NEW YORK – You might have just received the latest, dazzling digital camera as a holiday gift. I don’t want to spoil your fun, but predicting the “new, improved” -- and much cheaper -- version that’s almost certain to come down the pike in the next six to 12 months (a cycle of obsolescence that defines the consumer electronics business) is what this reporter is paid to do.
Recent conversations with digital camera experts offers helpful insight into what OEMs and chip vendors are jockeying to establish as the battleground for the digital still camera (DSC) market in 2011.
Five things to watch for in the new DSCs next year are:
1. Full HD video capture (already common in 2010)
2. 3-D video (the jury is still out but it may be inevitable)
3. A new-generation video codec (enabling a substantially smaller file size in a memory card)
4. Compact camera body with abilities to switch lenses (offering Single Lens Reflex camera-like performance)
5. Delivery of smooth, steady video under all conditions (i.e. bad lighting, CMOS rolling shutter effects)
The DSC market will become an even more cutthroat in 2011, with some camera manufacturers fighting literally for survival.
Market research companies are already predicting the decline of the DSC market in just a few years. “The point-and-shoot digital still camera—once a red-hot consumer item and a best-selling retail mainstay—has about three good years left before its shipments begin to decline, supplanted by newer technologies and the omnipresent camera in cell phones,” stated iSuppli market researchers in their recent press release.
With the quality of cell phone cameras growing dramatically better, iSuppli’s prediction is not so surprising. But for now, DSC manufacturers’ job is still to pick a spot, find a growth opportunity, flesh it out and define it a new fault line.
DSCs in 2010 have overwhelmingly gone video – or with a “hybrid” of still and video. In fact, 100 percent of DSCs sold today offer video shooting capabilities, according to Senya Pertsel, senior director of marketing at Zoran Corp.’s mobile division. But each DSC’s video recording capability, in quality and performance, substantially varies.
The biggest push DSC vendors are planning in 2011 is “Full HD” video recording. Hybrid cameras – integrating HD video recording and still-camera capability into a single device – will become more popular and affordable, according to iSuppli, “especially as advancements are made in silicon processing capability and as prices for flash storage decline.”
The market research firm predicts shipments of hybrid cameras to grow from “8.3 million units in 2009—representing about 7.6 percent of total camera units shipped—to 120 million units in 2014—accounting for about 89 percent of total cameras shipped.” With hybrid camera models already available from Eastman Kodak, Canon, Nikon, Samsung and Sony, etc., iSuppli said that “hybrid cameras at the $150-$200 price point represent the next big growth opportunity in the industry.”
The chip vendor last month (November, 2010) announced a new 3-D video pre-processor called S3D, designed to work with Ambarella’s camera SoCs, by pitching that the combination will enable full HD 1080p 3-D video recording and high-resolution 3-D photography. The goal is to enable a consumer-friendly 3-D digital video camera – priced at less than $200 – capable of shooting 3-D video “that looks really good on 3-D TV,” said Chris Day, vice president, marketing and business development at Ambarella.
Consensus on this topic seems to be more than on some of the other articles. Perhaps that's because no one here cares to argue with the fundamental limitations of getting light onto a sensor? ;-) But to my mind, packing endless features and gizmos into a cell phone is like building a Swiss army knife: lots of functions; the thing is rather too fat; and none of the functions are all that good compared to standalone functions. On the other hand, Swiss Army knives are pretty popular among folks who want breath rather than depth. So we are likely to see other goodies in the electronic grab bag as time goes on. FWIW: when I want to make a photograph, I use a Graflex 4x5 camera, and scan the transparency. Resolution to die for. Fast, too: I can make my first exposure in about 10 minutes from the time I show up with my 16 pounds of equipment, and take another photo about ever minute thereafter. Until I run out of the $4/shot film. ;-)
In the high-end amateur and professional markets, we'll see some amazingly high quality cameras. In those segments, users will look beyond the simple megapixels marketing number and buy on resolution, dynamic range, SNR, speed and optical characteristics, and the manufacturers will keep improving in all of those areas.
For the point and shoot consumer, one only has to look at the past popularity of the 110 and 126 film cameras to get a strong feeling that phone cameras are already pretty close to what will be mass-acceptable.
