What Makes a Filter Useful

Everybody, this is a LONG post. I just wanted to leave some thoughts + tips around.

FilterForge currently has a vast filter library with tons of user submitted content. It's fun and it's great, but ultimately it comes a couple issues. Some of the harshest critiques of FilterForge basically come down to a couple things. Aside "it's slow" and the blah blah blah obvious FilterForge issues, there are pretty critical ones like "1/10 filters in the library are actually useful." I actually recall that quote specifically. While I do find that mildly disrespectful, as a professional motion graphics designer, I do find truth in that.

To be fair, us authors are doing this on our spare time and... technically for nothing. It's not like everybody gets "High usage" statuses or editor's picks. The point here is that we're just having fun and being asked for "quality filters" sounds like asking somebody to do free work.

However, as a guy whose been getting at least 5 or 6 free copies of FilterForge, I really felt I should take the time to leave some tips on making "good filters".

First the definition of a "good filter": 1. The filter works. 2. The output is engaging in terms of design, follows the principles of design, has a solid basis for it's design and purpose, and is practical for everybody who wants to use it. 3. The filter has a clear goal both the author and user can see and enjoy.

I think that definition is as dry as it gets. Quite frankly I purposely left out "it's just fun to use it" because every filter is technically fun.

So here are some tips I thought I'd leave around.

1. If you intend on making a filter for the library, you should always make something that works for everybody, not just yourself.

The point here is that you're making a mini program people can download. Typically programs that fail to work have low and hideous ratings and leave a pretty low impression of the developer responsible for it. The FilterForge community has been very positive and respectful about that. You rarely see the harshness you see on websites like Youtube. But ok, as a person responsible for uploading stuff online for people's usage, the filter should really be carefully constructed, mindful of their user, and purposeful in terms of output.

2. Make the filter concise and succinct.

So let's say you're making a filter and you happen to like it, that you copy and paste and start making a new filter based on it. The output... ends up looking similar but not quite, so you upload that with a different name and yay it's in the library. Sure, nothing wrong with that.

In the past, there were authors here who'd upload 10 to 20 variations of the same filter... just with mildly different appearances. Again, no written rule, nothing wrong. But think about it this way. You're working in an office and your boss throws down a giant stack of paper work and asks you, "can you dig out 1 document that I really need amongst this pile? Sorry." Yeah, this is basically what happens with users and their giant filter libraries. It's hard digging out that one filter that works. So it's always been my suggestion: if you have multiple filters sharing the same structure, make it 1 filter, make it usable and fun, and don't bother uploading the rest.

Also... I really hope nobody here thinks this is some kind of downloads / numbers game. Your 100th filter? Congratulations, but it's really just another filter amongst the 10000s of filters already here. So don't make variations for numbers, it's considered flooding, generally rude, and technically it's against the unwritten rules of the internet.

Now there may be a need to MAKE a variation... but we really should avoid that anyway.

3. Have a goal for making something.

So when we do design, we dig out references. We look for what others have done in the past. We inspire ourselves to achieve the level of quality those references have. That's design. Have references for yourself. Help yourself to imagination.

A lot of editor's picks here have a very clear goal in mind. Filters that have a clear idea just looks better and are designed appropriately while maintaining a very very consistent idea.

70% of the filters in the library, probably don't have "goals". They're uploaded based on whether if it looks cool or not. That's ok, but that "cool" factor wears off in 10 minutes or so. As much as I find great filters here, I constantly find ones that look cool but don't go anywhere beyond that. It's not just about being cool, it's about making something that has a point.

4. Good design leads to happier people, more motivation, and ultimately makes you just better.

Since I picked up FilterForge, I think I've gained some level of insight into many other aspects. For example, designing a filter isn't just about making cool textures or making an image look even cooler in the output. It's about how you can engage with others, that is, the people who download your filter. It's not like everybody cares about these little things, but taking the effort to be considerate while bold in your imagination pretty much makes damn good filters.

