My Photo

Ordering Information

Tomi on Twitter is @tomiahonen

  • Follow Tomi on Twitter as @tomiahonen
    Follow Tomi's Twitterfloods on all matters mobile, tech and media. Tomi has over 8,000 followers and was rated by Forbes as the most influential writer on mobile related topics

Book Tomi T Ahonen to Speak at Your Event

  • Contact Tomi T Ahonen for Speaking and Consulting Events
    Please write email to tomi (at) tomiahonen (dot) com and indicate "Speaking Event" or "Consulting Work" or "Expert Witness" or whatever type of work you would like to offer. Tomi works regularly on all continents

Tomi on Video including his TED Talk

  • Tomi on Video including his TED Talk
    See Tomi on video from several recent keynote presentations and interviews, including his TED Talk in Hong Kong about Augmented Reality as the 8th Mass Media

Subscribe


Blog powered by Typepad

« Understanding Internet Access: How Much is Mobile, What Kind of Mobile; and How Much is Access by PC or by Tablet | Main | Filling In the Missing Bits: Kantar Numbers and World Smartphone Market »

March 14, 2017

Comments

Thomas Wright

1) PEAK CPU: We Hit "Peak CPU speeds" for production cpu's about 10 years ago with attempts to hit mid 3ghz and low 4Ghz, and now instead are focused on getting ram bus speeds to close in on the mid 2ghz speeds of most CPU's, and using SSD's and other optimizations.

2)PEAK Highway speeds: In the US, except in a few rare sparsely populated areas, most peak Interstate highway speeds have settled on around 75 mph.

3)Peak Retail store sizes: The advance of the superstores, warehouse stores, supermalls, and the like has basically stalled, and even places like walmart are experimenting with more small "neighborhood" grocery size and even convenience store type stores.

4) Peak Tiny/huge/wide smartphones: Tiny: There isn't really a market for phones the size of matchboxes, and Huge: larger phones seem to be bumping up against an upper limit, and width, my galaxy note 4 is narrower than my galaxy note 1, although it is longer approaching a form factor more akin to the classic and very old CHECKBOOK WALLET, rather than say that of a paperback book.

5. Peak Jambox: Some of us remember years ago when consumer electronics hit a "concord moment" with the increasing size of portable "jamboxes", and NOW when we complain of rude and antisocial youth it is fortunately NOT that they are loud and disturbing an entire neighborhood with the bass from a 50 watt box on their shoulder, but in being rather quiet and withdrawn, lost in a QUIET interaction with a smartphone while wearing earbuds...

E.Casais

There are other cases of IT technology that hit their peak performance recently.

A relevant one, but very subdued, is peak rpm for hard disk drives. 15000 rpm for server-grade disks and 7200 rpm for consumer-grade ones were reached years ago, and they stopped increasing ever since. Actually, newer consumer-grade HDD are often returning to slightly slower rotation speeds.

The reason is simple: HDD are now the components of choice for affordable large secondary storage capacity, whereas SSD is the technology of choice for fast, but more expensive, online storage. The performance objectives and how to fulfil them have shifted.

The peak performance in aviation hides the fact that other requirements came into play:

a) fuel consumption for commercial airliners; supersonic aeroplanes consumed _way_ too much fuel. And the various oil crises struck precisely just when those planes started flying...

b) new escape techniques for military aircraft; when 1970-era anti-aircraft missiles could already fly at mach 2.7 and current ones at mach 6, over-engineering a plane to fly at mach 3 is pointless -- better invest in electronic counter-measures.

I suggest that two trends have killed (or at least suspended) the evolution to higher-end cameras:

a) The craze for thin mobile phones -- which prevent advanced optics.

b) The lowering margins in the business, which means that manufacturers prefer to rely upon cheap, commodity hardware components -- rather than the expensive, smaller-batch premium optics, or the bespoke image processing electronics used in the Nokia 808.

Things may well change if AR takes up and brings genuinely innovative applications.

By the way: Apple did not introduce the dual-camera setup for variable zooming with the iPhone 7+; it was LG, with the G5 model released months before.

A Giz

Also the current Moto Z + the Hasselblad 10x zoom optional part. 41 Mpx Nokia might be the max but 10x zoom is still repeated.

