heres something i've been wondering. there seems to be a flurry of new releases every year of cameras with more megapixels, making the old models obsolete. there are two things that spring to mind reglularly.
1. what will happen to the images that we have all struggled and strived to take, many of which are in libraries, who seem to demand that megapixels are of a minimum, and that minimum i presume will keep increasing as the market provides cameras of greater capability. will we all have to go out and try and recreate the exact same images in five years time when 22 megapixel cameras are on the market, when the 6 and 8 megapixel bench standard of only a year ago is thought of as old and obsolete, and the images from them unsaleable. will this then create a continuum of supply and demand? if so then it would make the arguement that the digital era has made the market more difficult, somewhat false, as it would appear that the digital era might end up creating an endless demand for repetition at endlessly greater megapixels.
2. will there ever be a more accurate benchmark than the megapixel? there are many six and eight megapixel compacts on the market now, and i can only assume it will be a matter of time before there are a large number of 10 and twelve megapixel compacts. ok, compacts are handy, but it's misleading to state WOW MEGEPIXELS when the sensor is tiny< and the lense is tiny< how will the marketplace determine decent quality? we all know that the same image from a compact will never reach the standard of a decent SLR with a high end lense, but all we ever seem to hear is how many megapixels a camera has. is there a better yardstick we should all be using?
would love to hear your thoughts.
m
Posted by Mike Smith on October 3, 2006 at 6:39 PM
For 90% of the work, agencies are looking for a max of A3 printability being a double page magazine spread.
A couple of years ago the agencies seemed to band together and made the 1ds/dsII their preferred supported camera to keep their Pros onside, but those days are now numbered with the advent of the Canon 5D......
Can't really comment professionally as I have never sold a picture, but I suppose like the computer industry, to stand still the camera manufacturers will be dead in the water, Joe Public expects bigger and better with every release despite not knowing what to do with the pictures.
I think sensor size will peak over next 18 months and we will see more addons such as live view, foveon sensors and blue tooth downloading from here on in
Posted by Rob Ashton on October 3, 2006 at 6:42 PM
You've put across some very relevant thoughts here.
It may be a good thing if the libraries continue to change the minimum requirements as there will always then be a demand for new photos and will keep us photogs in work. However, I have been saying for the last 2 or so years that it is time the marketers started to come clean with the megapixel thing. It is an easy tool to use to pull in the punters and the gullible public keep on lapping it up
You are quite right that there are a number of factors that make up a good digital image. The libraries know this and so won't automatically accept an image just because it was shot on a 12MP camera. They do look at other criteria. However it is one way to keep an avalanche of images from wannabe pros clogging up the system by stipulating that they only accept images taken with a "pro camera".
I think the megapixel issue is no different to the film format issue.
Ignoring compacts for the moment. In the past if somebody had a stunning picture of a lesser-spotted-pink-rhino and nobody else had that image, then that image would sell. If however two people had an acceptable image of such, and all else was equal, then it is likely the person with the image on a larger format will be the reciever of the cheque. I think the same will happen with digital. Therefore, as sensor size increases, there will be a competitive advantage to be gained by those with images similar to their competitors - but captured on more megapixels. The same old limiting factor will return, that being the suitability of the glass availiable and the ease/portability of carrying the larger sensor camera verses the smaller sensor camera.
If we look at the pro market at the moment I think it is segmenting into three pro sections,
1) Leica ( and their clones) - Leica, King of Reportage and highly desirable by pros and those with ££££ in their pockets, have select a 10.4M chip for their, just released, M8. Given their clientelle, and at a body cost of £2600, they could have easily added £400 for a bigger sensor - they didn't. The truth is that they know their market does not need more than this size of sensor - it is the papers and candid market. Portability and being unobstructive are the key criteria. Bigger chip == more memory and more power == unneccessary and unwieldable. Yes I am sure that the chip in the M_next will be bigger but I would suggest not by much unless the price drops and the technology advances significantly.
2) Canon/Nikon/Sig/Pentax/Om/Fuji etc appear to he hovering around the 8-17M mark at the moment (Fuji's S5 is a bit of a odd decision) and I would expect these sensors to finally go up to about 22M. The criteria here is portability v features v speed and the ability to fit the chip into a camera with a 35mm lense mount.
