Digital Bolex weeks away

I would like to see the DB with some Zeiss Super Speeds and Linos MeVis C-mount lenses. The Zeiss' are my favorites for the Ikonoskop and the MeVis C-mounts are incredibly sharp compact lightweight lenses that have a very modern clean look.

Joe has been shooting with Zeiss CP.2's on occasion. The camera can be very sharp with good lenses and processing if you want that look.

10093163075_8f8efb977b_b.jpg
 
The D16 footage will look soft if not sharpened in post. But you can push quite a bit of sharpening without getting bad artifacts as you would with a BMC due to no aliasing issues. Of course people will criticize that as looking like video.
??????? In the next post, you say " The D16 resolves edges with a much smoother more organic quality ......."? Which is it?

If you use post sharpening (detail enhancement, aperture correction, call it what you will) I cannot see any way you can describe edge resolution as having "a much smoother more organic quality" - the two things are complete opposites! Start to use post sharpening and it's the quickest way towards "video look" that I can think of.

And this is the problem with Digital Bolex using a 1920x1080 sensor, as I said many posts ago. The best deBayering won't get full HD resolution from it, period, and since aliasing onset is mainly determined by inter-photosite spacing, if you don't want aliasing you need a very aggressive OLPF. Which will make it softer still.

Which is why the first prosumer full HD (3x1920x1080 chips) were so good. If you wanted the "cinematic" feel you dialled the detail level DOWN. That was what gave the "organic" look to edges - but because of the three full HD chips, there was still good natural fine detail. Start with a single 1920x1080 chip, and you're starting with one hand tied behind your back.
 
Joe has been shooting with Zeiss CP.2's on occasion. The camera can be very sharp with good lenses and processing if you want that look.

10093163075_8f8efb977b_b.jpg

Looks great, the lighting is not quite my taste but the image has a very nice quality to it. Its funny to see when any new camera comes out everyone complains that the first samples are not what they would shoot and why can't someone shoot x,y or z. Theres always a frenzy to be the first to yell success or failure first.

The feature I'm most jealous of that is missing on the Ikonoskop is more full featured audio recording.
 
Last edited:
Looks great, the lighting is not quite my taste but the image has a very nice quality to it.
But the attached image is only 1024x576. I don't disagree with comments about it looking good, but isn't that more down to the photography, lighting etc than intrinsic abilities of the camera?

If we're trying to make sensible judgements about the sharpness of the camera, I don't think they can be made on the basis of a 1024x576 downscale? What does a 1920x1080 or 2K frame grab look like?
 
Looks great, the lighting is not quite my taste but the image has a very nice quality to it. Its funny to see when any new camera comes out everyone complains that the first samples are not what the would shoot and why can't someone shoot x,y or z. Theres always a frenzy to be the first to yell success or failure first.

The feature I'm most jealous of that is missing on the Ikonoskop is more full featured audio recording.

What I love about both the D16 and the Acam DII is the rich refined color palette in raw. There is definitely a learning curve and the Kodaks are not the easiest sensors to calibrate or grade, but the range one has to work with is amazing.
 
But the attached image is only 1024x576. I don't disagree with comments about it looking good, but isn't that more down to the photography, lighting etc than intrinsic abilities of the camera?

If we're trying to make sensible judgements about the sharpness of the camera, I don't think they can be made on the basis of a 1024x576 downscale? What does a 1920x1080 or 2K frame grab look like?

OK

10093163075_9522d0bed9_o.jpg
 
??????? In the next post, you say " The D16 resolves edges with a much smoother more organic quality ......."? Which is it?

If you use post sharpening (detail enhancement, aperture correction, call it what you will) I cannot see any way you can describe edge resolution as having "a much smoother more organic quality" - the two things are complete opposites! Start to use post sharpening and it's the quickest way towards "video look" that I can think of.

And this is the problem with Digital Bolex using a 1920x1080 sensor, as I said many posts ago. The best deBayering won't get full HD resolution from it, period, and since aliasing onset is mainly determined by inter-photosite spacing, if you don't want aliasing you need a very aggressive OLPF. Which will make it softer still.



Which looks smoother and better resolved to you? Which looks more film like? Full frame the BMPCC looks superficially sharper, but it is illusion created by false detail from aliasing, not real resolution. This is just a simple diagonal line example. Lack of OLPF creates problems with some demosaic algorithms too that result in chroma aliasing you don't get with the OLPF. When chroma aliasing appears, your visible resolution is significantly lower than with the OLPF. The quality of the demosaic algorithm also has significant impact. These were processed with an advanced algorithm that is good at eliminating chroma aliasing. It won't run real time in video cameras as it takes about 60 times the processing power.

Not arguing the advantages of resolution oversampling for single chip bayer cameras. But the D16 is plenty sharp and well resolved for HD or 2k cinema. The D16 is a native 2k camera 2048 x1152 16:9. It resolves 2k cinema formats better than 1080p video upscaled and does a perfectly fine well resolved 1080p too. Bayer raw is not nearly as compromised for image quality as any AVC 4:2:0 8-bit compressed CMOS video camera, oversampled or not. About all you will find other than the BMC's in the D16's price range.

aliased diagonals.JPGBMPCC no OLPF

OLPF diagonals.JPG D16 with OLPF Note: there is a black line on the slate between the green and yellow bars
Both at 300%.
 
Last edited:
The day we start watching films at 300%, these types of comparisons will mean something.

There seems to be an essential point that gets lost in these magnified pixel debates - perceived image quality is image quality.
People perceive images...they interpret them just as they do audio, and all kinds of things get done to both audio and video that make them "appear" better to the person that experiences them.

That's all that matters - how our work is perceived by the people that experience our work - and no one views any film with magnification.
So all that matters is how things look to the public at 100%.

