The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Phase Focus - works - a Toronto Test

B in Toronto

New member
Hello -

Spent some time at B3K In Toronto and a few hours with the new Phase XF and the IQ3 TriChromatic Back.

1 - The focus works very very well. The camera and the 150mm f2.8 BR lens - spot on amazing as with the 80mm and the 110mm. Tried close shots, far shots and extreme recompose. The camera was dialed into pixel peeping perfection.

2 - The tri-chromatic: I was a bit meh at first. But - a close explanation and sample images with Walter showed me that the back now faithfully reproduces colours I had completely given up on ever photographing. Magentas, fuscias - ripping with saturation and difficult light - perfect detail rendition.

Presently saving up

Bruce
 

dougpeterson

Workshop Member
Great to hear your experience.

I've been away on my honeymoon for two weeks so missed out on all the launch fun.

But my one takeaway from reading all the threads is this: theoretical online discussion won't get you very far on evaluating these Phase One announcements (Focus+Recompose, HAP2, Trichromatic).

It will take hands on, in-person evaluation to see if these are marketing fluff or meaningful stuff. I know I'll be doing my own testing to form my own opinion.

In the coming months we'll be having open houses in our NY and LA offices, as well as additional cities via a road show; if you're interested in us coming to your city please email me at [email protected] so I can get a sense of where there is demand for that. Up in Canada B3K is a great resource for that kind of testing as well; I know Walter personally as well as professionally and he's a great guy.

Regarding the color of the Trichromatic, I think it's especially ill suited to evaluate based on armchair color science theory discussion. Over the last few years as my role has shifted more to R+D, I've learned a ton about color science (mostly related to our museum and library division, but also for our Contax Wedding 400H Profile and C1 Style Pack) and what I've primarily learned about color in that time is how much of a rabbit hole it is. I think this xkcd.com web comic summarizes up pretty well:



Is the color of the Trichromatic that much of a step forward? Ask me in a couple weeks after I've spent a lot of time taking actual pictures. I'm definitely not going to try to derive it by drawing charts.
 

Christopher

Active member
While I agree with some parts, I still think Phase One should be able to actually give us some decent technical information. The marketing stuff is just "bs" which on a technical standpoint doesn't make sense. (At least how it's presented. )
 

MGrayson

Subscriber and Workshop Member
Doug,

I haven't been following too closely, but most of the complaints I've seen color-wise are about the truly nonsensical claims of non-overlapping response functions. If the headline advantage being trumpeted is outrageously false, then that puts a damper on the whole release. Now I'm sure the truth is not nonsensical, and the damned thing works very well. It's just a bad idea (outside politics) to support a position with a black-is-white statement.

Even stranger, the "more like the eye" and "more disjoint response curve" claims are opposite. Perhaps "closer to orthogonal basis for eye response curves" is true, but doesn't make good ad copy. I'll give them that. :)

Best,

Matt

(Yes, you can't really get closer to an orthogonal basis without having negative response values, so even that is nonsensical. I don't know what they've done. I'm sure it's good.)
 

dougpeterson

Workshop Member
Doug,

I haven't been following too closely, but most of the complaints I've seen color-wise are about the truly nonsensical claims of non-overlapping response functions.

[...]Perhaps "closer to orthogonal basis for eye response curves" is true, but doesn't make good ad copy. I'll give them that. :)
I'm a nerd, so I like any and all technical explanation I can get. But, like every other camera company I've ever dealt with or seen (short of some extremely niche players that are primarily aiming at very technical users) the marketing team is composed of marketing folks and the technical team is composed of technical folks. God help us if they ever switch that up – we'll have extremely precise and technically accurate marketing of products that don't work! So if you're interested in such information you're not going to get it from the marketing team.

Fortunately you're not limited to reading about it from Phase One's marketing team. There's plenty of (non theoretical) ways you can make your own practical eval:
- Hear our view, once we've formed it. I like to think I have a good no-BS reputation here after nine years and meeting many of you in person. We'll share our view, and back it up with raw files that you can do your own evaluation of.
- Try yourself. Many ways (rental eval, at our LA or NYC office, at our road shows etc)
- Ask P1 R+D face to face. We host members of P1 R+D frequently at our events, including the Head of R+D and the Head Color Scientist. For best results I suggest waiting until after they've had a beer or two before you ask a question :).

