Randomizer bias

[kimbo]

I'm using gphotoshow pro 8.5.3 on Win 10. This is what I've been searching for for years. Terrific!
I've configured the slideshow for random folders, and sequential files
I have a bit over 1300 folders and about 35000 files (photos and videos).
I cannot stress how pleased I am with the software.
It appears that the randomizer has a bias that lasts about a week or so. Every morning I boot up and launch the program. For about a week we see what appears the same group of folders. Whether or not they are in the same order I do not know.
I assume the randomizer uses the system clock as a seed? Just wondering if anyone else has had a similar experience?
Thanks,
kim

[NeoAlucard]

Hi Kimbo! In the [File Options] section there's a box called [Global Weights], in there you can check to see if weighs are being assigned a certain amount for files modified in the last X amount of days. The information you're looking for may be in there, if not I'll see what I can find. (I'm not the creator of gPhotoShow but I have been using it for years)

[gpb]

It appears that the randomizer has a bias that lasts about a week or so. Every morning I boot up and launch the program. For about a week we see what appears the same group of folders.

Hello,
the gPhotoShow random function has no bias by itself, and it is initialized in a more complex way than using current time.
As NeoAlucard wrote check global weights in "File Options" page, or check the number on the right of each image folder. Weights are used by gPhotoShow to give a priority to some images.
Anyway if you are curious about the randomization function, gPhotoShow uses a variant of the Mersenne Twister algorithm implemented in the standard c++ library

[joShu]

I have been wondering about such bias also. One of my gphotoshow playlists/gss consists of six 50-minute videos , played at random start times , in 5 minute lengths , weight = 1, and i too have noticed that the random start points seem very similar to earlier random start points.

But perhaps this is just my erroneous perception : 6 x 50 / 5 ~= 60 segments , which is small , so in any given session it is likely that I would randomly see roughly the same material as before. Still , it does 'feel' like gps prefers some random start points more than others.

[gpb]

The randomization function has no issue but it's all about probability. It's like rolling a dice, if you consider a long enough sequence of rolls each number has the same probability but nothing excludes that the 6 could come out 3 times in a row