For instance, I just shot a film test using the BMPCC, of a model, that in every way important to the end viewer, looks as good as if it were shot on an Alexa.
It was very well lit, against a beautiful forest in a steady snowstorm. It simply looked amazing with the thing most of us like most about film - smooth but detailed imagery and organic color and DR...in 1080 @ 100%.

By contrast, the deli comparison here looks really not amazing.
It looks amateurish, poorly lit and unstable. It's hardly an example of what this camera can do, I'm sure.
There's so much that needs to be done for it to look good that comparative edge sharpness would rank infinitesimally small by any viewer.
It means as little to what we do professionally as does a 300% fabric comparison.

Wait till some professionals do some real composition and lighting with proper tools before comparing it to anything.
I really enjoy the debates and comparisons of new tech for a number of reasons, but you really need solid work to do that.
 
Last edited:
I've seen some footage that Joe shot in the desert, and it looked pretty stunning. Very rich, the studio stuff didn't really impress me that much but the landscapes looked very nice. I'm not sure why they don't post it. This camera has a lot of potential.
 
Razz16mm, your comparison is the stupidest thing I have seen today. If you point both cameras at the same object or chart and you crop 300% on exact same part, then you can make a point. Your comparison is completely pointless.
 
Last edited:
Razz16mm, your comparison is the stupidest thing I have seen today. If you point both cameras at the same object or chart and you crop 300% on exact same part, then you can make a point. Your comparison is completely pointless.

I chose these examples because the diagonals are at nearly the same angle. In life a diagonal line has infinite resolution, so at the same magnification you would see the same relative results between these two cameras regardless of what they are pointed at. you would see aliased distorted edges from the BMPCC and smooth well resolved edges from the D16. You will see chroma moire at critical spatial frequencies on the BMPCC and you will see none on the D16. People keep saying the BMPCC is higher resolution than the D16 based on superficial judgements of compressed web footage, this is absolutely false. As the above full frame 2k image I posted shows, the D16 with primo glass and high quality processing can produce very sharp well resolved images without incurring serious aliasing artifacts of any kind. Throw both of these cameras up on a large screen with a high quality projector and the differences are immediately apparent.
 
Last edited:
Throw both of these cameras up on a large screen with a high quality projector and the differences are immediately apparent.

Really?
I didn't hear about that large screen comparison, was there a digital timer gauging the speed at which it became apparent to everyone in the audience?

If what you meant to say was, "the differences would be apparent", then it would be good to clarify, because on forums like these, comments like those lead people to spread FUD.
Here's an example: "Hey did you see where the Bolex discussion over at DVX says that a big screen comparison of the BMPCC and the DB immediately showed the DB to be immediately superior".

I mean creative opinions are subjective by nature.
There's no need to pretend that's there's more to your opinion than there is.
 
Really?
I didn't hear about that large screen comparison, was there a digital timer gauging the speed at which it became apparent to everyone in the audience?

If what you meant to say was, "the differences would be apparent", then it would be good to clarify, because on forums like these, comments like those lead people to spread FUD.
Here's an example: "Hey did you see where the Bolex discussion over at DVX says that a big screen comparison of the BMPCC and the DB immediately showed the DB to be immediately superior".

I mean creative opinions are subjective by nature.
There's no need to pretend that's there's more to your opinion than there is.

I've had samples of both, along footage from other cameras in the general price range up on a 15 foot screen at the shop using a $60k 3DLP projector we had in for client demos. So in this case it is my own personal experience and opinion vis a vis the BMCC, FS700, BMPCC, Canon 5D3. Original footage samples downloaded from various people's posts, processed and graded by me to explore the limits of each. In my opinion the D16 can produce an image that is remarkably free of the range of common artifacts that plague most low end cameras these days. Aliasing and moire being characteristics I personally find extremely annoying along with the usual artifacts and lack of creative flexibility of highly compressed 8 bit codecs.
All of these cameras can produce pleasing video under good conditions, they all have their issues too.

I've ordered a D16 for myself. It best fits my intended uses, which are not commercial in nature. In some instances the final results won't even be encoded to common video codecs, at least nothing lower than full 10 bit 4:2:2 uncompressed broadcast HDSDI.
 
Last edited:
Having used 16mm film quite a bit in university and to make a few small films back in the days, I have to say that something seems to missing from the D16 footage.

Not sure what that is though, something intangible that I can't put my finger on. But it doesn't seem to capture the same look as the real thing, whatever it was. Would like to see this footage on a big screen to compare it though.
 
Having used 16mm film quite a bit in university and to make a few small films back in the days, I have to say that something seems to missing from the D16 footage.

Not sure what that is though, something intangible that I can't put my finger on. But it doesn't seem to capture the same look as the real thing, whatever it was. Would like to see this footage on a big screen to compare it though.

Haven't seen a digital camera yet, especially in this price range, that does "look like film". They just don't. Film is a different medium. The D16 and the Ikonoskop to me have more film like qualities, fewer artifacts, global shutter, a different more solid texture to the image, and richer color potential than most cameras I have worked with footage from. Digital Bolex still has much work to do with their firmware calibration, it is early days yet.
 
Haven't seen a digital camera yet, especially in this price range, that does "look like film". They just don't. Film is a different medium. The D16 and the Ikonoskop to me have more film like qualities, fewer artifacts, global shutter, a different more solid texture to the image, and richer color potential than most cameras I have worked with footage from. Digital Bolex still has much work to do with their firmware calibration, it is early days yet.

Yes, you're right. Hard to live up to the nostalgia that some old 16mm footage has too. Here's to hoping they sort out the firmware, and get the look they want.
 
Back
Top