ProfHankD

ProfHankD

Lives in United States Lexington, United States
Works as a Professor
Has a website at http://aggregate.org/hankd/
Joined on Mar 27, 2008
About me:

Plan: to change the way people think about and use cameras by taking advantage of cameras as computing systems; engineering camera systems to provide new abilities and improved quality.

Comments

Total: 473, showing: 21 – 40
« First‹ Previous12345Next ›Last »
On Tune in to DPReview Live this Thursday and Friday article (238 comments in total)
In reply to:

Kevin Coppalotti: Dynamic range.want.more. any thing coming?

Rishi, there are other ways to get much greater DR. For example, my TDCI research measures time-to-threshold independently on each sensel, so dynamic range is not inherently sensor-limited (it's bounded only by photon shot noise and the allowable range of integration times). Real scenes can have as much as 16-20 Ev DR. That said, 14 Ev is better than human eyes do. ;-)

Direct link | Posted on Oct 9, 2014 at 18:55 UTC
On Tiny fps1000 high-speed camera boasts 18,500fps article (137 comments in total)

There are a lot of high-speed industrial cameras out there, but the gap between sensor cost and camera cost is a lot larger than in this one, especially for a stand-alone camera. The biggest issue I see is that it doesn't seem to have a fast enough interface to support tethered continuous capture. (Actually, there are some software tricks from my research that might help them with that....)

Direct link | Posted on Oct 9, 2014 at 16:08 UTC as 30th comment
In reply to:

Jones R: The government really ought to take this beyond national parks. Look at Hungary and their recent photography law. Making images (static photographs and video) of people without their knowledge or consent should be a federal offense. Right now one could mostly take pictures or video images of pedestrians without their knowledge or consent - and in my opinion, this is not only wrong, but should be made illegal.

A person essentially consents to being imaged by the act of being present in an environment that has no expectation of privacy; of course, what you can do with those images is often subject to regulation. The fact that audio recording is treated very differently, and is much more regulated (and inconsistently regulated across USA states), probably has to do with the idea that people might reasonably convey private information in a conversation in a public place assuming privacy because other people with normal hearing wouldn't be able to overhear (although amplified recording equipment could). In any case, with cameras everywhere and cell phones always listening (for voice commands), I think the clear trend is going to have to be toward less regulation of capture in public places and more constraints on use of captured material....

This particular Forest Service proposal seems unconcerned with audio. I guess they figure conversations between wildlife have no private content? ;-)

Direct link | Posted on Oct 8, 2014 at 16:25 UTC
On Tune in to DPReview Live this Thursday and Friday article (238 comments in total)

Question to all: Digital cameras have the potential to support sophisticated, programmable (or at least scriptable), operation. There has long been an active community developing open source software to run in Canon cameras (CHDK and Magic Lantern) providing valuable improvements in camera performance and capabilities, but that community is unsupported by Canon and there is no process for validating or securely running such applications. Certainly, Apple has profited greatly from their support of in-device 3rd-party apps -- and most cameras now embed OS environments that easily could be augmented by an API and app certification process to provide secure in-camera execution of both simple user-written scripts and sophisticated 3rd-party apps. Do you have any plans to establish an officially supported, secure, method for powerful "in-camera apps" to be created by both commercial and open source 3rd parties? If not, why not?

Direct link | Posted on Oct 8, 2014 at 02:47 UTC as 65th comment | 2 replies
On 'See Impossible': Canon counts down to... something. article (1666 comments in total)

"At Canon, we see impossible." Ok. Now open your eyes, fire the ad guys who came up with that slogan, and make something new possible.

Direct link | Posted on Oct 6, 2014 at 11:01 UTC as 792nd comment | 1 reply

Take only pictures, leave only permit fees..

Direct link | Posted on Oct 2, 2014 at 02:56 UTC as 79th comment
In reply to:

Lee Jay: I'm having a hard time understanding how this could not produce lots of astigmatism, coma, CA and so on. Also, it would seem that aperture would be constant leading to either a very narrow zoom range, some crazy fast f-stops at the wide end, or some crazy slow f-stops at the long end.