Wayne Brady

As an avid and former professional wedding photographer - and owner of the new iPhone 7+ with dual cameras...

I would say we most definitely have NOT reached the Concord moment. I never had the 1020, but this iPhone camera is the first I've ever used that made me think "DSLR quality" about the images I'm creating. It has a "normal" lens -- meaning "the perspective your eye sees" which is what a 50mm lens gives you on a full frame camera. And while even longer (85mm to 100mm or so) are desired for a portrait lens...the "normal" view on the iPhone 7+ is a HUGE upgrade for portraits of people. It's not about megapixels or "zoom", it's about the different geometry a longer lens has on close ups of faces. No one shoots such photos with a wide angle lens in the real camera world, and that's all we've had up until now for smartphones (minus a few science experiments that never gained market acceptance and were basically cameras that had a phone glued on).

Realistic shallow DOF is the other missing ingredient from smartphones. With their tiny sensors, all smartphone cameras have near infinite depth of field. So does the iPhone 7+, but with the two cameras allowing "depth mapping", Apple has designed software that does a nice job of giving you that narrow DOF, creamy "bokeh" effect that you get from a DSLR and a nice wide aperture lens.

This is Apple's first attempt at these technologies. They are already producing superior results, but there is much improvement available. When in "portrait mode" (which turns on that bokeh effect), the camera performance is sluggish. There is room for better and smarter bokeh effect. While the results currently are pleasing, you can easily tell the difference with what a DSLR/wide aperture lens would have created.

There is room for longer optical zoom. There is room for more realistic bokeh. There is room for faster performance. There is room for more sensitive sensors. The normal lens lets in less light (this is true in the DSLR world too). There is room for better real and artificial noise reduction.

The iPhone 7+ is the first phone to make me desire an external flash. I'd love to take it on a portrait shoot, but I'm having to think through how to replace my DSLR flash ability. Some sort of bluetooth trigger that would fire off one or a set of external flashes would be terrific.

I upgraded to the iPhone 7+ specifically for the improved camera. I expect that Apple will be pushing the camera envelop as the reason to get the latest iPhone for a good few years to come. As are Samsung, Google and every player in the premium segment.

It's not about the amount of megapixels. The digital camera went through and and past the "mega pixel" race as well. 2mp, 3mp, 5mp, 8mp, 12mp, 18mp...up up and away. Each new camera had to have more megapixels than the last in order to sell. But in time, the manufacturers and the customers came to view the quality of the image in terms of dynamic range and low noise...as around 12mp or so more more than plenty. MP still increase over time, but only in step with the ability to maintain or improve image quality on the other vectors.

Isceald Glede

Concorde camera or sane camera?

If a hand held device has 10 pixels/mm the dots are small enough that you cannot see them. The same effect can be caused by a billboard with .02 pixels/mm 50m away, so there is no point in having more pixels for bigger displays as the viewer will be further away. Human eyes are about 1MP, but the pixels are not distributed evenly. You need 10 pixels/mm for the part of the picture you are looking at, but you would not notice if the other parts of the picture were at a lower resolution. As you gaze can move all over the picture, the picture needs 10 pixels/mm all over.

HD video (2MP) is good enough on a display that is 192x108mm (a BIG phone). Putting 2MP on a normal phone means the pixels are still small enough with the device held uncomfortably close to your eyes. Going to 4k (8MP) on an ordinary sized phones means you cannot see the pixels without a magnifying glass. Instead of carrying a magnifying glass about so you can look at a small part of a photo, you can get the same effect by asking your phone to expand a part of the image to fill the screen.

