When will full fram DSLR go mainstream?

[peter_phan]

<i>Field-of-view and DOF are close enough IMO that the differences aren't worth fussing over. Perspective is a function of subject-to-camera distance and is influenced by field-of-view rather than focal length. What's left? Weight, I guess. A 35mm f/1.4 is typically bulkier than a fast 50mm. This is worth considering. But reduced-image-circle lenses like the upcoming Sigma 30mm f/1.4 should take care of it.</i>

<p>

The Sigma 30/1.4 on a 20D, shot at f/1.4, focused at 6 feet, will have about twice the depth of field as a 50/1.4 shot at the same aperture and same distance on a full frame SLR. That's a pretty big difference and definitely has an impact on how your images are rendered. Try the calculations here: http://www.dofmaster.com/dofjs.html

<p>

I have the 35/1.4L. Yes, it's definitely bigger than the 50/1.4, but it balances very well with a 20D body. But, alas, it still doesn't have quite the same look as a 50/1.4 shot wide open. The 50/1.4 on a 35mm camera still does a better job with shallower depth of field and subject isolation. And the Sigma 30/1.4, with its shorter focal length, has even more depth of field than the 35/1.4L.

[mark u]

IMHO there are a couple of things to bear in mind over the full frame future:

 

Full frame becomes an economic proposition when the added cost of the larger sensor is less than the added cost of a (good) wide angle lens for APS crop cameras. The argument that the crop cameras have added reach with telephoto lenses becomes irrelevant, provided you have adequate pixel density to crop - the Nikon D2X shows the way, despite itself being a crop camera.

 

A full frame sensor will never be less than 3-4x the price of an APS one simply on grounds of relative area, but the cost of APS sensors (and full frame ones) has fallen dramatically over time. The higher cost of full frame sensors is dominated by lower yield factors. This is partly an issue of silicon wafer quality, and partly of the state of development of mask printing technology for larger chips. There will be an increasing drive to make improvements in both areas across the wider chip industry as computer chips are now moving toward multiple CPU, larger area chips since limitations are being reached with reducing the size of chip features.

 

Looking at the state of the market may help to provide some pointers:

 

http://www.pmai.org/mktrsrch/mrweb/pi2005.pdf

 

We have reached about 80% of cameras being sold being digital. However, images actually printed from film have only declined a little over 25% since 2000 when digital was a negligible minority sport, and is projected to be about 40% down on 2000 in 2005 (see p.7). The process of digital substitution still has a long way to go in terms of actual useage of cameras. Household penetration of digital cameras is around 40% and projected at about 50% by the end of 2005 in the US. Since sales are projected at over 20 million cameras, this implies almost half will be digital upgrade/2nd camera sales (there are a little over 100 million US households). Looking forwards, that proportion is likely to increase - at least provided manufacturers can offer an upgrade proposition.

 

The megapixel wars are beginning to reach an end point: for most there is little to be gained from a mere increase in resolution. This means that camera companies will have to turn to other selling propositions. Part of that will be ease of use and software. However, the limitations of small sensor size in terms of action freezing/low light higher ISOs and narrow dynamic range - now largely constrained by the physics of light rather than technological limitations - will offer a way forward. ISO 50 and image stabilisation doesn't cut it.

 

The biggest limitation on form factor for digital is the need for the lens to project the image almost at right angles to the sensor: this sets the lens to sensor size ratio. It is why we aren't going to see even APS sensors in a camera as compact as a (full frame) Olympus P&S film model any time soon. The form factor advantage for APS over full frame effectively disappears if you don't need to carry a large retrofocus extra wide angle lens and you tolerate a cropped image at the telephoto end (similar to that you'd get from the same lens on an APS camera). Meantime, full frame offers the prospect of greater DoF control and enhanced dynamic range (or resolution if you prefer that) as a consequence of gathering more light. It's interesting to note that APS never really made it as a film format despite some delay before the emulsions specially developed for it effectively became available to 35mm shooters.

 

We must also remember that the DSLR market is in its infancy - for most SLR shooters, just the cost of a DSLR body represents several years' worth of D&P costs. The economics look even worse if it precipitates a computer upgrade. For these lower volume shooters, joining the DSLR rat race doesn't make sense until they can see that the technology has matured and prices have come down. A sure sign of that would be affordable full frame DSLRs.