As for me, I've spoken with R+D and we're only at the early stages of our in house testing here at DT. I'm already starting to form opinions about what I'm seeing, and, based on talking directly to R+D (without the filter of marketing) I am starting to form a idea about why I'm seeing it. As those solidify I'll write more.

P.S. I don't know that I've seen "non-overlapping response functions" anywhere in the marketing put out, but I'm still reading through some of the documents so maybe I missed that. Of course they are not "non-overlapping"; the overlap is simply reduced and shaped in a way that is better suited to the end goal.
 

B in Toronto

New member
My experience is photo speaking purely practical - colour wise.

There are certain colours that cause digital sensors to 'explode.'

Standing in front of a group of dresses or children's clothes in certain shades of red and magenta knowing full well that nothing within my skill set or equipment will yield a pleasing result is a tad disheartening.

This new Phase approach has a solution for a problem that has dogged me personally for 15 years. And, these colours always show up at the worst possible moment with the most critical of clients.


I thought it was all marketing BS at first until I saw the heretofore unobtainable results: pleasing images.

And, I am happy with B3K and the generous opportunity to experiment liberally with the system.
 

MGrayson

Subscriber and Workshop Member
P.S. I don't know that I've seen "non-overlapping response functions" anywhere in the marketing put out, but I'm still reading through some of the documents so maybe I missed that. Of course they are not "non-overlapping"; the overlap is simply reduced and shaped in a way that is better suited to the end goal.
From Phase One's Site:



Alas!

I fully endorse the "See what it does before having an opinion" philosophy. I just hate marketing that makes no more sense than "we use antigravity from squirrel blood".
 

Shashin

Well-known member
How much of exploding colors is CFA and how much the color space? Still, I agree with Matt, less art, more matter.
 

MGrayson

Subscriber and Workshop Member
And not be even more pedantic, but the overlaps in the "Standard CMOS Sensor" picture will also have dead zones where different colors will appear identical in the RAW file. Honestly, if this is the "delve into the science" part of the website, their real scientists must be pulling their hair out.

Of course, I've never looked closely at Leica's marketing material. At least their vague generalities are obviously vague generalities. :ROTFL:

And to the OP's point, getting that coral pink in a sunrise is almost impossible - the Reds always blow out leaving it yellow. If the new CFA can help with that, I'd consider switching systems.

--Matt
 

dougpeterson

Workshop Member
From Phase One's Site:



Alas!

I fully endorse the "See what it does before having an opinion" philosophy. I just hate marketing that makes no more sense than "we use antigravity from squirrel blood".
Reading that cartoon representation as a technical specification chart of the new CFA's spectral transmission is probably taking it a bit too literally (spoiler alert: the response curves also aren't triangular!). Given that the average photographer has a very basic understanding of color science, I don't envy marketing people trying to take something as complex as a change in CFA transmission and it's effect on color and distill it down into it's most succinct and digestible form.

But in any case, I can see where it makes you twinge. It's the equivalent of describing "infinity" as "a really really big number". Absolutely technically wrong; though you can kind of see what it's getting at.

Bottom line: I don't turn to marketing people for deep technical information, and I would suggest you shouldn't either. They've given very little red meat for people with a more technical mindset, and I hope to fill in that gap in the coming weeks.
 

dougpeterson

Workshop Member
And to the OP's point, getting that coral pink in a sunrise is almost impossible - the Reds always blow out leaving it yellow. If the new CFA can help with that, I'd consider switching systems.
Let us know when you want to meet for a sunrise comparison!
 

MGrayson

Subscriber and Workshop Member
Reading that cartoon representation as a technical specification chart of the new CFA's spectral transmission is probably taking it a bit too literally (spoiler alert: the response curves also aren't triangular!).
So when I click through "Explore the Science" and then "Read More about the Science", I'm not supposed to take it as scientific. Check.

My apologies, I'm taking this way too seriously.

--Matt
 

GrahamWelland

Subscriber & Workshop Member
The coral pink vs yellow sunset/sunrise band of colour resonates with me. If it can accurately shoot bluebells in springtime without turning them into pink or purple bells then I'll be truly impressed.