The obvious benefit of 40MP on the camera is x20 digital zoom (picking 2MP from the middle of the picture and throwing the rest away). It is a real benefit: you can point the camera in vaguely the right direction, take a picture of a flying duck immediately without having to fiddle with the zoom then crop the image to just the interesting bits in your own time even the the duck is long gone. The problem is that all those extra pixels have several costs. Comparing a 40MP sensor to a 4MP sensor:

Each pixel in the sensor has a boundary to prevent charge leaking into neighbouring pixels (and each column of sub-pixels has a wire to control shifting the data out of the device). The width of the boundaries (and wires) depends on the manufacturing process, so the 40MP sensor will waste 10 times as much area on boundaries. Any light falling on the boundaries is lost, so the 40MP sensor needs a brighter flash to get the same image quality as a 4MP sensor. Getting the image off the sensor involves reading all the pixels on one edge, then shifting the entire image one pixel sideways and repeating. If shifting sideways is the speed limit, this will take 3.1 times longer on a 40MP sensor. If the speed limit is reading the pixels on the edge, the 40MP sensor takes 10 times as long. If you do not have a mechanical shutter, the sensor still detects light while the image is shifting. To some extent this can be fixed by image processing.

Light comes in fixed sized lumps (photons). You either get a photon, or you don't. This is fine when bright pixels have millions of photons per frame, but less good when bright pixels contain dozens. You can see the problem directly by turning out the lights. All of a sudden, the world goes becomes speckled. Each "frame" photo sensors in you eye focused on a uniformly dark wall will get either 2 or 3 photons (or occasionally 1 or 4). The result is randomly flickering speckles. A 40MP sensor suffers from this sooner than a 4MP sensor. The fix: process the image by replacing each 3x3 square of pixels by a single pixel containing the average value of the nine in the source square - or use a 4MP sensor.

In the real world, humans cannot hold their phones steady, and this becomes more and more obvious with increasing zoom. Post processing can to some extent reverse this (and also fix the image being out of focus). Extra pixels can help with the post processing, but only if even the dark pixels have plenty of photons. Just when you thought your 40MP sensor was going to do something useful you remember that human vision uses a logarithmic scale: successively brighter pixels are not 100, 110, 120, 130, ... photons, but 100, 141, 200, 282, ... photons. (About 100 shades of brightness looks smooth to a human, but you may need to put 1000 shades in to get 100 out of image processing). Photons get converted to charge by the sensor, but there is a limit to the amount of charge each square micrometer can hold. More pixels means smaller pixels and using a brighter flash or longer exposure just saturates the image.

A 40MP sensor with lots of post processing can produce a clearer 2MP image than a 4MP sensor under certain conditions. Under other conditions it _will_ produce a worse picture. You can arrange to get the better conditions with good lens, a mechanical shutter, a tripod, a bunch of appropriately placed Xenon flashes and experience. Thin phones in the shaky hands of amateur photographers almost never benefit from 40MP sensors.

E.Casais

"A 40MP sensor with lots of post processing can produce a clearer 2MP image than a 4MP sensor under certain conditions. Under other conditions it _will_ produce a worse picture. You can arrange to get the better conditions with good lens, a mechanical shutter, a tripod, a bunch of appropriately placed Xenon flashes and experience."

The designers of the Nokia 808/1020 were aware of that, since they not only endowed the phone with a 40mpx sensor, but also a Xenon flash, Carl Zeiss lens, and a mechanical shutter. In addition, the 1020 had hardware OIS.

As for experience, well, it is up to the end-user.

Pertti

@Tomi
Bugatti Veyron was not the end of top speeds. Top Gear (that you refer to) already made headlines of Koenigsegg One:1 which has top speed of 273mph compared to 268mph of Bugatti Veyron.

http://www.topgear.com/car-news/supercar/exclusive-more-koenigseggs-273mph-one1

tz

Physics time!

(cameras below)

The Concorde moment is one of cost and fuel efficiency, which is also why we aren't going to have very many electric vehicles outside of cities.

https://www.youtube.com/watch?v=n1QEj09Pe6k

You can afford to fly at 900 k/h. Even the rich have trouble going supersonic. Even the military jets, they have to carry weapons and fuel, be affordable to produce, and going to hypersonic wouldn't make them a more effective fighter (the Blackbird was high altitude recon, not a fighter or bomber and used special fuel).

The problem with mega/gigapixels is that light has a wavelength. As the pixels used in the sensor shrinks toward the wavelength of light, we first get diffraction problems, then it just doesn't work - the sensor area must be larger than the photon. And the larger, the more photons, for low-light. But you aren't going to put a two inch sensor into a phone, and the optics needed to focus onto it properly (I'm not sure if even a Fresnel lens would work).

