The Canon EOS R1 may not come until 2024

[Del Paso]

We don't all have the same DNA. At the basic level, we all have a vast number of differences called SNPs, single nucleotide polymorphisms (https://en.wikipedia.org/wiki/Single-nucleotide_polymorphism). These lead to different amino acids at different positions in proteins and so they can have different activities. It's these SNPs that the ancestry DNA sites use to tell you your genetic background, find your relatives etc. Your various traits may be linked to combinations of these SNPs - the science of GWAS (https://en.wikipedia.org/wiki/Genome-wide_association_study). These SNPs don't usually qualify as being medical abnormalities as they can be very widespread. Some rarer mutations, like those in cystic fibrosis etc can have a severe affect by themselves. These SNPs in the opsin genes count as a "medical conditions" https://www.snpedia.com/index.php/Colorblind.

This is what I really like about this forum: learning is not restricted to photography, but extends to many other highly interesting topics !

 

[Michael Clark]

We don't all have the same DNA. At the basic level, we all have a vast number of differences called SNPs, single nucleotide polymorphisms (https://en.wikipedia.org/wiki/Single-nucleotide_polymorphism). These lead to different amino acids at different positions in proteins and so they can have different activities. It's these SNPs that the ancestry DNA sites use to tell you your genetic background, find your relatives etc. Your various traits may be linked to combinations of these SNPs - the science of GWAS (https://en.wikipedia.org/wiki/Genome-wide_association_study). These SNPs don't usually qualify as being medical abnormalities as they can be very widespread. Some rarer mutations, like those in cystic fibrosis etc can have a severe affect by themselves. These SNPs in the opsin genes count as a "medical conditions" https://www.snpedia.com/index.php/Colorblind.

Of course our entire DNA sequences are not identical. But for those who have normal vision, the DNA that tells our retinal cones to grow during development is close enough to the same in all of our sequences. There are not multiple varieties of S, M, and L cones. Everyone's L cones, M cones, and S cones are all sensitive to the same wavelengths of light, respectively. (Give or take 3-5 nanometers.)

And yes, there are a few rare individuals, mostly females with northern European roots, who are tetrachromats instead of trichromats. But their numbers are small enough to consider them to be abnormal. And their overall visual range is not extended in any way. They have higher sensitivity in the space between the M cones and L cones which allow a very, very few of them who have trained their vision to distinguish between smaller differences between colors in the yellow-green portion of the spectrum than the rest of us can.

Likewise, those who have color blindness in any form would also be considered abnormal, though people with those genetic maladies are much more common than those who are tetrachromats.

Does everyone see the same colour when they see red?

Neuroscientists tackling the age-old question are coming up with some interesting answers.

scroll.in

"So, yes, we can determine colour by measuring what happens in the brain. Our results show that each colour is associated with a distinct pattern of brain activity."

"What is more, these similarity relationships are preserved across people."

"Physiological measurements are unlikely to ever resolve metaphysical questions such as “what is redness?” But the magnetoencephalography results nonetheless provide some reassurance that colour is a fact we can agree on."

 

[Michael Clark]

We can all have receptors for the same colour spectrum in our eyes, but we will never be sure how those colours are translated inside of our brains. That does not even work with pain. If we both hit a hammer on our thumbs, we will both feel pain, but we can't find out who of us feels more pain. The same food might even taste differently for different people, although we all have the same receptors. There is a scientific phrase for that problem, but I can't find it at the moment.

Up until the early 1990s you would have been supported by the medical literature. But then researchers found a way to independently measure the precise wavelengths of light to which each type of retinal cone is most sensitive by measuring the biochemical signals travelling along the optic nerve, which for all practical purposes is an extension of the visual cortex.

Recent developments have shown that we can measure the brain wave patterns that result from being shown specific colors. And they are the same for all of us, within the margin of error of the measurements, which is very small.

Does everyone see the same colour when they see red?

Neuroscientists tackling the age-old question are coming up with some interesting answers.

scroll.in

 

[Michael Clark]

Technically, there is a significant amount of perceived color variation among people. In the color industry we call it viewer metamerism. If color is your business you have to acknowledge and deal with it.

Metamerism is caused by differences in viewing conditions, not by differences in color perception by different people under identical viewing conditions.

The same individual will experience metamerism when the viewing conditions are altered.

 

[Michael Clark]

Maybe there are people who see more or less colour, but what I mean is what happens in our brains. This is the same photo, but with a different hue. Imagine someone see the world like on the right photo. Then he will still call the sky blue and the grass green. There is no way to find out that two people use the same word for different colours, if both colours are just different translations from the same wavelength.

Again, recent research has confirmed that the brainwave patterns that result from exposure to the same optical stimuli are essentially the same.

Does everyone see the same colour when they see red?

Neuroscientists tackling the age-old question are coming up with some interesting answers.

scroll.in

 

[Michael Clark]

Well, yes, unless they are among the ~4.5% of the world population who are at some level colour blind. The most common type is red-green (including me to a lesser degree), but there is also blue-yellow. So at least some of those people will absolutely 'see' the sky differently than the majority.

