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!

Trichromatic vs 3100

dchew

Well-known member
Link to the two raw files (color checker and fruit):

https://spaces.hightail.com/space/HtAMyc7o9y

Please don't judge sharpness in this comparison; I used the TC first and focused with that. The 3100 must have a slightly different registration since it is back focused a bit. Proof that Alpa wasn't crazy with the whole shimming thing.

Dave
 

yaya

Active member
Dave is there a particular reason for liking the Lumariver profile? It seems to desaturate some of the warmer tones and some, like Reds and Oranges look a bit Magenta to me...

Also note that when using any of these chart for generating profiles, setting the correct exposure is critical, as is avoiding reflections on their semi-matt surfaces (i.e. the Black patches) as this changes their readings...

BR

Yair
 

dchew

Well-known member
Dave is there a particular reason for liking the Lumariver profile? It seems to desaturate some of the warmer tones and some, like Reds and Oranges look a bit Magenta to me...

Also note that when using any of these chart for generating profiles, setting the correct exposure is critical, as is avoiding reflections on their semi-matt surfaces (i.e. the Black patches) as this changes their readings...

BR

Yair
Hi Yair,
Thank you for chiming in here. I wouldn't say I like the Lumariver profile any more than the other two. Actually right now my default for the 3100 is the Leaf profile. I’ve been using the Lumariver profile here because it appears to be the closest to the TC in everything except yellows and reds. That is certainly debatable...

There are some images, particularly fall leaves, sunrises/sunsets where I get better results from the Lumariver profile. I think both the P1 and the Leaf profiles push those warmer colors too far while I am making adjustments. But you are right in that the Lumariver profile is less saturated, especially with reds. What started me down the custom profile path was not colors but the curves. Film Standard was too much and Linear was too little. I hoped to get a result in the middle by making my own profile. In a way it helped because the Lumariver profile is less saturated in those colors. Think sunrises: I have more latitude using a brighter exposure when I switch to the Lumariver profile.

Regardless, you did make me think that using the Lumariver profile in these comparisons is not helpful to others, since they don’t have it.

There is definitely something going on with these 3100 images and that orange (I mean the edible thing - not the color checker squares). The TC renders it almost exactly as it really is. It is an organic orange, so the color is not in-your-face Buy Me Now Orange. Phase One Outdoor Daylight is not to bad. I see the magenta you refer to in the Lumariver profile. The Leaf profile blows it out like it is out of gamut. See the order I posted above for what I thought was best to worst, and here are some screen shots:

Trichromatic:


3100 P1 Outdoor Daylight


Lumariver custom


Leaf Landscape
 

dchew

Well-known member
Yair,
One other thing about the Lumariver profile: as Victor pointed out above, in my opinion the 3100 makes things, especially green things, yellow (for lack of a better way to describe it). The TC doesn’t appear to do that, and my Lumariver profile does it less than the other two profiles. Probably because those warm colors are coming across less saturated.

Sorry I didn’t mean for this to veer off into profiles, but I suppose that is inevitable when comparing these two backs.

Dave

Edit: in post #16 I made a mistake. Third bullet, the order from blue to green should be TC, 3100 Lumariver, P1 Outdoor Daylight, Leaf Landscape.
 
Last edited:

ErikKaffehr

Well-known member
Hi Dave,

I checked both images in Capture One. For me there is a great difference and that is the greens in the grass. The Trichromatic looks like green while the IQ3100 looks like brownish. Vegetation has very high near IR content, so this difference may be just caused by stronger IR filtering on the Thrichromatic.

It would be interesting to include a "traditional camera" in the test like a Nikon, Sony or Canon.

Capture.JPG
Top left, Lightroom with Lumariver profile. Top Right Capture One
Bottom Left, IQ3100 with Lumariver profile, Bottom Right Capture One.

I adjusted WB on second brightest gray patch while I adjusted exposure to L=51 to fourth brightest patch. Commercial verson of C1 does not have lab display, so final adjustment was done in Photoshop on C1 images. My understanding is that "Beta version" of C1 has Lab display. So, we may see that feature in a coming release.