Now, obviously, as these sensors progress upwards in Megapixels we will see the agencies raising the bar with respect to image pixel size for submissions, hence I think you are right we will be shooting the same subjects until the chip size stabilises. As it happens I think that will be about 22M due to the size of a 35mm sensor - but I could be wrong as I previoulsy worked out that it would be possible to bet 36M on a chip - but (currently) it is limited by acceptable noise, so really we ain't that far of it.
Obviously these size of cameras are what Sports and nature photographers are accustomed too and it is (more importantly) what our investiment (glass) fits. I expect to see a 22M Canon sometime next year for Nov 2007/ Feb 2008 release and if we do see a 36M (A2 straight) 35mm it would be in the next 4 years.
3) The medium format crew is where the sensors start about 22M and go up to 39M on the new Hassy H3D. These are studio/still life or landscape beasts - they always have been - this is what their glass is tailored to suit. Yes, needless to say, the likes of Heather Angel will use one for Macro, somebody else will stick one in a UW housing, and subsequently, if there are two similar shots presented to a buyer, Mr A.N.Other on a 12M chip will be beeten by A.Vendor using 39M, - and I am sure Heather and her ilk have always been aware of this, hence her choice of medium format film.
As for normal compacts (not Leica)... well they have their place, but I think you can discount them on the glass front. Sure somebody will be in the right place, right time and hit the button and get the prize that particular day but there will be a good bit of luck associated with it... but, I suggest, to get the prise consistantly your odds increase with the corretc tool for the job, namely, a SLR/Rangefinder/MF dependant upon what, where and when you shoot.
As to the yardstick, well IMHO thats simply price verses quality for useage (basically size of reproduction). This is where it could get sticky as there are hundreds of people now with decent digital cameras (with decent glass) that are capiable of A4, but then this was also the case with 35mm - lets just hope the status-quo continues and they don't know how to use them, cannot get where we go , don't have the time to build a library and don't stick with the pace and don't want to sit in their lightroom(darkroom) perfecting their craft to get the best from it.
PS there are signs of the compact market saturating, in that people are not moving away from their 6M compacts as they do the job adquately - I would suggest that this will be the case with most users of the lower end 8->10M SLRs.
The images already in the libraries will continue to sell as long as they meet the purpose of the buyer, simple as that. For newer images, libraries will continue to raise the bar of Camera Spec/MPix as a semi-effective way of controlling the noise level and keeping the submission levels manageable. In terms of cost required to obtain an image, some images will never get reshot as the time and cost involved in getting them in the 1st place are too high. Sure, all the pics of pretty robins etc will get reshot, but pics of snow leopards mating, Weddel seals under the antarctic ice etc won't - they cost too much to make in the 1st place.
Mpix will always be used as a benchmark until someone can invent a better way of doing it - it's a handy way for Joe Public to measure a camera against. As Tom pointed out above, there're MegaPixels and there're Megapixels - not all are created equal! The Point-n-shoot compact market doesn't need the Mpix it's already got. How many Mp's do you need either for 7x5 or 6x4 prints, or for a full-screen backdrop of your kids for heavens sake?!?! Certainly not 10-12Mp! For P&S cameras, the market is driven by the marketing people needing to regenerate sales and the bragging rights of Joe & Jane Public who need to feel secure in the feeling they've just paid megapounds for the latest and greatest.
Like mobile phones, I guess we'll see the P&S market start to branch into more 'fluffy' ideas - how about a P&S with iTunes, or with an integrated diary. An in-camera overlay to produce those seaside snaps that used to be done with painted backdrops and cut-outs to stick your head through . More sensibly, perhaps use Bluetooth to send direct to a printer? GPS to find the nearest store to get prints done at. How about wireless networking and an e-mail button as well as the print button? How about adding a mobile phone to send them to Granny? Oh, sorry - mobile phones are already nibbling at the P&S market. How long before they have 20x zoom lenses and 10Mp? Not long I bet! Could add always-on mobile (GPRS or whatever) to send each shot directly back to the picture sever on your home PC from where-ever you are in the world (no need for those 100Gb CF cards or 10Tb portable storage systems....)
Seriously though, we will begin to see more discrimination on things such as features of the camera (eg Noise processing) and resolving power of lenses. Why? Because, as with computer CPU's, there are process constraints with sensors that will keep Mpix values down. Sensor wafer yeilds drop as the sensor size increases (due to faults in the wafers), the larger the sensor, the greater the chance of it having a fault on the chip. If the physical size of the sensor is limited (ie to 35mm), then you need to make the pixel smaller to increase the Mpix value - and then you start to see problems with the wafer growing processes...and then with adding in the extra circuitry to make use of the extra pixels....etc etc. Not insurmountable, but expensive.