The smaller sensors for the ultra-thin phones simply can't resolve light any better. What they could do is improve spectral resolution, so instead of GRGB, they could do IR/R/O/Y/G/B/V/UV, or some kind of mini-spectroscope. Some wouldn't be visible but for image processing could reveal things like diseases in plants.

https://infogalactic.com/info/Daguerreotype

The above were clear enough you can see tiny details with a microscope, basically the photon level, but it was on a large plate.

Phil W

I'm not a camera expert, but the point with the 41 mpx sensor in the Nokia 808 was that each of the individual pixels were the same size as the individual pixels in the Iphone of the day. That meant the sensor was huge for a smartphone and led directly to the camera hump in order to allow the optics necessary to support the sensor. The intention was not to create 41 mpx photos, but to combine the output of seven pixels into one leading to a 5mpx photo. Each of the resulting pixels had 7 times the light available that that of the equivalent Iphone, leading to much lower noise photos.

The secondary benefit was that you could do a digital zoom, and as you zoomed you maintained the 5mpx photo by reducing the number of pixels being combined until you arrived at the maximum zoom level and no pixel combining. Of course at that level the light being gathered was no better than that for equivalent smartphones.

So yes if you take a standard sensor and keep increasing the mpx count without changing the sensor size, then all you are doing is trading resolution for increased noise. But that was not what Nokia did.

asd

HMD not having Carl Zeiss lenses isn't that surprising to me. When Nokia sold D&S and it became Microsoft Mobile, Zeiss' licensing deal went with Pureview etc. We know this is probably the case because MS Mobile still published phones like Lumia 830, with Zeiss lens, after the deal was closed and MS Mobile was born.

Now, how long the Zeiss-Nokia exclusivity deal was/is? I'm not aware that the length of the deal would have been disclosed, but the original deal from 2005 was renewed in 2012. That was seven years. Was it originally a deal for seven years? Maybe 5 + option of 2, or maybe a 10 year deal and Nokia just wanted to renew it early before any competitor became too interesting for Zeiss.

As I see it there are two possibilities; 1. There is an exclusivity deal for Zeiss phone lenses that still stands and it is held by MS. 2. There is no standing license deal with anyone. The license deal has either lapsed due to no lenses being used in last few years, or the term has just recently run out. If it was 5 years starting from 2012, the deal would run out this year.

Now, if we are talking about the first situation. Would it make sense for HMD to buy the remaining license? Apparently they didn't buy the trademark for Pureview or design IPs for smartphones, so it seems that they weren't interested in spending more than the bare minimum they had to in order to get feature phones and license for Nokia trademark on them. This would fit well with what you said about the low hanging fruit. Their best markets don't care that much about Zeiss branding. If they want to get Pureview and/or remaining Zeiss license, HMD can buy it later. Maybe they even have a non-disclosed option for that. Not having the correct branding doesn't prevent them from making good cameras. Pureview and Zeiss branding most likely aren't that important for MS, but maybe they won't mind having them just in case they want to use them on the Surface phone etc. Personally, if HMD would buy something extra from MS, I think they should go for some of the design and usability related patents first, such as Fabula or swipe.


If the second case is true and the license isn't valid for what ever reason, HMD might not want to pay for a new expensive license at this point and Zeiss might not want to give it to them, at least not before they have proved themselves.

In any case, whether or not HMD is interested in good cameras, we know that Nokia is. Juha Alakarhu moved from MS back to Nokia already ages ago. It might be OZO related, but I would not be surprised if Nokia would have some phone camera tech ready for HMD to license if they so choose. Once HMD goes for the high end I think we have no reason not to expect them to have good cameras. If they don't have a best in class camera this year, they probably will in a year or two. At that point they might want to do what Nokia was known for, bring out a crazy flagship camera.

So no, I don't think lack of Zeiss branded phones at this moment is sign of a peak in camera tech. There are other reasons for that lack of branding.

Wayne Brady

@Phil - yep. And you can simply put a camera on the back of a phone, complete with telescoping zoom lens as well. But short of making phones much larger one way or the other -- one is stuck with very small sensor chips.