"Radical aspherics" was what came to mind, but I don't see that working so well for continuous zoom. There is a possible way to do this with diffractive optics, but that's not what they said they're doing...? Actually, their WWW site says just about nothing.

Direct link | Posted on Sep 30, 2014 at 23:08 UTC
In reply to:

ProfHankD: Yes, Canon has been serious about mirrorless in that the EOS-M is grossly inferior to their DSLRs only in AF speed and ergonomics. The problem is that it isn't compelling against models from Sony, Olympus, Fuji, Samsung, etc. Canon was once an innovator, but not lately, and they have a long way to go to make a compelling mirrorless. For example, I don't think a 7D Mark II with an EF-M mount would be a very compelling mirrorless, and it has been years since Canon's sensors gave them an real edge over those made by Sony.

I buy lots of Canon cameras, mostly PowerShots to use with CHDK, but also an occasional higher model to use with Magic Lantern. Ironically, both those environments are conspicuously not aided or supported by Canon. I think it is long overdue that Canon should encourage, and actively leverage, this type of external innovation.

armandino, cameras are MUCH simpler than cell phones, but you are absolutely correct about inertia (and Sony cameras are not less usable than CaNikons, just less familiar -- and not that unfamiliar to us folks who grew up with Minoltas). Nikon and Canon are not sleeping giants... they're not that big. Samsung is the only real giant in this game, and unless something interesting happens, they'll probably rule the market in a few years. We'll see....

Direct link | Posted on Sep 30, 2014 at 11:39 UTC
In reply to:

ProfHankD: Yes, Canon has been serious about mirrorless in that the EOS-M is grossly inferior to their DSLRs only in AF speed and ergonomics. The problem is that it isn't compelling against models from Sony, Olympus, Fuji, Samsung, etc. Canon was once an innovator, but not lately, and they have a long way to go to make a compelling mirrorless. For example, I don't think a 7D Mark II with an EF-M mount would be a very compelling mirrorless, and it has been years since Canon's sensors gave them an real edge over those made by Sony.

I buy lots of Canon cameras, mostly PowerShots to use with CHDK, but also an occasional higher model to use with Magic Lantern. Ironically, both those environments are conspicuously not aided or supported by Canon. I think it is long overdue that Canon should encourage, and actively leverage, this type of external innovation.

"Nikon and Canon ... are the best in the business."?

Which business? Unfortunately, I think the answer is SLRs. Sony and Samsung are the leaders in sensor technology and camera electronics, which matters more now. Time will tell, but I think Canon in particular has proven to have excessively long development cycles for products with even the tiniest bit of innovation; they're starting to remind me of Graflex. Canon hasn't even been smart about leveraging a very active community developing firmware for their cameras (CHDK and ML).

Direct link | Posted on Sep 27, 2014 at 09:59 UTC

Yes, Canon has been serious about mirrorless in that the EOS-M is grossly inferior to their DSLRs only in AF speed and ergonomics. The problem is that it isn't compelling against models from Sony, Olympus, Fuji, Samsung, etc. Canon was once an innovator, but not lately, and they have a long way to go to make a compelling mirrorless. For example, I don't think a 7D Mark II with an EF-M mount would be a very compelling mirrorless, and it has been years since Canon's sensors gave them an real edge over those made by Sony.

I buy lots of Canon cameras, mostly PowerShots to use with CHDK, but also an occasional higher model to use with Magic Lantern. Ironically, both those environments are conspicuously not aided or supported by Canon. I think it is long overdue that Canon should encourage, and actively leverage, this type of external innovation.

Direct link | Posted on Sep 25, 2014 at 04:47 UTC as 66th comment | 19 replies
On Photokina 2014 Video: The Samsung NX1 article (98 comments in total)
In reply to:

Nukunukoo: I am so angry why Canon and Nikon are still thinking that they are still safe in their comfort zone. Sony, Panasonic, Fuji and Samsung are slugging it out with really good, cutting-edge, products.

