-
Welcome to the new forums! Please read this first. For known issues we are working to resolve, click here.
Sensor evolution stuck?
- Thread starter Jcpastor1
- Start date
John Crowe
Veteran Member
Everything you listed is not attributed to the sensor, but to the camera systems around it. So the sensor itself, is still stuck.
Actually, the advantages are a direct result of sensor evolution...sensors capable of moving more data to the processor much faster, and sensors with integrated AF points across nearly the full surface area. As a result of modern sensor design, mechanical shutters are no longer necessary (silent shooting), and autofocus is faster, more capable, and consistently more accurate. Burst rates are faster and the output image is better resolved.
Modern sensors made real-time EVF displays possible, displays that show the photographer the image being made, complete with depth of field and color rendering. Today's sensors make recording 8K video possible. They make 4K/120p video possible.
The dynamic range of today's cameras may be about the same or, in some cases, a skosh worse but, in so many other ways, the camera are just objectively, measurably better. It shows in the photos people make and it's a product of objectively, measurably better sensor tech.
John Crowe
Veteran Member
While these underlying improvements are interesting, the OP did specifically refer to the output imagery from the sensors. Generally when enthusiasts refer to sensor development we tend to focus on the image quality itself, rather than contributions to increased speeds, and improvements to AF, etc. Good to know that sensors are more entwined in the electronic process though.
dcstep
Senior Member
Thee technologies are in their infancy. Even so, the Sony a9 III's global shutter provides real world advantages in terms of photo accuracy (Not just DR and Resolution). Solving real world problems such as banding, rolling shutter distortion that is obvious in certain sports, like baseball and golf.
The a9III is aimed at sports, event, studio and wildlife photographers. I would be very surprised if Sony isn't working on solutions for other uses by applying technology similar to phone technology.
The next few years should be very interesting.
Welcome aboard.
I don't think sensor tech is stuck, it's just focused on a different format right now which is smart phones as 1) that's where the market is, 2) that's where improvements are the most rewarded
In theory let's say we could buy an affordable 1000mp full frame camera, what would the average photo enthusiast do with it and it's 250mb files other than keeping busy buying new hard drives every month to store the files that are so overkill when being shared on the internet and looked at on 3"x5" cell phone screens?
Last edited:
You confuse yourself with everyone. It is not true in the slightest. An enthusiast would have been following camera news and tech at some point in the last decade.
Stacked sensors are revolutionary and a massive leap forward. It is clear that you've not used one and don't know much about them.
Last edited:
WilyKit
Senior Member
- Messages
- 1,066
- Reaction score
- 1,314
Where is that "revolution"? Hardly any enthusiast noticed. Auto-focus speed and accuracy were very, very good already in top of the line cameras. Top of the line models which have stacked sensors are somewhat better in some cases, no revolution, and the improvements can hardly be attributed solely to the fact that they use stacked sensors. Burst shots - now you can take 120 pictures per second, OK, wonderful. How many people do that?
a9 is from way back in 2017, and RX100 IV is from 2015, stacked sensors in cameras are not a new thing at all, in technology lifecycle terms. Low light performance is somewhat better, enough that it may approach that of some DSLR cameras, but don't compare the venerable Sony a9 II with the much cheaper Nikon D780 (DSLR), it looks terrible compared to it in the higher ISOs. That's not a revolution. In MFT, differences at ISO 6400 (similar to ISO 25,600 in full-frame) are minimal between an E-M1X which has a "normal" sensor which doesn't even have BSI and OM-1 which has a stacked sensor with BSI. The OM-1's advantage (definitely not revolutionary) starts from ISOs even higher than that.
Rolling shutter is reduced, and that's great, but what's not great is that that's still not enough to make the electronic shutter capable of handling various problems with artificial lights, so if anyone thought "hey I'll get a camera with a stacked sensor, it will allow me to take silent burst shots in the events I photograph" they realized there is in fact no revolution. I've also seen examples of obvious rolling shutter artifacts with those sensors too, so it's reduced in some cases noticeably, in some cases not. Global shutter sensors? Yes, there's one camera (a9 III), it's a revolution for those who are lucky to be able to afford a $6,000 camera.
No, there's only incremental updates and the field of sensors is desperately in need of innovation and experts, and that's not just me saying it, it's also world-renowned experts who are pioneers in sensor technology like Albert Theuwissen who are saying it. It's only some "enthusiasts" in forums who pretend that the field of imaging sensors is doing well.
Last edited:
dcstep
Senior Member
Global shutter is at its infancy in that part of the revolution. For actual photography, it has already cured several issues that real photographers encounter. There's more to come, including trickle-down price competitiveness.
Oh, as we BS each other here, Sony is working on a three-layer stacked sensor that will increase DR and other advantages related to processing speed.