My 3MPixel phone camera isn't quite up to 110 film standards in all respects yet, but it's pretty close in most areas. The real key to the masses is a minimum quality level (fairly low by my standards) coupled with extreme ease of use.
I suppose we could debate the question of what is the mass-acceptable feature and quality set for quite a while. Time will tell.
J_Alan you hit it right on the nail!
Physics will be the underlying limitation to how good cameras can get. Whether it is shot noise, limitations in optical resolution, etc.
I don't see the Camcorder market imploding completely, though I expect it will shrink. My 10x lens with HD recording (and I can even zoom while recording) allows me to do most of what I need with a camcorder. MOST! ... Still cameras have far greater concerns w.r.t. distortion, vignetting and chromatic aberation versus a motion camera.
That said, I expect in camera processing to remove many of the optical irregularities to become standard. That would be something I could see in the short term and already happens on DSLRS to a degree.
I expect, though not this year, to see some optical tricks to improve resolving capability with small pixel sensors. It has not happened yet, but this will be a focus.
The comment about point and shoot cameras dieing sooner or later does not take reality into account. There are some basic physical constraints that cannot be avoided:
- Read noise can be reduced significantly and there are improvements recently that have resulted in CMOS sensors with significantly less read noise at high speed. That helps with noise in low light .... to a point.
- Shot noise is something that many who do not work with sensors are unaware of. Basically when you convert from photons to electrons the resultant noise is a function of the square root of the signal level. Currently there is no way around this.
- Hence, the only way to get truly clean pictures is with more light which means big lenses ... currently impossible with a cell phone.
Our ability to cram megapixels onto ever-smaller sensors has far outstripped the capabilities of the optics in front of the imager. A 10-megapixel sensor at DX (16x24mm, roughly) size is basically at the limit of the resolution of affordable lenses, especially zoom lenses. A 10-Mpix sensor that is micro 4/3 or smaller is well beyond what the optics will lay down in terms of image quality. You're kidding yourself if you think that small-sensor cameras, regardless of pixel count, will equal the image quality out of a DX or fullframe (24x36mm) sensor-based camera.
Granted, most point'n'shoot users don't care but there are those of us who grew up with real cameras and demand similar overall image quality out of our digital cameras. Nothing can substitute for original image size. Why else did Ansel Adams shoot 8x10 and larger negs?
I just wish I could afford a digital back for my old RB67 (60x70mm).
On the flip side of this discussion, my other objection to the "Swiss Army Knife" approach to cell phones that do everything is that I want a PHONE, not a gizmo that tries to do everything for me. Multi-purpose tools inevitabl comprome aspects of every function they try to cram in. Pretty soon, they're capable of doing everything poorly...and nothing well...
Both the size of the sensors and the quality of the optics make phone cameras and to a somewhat lesser extent point and shoots vastly inferior to current DSLRs. Not quantity of pixels. Light sensitivity and sharpness. Iphones and EVO's are about as good as it gets now and they are still just not up to task. I'd like to see a phone maker get serious the camera..
Maybe the most basic function that will need to improve in order to handle all of these technology wants/needs is the battery. Most smartphones currently have to be recharged daily (or more often) using only the basic function already present. All of these new technologies and improvements will add to the power drain.
Agreed. Despite cell phone camera improvements, the DSC camera features, processing power and ease of use just aren't there in any phone camera -- and probably won't be added any time soon, for cost reasons.
The phone camera is the one you always have with you when you simply need to take a picture, but it's not the camera you take with you when you plan to do some actual photography.
I also agree with your comment about CODECs. H.264 does quite a nice job of compressing 1080p and flash is cheap. H.264 is nowhere close to running out of steam.
I wonder if this will end the era of Digital Camcorders. so far, I believe makers like Sony have kept the features in DSC and camcorders such that they can sell both types. I guess DSC's getting better video and optical zoom capabilities is going to be inevitable and will eventually kill camcorders market. Camcorders are still on the expensive side maybe because there are only 2-3 major players there while DSC market is crowded leading to stiff competition and price war. Overall it is going to benefit the consumer.
Join our online Radio Show on Friday 11th July starting at 2:00pm Eastern, when EETimes editor of all things fun and interesting, Max Maxfield, and embedded systems expert, Jack Ganssle, will debate as to just what is, and is not, and embedded system.