While the cropped 100MP sensor from sony that is in a bunch of cameras is great, I'm not sure which systems have the larger 150MP variant besides the Phase One cameras. Does anyone know if there are never generations of this sensor size even planned? Seems like this one has been around for quite some time without any updates which sort of makes one wonder how stagnant that segment is. Seems to me like going from full frame to the larger of the medium format sensors would provide the most striking difference than just going up to the "cropped" variants.
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!
New "large" medium format sensor on the horizon?
- Thread starter larkis
- Start date
Paul Spinnler
Well-known member
This large sensor category is not being invested in because it is too niche and industry doesn't require this size. Machine vision needs a higher DoF, so for industrial quality control you would not need a large sensor - just global shutter and increased size.
Aerial also doesnt per se need a large sensor, but more global shutter and increased DoF.
Only photography requires this and given the economic environment it would need to come from Sony who pushes new technology to its customers (ie P1, Hassy) and not customers ordering a bespoke sensor which costs millions to develop. As a camera manufacturer you need to front the capital to commission a new design which you then need to amortize over time via back sales.
Since the pandemic sales of high end backs plummeted and as a result P1 decided to pour their capital in high RoI areas, namely drone imaging for industrial applications. Photography is just getting milked with existing products.
There is no belief that you can easily recoup a 5-10m sensor development investment via back sales in this market and no demand from the industry to make up for this lack in confidence in the photographic market.
So only hope is that Sony pushes a new tech for "free" ie for customers to order. It could happen, but probably is not a priority for Sony.
Befoe an IQ5 ships, we will first see an announcement by Sony and then 12-18m later a new back.
For the time being the IQ4 is and remains the best sensor available for high-end photography. To this day, its resolution, DR are unmatched.
Aerial also doesnt per se need a large sensor, but more global shutter and increased DoF.
Only photography requires this and given the economic environment it would need to come from Sony who pushes new technology to its customers (ie P1, Hassy) and not customers ordering a bespoke sensor which costs millions to develop. As a camera manufacturer you need to front the capital to commission a new design which you then need to amortize over time via back sales.
Since the pandemic sales of high end backs plummeted and as a result P1 decided to pour their capital in high RoI areas, namely drone imaging for industrial applications. Photography is just getting milked with existing products.
There is no belief that you can easily recoup a 5-10m sensor development investment via back sales in this market and no demand from the industry to make up for this lack in confidence in the photographic market.
So only hope is that Sony pushes a new tech for "free" ie for customers to order. It could happen, but probably is not a priority for Sony.
Befoe an IQ5 ships, we will first see an announcement by Sony and then 12-18m later a new back.
For the time being the IQ4 is and remains the best sensor available for high-end photography. To this day, its resolution, DR are unmatched.
Last edited:
P. Chong
Well-known member
I am curious. Aren’t the sensors made from wafers which are very large, and cut to size. Wouldn’t a larger sensor then be just cutting it to a different size? Understand that this may result in more wastage, and affect cost. Of course to be driven by demand of the photographic industry, which will also drive the need for a whole set of lenses.
Yes, wafers are expensive and the more usable sensors can be had from one wafer, the cheaper the production is. However, technology advances and changes. What was immensely expensive 20 years ago can be affordable today.
IF there would be technology breakthrough, that could make big sensors affordable. I would say there is a certain market for bigger sensors, but amateurs can not pay for even more expensive db:s. Therefore we need tech breakthrough.
Paul Spinnler
Well-known member
Wafers are just the running production cost side. With increased size the cost rises non-linearly, ie almost exponentially, as you can only use perfect sensors with no defects. The bigger the size, the bigger the risk of not being able to cut out a perfect sensor. Ie yields decrease significantly more than the linear size increase would imply. That's why for a large sensor like the IQ4 there is a natural low end cost barrier.
But we are talking about a new sensor here. This needs development, ie R&D and trial and error. It costs millions to develop a new sensor. Hence Arri only developed two sensors over the last 10 years. The old Arri Alexa sensor and the new one in the Arr 35. Even for a wildly successful company as Arri it is not an easy economic project to crank out new sensors.
If large photographic sensors were useful in other areas, ie military, industrial, it would be a different topic. But no one except photographers needs a 54x40mm sensor ... ! If you have a military drone you would want an optimal pixel count, low noise, etc., but definitely not a large sensor size and a huge optic attached to it. It is far more useful to invest in new sensor layering techniques which increase DR, noise performance and build compact photo systems to mount them on drones than huge sensors which create small DoF. You don't want to track an enemy with a 150 2.8 lens on a 645 sensor on a drone ...
Photography benefits from large sensor area because it allows you to isolate subjects with large aperture optics. E.g., portraiture, etc.
Paul Spinnler
Well-known member
No response, they milk the IQ4. That's it. If Sony offers them a new sensor design we might see an IQ5. If not, we will for the time being be on the IQ4. It is still the best of the best and works perfectly and does not at all feel outdated, contrary to some narrative here on the forum. It is the best sensor in the market and will stay it for the foreseeable future.
I’m going to print this quote in A0 size
Yes, I agree. Even if there will be a HB CFV-100c, at best a 44x33mm IQ4 equivalent. IMO, it would take a "global shutter" CFV to better the IQ4.
Men, don't be cruel, we got GPS, navigation, communication devices smarter than the ones in Start Treck, Teraflop computing on the desktop, laparoscopy, crazy AF systems,
etc
Best gegards,
If the market is there and maybe some Chinese company decided to go for it, we could have multichip Cameras combined by software. But I am not sure the market requires such a thing, it seems the market will be better served by computational imaging than any custom hardware.