Hi the megapixel race is already been won by the small point-and-shoot cameras if you to use the pixel density of some of the 10 megapixel point-and-shoot cameras you would end up with around 160 megapixel 35mm frame with a usable ISO range of 100-800, the only thing stopping the camera manufacturers is cost and computing power the actual processing power of the cameras is very slow compared to a modern PC they're running at something like 33 Hz when they can control the speed power consumption and heat generated you will see great improvements in pixel density and storage capacity.
The only reason the P&S can hit the pixel density is because the sensor size is small - so defects on the wafer affect less sensors, hence they can afford to dump the chips that don't make the grade from the wafer batch. At the sizes used by 35mm sensors, the wafer yeilds are too low to make them commercially viable. Go to the sizes used by people like Hassy and PhaseOne for their MF scanning backs etc, then the yeilds are low, but the price premium charged means that they can afford the lower yeilds.
P&S also don't need expensive cutting edge circuits to push the pixels around the camera and onto the memory cards - they have much slower frame rates - so they don't need to process things so quick.
Discussion
6 repliesheres something i've been wondering. there seems to be a flurry of new releases every year of cameras with more megapixels, making the old models obsolete. there are two things that spring to mind reglularly.
1. what will happen to the images that we have all struggled and strived to take, many of which are in libraries, who seem to demand that megapixels are of a minimum, and that minimum i presume will keep increasing as the market provides cameras of greater capability. will we all have to go out and try and recreate the exact same images in five years time when 22 megapixel cameras are on the market, when the 6 and 8 megapixel bench standard of only a year ago is thought of as old and obsolete, and the images from them unsaleable. will this then create a continuum of supply and demand? if so then it would make the arguement that the digital era has made the market more difficult, somewhat false, as it would appear that the digital era might end up creating an endless demand for repetition at endlessly greater megapixels.
2. will there ever be a more accurate benchmark than the megapixel? there are many six and eight megapixel compacts on the market now, and i can only assume it will be a matter of time before there are a large number of 10 and twelve megapixel compacts. ok, compacts are handy, but it's misleading to state WOW MEGEPIXELS when the sensor is tiny< and the lense is tiny< how will the marketplace determine decent quality? we all know that the same image from a compact will never reach the standard of a decent SLR with a high end lense, but all we ever seem to hear is how many megapixels a camera has. is there a better yardstick we should all be using?
would love to hear your thoughts.
m
A couple of years ago the agencies seemed to band together and made the 1ds/dsII their preferred supported camera to keep their Pros onside, but those days are now numbered with the advent of the Canon 5D......
Can't really comment professionally as I have never sold a picture, but I suppose like the computer industry, to stand still the camera manufacturers will be dead in the water, Joe Public expects bigger and better with every release despite not knowing what to do with the pictures.
I think sensor size will peak over next 18 months and we will see more addons such as live view, foveon sensors and blue tooth downloading from here on in
It may be a good thing if the libraries continue to change the minimum requirements as there will always then be a demand for new photos and will keep us photogs in work. However, I have been saying for the last 2 or so years that it is time the marketers started to come clean with the megapixel thing. It is an easy tool to use to pull in the punters and the gullible public keep on lapping it up
You are quite right that there are a number of factors that make up a good digital image. The libraries know this and so won't automatically accept an image just because it was shot on a 12MP camera. They do look at other criteria. However it is one way to keep an avalanche of images from wannabe pros clogging up the system by stipulating that they only accept images taken with a "pro camera".
Rob
Ignoring compacts for the moment. In the past if somebody had a stunning picture of a lesser-spotted-pink-rhino and nobody else had that image, then that image would sell. If however two people had an acceptable image of such, and all else was equal, then it is likely the person with the image on a larger format will be the reciever of the cheque. I think the same will happen with digital. Therefore, as sensor size increases, there will be a competitive advantage to be gained by those with images similar to their competitors - but captured on more megapixels. The same old limiting factor will return, that being the suitability of the glass availiable and the ease/portability of carrying the larger sensor camera verses the smaller sensor camera.
