Hi all! Probably a weird question but still:) According to this tweet Apple started used some new tech for their displays in M4 MBP’s. And this pic shows that new displays have peaks more evenly distributed across emission spectrum.
Imagine such displays will appear in iPad’s soon.
Hi there, welcome to the forum. There are people on this forum that know a great deal about light sources so I hope they see and respond to your post. Since this is the NLP forum I imagine that you are thinking in terms of copying colour negative and using NLP, and perhaps you are making the association with those clearly defined peaks on this new MacBook Pro and the concept of using individually controlled R,G & B LEDs with perhaps an integrating sphere.
Have you seen this thread?
Lots of information there although I don’t recall anyone using Macbook Pro screens as a light source, rather it might be a relatively recent Iphone or Ipad which can be very good I think. One possible disadvantage is that I don’t think that NLP is ‘tuned’ for separate RGB light sources but only Nate can really comment on that.
Edit: I see that the Ipad Pro has something similar.
Welcome to the forum @wsnake
We’re talking about OLED (organic LED) technology and the iPad Pro looks promising with a max. brightness of 1000 nits (1000 candela per square meter). Most of the light will have to be masked off, unless it is channelled onto the negative.
OLEDs are also used in smartphones priced well below that of an iPad Pro, but brightness will be lower than 1000 nits, I suppose.
Current recommendations are to use high CRI light and old “hot to the touch” light bulbs have produced captures that were easily usable and tuneable. Nevertheless, camera sensors use colour filter arrays (CFA) than make sensels (more or less → note 1) insensitive to other colours than the intended R, G and B hues (-> note 2). In the end, colour rendition depends on more than one link of the whole processing chain (backlight, film, CFA, rendering with NLP and Lightroom, screen gamut etc.) even though the backlight quality is important because it is the initial source of (im)perfection.
Notes
thanks for the explanation! my focus only on the light source and the original question is about displays that are used in the newer MBP’s from the tweet.
I understand that most likely they will not used in iPad’s ofc because they use OLED’s.
But still, maybe someone knows if there will be any benefits of using such displays since effect will be similar (or not? question mark) to narrowband RGB LED’s like from this project? https://jackw01.github.io/scanlight/
I’m asking since I even thought to replace led strip in NEEWER NL-192AI (or similar panels) to more expensive ones like SunWave CRI 98 or Waveform Lighting Absolute Series, but I’m not sure if it worth the hassle and maybe there are better options exists (like reusing iPad pro or this new MBP)
Assuming you are talking about colour negative I’m just wondering how you would be able to tell.
Imo, there is a potential issue with using any technological variant of narrowband R, G and B light sources. Assume that they don’t match the sensitivities of the CFA/sensor of your camera…creating a different balance between colours…that NLP will have to compensate. Camera sensor sensitivities are not standardised and can change between camera brands and models, see dxomark.com.
Moreover, “normal” cameras are made to take pictures in all kinds of natural and artificial light and without control over the light source. They will simply compensate any light mix when we white balance the takes.
Narrowband R, G and B light sources could make a lot of sense if we’d scan colour takes with monochrome cameras. Most of us don’t have such cameras…but I’ve run a PoC in order to see how things handled in such a case.
Now for accuracy: Think of it as an illusion. Variations of chemical processing and introduced by time and storage conditions make calibration a fool’s errand. Even one’s own ideas about how an image should look changes over time and seeing habits. Imo, it is best to judge the results subjectively rather than by measurement.
My question about how you might tell if one light source for scanning colour negative might be better than another was rhetorical really, though I really would like to know if there is a way. In the thread that I linked to @Richard1Karash has attempted to do this using is own personal test scene, starting here:
He has also more recently presented the results for the Cinestill CS-Lite and each of its different colour temperatures, if I can find it I’ll put in a link. Also very recently there is this other post where the very poor results were cured by replacing the cheap panel with a Kaiser Slimlite Plano:
So it seems it is possible to demonstrate the poor results obtainable from poor panels (provided you have a better one to compare it to) but I’m not sure that it’s quite so straightforward to distinguish between results from the better ones, particularly on colour negative of course. I don’t know how you do it really, even the manufacturers don’t seem to attempt to demonstrate the difference between their 95CRI and 99CRI offerings.
