Rumors say Olympus's modular camera system is now in progress and this opens a whole new world of opportunities, changes will be on every aspect of the camera, new lenses, new sensors, new AF...
I don't put much stock in rumors most of the time.
I'm always interested to see what Olympus is up to, though I've never owned anything of theirs except a digital voice recorder. The E-P2 is pretty nice.
I don't think Olympus will be dropping Four Thirds or Micro Four Thirds anytime soon, but doesn't every camera maker have new lenses, new sensors, improved AF, etc. etc. etc. every year or two? So even if your rumor is true, what is it telling us that we didn't already know?
I think it's a great idea in theory...hopefully the implementation works out. It would be cool to build a dSLR like you can build a desktop pc. Put your money into a better processor, better burst rate, better focus system, etc, depending on your needs.
Canon 7D and some stuff that sticks on the end of it.
Rumors say Olympus's modular camera system is now in progress and this opens a whole new world of opportunities, changes will be on every aspect of the camera, new lenses, new sensors, new AF...
What do you think?
I think that a rumor is a rumor. It has no basis in fact.
I think it would be a mistake to do that. It would make a system more expensive, it would make it heavier, it would make it more prone to problems, and I think it would be the one thing they could do that would put them out of the camera business.
Thom Hogan has been hoping for this for a while now. I think he makes some good points, and it will be interesting to see if this is a fact or simply a rumor.
Modular systems just don't make a lot of sense. In the scheme of things, bodies aren't that big of a cost. And your modular system will only be modular to a point and with the way technology changes it will be outdated too quickly to get any benefit.
For example, a new sensor may require a completely new logic board to handle some new high speed datapath... Guess what, you have to upgrade most of the guts. A new sensor could require more power than the system was originally spec'd for. Guess what, more new guts. A new sensor could be larger than the body was designed for, New mounts and body will be needed. New dust control technique could come along that requires a new mount.
It's all a mess.
Look at a PC. Most people just add ram or harddrives. But if you add a new video card you may need a new powersupply. If you get a new processor you most likely will need a new motherboard, ram, and powersupply.
Lets not even discuss voiding your warranty for opening your PC. The same would be true for these cameras. There is too much to go wrong and Olympus will get tired of dealing with warranty issues over customer mistakes.
Canon Gear: 5D MkII, 30D, 85 1.2 L, 70-200 2.8 IS L, 17-40mm f4 L, 50 1.4, 580EX, 2x 580EXII, Canon 1.4x TC, 300 f4 IS L, 100mm 2.8 Macro, 100-400 IS L Other Gear: Olympus E-PL1, Pan 20 1.7, Fuji 3D Camera, Lensbaby 2.0, Tamron 28-75 2.8, Alien Bees lighting, CyberSyncs, Domke, HONL, FlipIt.
~ Gear Pictures
Modular systems just don't make a lot of sense. In the scheme of things, bodies aren't that big of a cost. And your modular system will only be modular to a point and with the way technology changes it will be outdated too quickly to get any benefit.
SLRs are already modular, though. The companies have (at least internally) defined interface standards for the bodies to communicate with the lenses. You don't think they could define a standard for how the sensors communicate with the body?
Sure, power requirements could change, and a new processor might be required, but with firmware updates I'd guess that a logic board could last at least a couple of generations of sensor. If one bought a camera with a "sports/action" sensor and could buy an additional "portrait/low-light" or "landscape" sensor for maybe half the cost of buying an additional camera, I can see how it would be popular. And if maybe in a couple of years a new sensor comes out but it requires upgrading to a different battery or logic board, that could be feasible.
Is there a business case? I don't know. Would there be warranty issues? I don't know. I'm neither an MBA nor a lawyer, but as an engineer I think it's totally possible, and as someone who has no fear of building my own PC or removing an engine or transmission, I think it's totally doable. I don't know if it would work well enough or be idiot-proof enough for large-scale sales, though.
There is more to logic boards than firmware, and there is more to providing more power than just changing a battery.