For me, the grass stuff is a major improvement.

Best regards
Erik

More ramblings...

...
Tomorrow I will try to do some shifting with the 35 and 40 for you all.

Ciao,
Dave
 
Last edited:

vjbelle

Well-known member
Dave.... don't fret or drive yourself crazy with some colors being different/better/worse than others when comparing the mighty 3100 against the Trichro. I have just become used to the fact that the Phase backs when processed through C1 are warm compared to other digital sources. I am convinced that Phase chooses to make their files 'Warm' and if they wanted to could make those files look totally different. I don't have a Hassy equivalent but it would be interesting to see how the same scene would look/measure like when processed with Phocus vs C1 using the same 100MP chip. I've never heard of the Hassy crowd complaining about brownish green tones but maybe they will chime in.

Bottom line is that I have never been disappointed with the final results I get for print with my 3100 with regards to color.... never.

Thanks again for this incredible effort......

Victor
 

vjbelle

Well-known member
Not necessarily. I'm only saying that the 'look' of the file depends on the profile that Phase provides and that they have many choices. Conspiracy theories are, though, interesting.

Victor
 

dchew

Well-known member
Here are some raw files:
https://spaces.hightail.com/space/HtAMyc7o9y

Same link as before just more files. The first two are the same in case you already download them. Here is the legend:

I also posted this legend in excel so you don't have to go back and forth.

I apologize: there certainly are some comparisons that I should have done that I didn't do. These things are tough to manage correctly, and swapping the backs becomes tedious after a while, not to mention unnerving!

Hopefully this gets most people what they wanted. Most of the images are too dark on purpose. Several people wanted to see shadow recovery.

Ciao,
Dave
 

dchew

Well-known member
Dave.... don't fret or drive yourself crazy with some colors being different/better/worse than others when comparing the mighty 3100 against the Trichro. I have just become used to the fact that the Phase backs when processed through C1 are warm compared to other digital sources. I am convinced that Phase chooses to make their files 'Warm' and if they wanted to could make those files look totally different. I don't have a Hassy equivalent but it would be interesting to see how the same scene would look/measure like when processed with Phocus vs C1 using the same 100MP chip. I've never heard of the Hassy crowd complaining about brownish green tones but maybe they will chime in.

Bottom line is that I have never been disappointed with the final results I get for print with my 3100 with regards to color.... never.

Thanks again for this incredible effort......

Victor
Thanks Victor. I agree. Also note that these first two images were taken in daylight shadow, which is very blue. Color temp was raised 1800 degrees in the 3100's case. As I mentioned before, I did that to white balance the card. i would never want a "real" image to be that warm.

Dave
 

dchew

Well-known member
For me, the grass stuff is a major improvement.

Best regards
Erik
Erik,
BTW, the grass is a ground cover called vinca minor / periwinkle. Pretty blueish/purple flowers in the spring. They would be a great test for blue hues!

Dave
 

Abstraction

Well-known member
Not necessarily. I'm only saying that the 'look' of the file depends on the profile that Phase provides and that they have many choices. Conspiracy theories are, though, interesting.

Victor[/QUOTE

Yeah, so if Phase can make their files look like anything they want, then the whole idea of color being "better" as a function of hardware is moot because they are able to make the files look different already. The hardware already allows them to make the files anything they want them to be.
 

dchew

Well-known member
Not necessarily. I'm only saying that the 'look' of the file depends on the profile that Phase provides and that they have many choices. Conspiracy theories are, though, interesting.

Victor
Yeah, so if Phase can make their files look like anything they want, then the whole idea of color being "better" as a function of hardware is moot because they are able to make the files look different already. The hardware already allows them to make the files anything they want them to be.
That's a straw man. Victor said, "...they have many choices." Not, "... look like anything they want." As photographers, we also have "many choices" in how we render images. That doesn't mean all the hardware out there from all manufactures is a ruse.