If we look at the pro market at the moment I think it is segmenting into three pro sections,
1) Leica ( and their clones) - Leica, King of Reportage and highly desirable by pros and those with ££££ in their pockets, have select a 10.4M chip for their, just released, M8. Given their clientelle, and at a body cost of £2600, they could have easily added £400 for a bigger sensor - they didn't. The truth is that they know their market does not need more than this size of sensor - it is the papers and candid market. Portability and being unobstructive are the key criteria. Bigger chip == more memory and more power == unneccessary and unwieldable. Yes I am sure that the chip in the M_next will be bigger but I would suggest not by much unless the price drops and the technology advances significantly.
2) Canon/Nikon/Sig/Pentax/Om/Fuji etc appear to he hovering around the 8-17M mark at the moment (Fuji's S5 is a bit of a odd decision) and I would expect these sensors to finally go up to about 22M. The criteria here is portability v features v speed and the ability to fit the chip into a camera with a 35mm lense mount.
Now, obviously, as these sensors progress upwards in Megapixels we will see the agencies raising the bar with respect to image pixel size for submissions, hence I think you are right we will be shooting the same subjects until the chip size stabilises. As it happens I think that will be about 22M due to the size of a 35mm sensor - but I could be wrong
Obviously these size of cameras are what Sports and nature photographers are accustomed too and it is (more importantly) what our investiment (glass) fits. I expect to see a 22M Canon sometime next year for Nov 2007/ Feb 2008 release and if we do see a 36M (A2 straight) 35mm it would be in the next 4 years.
3) The medium format crew is where the sensors start about 22M and go up to 39M on the new Hassy H3D. These are studio/still life or landscape beasts - they always have been - this is what their glass is tailored to suit. Yes, needless to say, the likes of Heather Angel will use one for Macro, somebody else will stick one in a UW housing, and subsequently, if there are two similar shots presented to a buyer, Mr A.N.Other on a 12M chip will be beeten by A.Vendor using 39M, - and I am sure Heather and her ilk have always been aware of this, hence her choice of medium format film.
As for normal compacts (not Leica)... well they have their place, but I think you can discount them on the glass front. Sure somebody will be in the right place, right time and hit the button and get the prize that particular day but there will be a good bit of luck associated with it... but, I suggest, to get the prise consistantly your odds increase with the corretc tool for the job, namely, a SLR/Rangefinder/MF dependant upon what, where and when you shoot.
As to the yardstick, well IMHO thats simply price verses quality for useage (basically size of reproduction). This is where it could get sticky as there are hundreds of people now with decent digital cameras (with decent glass) that are capiable of A4, but then this was also the case with 35mm - lets just hope the status-quo continues and they don't know how to use them, cannot get where we go
PS there are signs of the compact market saturating, in that people are not moving away from their 6M compacts as they do the job adquately - I would suggest that this will be the case with most users of the lower end 8->10M SLRs.
Mpix will always be used as a benchmark until someone can invent a better way of doing it - it's a handy way for Joe Public to measure a camera against. As Tom pointed out above, there're MegaPixels and there're Megapixels - not all are created equal! The Point-n-shoot compact market doesn't need the Mpix it's already got. How many Mp's do you need either for 7x5 or 6x4 prints, or for a full-screen backdrop of your kids for heavens sake?!?! Certainly not 10-12Mp! For P&S cameras, the market is driven by the marketing people needing to regenerate sales and the bragging rights of Joe & Jane Public who need to feel secure in the feeling they've just paid megapounds for the latest and greatest.
Like mobile phones, I guess we'll see the P&S market start to branch into more 'fluffy' ideas - how about a P&S with iTunes, or with an integrated diary. An in-camera overlay to produce those seaside snaps that used to be done with painted backdrops and cut-outs to stick your head through
Seriously though, we will begin to see more discrimination on things such as features of the camera (eg Noise processing) and resolving power of lenses. Why? Because, as with computer CPU's, there are process constraints with sensors that will keep Mpix values down. Sensor wafer yeilds drop as the sensor size increases (due to faults in the wafers), the larger the sensor, the greater the chance of it having a fault on the chip. If the physical size of the sensor is limited (ie to 35mm), then you need to make the pixel smaller to increase the Mpix value - and then you start to see problems with the wafer growing processes...and then with adding in the extra circuitry to make use of the extra pixels....etc etc. Not insurmountable, but expensive.
Sorry for the ramble
Phillip.
P&S also don't need expensive cutting edge circuits to push the pixels around the camera and onto the memory cards - they have much slower frame rates - so they don't need to process things so quick.
Film was so simple in comparison