Why did Canon come up with APS-H and why did the top-of-the-line 1D's *still* use it and 1.3x crop, when FF existed for four years (the 5D)?

Question

With the 1.5/1.6X (APS-C/DX) crop factors and full frame just about ubiquitous these days in the digital world -- why hadn't Canon's top-of-the-line 1D's moved to FF sooner? It took four years!

Is there anything special about the 1.3x crop factor? Or is Canon afraid of causing old 1D users who are upgrading to the latest Mark XX to have to re-adapt?

Or is there really an advantage?

You take a look at the Reuters photoblogs, they seem to use 1D's a lot.

Is it an accident of history, or is there a practical reason here? I would be an interested to see an answer from a late-model 1D-user!

Answer 1

When Canon released the first 1D, APS-H was simply the largest sensor they could get away with, economically. They followed it up with the 1Ds which was full frame. However the 1Ds was slower than the 1D, and offered less reach with telephoto lenses, so was less popular with sports and wildlife photographers. For this reason Canon chose to continue offering a faster, lower resolution 1.3× crop body in the 1D line.

For whatever reason, Canon decided to end the 1D line with the introduction of the 1DX. They were able to exceed the speed of the last 1D whilst improving resolution (slightly) and offering a full frame sensor. Even so the decision wasn't universally popular, especially with photographers that appreciated the extra reach from the 1D crop factor.

Would be an interested to see an answer from a late-model 1D-user!

I had a 1DmkIV for 2 years. I was a 5D user but was never fully satisfied with either the speed or more importantly the autofocus with the outer points. I waited for the 5DmkII to come out and when it did I was very disappointed that they addressed neither the speed or the AF performance.

The 1DmkIII was a 2MP upgrade from its 8MP predecessor, whereas the 1DIV jumped to 16MP. This prompted me to switch from full frame to APS-H, as I could still get large prints if I needed to. I found no fault with the camera in terms of its AF, speed, or features, but I had to convince myself I could live with the crop factor.

However over time I began to find the crop factor annoying. Ultra wide angle lens options were very limited, I ended up using the EF-S 10-22 which I modified to fit the EF mount and was usable without vignetting from 13-22mm. The only problem with this was that the mirror would hit the back of the lens if I accidentally zoomed right out. I also missed the 85mm FOV, my 85mm was a bit too long and my 50mm was a bit too wide.

In summary I don't think there's anything magic about the 1.3× crop, unless you need the reach I think it can be quite annoying.

When the 5DmkIII came out with top-of-the-line AF and 6.5 fps, the decision became clear and I sold the 1DmkIV. Fortunately they had become sought after since their discontinuation so I sold it for a fraction less that I had paid for it!

Answer 2

Everything in product design is a comprise and Canon wanted to provide a solution to combine high-quality and high-speed for sports photographer. It did so with the 1D series. Its APS-H sensor and relatively large pixels make it sensitive to light and possible to shoot at high-speeds, up to 10 FPS with the 1D Mark IV. At the same time, the full-frame 1Ds Mark III stopped at 5 FPS.

As you know, there are no more APS-H camera in production. The 1D and 1Ds lines were fused with the introduction of the 1D X which brought high-speed to full-frame. The resolution was only a little down compared to the 1Ds Mark III (18 vs 21 MP) but the frame-rate exceeded the 1Ds Mark IV (12 vs 10 FPS).

The reason I suspect that the APS-H line was ended, is to reduce complexity. It is something less to support and they wont have a system which lacks ultra-wide-angle coverage from lenses.

Answer 3

If you put the same lens at the same aperture on a full frame and APS-H camera, you will need to get closer to the subject with the full frame to have the subject the same size. As such, the depth of field in the resultant image will be shallower.

If you take both shots from the spot the depth of field will be identical, but the subject will be smaller on the full frame.

This APS-H sensor was put there on purpose because it was made to be a sports camera and for fast moving objects that's why it has 40 tracking points . using a crop sensor allowed the user to be closer to the action with the same mm lens with out moving closer to the subject and not having to use a bigger lens that used a lower f stop like 4.0 instead of 2.8 . Canon wanted to keep as much light coming in as possible but also keep the iso down to keep the noise out and still have a fast aperture and higher shutter speed to make moving objects clear with our blur and be able to see the faces of the players on the football field from the sidelines with out loosing a lot of light . So by in camera sensor cropping to 1.3 the photos came out clear and sharper than they did when they used a program to do the cropping on the full frame cameras to get the same closeness as the crop sensor .

Answer 4

Your question asks, "Why did Canon come up with APS-H and why did the top-of-the-line 1D's still use it and 1.3x crop, when FF existed for four years (the 5D)?"

1) Canon did not come up with the APS-H size format.

2) Only half of the top of the line 1D series still used the 1.3X crop format for seven years after the 5D was introduced in 2005, the other half of the line, the 1Ds had been using full frame sensors since 2002.

The 1D line actually had two series: The 1D series that used APS-H sized sensors and the Full Frame 1Ds series. They were both available in various successive models between 2002 and 2012 when the 1D X supplanted both the 1D mkIV and the 1Ds mkIII. The original 1Ds was introduced in 2002, three years before the 5D was the first mid-line full frame body offered by Canon. So half of the "top of the line 1D cameras" were already using full frame three years before the 5D was introduced, not four years after.

As Stan Rogers mentioned in a comment to the question, APS-H was one of several sizes that images could be formatted in on the Advanced Photography System film that was supposed to eventually replace 35mm film. The reason APS-H was used in the earliest Pro-level digital bodies by Canon was probably due to the limits of available processors at the time and the fact that Canon probably had some designs in the pipeline that were optimized for APS-H sized film. By using a lower resolution sensor they could make the camera's frame rate faster and squeeze more photos onto much smaller memory cards than we use today. By making that sensor smaller than 35mm film, it also gave the photographer more reach for any given lens in exchange for the lower resolution. Keep in mind that when the original APS-H 1D was introduced it was still a huge sensor compared to the sensors in most digital cameras at the time.