I'd consider those with color blindness to be abnormal, because it is due to a genetic mutation which is considered a defect. Abnormalities are addressed in my original comment above.

 

[Michael Clark]

I yearn to know for sure, if it isn’t coming I would be leaning toward Fuji 100s.

If it isn't coming we'll never get any official word about it from Canon. It will be as if the idea of an R5s has never existed in their public corporate history.

 

[Michael Clark]

Wow. I would love to learn more about this. I have often wondered about this.

Metamerism is due to differences in viewing conditions (i.e. lighting conditions; brightness, color temperature, tint, CRI, etc.) affecting how disparate things can appear to be the same color under one kind of lighting and appear to be different colors under another lighting condition. For example, a piece of green glass and a piece of green plastic that are coated by different types of paint can appear the same color under bright D65 lighting, but appear to be two significantly different colors under dimmer tungsten lighting.

The same individual will see the same objects as different colors under different lighting conditions. That is what metamerism deals with. It does not deal with different individuals seeing radically different colors under the same lighting conditions.

 

[Michael Clark]

Maybe there are people who see more or less colour, but what I mean is what happens in our brains. This is the same photo, but with a different hue. Imagine someone see the world like on the right photo. Then he will still call the sky blue and the grass green. There is no way to find out that two people use the same word for different colours, if both colours are just different translations from the same wavelength.

Slight differences, sure. But even someone with severe color blindness would not perceive the same scene as radically different from those with normal vision as your two examples. That's absurd.

 

[Michael Clark]

Well, yes, unless they are among the ~4.5% of the world population who are at some level colour blind. The most common type is red-green (including me to a lesser degree), but there is also blue-yellow. So at least some of those people will absolutely 'see' the sky differently than the majority.

But nowhere near as differently as the example image he posted. If someone is blue-yellow deficient, they do not see blue-yellow objects as fluorescent magenta. They see them as unsaturated shades of grey (when they are not mixed with other colors).

 

[Michael Clark]

Although I don’t know what you look like, I am certain we are not identical twins. Therefore, we don’t have the same DNA.

The diverse genetic polymorphisms that result in people having different physical appearance also result in people having different spectral sensitivities in their cone opsins (and in rhodopsin, as well). As one example, a serine instead of alanine for the 180th amino acid in the L cone opsin (long wavelength, i.e., red) shifts the λmax 3-4 nm to the red side and results in greater sensitivity to red. That particular polymorphism is common, ~60% of the population has a serine, ~40% has an alanine at that position.

Polymorphism in red photopigment underlies variation in colour matching - Nature

GENETIC variation of human senses within the normal range probably exists but usually cannot be investigated in detail for lack of appropriate methods. The study of subtle perceptual differences in red–green colour vision is feasible since both photo-pigment genotypes and psychophysical...

www.nature.com

Yes, there can be minor differences in sensitivity to certain colors, especially longer wavelengths. But here's the comment I was responding to:

"We might never know if one person sees red as the same colour as another person sees blue. We have all agreed that the colour of blood is called red, but we will never know how other people perceive that colour. Perhaps it might look blue to them, but they still call it red."

We can scientifically demonstrate that no human perceives red the way others perceive blue. No human perceives blue the way others perceive red.

 

[David - Sydney]

I think the R1 would be delayed because Canon is developing a new higher resolution (+45 MP) stacked BSI sensor because they need it to compete with Nikon and Sony cameras, and they're not going to reuse existing sensors for it.

Although the colour topic is very interesting... back to the R1

I believe that there will be a high initial volume of R1 as it will be the body that will move 1DXiii users to mirrorless who haven't already migrated to R3.

The 1DXiii was announced in Jan-2020 so 2024 is not unreasonable although the ~4 year refresh cycle is in flux. Could be in 14 months to make it into 2024.

It would be interesting to know what the wafer yield is for stacked/BSI full frame sensors. A problem in design/fab or fab capacity constraints can easily shift release dates.

 

[Michael Clark]

Although the colour topic is very interesting... back to the R1

I believe that there will be a high initial volume of R1 as it will be the body that will move 1DXiii users to mirrorless who haven't already migrated to R3.

The 1DXiii was announced in Jan-2020 so 2024 is not unreasonable although the ~4 year refresh cycle is in flux. Could be in 14 months to make it into 2024.

It would be interesting to know what the wafer yield is for stacked/BSI full frame sensors. A problem in design/fab or fab capacity constraints can easily shift release dates.

Fourteen months from now will be in 2024.

 

[David - Sydney]

Fourteen months from now will be in 2024.

yep, that is what I said

 

[dcm]

But nowhere near as differently as the example image he posted. If someone is blue-yellow deficient, they do not see blue-yellow objects as fluorescent magenta. They see them as unsaturated shades of grey (when they are not mixed with other colors).