In the meantime I'm sticking with my pebble of an old worthless IQ3100 :thumbup::ROTFL:
 

jdphoto

Well-known member
Doesn't Sigma's Foveon sensor basically do the same thing in regards to color? I find most clients want a film look with muted colors or just B/W. Any other color adjustments are done using a color checker passport.
 

cunim

Well-known member
I have always preferred color wheel and multishot images to those produced by matrixed systems. With the wheels I think it is not just saturation, but that SN can be optimized for each of the discrete wavelengths before fusion. Multishot benefits are more subtle but still obvious.

I wonder if this matrix variant better approximates the multishot "look". That's the real comparison, at least for me. Anyone care to post some relevant images?
 

Paul2660

Well-known member
Foveon, as I understand it, uses 3 separate photodiode layers, red, green, and blue thus there is no interpolation for color. The standard Bayer pattern captures 1 color per photodiode, then interpolation is done to figure out the other colors. So if Red is capture, the camera will interpolate for blue and green. Downsides are that the Foveon layers are stacked, thus light transmission is limited and so the sensor so far tend to work best in the base to lower ISO ranges. This as far as I know has never been upgraded or fixed. There is no doubt to me that the Foveon chips can take amazing images, however the raw support for Foveon/Sigma is still very very limited.

This new tech from P1, as I can understand it from the limited amount of literature, still uses only one photodiode layer which appears to be the same as the current IQ3100. The difference is the color filter array which seems to allow for more precise color captures and no overlapping which can cause cross talk.

The need for interpolation still would be there as far as I can see since the P1 design is still only using 1 layer of photodiodes not 3 as Foveon does. Only 1 of the 3 basic colors is being captured, but that color per the claims of the marketing literature is supposedly more accurate.

It will be interesting to see some side by side shots from the existing IQ3100 and this new back.

Since the base ISO has been lowered to 35, I am also curious if this back will be more like a CCD back, in that it will best perform at base ISO to maybe 2 stops higher due to the difference in light transmission.

Paul Caldwell
 

0beone

Active member
Foveon, as I understand it, uses 3 separate photodiode layers, red, green, and blue thus there is no interpolation for color. The standard Bayer pattern captures 1 color per photodiode, then interpolation is done to figure out the other colors. So if Red is capture, the camera will interpolate for blue and green. Downsides are that the Foveon layers are stacked, thus light transmission is limited and so the sensor so far tend to work best in the base to lower ISO ranges. This as far as I know has never been upgraded or fixed. There is no doubt to me that the Foveon chips can take amazing images, however the raw support for Foveon/Sigma is still very very limited.

This new tech from P1, as I can understand it from the limited amount of literature, still uses only one photodiode layer which appears to be the same as the current IQ3100. The difference is the color filter array which seems to allow for more precise color captures and no overlapping which can cause cross talk.

The need for interpolation still would be there as far as I can see since the P1 design is still only using 1 layer of photodiodes not 3 as Foveon does. Only 1 of the 3 basic colors is being captured, but that color per the claims of the marketing literature is supposedly more accurate.

It will be interesting to see some side by side shots from the existing IQ3100 and this new back.

Since the base ISO has been lowered to 35, I am also curious if this back will be more like a CCD back, in that it will best perform at base ISO to maybe 2 stops higher due to the difference in light transmission.

Paul Caldwell
Interesting the comparison with the Foveon chip. As I understand it (and believe me I could be so ooo wrong) I believed the Foveon chip was meant to imitate the layers in film... Schneider in their lens literature of their digital capable lenses clearly indicate that their Digital capable lenses were designed to produce an image at the film plane that was essentially "flat" (color wise) as opposed to a lens designed for film which was layered, thus producing less "fringing" in a digital system as compared to using lenses of the earlier film type.
This new sensor IMHO is a welcome addition to the technology in image capture and for me I welcome it and have mine on order...

Best
Frank
 

Boinger

Active member
As far as I know the only way you can "change" the color reproduction of a sensor is to directly alter the photo sites.

A bayer color array will always produce bayer color.

It's why fuji camera's look different than others and they struggle with different color issues.

The ONLY way to produce a "true" color image is if you capture all the colors. Which is what the foveon sensor does. Any other form of technology will throw out colors and use an algorithm to interpret what the image should look like. Hence we have different looks from different cameras as they all use different flavoring of their own sensors. But that is exactly what it is flavoring.
 
Top