Most photojournalists chose the APS-H 1D models. In photo journalism resolution has not been the prime concern, especially when it was still primarily a newsprint industry. Newsprint photos are very low resolution compared to even web sized photos. Getting the shot using gear that can survive abuse under the most demanding conditions was the top feature desired by photojournalists. The fast handling and longer reach of the 1.3x APS-H 1D series was preferred by many photojournalists over the higher quality images of the slower 1Ds series.

Many studio photographers used the Full Frame 1Ds because they considered the image quality of the FF sensor to be more important than the faster handling speed allowed by the APS-H sensor.

In addition to newspaper photographers and other photojournalists, wildlife specialists also enjoyed the 1.3x focal length factor and the faster frame rate of the 1D series versus the 1Ds series. They were probably the biggest group that was none too happy when the 18MP 1D X became the replacement for the 16MP 1D mkIV. By the time you crop the 18MP of the 1D X to APS-H size to regain the lost reach, you are left with roughly 15-20% fewer pixels than the 1D Mark IV had. This coupled with Canon's decision to turn off all AF on the 1D X when lens/extender combos with f/8 as the widest possible aperture were attached to the camera was met with howls of protest from the wildlife/birding community. Canon relented and released a firmware revision that allows AF to function on the center focus point of the 1D X at f/8.

Answer 5

The historical reason for this is the nature and cost of manufacturing full-frame sensors, given the technology available at the time the EOS-1D was first introduced. APS-H persisted even after high-speed full-frame cameras became viable likely because of user familiarity.

Pages 11-12 of an old Canon whitepaper detailing their full-frame sensor technology describes the economics of manufacturing image sensors of different sizes. Image sensors are semiconductor devices, like the CPU in a computer or the SoC in a smartphone or tablet, and are manufactured in batches by the wafer. The cost of manufacturing any semiconductor device increases dramatically as the size of the die (chip) increases. Full-frame image sensors are extremely large compared to the typical chip, which has several significant implications that make them far more expensive to manufacture than smaller sensors:

• Fewer dies can fit onto the silicon wafer. Obviously, this means that fewer full-frame sensors than APS-C or APS-H sensors can be made in each batch, which dramatically drives up cost. (200 APS-C sensors will fit on a 8" wafer but just 20 full-frame sensors will fit in the same space.)
• A larger portion of the wafer is wasted. More of the space near the edges of the wafer cannot be used due to the larger size of the die; each die on the wafer must be completely intact, leaving the areas on the edge of the wafer unusable.
• The manufacturing process is more sensitive to defects, necessitating higher precision. Defect densities that would render only a small portion of a wafer with smaller chips unusable can potentially ruin an entire wafer of full-frame sensors:

  Consider, too, that an 8" silicon wafer usually yields 1000 to 2000 LSI (Large-Scale Integrated) circuits. If, say, 20 areas have defects, such as dust or scratches, up to 1980 usable chips remain. With 20 large sensors on a wafer, each sensor is an easy “target.” Damage anywhere ruins the whole sensor. 20 randomly distributed dust and scratch marks could ruin the whole batch. This means that the handling of full-frame sensors during manufacture needs to be obsessively precise, and therefore they are more expensive.

• Three photolithography passes were required to manufacture a full-frame sensor, which further increases manufacturing cost. It so happened that APS-H was the largest sensor size Canon could make with a single photolithographic exposure, which alone made them much less expensive to manufacture than full-frame sensors.

Therefore, the cost of manufacturing a full-frame sensor (and associated mechanical parts like the mirror and shutter) capable of operating at the speeds needed for sports photography would probably have been prohibitive at the time the original EOS-1D was released. Nikon did not have a full-frame sports camera until 2007, when the D3 was announced. Canon did not release a high-speed, sports-oriented full-frame EOS-1D camera until 2012 likely because of user familiarity with the APS-H sensor size, instead waiting until its userbase was ready for a major, breaking change.

• Bear in mind that EOS-1 cameras are professional tools, not merely picture-taking devices. Press and sports photographers who use these cameras daily need to be able to work the controls quickly and consistently with minimal fuss, so even slight changes to the camera's UI/UX can be problematic. This is evidenced by the control layout remaining nearly unchanged over all four generations of the EOS-1D prior to the EOS-1D X.

• The full-frame EOS-1Ds models were meant for studio and on-location photography (e.g. weddings). The EOS 5D Mark III has since inherited many of the capabilities of the EOS-1Ds series, including high-resolution sensor, tougher construction and weather sealing, and high-density AF system, rendering the EOS-1Ds series mostly redundant. It is probably no coincidence that the EOS-1D X was announced at nearly the same time as the EOS 5D Mark III.

Note that advances in semiconductor process technology has eliminated the need for three exposures, making less expensive full-frame cameras viable, but the other factors remain true. It is and will always be more expensive to manufacture a full-frame sensor then an APS-C or APS-H sensor.

Answer 6

Canon and Eastman Kodak worked together to build the famous 1D what no one seems to understand in my reading the responses is that the 1D sensor is a (CCD) sensor unique to this last partnership with Kodak. The CCD is a higher quality and expensive sensor and is approximately equal to 10 MP at 4.5 MP as to resolving power. The CCD is used by NASA and for high end commercial settings. This is why the 1D not the 1D2 is a prized camera in good condition. The later 1D line - 1D-2- 1D MKIII, 1D MIV - 1DX use (CMOS) sensors.