Jon Stern

Lives in United States Mountain View, CA, United States
Works as a Electronics Engineer
Joined on Apr 21, 2003

Comments

Total: 126, showing: 21 – 40
« First‹ Previous12345Next ›Last »

What's missing from this tour is the coating. You go straight from polishing to assembly (actually, I'd expect an annealing stage too).

Did they show you the coating chambers? I'm particularly interested in the nano-coating technology and how this is grown.

Link | Posted on Mar 21, 2017 at 00:02 UTC as 41st comment | 3 replies

Among the many questions I would have asked is, "What are there so few DO lenses?"

Link | Posted on Mar 20, 2017 at 23:57 UTC as 14th comment | 3 replies
In reply to:

justmeMN: "I can say it use to take about 70 people to make a lens like that prior to automation, now we need about 6 or 7."

In the USA, there is political discussion about "bringing back" manufacturing jobs. Increasing automation is why it's not going to happen.

83% of the 5 million US manufacturing jobs lost in 1995 have been lost to automation. Only 17% have been lost to globalization.

Link | Posted on Mar 20, 2017 at 23:54 UTC
In reply to:

krassphoto: "it use to take about 70 people to make a lens like that prior to automation, now we need about 6 or 7."
...yet the prices are going up.

Stronger Yen.

Link | Posted on Mar 20, 2017 at 23:53 UTC
In reply to:

Jon Stern: I hope you had gyoza and beer while in Utsunomiya. It's one thing it's known for.

That's a shame. I did once manage to fit that in when on a business trip to Ikegami, the renowned Japanese broadcast camera manufacturing who is headquartered in Utsunomiya.

Link | Posted on Mar 20, 2017 at 20:29 UTC

It's interesting that they're testing lenses on modified cameras, rather than using a Trioptics machine.

Link | Posted on Mar 20, 2017 at 18:41 UTC as 61st comment | 1 reply

I hope you had gyoza and beer while in Utsunomiya. It's one thing it's known for.

Link | Posted on Mar 20, 2017 at 18:37 UTC as 62nd comment | 2 replies
In reply to:

orion1983: Visionary approach! However, I assume that their "image fusion technology" to render a 3x optical zoom a 5x "optical" zoom is just few more than a normal digital zoom until I a proven wrong ;)
Also the chip seems to be not a 1/2.3" one which I for myself see as absolutely crucial for great phone pics

Super resolution genuinely provides additional resolution (by fusing the two camera images). The 5x claim is sound.

https://en.wikipedia.org/wiki/Super-resolution_imaging

Link | Posted on Feb 27, 2017 at 20:33 UTC
In reply to:

chaos215bar2: Can someone correct the headline, or is OPPO actually calling this "5x optical" zoom? That's certainly not what's described here.

Super resolution allows the system to deliver equivalent performance of a 5x optical zoom from a 3x focal length ratio.

It's hard to market without it sounding like BS, but it's real.

https://en.wikipedia.org/wiki/Super-resolution_imaging

Link | Posted on Feb 27, 2017 at 20:31 UTC
In reply to:

cosinaphile: its cool hope its a better sensor than usually seen.... also 3x not 5 optical

The additional resolution is from super resolution techniques.

https://en.wikipedia.org/wiki/Super-resolution_imaging

Link | Posted on Feb 27, 2017 at 20:30 UTC
In reply to:

wetsleet: "OPPO announces dual-cam 5x optical zoom technology for smartphones"
Really? I thought the optical zoom was 3x.
Who is responsible for the marketing inflation here, Oppo or DPR?

These systems get an additional boost in resolution from super-resolution techniques (combining images from each camera). This is a real over-sampling technology that allows a 5x equivalent optical zoom performance from a 3x focal length ratio.

This is not marketing BS.

https://en.wikipedia.org/wiki/Super-resolution_imaging

Link | Posted on Feb 27, 2017 at 20:30 UTC

My father had one, but sold it for next to nothing, not realizing how much it was worth.

He had the rare roll-film back on his too.

Link | Posted on Jan 31, 2017 at 20:44 UTC as 19th comment
In reply to:

noflashplease: $1,000 for a cellphone from a Chinese manufacturer I've never heard of? What? For some reason, Samsung, Vivo and Oppo are selling phones with the same processor and 6GB of RAM, so this thing fits into some sort of odd consumer category in China. It just doesn't make sense to the outside world.