Dave
 

vjbelle

Well-known member
Dave..... out of the box it looks, to me, like the Trichro does a better job with an LCC than the 3100 for a 35XL file. Your water scene is a very good example as it has lots of (should be) neutral water areas that the Trichro handles better than the 3100 - especially in the lower left hand quarter. I was able to get the 3100 to match really well by adjusting white balance to 4400 and tint to -11. I tried to use color balance but never achieved the results I got by just adjusting overall white balance.

You deserve a really nice bottle of red wine for this effort......

Cheers......

Victor
 

dchew

Well-known member
Dave..... out of the box it looks, to me, like the Trichro does a better job with an LCC than the 3100 for a 35XL file. Your water scene is a very good example as it has lots of (should be) neutral water areas that the Trichro handles better than the 3100 - especially in the lower left hand quarter. I was able to get the 3100 to match really well by adjusting white balance to 4400 and tint to -11. I tried to use color balance but never achieved the results I got by just adjusting overall white balance.

You deserve a really nice bottle of red wine for this effort......

Cheers......

Victor
Victor,
I added four more images this morning. Direct comparison 3100 vs TC on the 35xl with a big blue sky. I agree; I think there is some improvement. This sky is a harsh test, and there is naturally some lightening of the sky as the angle from the sun changes. Here is the revised list:


Wine sounds good. My wife's family all migrated to Cookeville. Still a bit of a drive to where you are, but maybe someday I can take you up on that. It would be fun to shoot together somewhere in TN.

Dave
 

dchew

Well-known member
Now that the Trichromatic is back in the barn, aside from satisfying my curiosity I am not sure what I accomplished. I had this idea in my head that development of the TC was for applications that had little to do with what I shoot. Like cultural heritage or some obscure military application. I neither verified or dispelled that notion in my head.

I definitely like the OOC color from the TC better. At least, better than the three 3100 profiles I compared, with the lighting I had. Capture One gives you "many choices" [ :-} ] of profiles and curves to experiment with. I tried a fraction of those. In general, I like the TC's color because I think it beautifully handles and maintains greens as the white balance is adjusted. Victor and I agree on a lot of things, and apparently this is another one. It is something I've struggled with for a long, long time. Anyone remember using an 81A with Velvia on a cloudy day? I didn't love those results either.

Although I do struggle with C1's white balance, the results I get from that struggle are better than what I get out of LR. My worflow is weird: I import everything into a LR WIP catalog. I try to work on images I like in LR, but usually end up opening that image in C1 and getting a result I prefer because of color nuances. Then I export a tiff into my LR MASTER catalog. Often I do further processing in LR (or PS) after that import. I am pretty agnostic when it comes to LR vs C1. I use them both in a way that suits my personal needs and preferences.

Victor, Erik and I have been bitching about those greens going yellow. But the 3100 is better than anything else I've previously tried in either C1 or LR. Just because the TC is better doesn't make the 3100 a dog. Erik, you are right: it would have been interesting to add the a7rii in the colorchecker images. But the lens would be different, so I am not sure what it would have told us. All I can say is in the past when I've done P1/Sony comparisons for myself, I like the P1 results better. I will not go there since it is easy for others to chalk my preference up to wishful thinking.

I am happy I went through this exercise. It helped me hone in on that difference, and come up with a way to manage it with the tools I have. Here are the two backs' rendering of the image I previously posted:


I think they are pretty close, and I did not even need to go into C1's excellent color tools to do it. Just a subtle combination of temperature and tint that I couldn't find before. Now, Eureka! **

Comparing shadow recovery is even more fraught with "what ifs." The backs don’t expose precisely the same, so comparing is tricky. I do have several different exposures from those water/hillside series, so if anyone wants to see other exposures let me know. For all intents and purposes, I think shadow recovery is identical, or at least within the limits of anything I can tell from a cursory look. I am sure the 35 ISO has some benefit. Maybe DXO will find it in there somewhere…

As for wide angle color cast, that jury is till out in my head. The TC may be marginally better with the 35xl. Hard to really tell because the exposures are not exactly the same between the backs. The LCC's look different, with a slightly different cast in different places.

