Ramlyn
![]() |
This period I checked many times how the filters go up/down..
Like many other filter creators, I would like to make filters that the people can like and use. Looking at the "Popular" and "Recent" tab in the site, I could check everyday what the users liked to download and what was new. So I thought..... "And the High rank filters?". Mystery. FF says that, if we do good filters and the people use them, we can get Reward Points. But........ if there is no way to understand what filters the people generally use, we have no idea about what kind of filters we should do. Let's make a practical example. ( Sorry if the example is very... materialistic ). Let's suppose that I didn't buy FF and I want to get Reward Points to win the program. I arrive here for the first time and "Oh! Very nice program. Oh, if I do good filters, I can get it free". I do 10-15 filters and.......................................... ???? There is no place for understanding if my filters are going well or not. Please, don't say that there is the High/Average/Low ranking, because there are about 3000 filters in the average line. If I had to take it as reference for understanding if my filters are going well or not, if I had to change/update anything, if I had to make new filters, how can I understand? The top of the 3000 and the last are all in the same group with no way to see points or if they are going up or down. So.................................. What if we consider only the downloads? 1. We exclude the filters that already got a reward ( let's put them in the "Hall of Fame" ) 2. We check what are the filters that succeed to stay on the top 10 for at least 2 weeks. Those filters get a reward point and they get out from the list. If we do like this, it is a clear way for everybody. When our filters go down, we know that we have something to update. When our filters don't even appear on the top 10, we know that they aren't enough good or attractive. Now, we understand nothing. Even it was a random pick, like a lottery, it would be the same. |
|||
Posted: September 29, 2014 4:39 am | ||||
Ramlyn
![]() |
Hummm..........
Nobody said anything. ![]() So..... let's take a real example. My case. I made 47 filters. One got a reward. All other 46 are on the average level. Let's suppose that I want to try to get a second FF license. To do this I should have an idea, at least among my filters, about what ones are more appreciated and what ones aren't. Is there any place to check it? I don't see any. Maybe I simply don't know where to look. That's why I thought that probably considering the number of downloads would be easier to understand. It would also not require to add anything, simply to consider a list instead of another. Maybe nobody is interested. Who knows. |
|||
Posted: October 3, 2014 3:13 pm | ||||
Indigo Ray
![]() |
Well, doesn't data like that usually cost money? But maybe you can get around that.
With lack of data on your own filters, you gotta look elsewhere. Do other people's high-usage or editors'-pick filters stand out in any way? What filters get a lot of comments? Look at the Popular Filters list: There are newer filters there, but which older filters consistently make the first few pages? What filter ideas do people ask for in the forums? What kinds of filters are featured by FF on their facebook page? Which filters are used by other artists in the gallery or elsewhere? Maybe the legendary Crapadilla's FF self-help guide ![]() Anyway, free licenses are nice, but art means more to the creator than its popularity, especially when you aren't making money off of it (others using your filters). ![]() Side note: Anyone else remember, years ago, FF had a "most downloaded" list instead of a "popularity" list? Or am I just making this up? |
|||
Posted: October 3, 2014 5:26 pm | ||||
Skybase
![]() |
Here's just a thought:
Downloads would work as a quantitative set of data. So the more downloads = more popular seems reasonable until you think about what FilterForge is. The trouble is, downloads, in this case, doesn't quantify "quality". When you're downloading something on FilterForge, you're basically downloading a little program. While each download has a preview, you can only know the outcome by actually testing the filter on an image. Therefore, you really can't appreciate "most downloads" as a qualitative assessment of "good filters" since users download based on a preview image and not on a try-out basis. Downloads WOULD work we were just uploading pictures or video since you can assess the quality of those directly online. That's just not possible with FilterForge. |
|||
Posted: October 3, 2014 7:25 pm | ||||
xirja
![]() |
Considering that people have their own armies of fan boys, spam bots, or whatever we may call them, I don't even know how high usage rewards even makes sense in the first place, except to benefit the word of mouth appeal of FF as a generous group of talented people. But as this is a moot point, yeah I agree with what Indigo and Skybase said. Editor's picks can be predicted with greater frequency after a few years of careful aesthetic and technical analysis. Occasionally I'm surprised by the choice, but not often. Most high usage items on the other hand leave me dumbfounded as to how it can be.
![]() _____________________________________________________
http://web.archive.org/web/2021062908...rjadesign/ _____________________________________________________ |
|||
Posted: October 3, 2014 11:30 pm | ||||
SpaceRay
![]() |
It can take many months until a filter is changed to high and get a reward point, although this is not a real measure, because it all depends on the usage frequency and how many people is using it, so there is really no time measure, it could be 2 or 6 months or 1 year, or never (if it is not used much and enough)
I agree with Ramlyn that there is NO way to know the evolution of a filter and know if this filter is being used frequently or not as the problem is that the ranking is like a light switch, there is ONLY available Average and High, because low takes a lot of time to be considered to be assigned. So after submitting the only real state of most filters is Average and there is no way to know how a filter is getting used and how near or far is from the High Rank HAVE MORE THAN 2 RANK LEVELS To be able to have a better understanding and now better the evolution, success and popularity of a filter it would be possible to have at 3 rank levels (Average, Medium, High), or even better have 4 levels (Average, Medium, Popular, High) so this way with 4 levels it would be much better to know how a filter is going. DOWNLOADS COUNT As Skybase said, downloads is not a measure of quality or popularity for knowing the usage of a filter, as it could be downloaded 1000 times but users just use it 3 or 4 times to see it and then leave it. |
|||
Posted: October 3, 2014 11:47 pm | ||||
Ramlyn
![]() |
Thanks to everybody for the answers!!
