 3.11 32 bit, Windows 10 Home 32 bit

[nonus]

What is the problem? I can't install GeigerLog on Windows 10 32 bit operating system because python modules don't have current version for 32 bit systems.

There is a similar case here

[ullix]

Your screenshots do not capture the full output. If this is needed again, then please do:

GeigerLog.bat setup > mysetup.txt

which will redirect all output to file mysetup.txt . Then post this file.

However, current info points to a problem with package matplotlib. Let's try this outside of GeigerLog: issue these commands in succession and post each output:

pip install -U pip
pip install -U setuptools
pip list
pip install -U matplotlib
pip index versions matplotlib

[nonus]

I sold a computer with Windows 10 32 bit. I will be able to try your recommendations in a month, when I will have access to another 32 bit system. I think it depends on third-party modules. New versions of Python modules do not support the 32 bit version, and old versions do not support GeigerLog. It's a vicious circle.

[ullix]

Even a Raspi4 is good enough to run GeigerLog with GMC counters on Python 64bit!

[nonus]

It's not about hardware. Python 32 bit is the reason. If your computer does not support 64 bit operating system (at BIOS level), then you will not be able to use GeigerLog. There is an option to use Linux 64 bit, but Linux does not have an application for radmon, only a script for python.
Look at these Chinese computers. They are much more interesting than raspberry pi, but they don't work with geigerlog if the operating system is 32 bit
Pipo

[ullix]

GeigerLog has NO(!) restriction with respect to 32bit or 64bit! I also do not see that the required modules do have such a restriction, or can you provide a link where it specifies such restrictions? I suspect the problem is elsewhere!

[nonus]

Try running GeigerLog on Windows 10 32 bit. You will encounter the same problem with old modules.
problem

[ullix]

I have no access to Windows 10 32 bit. But after some searching I have the impression that running matplotlib on it is simply not possible anymore. Probably not even if you compile it yourself. Likewise, numpy and scipy are also suggesting to drop 32bit support! This trio of packages is at the core of GeigerLog (and a whole lot of other software)!

Given that Python itself is still producing 32bit releases, I really don't understand their decisions. But, alas, this is voluntary work, and it is a very big job.

It looks, though, that 32bit packages still work on Linux, but this won't help you, I am afraid. But due to the lack of matplotlib on Windows 10 32 bit the use of GeigerLog will no longer be possible!

Alternatives:
If you have a second computer, which can run Win 64bit, you could use your 32bit computer as dataserver to send data into your local LAN, to be picked up by the 64bit computer. You find this software in the raspi folder of GeigerLog. The dataserver-computer must be connected by USB-cable to the counter. It does need neither matplotlib nor PyQt5.

Last resort is to use your 32bit computer as a stand-alone system, connected to a counter, and saving data on internal file. Use this software:
https://sourceforge.net/projects/geigerlog/files/geigerlog_simple_500plus-v0.2.2.zip/download

[nonus]

Thanks for the help and development of GeigerLog. I found a replacement for a 32-bit computer from the same manufacturer with 64 bits. GeigerLog's support for cheap Chinese dosimeters like Fnirsi gc-01 opens up many possibilities. I successfully installed a mica sensor (sbt-10, si-8b) into a Chinese dosimeter (radpro). A cheap computer + a Chinese dosimeter costs ~60-70 $. Excellent result!

[ullix]

I am pleased to hear!

The Fnirsi gc-01 with the custom firmware RadPro indeed opens up many possibilities. One particular aspect is that the anode voltage can be raised up to 1500V! Unheard for most - if not all - other modern Geiger counter. This allows to use even some strange old, post-war tubes!

I am not sure what GL version you use, but the latest devel versions do have some more improvements, You find them here:
https://sourceforge.net/p/geigerlog/discussion/devel2/

Always pick the latest one.