I think rather than putting large humps on the backs of phones, we'll see more multi-sensor arrays like the iPhone and others that are using multiple cameras. But maybe we will see more humps, who knows. Apple certainly caved and put a camera hump on iPhones...not a huge one, but the phone back is no longer flat.

I think the Oppo 5x "periscope" technique will be interesting. Apple has a patent on a similar idea. Bend the light 90% and you can increase the lens focal point without extending the lens out of the phone.

There will be a point of diminishing returns for a "hand holdable" magnification level. I have an 8x lens attachment for my iPhone. It's practically unusable without a tripod. And if I'm going to carry a tripod, I might as well carry my real camera.

That's one of the limits of phone camera's....there is only so much you can do before someone decides to just carry a real camera (point and shoot or DSLR).

Apple will attack the problem with a coordinated, integrated set of chips, software and optics. The amount of processing to get the fake bokeh is tremendous. We have been in the dark part of the year where I live. Can't wait for Spring and summer light to come about. But for now, I can really feel the camera struggle to perform those billions of operations per second to give that effect.

The "telephoto" camera on the iPhone 7+ is not image stabilized, so hopefully the next one will be.

I'd also like to see zoom lenses that are "inside" and slide over when needed.

Phil W

@Wayne

I agree with you, the hump is not acceptable for most people, that is why the 1020 had a slightly smaller sensor and a thinner hump. The main problem with the 1020 was the lack of a dedicated processor to handle all the computations. The shot to shot time (4s) was unacceptable, I think. I beleive that was a problem created by MS.

I am just finally getting round to the conclusion that I will have to replace my ageing and failing N8 and what is apparent to me is how much improvement has been made over the years. Most of the top of the range smartphones produce results which are very close to each other and are at a level that most people are perfectly happy with, even with the small sensors and thin formats.

I am actually more bothered about replaceable batteries as I buy my phone outright and will therefore want it to last a long time. That means I want the option to renew the battery if I sense the performance tailing off. That is probably going to drive me to LG. (I would have liked to buy one of the new Nokias)

SDS

Not such a good analogy. Phone cameras are already probably more than good enough for the vast majority of phone owners, and those who want more will invest substantial amounts in specialized equipment anyway. On the other hand, bearing in mind what a miserable experience that flying in airliners has become, flying between London and San Francisco in less than four hours, and at a reasonable price, would be a dream.

zlutor

@Tomi: some clarification about Carl Zeiss vs. Nokia/HMD
http://mynokiablog.com/2017/03/15/clarification-zeiss-not-ruled-out-nokiamobile-says-current-line-up-doesnt-but-stay-tuned-for-more-announcements/

Long story short, they clarified CURRENT models does not have CZ tech...

Tomi T Ahonen

Hi zlutor

Thanks! That is welcome news and gives us some hope. I also shared the MyNokiaBlog news to my Twitter followers and thanked you zlutor for spotting this story.

It would make sense to include Carl Zeiss in some higher-end Nokia/HMD smartphones. Its possible they would do a 'Cameraphone' special device that pushes the camera aspect but is not necessarily a full 'flagship'. This could be then easily branded say an N-Series device and have Carl Zeiss optics and various cool camera tech otherwise. To be what say the N93, N82, N86 etc were to the N-Series lineup, the specialist camera-optimized phone (what also later 808 Pureview was).

Separately either with the above or perhaps later following it, there could be Carl Zeiss optics also in the 'flagship' class phone like the now-rumored Nokia 8. But these are not necessarily linked. HMD could do a more 'generic' camera into the Nokia 8, then do a separate N-Series 'cameraphone' with top Carl Zeiss camera alongside the Nokia 8, where some top specs are perhaps sacrificed but the camera is optimized. And then perhaps.. over time.. say the Nokia 9, next year's flagship, might also have Carl Zeiss on it, if HMD find that this aspect still works in helping sell Nokia premium phones (it may be that it doesn't really matter anymore) and they could move slowly on that path, test it with one phone model in late 2017, and only if it adds the ability to charge more for a phone, then do more Carl Zeiss later into next year etc...