![]() I agree about many things that have been written. But still...... hummmm....... We have 4 "rankings". I call all them "rankings", in very broad way, just to use a single word. 1. "Featured" / "Editor's Pick". This is just FF choice. It depends by their taste and opinion, and it doesn't give any point. They surely have their own criteria to choose and as the name says, it's just up to them. Clear and no problem. 2. "Recent". It only shows the most recent or recently updated. It is well ordered and grouped. Also here no problem. 3. "Popular". The most downloaded. Also this ranking is well made. The beginning it was a bit "???" to understand why recently updated filters suddenly jumped on the top, but it is clearly explained with the fact that everybody update them. Anyway there is a clear position of every filter and a clear number indicating the downloads. 4. "Usage". This is not clear at all. What position is every filter? No idea. How many people used a filter? No idea. How many people should use a filter to get high rank? No idea. If the same person uses a filter 300 ( just example ) times, does it make the filter high rank? No idea. And the problem is that this is the only one ranking that gives the points. It should be the most understandable one. And, sorry, but at the moment it is very hard to understand what happens looking at : - Downloads : really, there are A LOT of filters that got hundreds or thousands of downloads, but no rewards. And others with many less downloads but that got rewards. - Rewarded filters : first of all because they are many. And second, and even more important factor, because many of them were done long time ago. Even I make a filter similar to what somebody made last year, it doesn't represent what the people use now. I don't mean to do new things. As it was said, setting more accurate information about how many times the filters are used may require costs ( even if........ FF should have this data easily available, otherwise how they can calculate the usage rank? ). My question is: why don't we simply cut the usage rank and we simply gives the points on the "Popular" rank? There is no more work to do and it would be easier to understand for everybody. |
|||
Posted: October 4, 2014 6:07 am | ||||
Sharandra
![]() |
But the usage rank is what shows if a filter is usefull to ppl. It´s how often it has been used, I don´t understand what your problem is with that? Giving the points to most downloaded filters would be bad, because as Skybase said, you have to download and use a filter first before you know if it is any good. And it would be unfair to award points to a filter that is crap, just because ppl downloaded it out of curiosity. Indigo already gave you good advice on what to consider if you´re after reward points. You need to make quality filters, preferably something that hasn´t been made before and is in demand. With 10k Filters in the library, it´s not easy. Don´t expect to do it within the 30 days trial unless you make some really outstanding filters. It took between 6 months to a year for my filters to reach high usage (the ones that did ![]() Good luck and happy forging! ![]() |
|||
Posted: October 4, 2014 9:41 am | ||||
Ramlyn
![]() |
The problem is, as I wrote, that there is no way to see if a filter is used or not.
And, just because we cannot expect a result in 30 days, it would be fair to have an idea about it. Or not? About "usage vs downloads" in the ranking, I don't see any problem in choosing the usage. I suggested the downloads just because there is already a well defined ranking ready and FF wouldn't have to do any further work on it. Anyway it is fine, probably most of the people aren't interested. ![]() |
|||
Posted: October 4, 2014 10:31 am | ||||
Indigo Ray
![]() |
My interpretation of FF Usage Stats and ranks have changed several times (I wonder if the text has ever changed...)
Currently, I think usage works like this: Usage stats are collected from each user that allows it. The stats are a most-used list: The filter you used the most during some time period is ranked #1, the one you used second-most is #2, and so on. FF doesn't collect actual # of uses, just ranks. Somehow, individual most-used lists are combined, and the result is the "popular filters" list on the website. I don't know how often this list is updated, but I think it's multiple times per week. "Usage" = "popularity" NOT = "most downloaded"! Say you have a new filter that makes near the top of the list consistently. If it keeps this up for some period of time (months?), it will get promoted to High Usage. How "near top" a filter has to be, I don't know. Filters are already ranked by popularity. You can already find the exact rank of your filter by going through page after page after page in the popular filters list every week (or more often)! It would be nice if that was made a bit easier! Maybe exact usage rank # could be displayed, publicly or privately to the filter author. SpaceRay's idea of extra rank levels could also help. |
|||
Posted: October 4, 2014 5:41 pm | ||||
Skybase
![]() |
I just think it's overall just intended to be vague indicators of performance.
I also personally think if we had some kind of rating system some folks would find it very unsatisfactory. ![]() |
|||
Posted: October 5, 2014 3:45 am | ||||
xirja
![]() |
As FF isn't licensed per machine, why couldn't I have 100 machines with different IP addresses running my filter to get free copies of FF?
![]() oh yeah, ![]() _____________________________________________________
http://web.archive.org/web/2021062908...rjadesign/ _____________________________________________________ |
|||
Posted: October 5, 2014 4:36 pm | ||||
Ramlyn
![]() |
You don't probably need 100 machines. If you use a router to mask your IP, save your filter, then change location, save your filter, then change location, etc etc....... it may work. Who knows. 555+
I have never tried, so I can't say. ![]() |
|||
Posted: October 6, 2014 4:39 pm | ||||
Ramlyn
![]() |
We all have our opinions.
![]() |
|||
Posted: November 29, 2014 6:26 am |
Filter Forge has a thriving, vibrant, knowledgeable user community. Feel free to join us and have fun!
33,712 Registered Users
+19 new in 30 days!
153,534 Posts
+27 new in 30 days!
15,348 Topics
+72 new in year!
5 unregistered users.