Tool Others F95Checker [WillyJL]

[Dudlbert]

I couldnt find it but i always wondered: If i manually get games or threads on my list, is it possible to also manually set a picture for it? All i could find was "Run a full refresh..."

Thx in advance!

 

[WillyJL]

I'm thinking of moving to Zorin OS and I wonder how do I do some kind of backup to my games and settings for the app to have it on the new system.

copy %appdata%\f95checker to ~/.config/f95checker

 

[WillyJL]

I couldnt find it but i always wondered: If i manually get games or threads on my list, is it possible to also manually set a picture for it? All i could find was "Run a full refresh..."

Thx in advance!

right click the image, or the space where the image would be

 

[Dudlbert]

right click the image, or the space where the image would be

Maaan! Im dumb and or just blind! Thx!

 

[Mayhap]

WillyJL Umm, not sure it's the right place to ask, but I'll try: you use C=0 in your bayesian_average function, that is - assuming the expected average site-wide score to be zero. I did some tests and I believe setting C=3 (assuming 3 stars being an average score) makes sorting slightly better. With C=0 less popular titles are penalized way too harsh, and popular but not so good ones still get their high ranks just because of the number of votes. Or maybe you could make C a setting - like, for power users who are not quite happy with the default 0? As for W=100 - this one looks sane, no complains here.
(Ideally, C should be a true global average score, but it's too much hassle finding it, even if we try some random sampling as an approximation. That's why I humbly suggest 3 as the next best thing.)

 

[FaceCrap]

(Ideally, C should be a true global average score, but it's too much hassle finding it, even if we try some random sampling as an approximation. That's why I humbly suggest 3 as the next best thing.)

Actually, if I'm not mistaken, the bayesian_average function existed already before the changeover to using a centralized caching server. Amirite WillyJL? If that's true, then the cache should contain enough score/vote data to get a fairly accurate global average... The only problem is, it's also hardcoded.

Otoh...

With C=0 less popular titles are penalized way too harsh, and popular but not so good ones still get their high ranks just because of the number of votes.

There's also a certain amount of subjectivity at play here... If a game is popular enough to end up with a higher weighted score, who's to say it's not that good?

 

[WillyJL]

WillyJL Umm, not sure it's the right place to ask, but I'll try: you use C=0 in your bayesian_average function, that is - assuming the expected average site-wide score to be zero. I did some tests and I believe setting C=3 (assuming 3 stars being an average score) makes sorting slightly better. With C=0 less popular titles are penalized way too harsh, and popular but not so good ones still get their high ranks just because of the number of votes. Or maybe you could make C a setting - like, for power users who are not quite happy with the default 0? As for W=100 - this one looks sane, no complains here.
(Ideally, C should be a true global average score, but it's too much hassle finding it, even if we try some random sampling as an approximation. That's why I humbly suggest 3 as the next best thing.)

scores="$(for thread in $(redis-cli scan 0 MATCH 'thread:*' COUNT 10000000000000 TYPE hash | tail -n +2); do redis-cli hget $thread score; done | sed '/^$/d')"
total="$(echo "$scores" | paste -sd+ | bc -l)"
count="$(echo "$scores" | wc -l)"
echo "$total / $count" | bc -l
# 1.76286776714152910730

so C=1.76

fun fact theres 40363 threads cached, of which 39801 have a score associated with them (the other odd 1200 are probably bogus IDs that never existed on f95 and someone hammered the cache api with, fairly normal in the age of bots)

 

[Mayhap]

so C=1.76

Heh... Well, I expected it to be higher than 3, actually - didn't know there were so many poorly-rated titles. But as FaceCrap noted, "there's also a certain amount of subjectivity at play here", and making the sorting score gravitate to that "global average" is not really a dogma. So, my point is... if you decide to change C, please consider making it tunable for the end user, if it's not too much work. With the default of 0 ("as it was") or this new 1.76 (probably won't change much in the foreseeable future), or something close and rounded, like 2.

P. S. One more thought - I'm not quite sure the site's scores are not already weighted somehow. So, it's all heuristics and "what seems to work better".

 

[sigvar]

this is mostly a very basic parser, it just looks ffor "download" or "downloads" and then keeps grabbing text until it finds an image, so most walkthroughs and ports nd things like that will be picked up. but, this only accounts for the OP and most mods and walkthroughs are usually links to comments or other posts on f95...

This post was quite a while back and I was wondering if anymore thought was put into how to add walkthrough/cheat/etc mods that are posts in the main thread instead of having their own thread.

 

[Fauno36]

First of all, thank you very much for your contribution.

My question is, is an extension for other browsers possible? I use Vivaldi.

 

[ascsd]

First of all, thank you very much for your contribution.

My question is, is an extension for other browsers possible? I use Vivaldi.

from a quick google search (so dont quote me on this) vivaldi is chromium-based
so the chrome extension should work with it, just follow the steps in the first post

 

[WhiteVanDaycare]

from a quick google search (so dont quote me on this) vivaldi is chromium-based
so the chrome extension should work with it, just follow the steps in the first post

Nah, Vivaldi uses the original Opera engine Presto (before Opera moved over to Chromium). Vivaldi is about 0.5% of the market, very niche.

 

[WillyJL]

Nah, Vivaldi uses the original Opera engine Presto (before Opera moved over to Chromium). Vivaldi is about 0.5% of the market, very niche.

this is factually wrong, see below.

First of all, thank you very much for your contribution.

My question is, is an extension for other browsers possible? I use Vivaldi.

as ascsd said, it is chromium based and chrome extensions will work on it. it is explicitly explained on vivaldi's own website:

You must be registered to see the links

since this program is nsfw, google does not allow the extension on chrome webstore, so it has to be installed manually on chromium based browsers. just click the chrome button in the extension section inside f95checker sidebar and it will give you a step by step.

 

[WhiteVanDaycare]

this is factually wrong, see below.


as ascsd said, it is chromium based and chrome extensions will work on it. it is explicitly explained on vivaldi's own website:

You must be registered to see the links

since this program is nsfw, google does not allow the extension on chrome webstore, so it has to be installed manually on chromium based browsers. just click the chrome button in the extension section inside f95checker sidebar and it will give you a step by step.

Well that is just sad, it was started by disgruntled Opera devs to target:
former

You must be registered to see the links

users disgruntled by its transition from the

You must be registered to see the links

You must be registered to see the links

to a

You must be registered to see the links

-based browser, which resulted in the loss of many of its distinctive features.

You must be registered to see the links

(not that Wikipedia is a great source, but I am lazy).

Que sera, sera.

 

[px_vds]

This game is giving me parsing errors in the checker. If this was a sign to remove it I took it.

 

[Popirpo10]

This game is giving me parsing errors in the checker. If this was a sign to remove it I took it.

Same bugs for me

 

[WillyJL]

This game is giving me parsing errors in the checker. If this was a sign to remove it I took it.

doesnt for me.

 

[embed]

PARSING_FAILED

 

[WhiteVanDaycare]

This game is giving me parsing errors in the checker. If this was a sign to remove it I took it.

PARSING_FAILED

These gave me errors last night, but are cleared this morning.

 

[WillyJL]

PARSING_FAILED

the issue did not exist at the time you posted this comment, nor does it exist now.