But anyway, thanks zlutor! This is very welcome news. I don't think Nokia/HMD would have bothered to 'clarify' the denial, if they were 'totally committed to never using Carl Zeiss again'. If they really had ended totally CZ association, then they would not bother now to go 'backpedal' that Twitter comment. We may well see Zeiss on some premium Nokia phones again... and you guys know I would love to see that!

Tomi Ahonen :-)

Wayne Brady

Whither Carl Zeiss -- I'm not sure how much of that is marketing. Like having "Harmon Kardon" sound on a laptop. In photography, quality lenses certainly matter a great deal. But there is a pairing of the resolving power of a lens and the sensor. I'm not sure there is currently a great mismatch that the lens quality is holding back the final result.

For example - quality lenses typically help deal chromatic aberration, deliver sharp results across the entire frame (and not just the center), provide wide aperture (let in much more light, allow for depth of field control).

I'm not sure that today's phone camera deficiencies (among the premium smartphones) are along any of these axis.

Crispin

@Tomi: "Nokia Threat intelligence Report 2017" - still Elop's bad PR habits?
Could you, please, comment on this?
https://is.gd/threat2017 (Nasdaq GlobeNewswire, 27 March 2017)

It is presented through media as a relevant REGULAR PERIODIC (half-annual?) research.
By googling the subject I found that there are only few of those "periodic researches", only for last 3 years. Actually, there is one more set of 4 reports, for 2012 - but those are quarterly released and there is a gap of 1,5 year between this set and actual one:
https://networks.nokia.com/products/malware-reports
Somehow, time-lap collides with "King Elop reign". If you recall those times, it seems that microsurf wanted to conquer the market by PR, with fake reports (e.g. "Lumia is selling better than iPhone in China").
So, now it looks like legacy of bad habits.
What is the purpose of that report?
Beside legitimate "guerrilla marketing" (e.g. re-positioning of Nokia as trustful and relevant factor on mobile scene), I fear that it could be something with conclusion of this report:
"smartphones (and smart gadgets at all) are responsible for most malware infections".
It looks like this report tries to sustain selling of new Nokia 3310 and other feature-phones (dumb-phones) in India, for example.
Although approach like this is quite a legitimate, I think it is substantially wrong.

Maybe I'm missing something more important, but this report is nonsensical to me.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Working...
Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.

Working...

Post a comment

Your Information

(Name is required. Email address will not be displayed with the comment.)

Available for Consulting and Speakerships

  • Available for Consulting & Speaking
    Tomi Ahonen is a bestselling author whose twelve books on mobile have already been referenced in over 100 books by his peers. Rated the most influential expert in mobile by Forbes in December 2011, Tomi speaks regularly at conferences doing about 20 public speakerships annually. With over 250 public speaking engagements, Tomi been seen by a cumulative audience of over 100,000 people on all six inhabited continents. The former Nokia executive has run a consulting practise on digital convergence, interactive media, engagement marketing, high tech and next generation mobile. Tomi is currently based out of Hong Kong but supports Fortune 500 sized companies across the globe. His reference client list includes Axiata, Bank of America, BBC, BNP Paribas, China Mobile, Emap, Ericsson, Google, Hewlett-Packard, HSBC, IBM, Intel, LG, MTS, Nokia, NTT DoCoMo, Ogilvy, Orange, RIM, Sanomamedia, Telenor, TeliaSonera, Three, Tigo, Vodafone, etc. To see his full bio and his books, visit www.tomiahonen.com Tomi Ahonen lectures at Oxford University's short courses on next generation mobile and digital convergence. Follow him on Twitter as @tomiahonen. Tomi also has a Facebook and Linked In page under his own name. He is available for consulting, speaking engagements and as expert witness, please write to tomi (at) tomiahonen (dot) com

Tomi's eBooks on Mobile Pearls

  • Pearls Vol 1: Mobile Advertising
    Tomi's first eBook is 171 pages with 50 case studies of real cases of mobile advertising and marketing in 19 countries on four continents. See this link for the only place where you can order the eBook for download

Tomi Ahonen Almanac 2009

  • Tomi Ahonen Almanac 2009
    A comprehensive statistical review of the total mobile industry, in 171 pages, has 70 tables and charts, and fits on your smartphone to carry in your pocket every day.

Alan's Third Book: No Straight Lines

Tomi's Fave Twitterati