Even if you do art just as a hobby, just try studying what qualifies "good design" in this world. Go to websites like Behance and look at the top notch works by other designers there. Maybe you won't make it, maybe it just isn't your thing, but my point is that these people have achieved something regardless of how you think about it.

Good design is always rewarded and that reward grows on you even if it isn't your goal. It's fun doing good stuff for others.

==========
There, that's about it for now. I definitely have more, but let's just keep it at this. I'm sure it's already too much reading material. But I hope this kinda helps see things.

Two elaborations on points you made come to mind.

Firstly, as you said, "variations for numbers, it's considered flooding, generally rude, and technically it's against the unwritten rules of the internet." This I see as a sort of perverse incentive on the part of FF when any member can submit and does so with great frequency hoping to hit the bulls-eye sooner or later shooting in the dark so to speak.

Secondly, "Good design leads to happier people, more motivation, and ultimately makes you just better. " I totally agree, and would go further to suggest the repeal of anonymity so that users who do flood for malicious and selfish purposes are shamed publicly.

How's that for dry?

Goodness, I thought there were only 1000 or fewer things one could do on a smart phone. My take away fr om that article is that quality standards are wh ere a good editor comes in. Here too its helpful for those who don't have the time to pick through the pile, to get to the gems easily via the 'Featured Filters'.

Quality over quantity sounds right to me. Another issue is novelty or uniqueness. That's my guide. Of course knowing what is unique is not an easy task, and the ever shifting sands.

Simple and concise is also paramount. I'd still be producing music if it was NOT for the " I just need one more piece of equipment " mentality. First Logic, then a DX200, then Reaktor, then Ableton. 4, 16, 128 tracks oh god I can't even turn on the power anymore without feeling sick due to the over complexity of the system I've built. Same goes for the number of controls here in some of the filters!!!

I'd appreciate a thread here on poaching high quality clients, if that's not oxymoronic.

Quote
I'd appreciate a thread here on poaching high quality clients, if that's not oxymoronic.

edit:

in other words:

I want to profit off of a piece of string here in the hunt for rich bosses that aren't fools with too much oxygen.

or possibly according to google translator:

If it is not the oxymoron, I, I think that you thread here in poaching customers of high quality.

Hello John Worthing.

Good advice, man. Ramlyn, don't underestimate the popularity of complex, specialized filters. The windows of a skyscraper, or a bloody sawblade, or putting an image in a fish tank... apparently lots of people need these things!

Quote
...I wouldn't mind putting down some ideas anybody can easily do to just make filters simpler and concise...

Sounds like prime Wiki material. If we actually used the FF Wiki...

Hmmth ... a bit off topic.

After reading what DJI left here:

Quote
"My dear Ernest,to set next to a man who has spent his whole life in trying to educate others!"

It made me think: Am I really just being Ernest here?

In all seriousness, I teach general 3D texturing, lighting, and rendering part-time. I lately don't have too much time but I try my best to stay around and help folks out with all those issues. I also don't like ignoring issues. So I kinda see my life just doing all that, practicing, teaching and learning art for a living.

I just didn't expect to see The Importance of being Earnest here.

Probably took that too seriously but it's just a thought I had.

Back on topic please!

Thanks people! I probably would need to clean up a ton of thoughts. I just hope people look at the wiki.

One reason why I've been kinda writing a load of these things is because over the last couple years I noticed this trend of filters that appear to have something going on... but somehow lacking in technical design. Not that these are bad! Some folks are just starting! It's more so that they get uploaded to a PUBLIC library.

The key difference is simply that a public library kinda has higher expectations. Users who have purchased FilterForge DO expect high quality content from the library.

There's also another point: that if base standards are followed, constructing filters would actually become easier, more intuitive, and higher quality. It's ok to have your own ways going, but a program works only in certain ways anyway.

I think a load of filters basically just lack technical and aesthetic depth. I mean... a good number of recent filters in the library just look basic. While it's no surprise that there's a huge number of those "basic" filters it always seems so dominant. Lately, there are more "basic" filters sitting in the library recent list than those which exhibit a bit more technical understanding.

Ha! Ha! Ok Skybase, I'll let you know if there is any chance of business.