The text provided here (A Better Light Source For Scanning Color Negative Film | A Better Light Source For Scanning Color Negative Film) contains notes about “better” colour, and I’d like to emphasise the passage here:
I think it’s safe to say that most photographers shooting film in 2024 are looking for aesthetically pleasing colors, not technically perfect colors. If you have a more technically accurate way to process your scans, then by all means use it ... but using a narrowband light source instead of a white light source will still get you closer to the results from a professional scanner.
Key takeouts:
Thanks! I’m no stranger in photography (professionally/commercially and as a hobby) so that’s why I’m leaving out perception out of equation and only focusing on the subject of the light source paired with modern digital camera sensors.
I already seen comparisons between negative supply’s cri95 and cri99 and indeed there is almost no difference in the end results. But from what I understand from the article I shared (A Better Light Source For Scanning Color Negative Film | A Better Light Source For Scanning Color Negative Film) than narrowband LED’s are better than monochromatic ones, hence my question.
And just to avoid misunderstanding, I’m not talking about “perceived results” and other “artistic” aspects. I’m just curious purely on technical side of things as I don’t really investigated it, so wondering if someone else already done.
I guess that you may well be right, the article that @Digitizer linked to seems quite clear on this, the avoidance of crossover as it were. This post from 2021 includes comments from Nate on the subject, and I see that @Digitizer contributed to that also:
Just a general remark: What does “better” mean?
That word is used very often but is hardly ever defined. Better only exists in comparison to something else, and it often also relates to just a few of many aspects that can be compared.
Experimenting with “different” tech is certainly interesting from an engineering point of view and I’ve been tempted to explore discrete light sources like this and in testing RGB/monochromatic capture as a proof of concept. From what I learned, we’re dealing with the flat end of the 80/20 curve where big effort brings small benefits.
Then again, building a lighting device can be fun and the solution documented in Github looks fairly mature (and expensive - but we usually don’t add our own time to the TCO).
I hope I eventually can add something to the discussion once I finished my current project. I use tree-laser Optoma ML1080 video projector with actual RGB lasers. I switched projector in the mode where I can control each channel independently. I wrote application for PC which presents user with the panel with five sliders RGB, Alfa channel and vingnette simulation mode. So I can use video projector with narrow band lasers and control each channel independently. All of the above is done.
Looking forward to your report.
For those who use white light: As hot bulbs are hard to get, a beamer replacement bulb/lamp/thing could be useful. Some of the modules come with an integrated blower.
I had never thought of projectors as a backlight but with their intensity they could definitely prove quite useful. I’m curious if anyone has tried using the Intrepid compact enlarger head as a scanning backlight (I’ve only ever found one Youtube video where this is shown but its an old version of NLP and the guy forgets to white balance on the film base) - supposedly it’s not that great for its intended purpose as an enlarger as the steps are too large to consistently dial in prints but as a film backlight it might be one of the only RGB scan-ready backlights you can buy off the shelf. In a different Reddit thread someone said they asked Intrepid for the spectral date on the LEDs but they never got a reply
What sort of device is that?
Just the lamp…instead of a complete beamer.
Haven’t found a replacement for the RGB laser for the Optoma though.
as far as i understand ml1080 | Optoma USA this is the only projector with three separate lasers and under $1000. And here is the control panel i wrote for it.
…still a lot of money, more than what most of us will want to spend for “a backlight” 
Searching for a replacement lamp, I found that most of them are below $ 100 or well below that, but probably still to high for the casual scanner - if we count the DIY part to accommodate the lamp.
of course you don’t buy projector for back light , but if you consider buying it anyway for home theater , you may consider it. Projector does not require any modifications. Of course it yet to be proved that it brings in radical change