A color disc is an interesting way to illustrate this shift to grey. This occurs there are different types of receptors in our eye. Red, green, and blue cones register color (on not). Rods and ganglion give help with brightness, black/white and night vision where there isn't sufficient light to register color. These simulations show normal vision and green/red/blue deficiencies.

 

[David - Sydney]

I can only see a 85+ MB sensor if they go internal recording, similar to the recent Hasselblad. Take 30 fps x 85 MB = 2.55 GB/sec which is above the spec of the Type B CFExpress cards and well above the observed performance.

So, either they are waiting on a Type C CFE card or it is internal recording.

I can't see Type C CFe cards being released. There are very few use cases for Type C / 4 PCI lanes and no cards have been released.

The CFe A/B/C 2.0 specs were released in 2019 and the first CFe A cards were from Sony a year later for the A7Siii. They can make their own cards and set a price for them but Canon can't. I can't imagine a scenario where Sony makes the only cards that a R1 can use.

CFe B cards are now ubiquitous albeit with a wide range of write bitrates.
The simplest option if the R1 has 12k video (80mp video/100mp stills) is to only offer internal recording to fast CFe B cards with raw light compression.
8k/30 DCI raw needs ~320MB/s write speed. Well within the CFe B's theoretical 2GB/s
12k/30 etc would be ~1300MB/s which could still be within the current fast type B cards (Sony/Delkin/Prograde cobalt etc high capacity cards with ~1500MB/s write speeds)
=> Note that massive capacity cards would be needed just to record short clips @12k!

 

[David - Sydney]

I assume that with a high-res sensor they could easily employ a binning mechanism to shoot images at 1/4th or 1/9th of the full resolution.
My PhaseOne IQ180 back has a similar mechanism (it can deliver 80 or 20mp images) and that thing is old
Many current flagship phones employ similar mechanisms in their main camera sensors (my Samsung Galaxy S21 Ultra has a 108mp main camera but normally shoots 12mp images (1/9th)

I believe that the only option will be downsampling the sensor on the fly rather than lines skip/binning algorithms. Best of both worlds!

If the high resolution rumour is to be believed then 12k video in a hybrid camera (another first for Canon) is 80mp with ~100mp stills at 3:2.
12k/100mp downsampling is then a simple processing algorithm giving multiple "raw" resolutions L/M/S/very small with HEIF equivalents.

 

[David - Sydney]

Yes, in reality colours do not really exist. That is quite mindblowing if you think of it. We might never know if one person sees red as the same colour as another person sees blue. We have all agreed that the colour of blood is called red, but we will never know how other people perceive that colour. Perhaps it might look blue to them, but they still call it red. Since I know that fact, I care less about the colour of my clothes.

To be more specific, red is the colour of blood where hemoglobin is used to transport oxygen.
Plenty of examples in the animal kingdom of blue blood where hemoglobin (iron) is replaced with hemocyanin (copper) eg squid/octopus/snails/spiders/lobsters some scorpions etc.
3 creatures use hemerythrin instead giving purple blood
Some have white blood where there is no hemoglobin and the oxygen is dissolved into the plasma
Some sea creatures have yellow blood where vanabin is a pigment even though hemoglobin is used to transport oxygen.

 

[David - Sydney]

Hm...wild speculation incoming....

1. If the specs are correct, you'll be able to buy an R3 and an R5 and have money left over compared to the price of an R1.
2. As pointed out above, you'd probably have to use an m.2 drive instead of memory cards to match the data rates.
3. Is it possible that we could see an R5s be a 'mini' R1 - same 85mp sensor with much lower FPS and a feature set closer to the R5? If so, no R5S until after 2024. Might fit the 1DXIII/R6 and R3/R6II relationships.
4. I had hoped to have an R1 or 1 series body some day. But seems like it will be out of reach, as I bet they will push the price point on this one to $10k. I could half imagine saving up $6500 for a 1 series, and have been working on it for over 4 years so far. But 10k isn't going to happen. Maybe by the time the R1 hits, the R3 will drop in price, or there will be a stacked sensor R5II that will be more attainable.

Brian

While CFexpress Type B cards are pricey, they're essentially a small M.2 2230 NVMe 1.2-compliant SSD with a PCIe 3.0 x2 interface inside a special housing. You're paying a premium for a genuine CFexpress Type B card as you can get kits to build one yourself.

Given that 8k/30 raw video (33mp sensor resolution) "only" needs ~300MB/s sustained write bitrate, then a higher resolution could be handled by current high speed CFe B cards.

 

[Michael Clark]

A color disc is an interesting way to illustrate this shift to grey. This occurs there are different types of receptors in our eye. Red, green, and blue cones register color (on not). Rods and ganglion give help with brightness, black/white and night vision where there isn't sufficient light to register color. These simulations show normal vision and green/red/blue deficiencies.
View attachment 206333View attachment 206334View attachment 206335View attachment 206336

And nowhere in either example is red and blue reversed, which is what the comment I initially responded to proposed. No one, not even severely color blind individuals, see blue things as red or red things as blue. That is what the other commenter claimed.

I also specifically addressed those with abnormalities in my first comment.