[mark u]

I should have added that another important benefit of full frame is a rather more useable viewfinder. The poky view through a 300D decided me in 30 seconds that it was not a camera for me.

[david_kieltyka1]

>> The Sigma 30/1.4 on a 20D, shot at f/1.4, focused at 6 feet, will have about twice the depth of field as a 50/1.4 shot at the same aperture and same distance on a full frame SLR. That's a pretty big difference and definitely has an impact on how your images are rendered. <<

 

Note that I was refering to a 35mm f/1.4, not the new Sigma 30mm. I mentioned the 30mm as a lens scaled down somewhat in size for the digital APS format. Even so, twice the DOF with the 30mm is one stop...IMO still not a big deal. When you go from medium format to 35mm you have the same issue: increased DOF for a given field-of-view. It didn't stop widespread acceptance of the 35mm format and isn't likely to impede digital APS either.

 

Note also I'm in no way knocking the 35mm format. It has and will continue to have its place in the digital world, just as various film formats have co-existed for many decades. But that place isn't, and I suspect won't be, in consumer-level cameras.

 

-Dave-

[ken_jeanette]

Film had it's time of prevalence, and during that time, there was constant tweaking of the image size. Film started very large, and became smaller, and smaller until it reached a point of diminished return that was not acceptable, not even to the snapshooter. Hence, the demise of APC, 110, 16mm and disc. Digital on the other hand has taken the opposite route, starting with tiny sensors, and working their way up as technology and reduced production costs permit. Eventually, the market will dictate what is the optimum size for a sensor, or at least what is the smallest that will be tolerated in mainstream use. Film took the better part of 100 years to discover its absolutes, Digital will take less, but it will be a few years before the parameters are assembled and cemented.

[patricks]

I guess I'm framing my initial question the wrong way, it should rather be "When will the manufacturers start producing fast primes made for APS sized sensors?"

 

I like working with small, compact, well made and high quality primes with apertures towards f/1.4, both for their low light capabilities and to be able to isolate subjects/blur the background.

 

If Canon/Nikon would produce APS-sensored top quallity primes I guess my issues would all be solved...

[chiswick_john]

""When will the manufacturers start producing fast primes made for APS sized sensors?" - sigma have just anounced a 30mm 1.4 prime - more are likely to follow. Many other independent lens makers are producing zooms for DSLR's with smaller than 24x36 image areas. Soon there will be enough choice of lenses for people to choose and debate over and the whole 1.6 vs full frame debate will thankfuly fade away.

[min_wong1]

5-10 years.

[tom_karlo]

The simple answer here is Nikon and Canon will start making prime lenses for APS sized sensors when it's worthwhile financially. Right now, except for the wide end of the range, the marginal size improvement isn't enough to justify a whole seperate SKU and the risk and cost it entails. A normal 50mm 1.4 will work just as well as an APS 50 1.4 and is good for both groups -- there isn't really that much marginal value for the APS version, and you can't charge more since the APS user can always go buy the old one.

 

Zooms, especially $1000 ones like the Nikon 12-24, have reached a point where the only thing you really lose going from the prime to the zoom is a stop or two of speed (maybe) and marginal quality not noticeable to 95% of buyers. If you're Nikon or Canon, you're going to make the single wide APS zoom, route all the buying volume to that model, and maximize your returns.

 

I'm very interested in the Sigma 30 f1.4, but I think the reason they can serve that specialty need is by serving 3 different camera mounts and spreading the cost of the design across a wider market. That's not an option for Nikon or Canon, and it detracts from the sale of the very profitable wide zooms.

 

Moving to full sensor? I don't think it will ever happen for consumer dSLRs. Your average dSLR buyer (and I mean the middle 75%) just don't see this as a problem as long as they're getting a 17mm kit lens, which is wider than the standard 35mm you used to see on film "FF" models. That's "too wide" for most people. The marginal value of promoting a more expensive to build FF model in the sub-$1500 category, versus simply reducing the cost and price (and size) while keeping the APS sensor, is the logical route for manufacturers targeting the consumer (sub-$2000) market.