Can they make something modular? Sure. Is it good for us? NO.
Camera makers would release limited modules because sales go down as modules get more niche and as things become more niche prices go up. They would work hard to limit any compatibility with third party modules to ensure they can retain sales. They could only maintain compatibility for such a limited time due to the speed of advancements that your modules would only work on 3 generations AT MOST in this sort of pattern; Previous generation compatibility at reduced capacity mode, current generation at full capacity, and next generation at reduced capacity. The next generation body and modules will always be held back in one way or another to allow for backwards compatibility so they don't alienate too many customers with large investments. Heck, they currently don't even keep the batteries compatible beyond a generation.
And don't get me started on the crap the manufacturers would do to get extra money out of us... Right now we have great general purpose sensors. If we can swap sensors easily, manufacturers may 'nerf' the generally purpose sensor that comes with the body so it doesn't perform as well as it should in low light in order to force you to buy the low light sensor. Don't believe that? Look at the 'nerfing' they do to the rebels. Do you think they can't turn on ISO 50 or 3200 on a rebel? Do you think it would cost too much to put a sync port on a rebel? And they do it in reverse too. What took so long to put a flash commander in the camera body? Why did they feel the need to not allow the IR shutter remote to work on the better bodies to force you to buy their very expensive wireless shutter remotes? They are controlling profits.
More points of failure, more chance of damage in the field, more crap to carry, etc. Too high of a risk.
And if you build PC's tell me how many times you have changed the processor without having to upgrade other components? Heck, there are three different i7 processors right now that each require different motherboards. And transmissions? Tell me you don't match the transmission, not only to the car make, model, year, but even the particular engine in the car.
Canon Gear: 5D MkII, 30D, 85 1.2 L, 70-200 2.8 IS L, 17-40mm f4 L, 50 1.4, 580EX, 2x 580EXII, Canon 1.4x TC, 300 f4 IS L, 100mm 2.8 Macro, 100-400 IS L Other Gear: Olympus E-PL1, Pan 20 1.7, Fuji 3D Camera, Lensbaby 2.0, Tamron 28-75 2.8, Alien Bees lighting, CyberSyncs, Domke, HONL, FlipIt.
~ Gear Pictures
Is there a business case? I don't know. Would there be warranty issues? I don't know. I'm neither an MBA nor a lawyer, but as an engineer I think it's totally possible, and as someone who has no fear of building my own PC or removing an engine or transmission, I think it's totally doable. I don't know if it would work well enough or be idiot-proof enough for large-scale sales, though.
That's the problem. The average consumer may buy a PC that in theory allows him to upgrade memory or change video cards, but in general he never takes advantage of that because he isn't technically-inclined enough to open up the case and learn how to mess around with things. I doubt the average camera purchaser is any different. Changing lenses is one thing, but switching out the sensor is a different proposition, especially considering how precisely the sensor has to be positioned for images to be in focus. How many SLR purchasers even know how to swap out a focusing screen?
To make the guts of a camera modular and user-swappable, you'd have to make it no harder than changing lenses or memory cards. It's probably not impossible but I don't really see the point, because a DSLR body basically IS its sensor. Everything else is just support for that.
There is more to logic boards than firmware, and there is more to providing more power than just changing a battery.
I wasn't trying to imply that a new battery was the only thing required to provide more power, nor that logic is only firmware. What I was trying to say was that a power module and logic board/processor could in theory also be modular, so that you could upgrade them as well. Is that practical? Probably not, I'm just saying that the hurdles you mentioned aren't insurmountable.
And if you build PC's tell me how many times you have changed the processor without having to upgrade other components? Heck, there are three different i7 processors right now that each require different motherboards. And transmissions? Tell me you don't match the transmission, not only to the car make, model, year, but even the particular engine in the car.
All your points are well taken. I'm just not sure that a modular camera system is a bad idea. Is it mainstream? Probably not, for the reasons you and Craig mention. My points on building PCs and working on cars was intended to mean that I'm probably atypical in that respect, and that it wouldn't scare me off, but it might some. Many if not most people have a minor freakout when presented with the idea of adding RAM to a computer, and I don't know how many impressed reactions I've gotten when I say I do my own brake jobs. People don't tend to get their hands dirty like that anymore.
