I think the replacement cycle will start to grow, replacing models every 18 months or so when the competition don't is a little pointless if you ask me.
Yeah, and remember to dump the 20D ahead of the curve if you like to forecast. Reminder of what happened to 10D owners when the 20D came out. Many lost $400-500 by selling too late.
I guess Canon now has 8 Mega Pixel (Rebel and 20D) consumer DSLRs, an 8 Mega Pixel Pro Sports DSLR (1DMk II), a 11 Mega Pixel Pro SLR (1Ds), and the 16 Mega Pixel Pro DSLR (1DsMk II).
Speculating, they might bump the sensor on the successor of the 20D to a 11 or 12 Mega Pixel. But I don't see this happening before the successor of the 1DMk II.... the 1DMk III....maybe a 12 or 14 Mega Pixel Sports Pro DSLR comes out. Then they might upgrade the 20D to the next model at 12 Mega Pixel and the 1DsMk II to a 20 or 24 Mega Pixel. Something tells me Nikon might not quite be able to keep up with Canon on the D-SLR range.
I am quite content with two 8 Mega Pixel and one old 6 Megapixel D-SLR’s and probably will not upgrade to a new Canon until they bring out a 11 or 12 Mega Pixel camera in the same price range as the 20D… which probably will happen in 3 years.
robbie20D wrote:
I think the replacement cycle will start to grow, replacing models every 18 months or so when the competition don't is a little pointless if you ask me.
The cycle may not get any longer, but the level of upgrade will get shallower--my prediction.
Up until this year, Canon had a long way to go to get the image quality and handling characteristics of DSLRs to match those of film SLRs, so they pretty much had to crank out improved models as quickly as possible to keep the "digital locomotive" from stalling. They got there across the board with the release of the 350XT. All of their DSLRs are equal or better than 35mm and equal to each other up to 11x14, with the top models being good enough to take an art director's cropping. They all now handle quick as a bunny and have long enough battery life to be turned on in the morning and turned off at night--yet are instantly responsive.
If I had owned a D30, I'd surely have upgraded to a D60, and then to a 10D--the earlier models simply weren't satisfactory. The 10D was just about satisfactory image-wise, but lacked in handling quickness. From here on in, there really isn't pressure to upgrade for people who were satisfied with 35mm in the first place.
Sure, there will always be people who want to stay on whatever cutting edge is out there, but I think the next upgrades will be just maintaining step with the standards--Firewire across the board, USB 3.0, Digic III. But no real reason to for users to upgrade every version.
My view is pixel density won't change as the 20D and 350 already have the smallest photosite/pixels in the Canon dSLR line up - if I have read the info correctly not all of the area of a CMOS photosite is available for 'light' collection some of the area has the circuitry so on a smaller site the relative proportion of this area of the circuit becomes larger.
LIkely improvements in my opinion will be a range of cameras (aka film bodies) with improvments across the range on fps speed, focus, ruggedness etc. So you can chose if you want the entry level or the Eos 3 equivalent but the number of pixels will be the same for a given sensor size.
That said, the 20D and 350D are pretty fantastic and there's probably no real need for anybody who buys one to upgrade anytime in the next 4 or 5 years, if they are happy with 35mm style photography.
JVasek wrote:
My view is pixel density won't change as the 20D and 350 already have the smallest photosite/pixels in the Canon dSLR line up - if I have read the info correctly not all of the area of a CMOS photosite is available for 'light' collection some of the area has the circuitry so on a smaller site the relative proportion of this area of the circuit becomes larger.
Mr.Iwashita: CMOS has both advantages and disadvantages. The first advantage is superior power consumption. Secondly, although CMOS is known to be prone to image noise problems, we have fixed that. But the third point is that with CMOS, each pixel site on the sensor is surrounded by a circuit. The width of this circuit can not be smaller than a certain size. This means that even if the pixel size is reduced, the area required cannot currently be reduced. So it is not yet possible to make very small, high quality, CMOS sensors for compact digital sensors.
JVasek wrote:
My view is pixel density won't change as the 20D and 350 already have the smallest photosite/pixels in the Canon dSLR line up - if I have read the info correctly not all of the area of a CMOS photosite is available for 'light' collection some of the area has the circuitry so on a smaller site the relative proportion of this area of the circuit becomes larger.
LIkely improvements in my opinion will be a range of cameras (aka film bodies) with improvments across the range on fps speed, focus, ruggedness etc. So you can chose if you want the entry level or the Eos 3 equivalent but the number of pixels will be the same for a given sensor size....Show more →
I must disagree here. The name of the game in the collective mind of the camera-buying public is "megapixels." Canon knows this. They absolutely must keep increasing the number of megapixels to continue dominating the market. Joe consumer does not know CCD from CMOS from anything else. They only know megapixels.
The 20D replacement will be at least 10 megapixels, perhaps 12. No doubt whatsoever on that. Whatever technical limitations exist will be worked around, perhaps to the detriment of pixel-by-pixel image quality. But it will happen.
Mr.Iwashita: CMOS has both advantages and disadvantages. The first advantage is superior power consumption. Secondly, although CMOS is known to be prone to image noise problems, we have fixed that. But the third point is that with CMOS, each pixel site on the sensor is surrounded by a circuit. The width of this circuit can not be smaller than a certain size. This means that even if the pixel size is reduced, the area required cannot currently be reduced. So it is not yet possible to make very small, high quality, CMOS sensors for compact digital sensors.
Um, can't they just put the circuitry under the sensor array? I mean circuitry is circuitry, and can be moved at the expense of nothing other than increased resistance (which means more heat, which means more battery consumption). But dropping the circuitry to below the sensors should only increase resistance marginally, I would think. Perhaps there's somthing about CMOS circuitry which makes it more finickey than others, who knows.
Besides, I'm sure people didn't think they could break the GHz barrier with the size chips that we currently used - becuase circuits took up a finite amount of room. Well, they kept and still keep getting smaller.
Yep, they'll find a way to increase the MP of pretty much all replacement cameras in a given line.
Russ,
I also agree that the MP drive will continue. For the record, I have been predicting that we will see 100MP DLSRs. It has little to do with what we need. It has much to do with the Semicon Fab's desire to reduce the cost. And of course, it also aligns well with what Marketing wants
Jon,
I may be wrong but it probably is difficult to fabricate layered sensors as you describe. However, I agree that the peripheral circuits in each pixel site will be reduced in size. I don't understand why its size has limit.
Multiple layer set integrated circuits are extremely expensive (ICs have multiple layers, but forming one set for circuitry). So to be able to layer transistors over transistors in order to put the amplifiers / collectors behind the sensor will not be cost effective. You can't do it on two chips either because the circuitry needs to be right there with the pixel sensor.
Going from 8 to 10, 11, or even 12 MP may well be possible because when you take the square root of the area increment, the linear decrease in device line width is not that great. Going from 8 to 11 MP is just a 15 percent linear reduction.