People on this thread seem to belittle the HUGE improvements in AF since 2018, when Sony introduced the a9. An hour ago, I was shooting a hummingbird at 120-fps, with over 90% of the shots in sharp focus. Try that with your five year old Nikon or Canon. AF improvements have not merely been "incremental". Instead, huge strides have been made with initial acquisition, tracking and things like Bird-eye detection.
bastibe
Senior Member
- Messages
- 1,997
- Solutions
- 4
- Reaction score
- 2,999
Myles Baker
Forum Enthusiast
Image quality, especially at low ISO's, has not improved markedly in the last 10 years. But as you alluded to, the imaging pipeline has. Better processors, faster read-out speeds, higher fps, better af, they have all contributed to making today's cameras better all round tools. The biggest change in image quality is at high iso"s, the improvement in processors plays a big part in that. You can see that in "recycled" sensors used in new bodies with better processors.
As an example, one of the things that held Canon back for so long was it's processors. I think both Sony & Nikon had an edge there for quite a while and I think a lot of the complaints of Canon "crip..ing" their camera's, was less that they wouldn't & more that they couldn't. The RP v the R8 is a case in point. The RP was panned by a lot of people as being hit with the "cri..le hammer". I don't think that was the case. I think the processor available at that time simply couldn't provide the power to drive further enhancements.
And I would argue that the image quality between the RP & the R8 is negligible in most circumstances, but the improvements at high iso, rolling shutter, speed & af, is night & day.
Without the new processor those improvements don't happen.
I guess my point is, better processors have driven camera technology forward far more than improved image quality, in fact with the focus on video has come a slight reduction or stagnation in some measurements of IQ, in particular DR and higher resolutions, in favour of faster read-out speeds & higher fps all of which need higher throughput of data, which can only come from better processors.
I could be wrong of course, it's been known to happen
As an example, one of the things that held Canon back for so long was it's processors. I think both Sony & Nikon had an edge there for quite a while and I think a lot of the complaints of Canon "crip..ing" their camera's, was less that they wouldn't & more that they couldn't. The RP v the R8 is a case in point. The RP was panned by a lot of people as being hit with the "cri..le hammer". I don't think that was the case. I think the processor available at that time simply couldn't provide the power to drive further enhancements.
And I would argue that the image quality between the RP & the R8 is negligible in most circumstances, but the improvements at high iso, rolling shutter, speed & af, is night & day.
Without the new processor those improvements don't happen.
I guess my point is, better processors have driven camera technology forward far more than improved image quality, in fact with the focus on video has come a slight reduction or stagnation in some measurements of IQ, in particular DR and higher resolutions, in favour of faster read-out speeds & higher fps all of which need higher throughput of data, which can only come from better processors.
I could be wrong of course, it's been known to happen
dcstep
Senior Member
Moving from Canon to Sony in 2018 was a big enough leap for me to sell $30,000 in Canon equipment and replace it with a Sony a9 with a 100-400mm GM lens. Even for landscape, the 50.1mp and 60mp Sony sensors blow away my old 50mp Canon sensor in terms of DR and noise.
Landscape photographers can manual focus, like I did with my Yashica 44 in 1960 and the image quality will not be much better, if at all. Those 44x44mm files jumped off the screen! If landscape were my main gig, I'd be looking at one of those 100mp Hassy or Fuji. People rave about those files, but I have no personal experience with those. I AM consider a 60mp Sony as a gift for a talented friend.
Shooting wildlife, like a leaping coyote and a white-tail fawn running through tall prairie grass is much easier with Animal Eye detect and Tracking locked onto the eye. If I lock onto the eye, the a9III and the a1II stay locked on, even when the eye momentarily goes behind the grass. That was unheard of five years ago. Oh, and it hasn't stopped getting better.
--
Dave
Last edited:
John Crowe
Veteran Member
I am an enthusiast photographer, not a scientist. I am sure all the developments you refer to are wonderful. However, all one has to do is look at the studio comparison photos here on dpreview to see that the images made by all these different sensors, of similar megapixels, over the past ten to twelve years, have not changed considerably.
There may have been some advancements in dynamic range and ultra high ISO performance, but these will only provide nominal improvements in real world use.
There may have been some advancements in dynamic range and ultra high ISO performance, but these will only provide nominal improvements in real world use.
sportyaccordy
Forum Pro
Im too lazy to pull up the charts but if you compare sensors to "Ideal" on PhotonsToPhotos you can see we have pretty much reached the physical limits of dynamic range.
A modern MILC destroys something like a D750 on pretty much everything outside of IQ, so IMO an upgrade is still worth it.
A modern MILC destroys something like a D750 on pretty much everything outside of IQ, so IMO an upgrade is still worth it.
Gato Amarillo
Forum Pro
By chance I read Thom Hogan's take on sensor evolution earlier today. Now I see new action on this thread.
bythom.com
As he says: Sensor evolution is not stuck, it's been diverted"
Thom's point seems to be that we could have much higher MP, if people could use it and the camera makers thought people wanted it. Instead the research is going into improving speed and processing.