I just think it would be very cool if they (I'll use Nikon since that's what I shoot) could build a D700/D300 class body that could take a DX sensor for the extra reach, a D3s sensor for low noise, or a D3x sensor for the high-MP. Yes, the body itself would likely cost a fair amount more than they do today (maybe $3.5k for "modular-compatible" body with D3s sensor; roughly $1k more than the current D700), then an additional $1k for DX sensor and $3k for high-res sensor. That's still $7.5k, which is decidedly not in mainstream territory, but it's a whole heluva lot cheaper than buying a D300s, D3s, and D3x. I don't think that body or its technology would last forever, but one could probably replace it every 6 years instead of maybe 3-4.
Again, I'm thinking like an engineer and how cool (and possible) it would be, not how practical or how much ROI they'd get if they went down this road. That's why I work in a lab and not in the corner office.
Comments
I'm always interested to see what Olympus is up to, though I've never owned anything of theirs except a digital voice recorder. The E-P2 is pretty nice.
I don't think Olympus will be dropping Four Thirds or Micro Four Thirds anytime soon, but doesn't every camera maker have new lenses, new sensors, improved AF, etc. etc. etc. every year or two? So even if your rumor is true, what is it telling us that we didn't already know?
Got bored with digital and went back to film.
I think that a rumor is a rumor. It has no basis in fact.
I think it would be a mistake to do that. It would make a system more expensive, it would make it heavier, it would make it more prone to problems, and I think it would be the one thing they could do that would put them out of the camera business.
Well, you asked.
www.spanielsport.com
My site 365 Project
For example, a new sensor may require a completely new logic board to handle some new high speed datapath... Guess what, you have to upgrade most of the guts. A new sensor could require more power than the system was originally spec'd for. Guess what, more new guts. A new sensor could be larger than the body was designed for, New mounts and body will be needed. New dust control technique could come along that requires a new mount.
It's all a mess.
Look at a PC. Most people just add ram or harddrives. But if you add a new video card you may need a new powersupply. If you get a new processor you most likely will need a new motherboard, ram, and powersupply.
Lets not even discuss voiding your warranty for opening your PC. The same would be true for these cameras. There is too much to go wrong and Olympus will get tired of dealing with warranty issues over customer mistakes.
Other Gear: Olympus E-PL1, Pan 20 1.7, Fuji 3D Camera, Lensbaby 2.0, Tamron 28-75 2.8, Alien Bees lighting, CyberSyncs, Domke, HONL, FlipIt.
~ Gear Pictures
SLRs are already modular, though. The companies have (at least internally) defined interface standards for the bodies to communicate with the lenses. You don't think they could define a standard for how the sensors communicate with the body?
Sure, power requirements could change, and a new processor might be required, but with firmware updates I'd guess that a logic board could last at least a couple of generations of sensor. If one bought a camera with a "sports/action" sensor and could buy an additional "portrait/low-light" or "landscape" sensor for maybe half the cost of buying an additional camera, I can see how it would be popular. And if maybe in a couple of years a new sensor comes out but it requires upgrading to a different battery or logic board, that could be feasible.
Is there a business case? I don't know. Would there be warranty issues? I don't know. I'm neither an MBA nor a lawyer, but as an engineer I think it's totally possible, and as someone who has no fear of building my own PC or removing an engine or transmission, I think it's totally doable. I don't know if it would work well enough or be idiot-proof enough for large-scale sales, though.
My site 365 Project
If Fuji makes one of those modular sensors I might fall head over heels for it. Hopefully the price isn't more than I can get for the D700.
Can they make something modular? Sure. Is it good for us? NO.
