Your blog post has a questionable video by a manufacturer of a product as its main source, all the useful information looks very much like a rewording of an article on the URPro website. The rest has very little substance, the Sony rx100 is a good camera but using it for what you are is quite laughable and shows how much credibility your scientific research has. Beyond testing with your calibrated scientific grade rx100 lol what is there in that post that tells me anything of importance that I can't get from the urpro filter article myself.
URPRO Patented Underwater Correction Filters
Beyond that you have opinions based on your own theories with no practical tests or results shown anywhere in the blog, again like I said an AWB gopro will give random results as the algorithim used is for above water full colour spectrum and freaks out at times when underwater due to the diminishing colours with depth. This is greatly improved with a filter, you can see any of my results or the millions of other users out there. Even though filters aren't intended to be used on awb cameras the go pro without a filter will really struggle very shallow.
You will get better overall results with the higher grade materials as you know when it comes to optics, results may favour an overly dark filter in some poor light conditions due to it forcing more red by filtering others out much stronger. This will lead to more light loss and in good ideal conditions where filters are best this will degrade the image. The filter colour is one thing and the material is another, filters are not really intended to be used with awb cameras but surprisingly help the go pro quite a bit to set a better wb in auto.
Like I told the guys at polar pro I could easily post video favouring one filter over the other if I cherry picked the footage to where one gets awb issues and the other doesn't. This has nothing to do with the filter but the cameras awb, but if you actually look at the results from each you can tell where the better filters give better results in terms of overall colour. The magic filters are great at shallower levels and the urpro works to slightly deeper depth some get close to these but many are struggling more due to colour and cheap materials used.
For what its worth the polar pro videos I have seen show way too much red at normal working depths and so this will likely help the gopro wb in poor conditions but hurt the image when its good.
I think the PP gel sheet would most likely be a much better option then the clip on they made simply due to materials. Gel sheets are proven to work and are a common low cost option for filters. A good one may be as good as an external premium filter. Disadvantage is you need to put it inside the housing which is not convenient and can't be easily added or removed even on land. Going for cheap injection molded cubed plastic hasn't been used by anyone for good optics for a reason. Its a cheap option, I would rather use better equipment but many people just want the cheapest and if its good enough for them then everyone is happy.
The main purpose of a filter is not to help with wb but let the camera sensor capture more colours that it doesn't due to the strong and dominated blues in the water, a manual wb camera will give the best results every time as white balance and CC are two different things. You can get a great wb at depth with no filter but the image will lack colour combine the 2 and you get the best results. That's the main purpose of a filter, our eyes are better then any cameras sensor yet we lose colour at dept if we can't see it what chance has the sensor got of picking it up. Add a filter you reduce the dominant blues and you will pick up better colours then without. Pretty simple really but you need to get a good wb also and this is the trick with awb in the go pro that produces such inconsistent results at times.
Original GoPro's only did a full auto white balance and as anyone knows this sucks beyond 5m in most cases, using a filter will improve this greatly and there are a few to choose from. The main reason I would have liked to try the polar pro is to see how it handles the corners as that design is likely going to have an effect on the edges in full wide modes.
There are still very few user videos showing the pp and even the 5 they gave away haven't produced many videos. Heck if they sent the one to me they said they would publicly I would have tested it at the first good opportunity. I would have bought one just so I can test it and post some findings as I did with all the other things I have bought or been sent at times.
---------- Post added January 7th, 2013 at 12:56 PM ----------
If possible, all of this would be so much easier to compare if the subjectiveness was removed by doing a simple video with a Macbeth color checker or even something as simple as a box of pool balls. Shoot the chart or the balls above water in full sunlight then get in the water, take it down slowly with a depth gauge in the shot showing the depth as you slowly descend and watch the show. You could easily see the results of these filters as apples and apples since you're comparing known colors instead of guessing what a fish, coral, sand should look like....
This was the way we used to check film way back in the pre-digital age. Photographers would even check batches of film this way. For example before a National Geo Graphic photographer would go out on a 3 month photo expedition he would shoot test rolls of different batches of film and take 400 rolls of the batch number of the one he liked the color rendition the best.
This simple method would demonstrate these filters much more effectively.
Mike that is a good idea but only if you have a fixed WB mode not auto as the shifting WB makes comparing colours of filters fairly pointless side by side. If you dont have a consistent camera that is constantly shifting colour balance trying to compare 2 filters side by side when a camera without a filter can be so different is as I said pointless. You can cherry pick either one to look better then the other, if you pick the best shots from each then you can get a fairer comparison comparing them at their best not when one has AWB issues and the other doesnt.
I should have posted this earlier on but this is where I first noticed how bad the GoPro's AWB is. This was my first 3D dive using the EOM 3d housing, its only very shallow at Rye pier around 6m max but no filters or colour correction has been added. 2 Cameras side by side in a 3D housing produce such different results and I saw the same things in my 6 camera test at Fish Rock which I have just finished rendering and will up soon also.
[video=youtube;T35aPmaQDWM]http://www.youtube.com/watch?feature=player_embedded&v=T35aPmaQDWM[/video]
Its a 3d video but if you set view settings to side by side you will see how much the images vary, the AWB is almost random and has always been my biggest gripe about the gopro cameras. Interceptor you said my newer videos are getting much better and the surprising thing is it has been getting better since using the SRP gear and the reason I am a big fan of them. Even if you say its so terribly engineered that it should be casing me all sorts of issues compared to an alternative setup. Id like you to post me a link of what you would consider a setup that will give me better results at a much cheaper price and why.
The new 3 Black with cam raw has given me consistent workable results in every single frame, the auto mode without filters will give this sort of results on the good shots and some end up completly useless. Once I started using filters videos improved greatly and I was impressed how well the hero1 did in Thailand with a URPro CY filter to 30m depths in comparison to without a filter at 5m. That is why since using filters I have never done a daytime dive without one. The new 3 Black will actually do quite well without a filter too as you can set WB easily but it lacks colour at depth which the filters bring back.
I tried to do some more testing with protune on the hero2 to compare filters but found the exact same issues, I will do some filter comparisons with the 3 in cam raw though at a later date which will give results with much more meaning.