Some here have discounted the improvements, but even to a portrait guy like me they are real. Quick and accurate eye detect is a very real upgrade to me over the cameras of 10 years ago, maybe even five. Higher ISO means I can now shoot 3200 where I once felt limited to 1200. Or in a more realistic case, I can happily shoot 400 or even 800 for indoor portraits, meaning I can do more with lighter, less expensive lighting equipment. The difference in cost and weight between a 200WS flash and an 800WS is very real for a location photographer.
So for me the improvements are real, and I say keep 'em coming.
Gato
Is That All There Is? (Image Sensors, That Is) | Cameras and Photography Explained | Thom Hogan
where are we with image sensors?
As he says: Sensor evolution is not stuck, it's been diverted"
Thom's point seems to be that we could have much higher MP, if people could use it and the camera makers thought people wanted it. Instead the research is going into improving speed and processing.
Some here have discounted the improvements, but even to a portrait guy like me they are real. Quick and accurate eye detect is a very real upgrade to me over the cameras of 10 years ago, maybe even five. Higher ISO means I can now shoot 3200 where I once felt limited to 1200. Or in a more realistic case, I can happily shoot 400 or even 800 for indoor portraits, meaning I can do more with lighter, less expensive lighting equipment. The difference in cost and weight between a 200WS flash and an 800WS is very real for a location photographer.
So for me the improvements are real, and I say keep 'em coming.
Gato
dcstep
Senior Member
Wow! You're totally missing the point.
Sensor technology has way more to do with photography than taking pictures of static setups, on a tripod, in a studio. If you read the thread, many of us are enjoying HUGE advances in AF technology. If you compare the high-ISO performance (3200 and above) you'll see great advances in DR (greatly lower noise) that are very useful outside of a studio on a tripod. Read out speeds give us another set of very useful advantages for taking actual pictures.
Global shutter, in its infancy, has a whole bag of additional advantages and who knows what we'll enjoy once Sony gets its three-layer stacked sensor into a camera.
tbcass
Forum Pro
The OP, Jcpastor1 was referring to image quality alone. While there have been some advances in IQ over the last decade they are marginal at best and require careful examination at 100% to see. In some cases IQ has gotten a bit worse because on sensor PDAF actually has a slight negative effect on IQ while greatly improving the speed and photographic experience.
dcstep
Senior Member
The thread titles is, "Is Sensor Technology Stuck.." The answer is no, if you actually take pictures. If all you do is look at static images, shot on a tripod, at base ISO, with artificial lighting, then maybe only a little.
https://www.dpreview.com/articles/5...-quality-but-cameras-are-still-getting-better
As has been the case for a long time, a lot of the efforts are going on in smartphone sensors, because the market is bigger and the demands for improvement are greater.
Modern large sensors are already exceptionally good, and (though I'm sure history will prove me wrong in this), I'm not sure whether most people would benefit from, say, high-res full-frame moving from 60MP to 100MP and the base models moving from 24 to 40 or whatever.
There might be room for some more DR, but that might sensibly be exposed as a lower ISO, so may be primarily a landscape/controlled lighting benefit. It's hard to see, short of moving beyond colour filtration somehow, how low light performance is going to improve.
So it's mainly speed, which allows better video and better AF. And those improvements in AF, along with the major steps forward in lens design (and materials, and production), mean that more shots are in focus and look more detailed, despite the sensor IQ not necessarily changing that much.
Richard - DPReview.com
As has been the case for a long time, a lot of the efforts are going on in smartphone sensors, because the market is bigger and the demands for improvement are greater.
Modern large sensors are already exceptionally good, and (though I'm sure history will prove me wrong in this), I'm not sure whether most people would benefit from, say, high-res full-frame moving from 60MP to 100MP and the base models moving from 24 to 40 or whatever.
There might be room for some more DR, but that might sensibly be exposed as a lower ISO, so may be primarily a landscape/controlled lighting benefit. It's hard to see, short of moving beyond colour filtration somehow, how low light performance is going to improve.
So it's mainly speed, which allows better video and better AF. And those improvements in AF, along with the major steps forward in lens design (and materials, and production), mean that more shots are in focus and look more detailed, despite the sensor IQ not necessarily changing that much.
Richard - DPReview.com
sportyaccordy
Forum Pro
Not trying to take a swipe at you when I say this, but if your gear list is current it sounds like you haven't used a modern MILC which explains why you just judge through pictures. The shooting experience is substantially upgraded, even if the raw IQ hasn't changed much. Having AF coverage across the full sensor, being able to preview exposure through the viewfinder, Eye AF etc. Hell, just the accuracy of AF with fast thin DoF glass (no micro AF adjustments necessary) is huge- SLR glass often works better adapted to MILCs than the SLRs they were designed for. It's substantially easier to nail a shot with an MILC than a DSLR.
Similar threads
Keyboard shortcuts
- f
- Forum