The flaw in the IR-Cut Filter of BMD cameras

So what I learned so far:
1- The original IR-Cut filter from BMD cameras is susceptible to corrosion, under heat and humidity (not a lot actually).
It might be because it's a plastic layer (this part just a theory), and they cut costs with it.

2- You can buy a replacement from eBay (some Chinese manufacturers of optical filters specifically for the BMD's cameras), so it might be the same or similar to the original, with the same similar corrosion weakness? (link to it: https://www.ebay.com/itm/273972523800)

3- You can also buy a replacement, that as an additional benefit fix the moire and do a better job at IR noise (https://rawlite.com). And as an additional benefit, it has an extra layers of glass in between. I'm quoting the manufacturer at RawLite, Hans Hijmering, "Most likely surface corrosion due to moist. Blue glass is susceptible to corrosion. The blue IR glass of our Rawlite OLPF is sandwiched between two optical glass layers and the coating also protects against corrosion."
You see, that's a very good reason to buy it.

But my question still remains... why BlackMagicDesign didn't foresee this problem? I'm sure their optical engineers knew about this weakness. And why I need to pay for this flaw? =/

BTW, if someone at Blackmagic Design is reading this... you can still reimburse me, at my PayPal account, if you feel like.. you know, doing the right thing

Oh, and yes, I tried to contact support service and never got a reply

Well, it seems that there is either sample variation or it depends a lot on the environment.

In my old BMPCC (first gen.) after all those years there is just a hint of the damage around the edges, not yet affecting the image, even if that camera has been used a lot in a tropical country. But then, it was stored in a dry cabinet when not in use. In that environment I've seen the Canon lens of a family member getting mould stains after less than one year while not stored adequately. So, I can't really blame BMD.

That said, I can't recommend the RAWlite filter enough and BMD should get together with Hans and offer it as a paid option from the start.

I can also recommend the RAWlite filter to everyone

This is all about camera price, developers practical knowledge, available supply resources and company marketing. But i agree, at least BM could change filter coating over this time. Problem with sensor glass corrosion was known from first BMCC camera release. Also as i remember one camera generation was delayed about 6 months or so due problems with IR glass suppliers and quality control. Some people reported that BM support just send them replacement glass for request.

In optics mechanics and electronics there is always huge amount of accident things. Some problems are easy to predict some don't. For example recently i got Rawlite OLPF filter for secondary Micro Cinema Camera. As it appears i was the first person shipped with new version of the filter. New version of filter was sharper and was optimized for ProRes, but same time it provided not enough filtration for RAW. Communication with Rawlite was quick and easy so they offer to ship old softer version for additional testing for free. During further testing by accident i find another problem - purple light leak reflection from internal edge of the filter glass when camera pointed to the light under some very specific angle. Now Rawlite is working on complete frame redesign for BMMCC filters. As you can see even basic things may provide hidden problems. Unfortunately we are always potential betatesters for our own money

By the way it appears that Rawlite In addition to Moire/aliasing filtration and IR pollution filtration adds color temperature shift to improve colors and provide less noise under tungsten light. So it is really recommended option.

If folks at BMD continues to use low quality coated glass, backlash will be in for sure. Had the same issue with the 1st BMPCC and now with the 6K after 13-14 months of usage. Stored in a dry cabinet in between shoots. Looks like I can only use indoors with controlled environment based on the replies by tech support.

Not to be to contrarian. If my camera sits in a studio shooting products do I really need the added expense of that filter? But I agree it would be nice to have it as a seperate SKU at purchase time, especilly if BMD could source them at a discount.

Now Rawlite is working on complete frame redesign for BMMCC filters. As you can see even basic things may provide hidden problems. Unfortunately we are always potential betatesters for our own money

This is both great to hear and disappointing since I bought two Rawlite filters for my own two BMMCCs within the past year. I wonder if Hans will offer a discount to past customers, although so far I've been happy with mine (I used to shoot ProRes but am now shooting Raw 3:1).

So one more box with RAWLITE OLPF for BMMCC is finally here. Done very quick and not too accurate tests:
- It fits to BMMCC.
- New body seems slightly thicker, so there is slightly less space for rubber ring, and seems ring clamps slightly less secure compare to older filter with plastic body. But overall filter holds well.
- Seems no more flares.
- Statement from Hans was: "stronger than the first one you received. No more moiré with .dng". But subjectively from quick test i can't see difference compare to earlier version. It still skips some moire when shoot DNG. (but noise reduction and some post processing in Resolve usually allow to clean patch 9 very well)
- It is way sharper than Mosaic OLPF. ProRes with RAWLITE OLPF is sharper than DNG with Mosaic OLPF.

raw lite is not so effective against aliasing and moiré are still visible?
If I can still see moiré at 200% (as in your example image) then i can also see it at 100%. For that little bit of moiré reduction, rawlite is just too expensive.

Reducing moiré afterwards in Davinci Resolve is also a possibility.... it costs nothing.

My RAWlite is doing much better, this is a DNG out of my UMP 4.6 G1:
https://www.dropbox.com/s/32k69wow5peauvl/DNG_mit_OLPF.png?dl=0

And this is a BRAW before I got the filter:
https://www.dropbox.com/s/s24mxreh57xy6fx/Braw_ohne_OLPF.png?dl=0

Is there a similar comparison with micro or pocket?
On the example image from Dmitry it looks different. Moiré is filtered out less than with Ursa mini Pro camera.
The Ursa mini Pro has little aliasing and rawlite reduces moiré better than the lower resolution micro and pocket cameras. The aliasing on micro and pocket (not 4k) is bigger. I think it only works when the sharpness is reduced a lot?

Yes, a larger raster needs stronger filtering or a different balance between moiré and resolution.

can definitely recommend the RAWLITE. I have it on my G2 and love it.

Also contemplating getting it for my P4K if we decide on using it as a B-Cam on our feature. Been using the Tiffen IRVND for a long time and I'm pretty sure they were discontinued, might go Rawlite for the pocket, not sure yet.

Yep, it is known that RAWLITE on Ursa produce way stronger filtration at 1:1 pixel level because resolution is not a problem when use 4.6K sensor.
Filtration on BMMCC/BMPCC is weaker because at 1080p native resolution there is always a compromise between sharpness and moire filtration strength.

My filter and BMMCC still on the shelf and i still didn't test it side by side with Mosaic because i had some other things to do now... I ordered RAWLITE filter for project in Summer, but due stupid bugs with flares and wrong filter frame sizes it was not happen in Summer...

That RAWLITE is too expensive, any other solution that it longer lasting? Not exactly fun to replace the IR filter every 6 months.