CI will allow stuff like smaller lenses that perform better than the best we have today via digital corrections, increased dynamic resolution, better dynamic range, smart sharing and watermarking, key moment recovery etc.
What we can dream in terms of development will be. a 44x33mm Stack CMOS sensor that will allow Nikon to Enter the market using the Z mount, computational imaging and the same lenses they have today with decent AF. (Nikon's software limitations force them to use fats readout sensors to get modern AF, but even a faster readout sensor from sony could lead to a Nikon Z with Z7 II peformance).
Best regards.
There's no new larger sensor on the horizon, so if an IQ5 does appear it will probably be the old sensor + upgraded electronics.
Sony is doing some interesting stuff around a 128MP MP chip (~57mm diagonal, 4:3 ratio) which features super fast global readout speeds - fast enough to deal with moving subjects, but whether that will translate into a 'prosumer' chip and whether or not P1 will go there if it did, remains to be seen. Probably not given it's 'small' size, more likely to find its way into a Fuji or HB mirrorless offering.
Back in the day high end photography was P1's raison d'etre, today it feels more like a niche part of their business and I think they'll just keep flogging the 150MP chip for as long as they can. More so given the myopic fixation people have on MP count (despite the fact that outside of a tightly controlled environment I doubt anyone ever actually gets 150MP of real data given just how small the sensels are and how much the camera moves during even an extremely short exposure).
If you're not using a P1 DB on something like a tech cam I don't see any reason to look beyond Fuji and HB at the moment. What they're doing with 100MP is much more interesting IMHO. Compact cameras, tilt screens, great lenses, faster readout times with the ES etc. And all without costing $50k+ and the massive depreciation that comes on the back of it.
I'm invested in P1 right now, and it's been a great ride over the years, but when the time comes I'll be off down the Fuji rabbit hole. 150MP over 100MP just isn't enough of a difference anymore now the new generation of AI uprezzing tools are coming on stream. ES aside, the great advances now are in software, not hardware.
BTW, anyone notice that Canon just announced a sensor with ~25 stops of DR?
Jim
Sony is doing some interesting stuff around a 128MP MP chip (~57mm diagonal, 4:3 ratio) which features super fast global readout speeds - fast enough to deal with moving subjects, but whether that will translate into a 'prosumer' chip and whether or not P1 will go there if it did, remains to be seen. Probably not given it's 'small' size, more likely to find its way into a Fuji or HB mirrorless offering.
Back in the day high end photography was P1's raison d'etre, today it feels more like a niche part of their business and I think they'll just keep flogging the 150MP chip for as long as they can. More so given the myopic fixation people have on MP count (despite the fact that outside of a tightly controlled environment I doubt anyone ever actually gets 150MP of real data given just how small the sensels are and how much the camera moves during even an extremely short exposure).
If you're not using a P1 DB on something like a tech cam I don't see any reason to look beyond Fuji and HB at the moment. What they're doing with 100MP is much more interesting IMHO. Compact cameras, tilt screens, great lenses, faster readout times with the ES etc. And all without costing $50k+ and the massive depreciation that comes on the back of it.
I'm invested in P1 right now, and it's been a great ride over the years, but when the time comes I'll be off down the Fuji rabbit hole. 150MP over 100MP just isn't enough of a difference anymore now the new generation of AI uprezzing tools are coming on stream. ES aside, the great advances now are in software, not hardware.
BTW, anyone notice that Canon just announced a sensor with ~25 stops of DR?
Jim
Sony's 128MP IMX661 (Type 3.4, 44x33mm) has global shutter. A diagonal 66.7 mm (Type 4.2) CMOS version thereof surely can't be too far away.
Global shutter would be a good feature for releasing a new DB. Many would upgrade for that feature.
----------
www.framos.com/en/products/imx661aaqr-c-24687
The IMX661 is a diagonal 56.73 mm (Type 3.6) CMOS active pixel type solid-state image sensor with a square pixel array and 127 M effective pixels. This chip features a global shutter with variable charge-integration time. This chip operates with analog 3.3 V, digital 1.2 V, and interface 1.8 V quadruple power supply.
Paul Spinnler
Well-known member
The main reason to own an IQ4 is to shoot on a tech cam with Digarons and Digitars - the ability to stitch and image quality is still top of the pops.
It is quite something to have an edge to edge sharp image with a Rodie HR lens.
It is a quality I am happy to pay for, especially combined with a tech cam like an Alpa. Fuji for me is not comparable at all in terms of of both joy of process and handling and possible output.
A three way stitch with a 90 HR for example is unbeatable in terms of sheer detail and IQ on a P1 back. Of course, it requires a lot of money, but if you want the truly best, there is no alternative.
It is quite something to have an edge to edge sharp image with a Rodie HR lens.
It is a quality I am happy to pay for, especially combined with a tech cam like an Alpa. Fuji for me is not comparable at all in terms of of both joy of process and handling and possible output.
A three way stitch with a 90 HR for example is unbeatable in terms of sheer detail and IQ on a P1 back. Of course, it requires a lot of money, but if you want the truly best, there is no alternative.
Probably more accurate to say its IQ is unmatched since the Samsung S23 smartphone has a 200MP sensor
In all seriousness though, I do wonder if the demand for DBs is still high enough to push P1 to produce an IQ5 in the near future. Back in the day their business model was virtually predicated on releasing a new DB every couple of years in order to keep the upgrade merry-go-round spinning, but since they've abandoned that and are now heavily invested in the industrial side of imaging, it's not clear anymore what their long term goals are wrt people like us.
Sony seems to be doing lots of interesting stuff with the 44 x 33 chip, so perhaps the H/B CFV line is more futureproof right now.