Camera makers would release limited modules because sales go down as modules get more niche and as things become more niche prices go up. They would work hard to limit any compatibility with third party modules to ensure they can retain sales. They could only maintain compatibility for such a limited time due to the speed of advancements that your modules would only work on 3 generations AT MOST in this sort of pattern; Previous generation compatibility at reduced capacity mode, current generation at full capacity, and next generation at reduced capacity. The next generation body and modules will always be held back in one way or another to allow for backwards compatibility so they don't alienate too many customers with large investments. Heck, they currently don't even keep the batteries compatible beyond a generation.
And don't get me started on the crap the manufacturers would do to get extra money out of us... Right now we have great general purpose sensors. If we can swap sensors easily, manufacturers may 'nerf' the generally purpose sensor that comes with the body so it doesn't perform as well as it should in low light in order to force you to buy the low light sensor. Don't believe that? Look at the 'nerfing' they do to the rebels. Do you think they can't turn on ISO 50 or 3200 on a rebel? Do you think it would cost too much to put a sync port on a rebel? And they do it in reverse too. What took so long to put a flash commander in the camera body? Why did they feel the need to not allow the IR shutter remote to work on the better bodies to force you to buy their very expensive wireless shutter remotes? They are controlling profits.
More points of failure, more chance of damage in the field, more crap to carry, etc. Too high of a risk.
And if you build PC's tell me how many times you have changed the processor without having to upgrade other components? Heck, there are three different i7 processors right now that each require different motherboards. And transmissions? Tell me you don't match the transmission, not only to the car make, model, year, but even the particular engine in the car.
Other Gear: Olympus E-PL1, Pan 20 1.7, Fuji 3D Camera, Lensbaby 2.0, Tamron 28-75 2.8, Alien Bees lighting, CyberSyncs, Domke, HONL, FlipIt.
~ Gear Pictures
That's the problem. The average consumer may buy a PC that in theory allows him to upgrade memory or change video cards, but in general he never takes advantage of that because he isn't technically-inclined enough to open up the case and learn how to mess around with things. I doubt the average camera purchaser is any different. Changing lenses is one thing, but switching out the sensor is a different proposition, especially considering how precisely the sensor has to be positioned for images to be in focus. How many SLR purchasers even know how to swap out a focusing screen?
To make the guts of a camera modular and user-swappable, you'd have to make it no harder than changing lenses or memory cards. It's probably not impossible but I don't really see the point, because a DSLR body basically IS its sensor. Everything else is just support for that.
Got bored with digital and went back to film.
I wasn't trying to imply that a new battery was the only thing required to provide more power, nor that logic is only firmware. What I was trying to say was that a power module and logic board/processor could in theory also be modular, so that you could upgrade them as well. Is that practical? Probably not, I'm just saying that the hurdles you mentioned aren't insurmountable.
All your points are well taken. I'm just not sure that a modular camera system is a bad idea. Is it mainstream? Probably not, for the reasons you and Craig mention. My points on building PCs and working on cars was intended to mean that I'm probably atypical in that respect, and that it wouldn't scare me off, but it might some. Many if not most people have a minor freakout when presented with the idea of adding RAM to a computer, and I don't know how many impressed reactions I've gotten when I say I do my own brake jobs. People don't tend to get their hands dirty like that anymore.
I just think it would be very cool if they (I'll use Nikon since that's what I shoot) could build a D700/D300 class body that could take a DX sensor for the extra reach, a D3s sensor for low noise, or a D3x sensor for the high-MP. Yes, the body itself would likely cost a fair amount more than they do today (maybe $3.5k for "modular-compatible" body with D3s sensor; roughly $1k more than the current D700), then an additional $1k for DX sensor and $3k for high-res sensor. That's still $7.5k, which is decidedly not in mainstream territory, but it's a whole heluva lot cheaper than buying a D300s, D3s, and D3x. I don't think that body or its technology would last forever, but one could probably replace it every 6 years instead of maybe 3-4.
Again, I'm thinking like an engineer and how cool (and possible) it would be, not how practical or how much ROI they'd get if they went down this road. That's why I work in a lab and not in the corner office.
My site 365 Project