Gionee is a well-know brand in China. The fact you've never heard of it doesn't change the fact that probably close to a billion people have.

Link | Posted on Dec 27, 2016 at 23:10 UTC
In reply to:

M Chambers: Freezing a GoPro still won't give it video output or the other basic features they lack.

GoPro cameras have HDMI output.

Link | Posted on Dec 21, 2016 at 19:44 UTC
On article Real world test: The Panono is a 108MP spherical camera (156 comments in total)
In reply to:

Jon Stern: Sensor performance is all about area, so a 1/4-inch sensor isn't really, "a bit smaller than the sensor likely found in your smartphone", it's a lot smaller. Half the area of those in mid-tier phones, smaller still than the sensors that are being used in current flagship phones.

For cost reasons, and probably depth of field, it's clear why they'd go for 1/4-inch, but it's a disappointment when it comes to image quality.

Your sensor sizes are slightly off. A 1/4-inch sensor is 3.7mm x 2.7mm. A 1/3-inch is 4.7mm x 3.6mm, meaning 1.7x area vs 1/4-inch.

If both have 1.12µm pixels, the per-pixel SNR will remain constant, but the SNR for a constant viewing condition will be nearly 1-stop higher for the larger sensor.

As both have the same pixel, the diffraction-limit issues are the same for both. In any case, Panono is probably using one of the many mobile lenses designed for 1.12µm sensors. Many of those are F/2.0, which is a good fit. Sure off-axis MTF@Ny/2 sucks for those mobile lenses, but on-axis they're close to the diffraction limit.

Incidentally, even though it has nothing to do my sensor area ratio point, Panono probably has a 20% overlap for stitching. That's pretty typical, and if I did my solid angle maths right, would mean that using mobile lenses with a DFOV around 76° (minus overlap) would be a perfect fit for 36 cameras to fill the spherical view.

[p.s. I'm a Dr Jon from the UK too!]

Link | Posted on Jul 27, 2016 at 05:34 UTC
On article Real world test: The Panono is a 108MP spherical camera (156 comments in total)

Sensor performance is all about area, so a 1/4-inch sensor isn't really, "a bit smaller than the sensor likely found in your smartphone", it's a lot smaller. Half the area of those in mid-tier phones, smaller still than the sensors that are being used in current flagship phones.

For cost reasons, and probably depth of field, it's clear why they'd go for 1/4-inch, but it's a disappointment when it comes to image quality.

Link | Posted on Jul 25, 2016 at 05:15 UTC as 29th comment | 3 replies
On article Lens shootout: Sony RX10 III destroys the competition (502 comments in total)

DPReview destroyed by writers feeling that it's necessary to create clickbait headlines.

Link | Posted on May 26, 2016 at 20:10 UTC as 131st comment | 14 replies

A camera need not perform pixel shift to extend dynamic range. HDR simply needs a sequence of exposures. Preferably taken in quick succession to minimize motion effects.

Pixel shift is all about using super resolution techniques to increase MTF. HDR is all about combining multiple frames.

Link | Posted on May 9, 2016 at 16:48 UTC as 93rd comment

About time too!

I wondered how long it would take for them to produce AF lenses.

I hope they'll do Canon AF lenses next.

Link | Posted on May 4, 2016 at 04:46 UTC as 35th comment | 6 replies
In reply to:

Jon Stern: Strange how the press release text doesn't mention Micro Four Thirds. While followers of Olympus digital cameras and the digital Pens, in particular, would know this, it's almost as if they are trying to avoid talking about the sensor size.

pocoloco, that would be nice, yes. But given how difficult it is to define IQ (I know, I work for an image sensor company) so proxy for performance is needed. While the industry has long tried to push megapixels for that, it's really sensor size which is the biggest factor. The laws of physics mean that will always be the case.

When sensors had low quantum efficiency, low dynamic range, and high read noise, you could look to improvements to the sensor to deliver ever better IQ. But now sensors have reach maturity and only small incremental improvements are available. As a result, sensor size makes the most difference.

And then there's the bokeh factor on top of that.

Link | Posted on Jan 28, 2016 at 04:21 UTC
Total: 126, showing: 21 – 40
« First‹ Previous12345Next ›Last »