Am I going to trade in my 3100? Nope. As a follow up to Doug's question of price, indirectly I heard the delta is something around $15k or less. If this was a hardware retrofit option for a few thousand I might do it. Especially if sometime in the future I had to send the back in for other service and I could piggyback the work. But that is not the option. I will continue "indefinitely" (thanks Adobe) into the future thoroughly enjoying the 3100.

Ciao,
Dave

**If you want to know which is which, I'll give you a clue: let's just say my 3100 sensor needs cleaning...
 

Geoff

Well-known member
Dave -

Thanks for all this work, and the very very clear explanations. Not commonly found, and much appreciated.
 

vjbelle

Well-known member
Dave..... I too would never part with my 3100 for a Trichro...... and certainly not for the price of admission that Phase is asking for. To move on..... I was absolutely floored by how clean the LCC shot was for the Rody 40. Its very impressive how that lens design mitigates light ray angles. The 35XL, as you know, needs much more software work for an LCC but is do-able for Phase. The 40mm Rody is the only other wide I would 'ever' consider as I just don't use wides that often. It has its faults (high distortion) but that can be somewhat mitigated with software - even though I'm really against that kind of file manipulation. The 35XL really shines with respect to distortion which is why I have never considered parting with it.

Thanks again, and again, and again for all of this effort.

Cheers.....

Victor
 
Last edited:

ErikKaffehr

Well-known member
Hi Dave,

To begin with, thanks for doing this comparison and for sharing the raw images. The reason I would like to have a "traditional" camera in the comparison was that I was interested in rendition of greens.

I am not sure about the role that lenses play. Having a ColorChecker in the image makes it possible to build a colour profile for the combination of illuminant, sensor and lens. The "Erik Kaffehr" test you refer to, I was using three different lenses on three different sensors. A Zeiss lens made for the Contax RTS, a Zeiss lens for the Hasselblad and a Sony G lens and I generated a colour profile for each, using Lumariver. My guess is that white balance takes care of colour caused by lens. I will go into retirement at new year, so I will look into that a bit.

Regarding highlight and shadow recovery, I would guess both sensors are the same. The samples you provided are pretty good.

Personally, I am interested in both photography and the science behind it. That said, I am not a scientist, just an engineer working in a quiet different area, namely reactor physics.

Hopefully I take some decent pictures, much of my photography is shown here: https://echophoto.smugmug.com/, but I don't see myself as a creative artist.

Now, that I will go into retirement, I may do more photography.

Best regards
Erik


Erik, you are right: it would have been interesting to add the a7rii in the colorchecker images. But the lens would be different, so I am not sure what it would have told us. All I can say is in the past when I've done P1/Sony comparisons for myself, I like the P1 results better. I will not go there since it is easy for others to chalk my preference up to wishful thinking.

I am happy I went through this exercise. It helped me hone in on that difference, and come up with a way to manage it with the tools I have. Here are the two backs' rendering of the image I previously posted:


I think they are pretty close, and I did not even need to go into C1's excellent color tools to do it. Just a subtle combination of temperature and tint that I couldn't find before. Now, Eureka! **

Comparing shadow recovery is even more fraught with "what ifs." The backs don’t expose precisely the same, so comparing is tricky. I do have several different exposures from those water/hillside series, so if anyone wants to see other exposures let me know. For all intents and purposes, I think shadow recovery is identical, or at least within the limits of anything I can tell from a cursory look. I am sure the 35 ISO has some benefit. Maybe DXO will find it in there somewhere…


Ciao,
Dave

**If you want to know which is which, I'll give you a clue: let's just say my 3100 sensor needs cleaning...
 

dchew

Well-known member
Erik,
Well, you hit my curiosity button, so in the next few weeks I will try the 3100 and a7rii with both the 55 FE and the 50 Apo Summicron. We will see if lenses make any difference once white balanced. I could also add a Canon lens into the mix.

I've been to your website; very nice images there, especially the "Winter" gallery and your travel images. I've never been good at documenting what a trip is like (my wife frequently points that out). You do that very well.

Dave
 
Top