Samsung probably is going to be more than ok. ;-)

Direct link | Posted on Sep 19, 2014 at 11:17 UTC
On Photokina 2014 Video: The Samsung NX1 article (98 comments in total)
In reply to:

Nukunukoo: I am so angry why Canon and Nikon are still thinking that they are still safe in their comfort zone. Sony, Panasonic, Fuji and Samsung are slugging it out with really good, cutting-edge, products.

Nothing to get angry about... we are simply at a point where the concept of a camera is quickly changing and which companies lead the industry has changed in response. Consumer electronics manufacturers took the lead a while back, and the better they are at making chips and systems, the stronger their position is and will be.

Direct link | Posted on Sep 19, 2014 at 11:01 UTC
On Photokina 2014: Sony stand report article (78 comments in total)

Actually, photo 4 makes a lot of sense to me -- the QX1 seems like a great choice of remote-camera-back to stick behind a lens... and a remote pan/tilt mount often would be the right thing to sit it on. I can also see it as a microscope or telescope camera. Or on a UAV. The thing that doesn't make sense to me is sticking it on your cell phone (or your cell phone on it). ;-)

Hopefully, next year's QX model will use a FF sensor. :-)

PS: People who've been complaining about lack of Sony lenses should be happy with Sony this year, although I've long been happy merely having the largest choice of lenses of any camera despite very few being Sony branded.

Direct link | Posted on Sep 18, 2014 at 00:03 UTC as 6th comment
On Canon EOS 7D Mark II First Impressions Review preview (2096 comments in total)
In reply to:

Charrick: I don't want to be "that guy", but I guess I will be. I'm not terribly impressed. The autofocus system is a huge step up (from its predecessor, though not the 70D). Its sensor has the same number of pixels as the 70D, but I hope that it's a very different one and not just a slightly upgraded version. However, my gut feeling is that the Sony A77 II's performance will be better on DXOMark (even with its "translucent mirror"). Canon sensors lag behind (often more than a generation) what other cameras have in terms of color depth, dynamic range, and even high ISO ability. As usual, the "higher-level" camera lacks a flipping screen (because "pros" don't want that...or so Canon tells themselves). For some odd reason, some of you would-be pros say you don't want touchscreens or pop-up flashes (even though both are just extras and you don't have to use them!), and Canon has listened on the touchscreen front. Also, 1080p is aging and 4K would be impressive. The 7DmkII is OK but nothing special.

This is such an impressive camera five years ago! Oh wait. It's now. Maybe somebody should tell Canon...?

I guess I'm "that guy" too. Canon just keeps looking older every year. I buy lots of Canons for my research, but I wouldn't buy any of their cameras if it wasn't for the ability to program them using CHDK and ML -- which are externally-developed environments that Canon does not support. On the bright side, at least the lack of new stuff inside most new models makes it easy to port CHDK and ML to them. ;-)

Direct link | Posted on Sep 16, 2014 at 00:33 UTC
In reply to:

riskinhos: for that price just get full frame.

I don't think you appreciate just how tough it is to do BSI on a sensor this large... I'd bet it costs more to make than a FF non-BSI sensor. Uncompressed 4K and 15FPS full res too. In sum, pretty darn impressive, Samsung.

Direct link | Posted on Sep 15, 2014 at 11:33 UTC
On Sigma announces dp1 Quattro article (199 comments in total)
In reply to:

Tungsten Nordstein: The loupe is interesting. Let's hope it both annoys and alienates the traditionalists – especially the ones who crack lame 'lol' type jokes.

I've been using a similar device for years on the back of my NEX-5 -- a rigid-frame magnifying finder that cost under $20 via eBay (and unfortunately didn't have a diopter adjustment). It doesn't compete well with the NEX-7 or A7 EVF, and it's big enough to attract attention (which could be good or bad), but there is no question that it does work quite well. Given the odd form of the Quattro, I think the LCD eyepiece actually makes more sense than an add-on EVF.

Direct link | Posted on Sep 14, 2014 at 00:42 UTC

I don't see much technical detail here. Fuji has a very long history of making lenses with add-on apodization of various kinds, from sink-strainer-like things to circular gradient filters that remove central hot spots in wide-angle large-format lenses. The DPReview descpription "filter is a radial gradient filter designed to exactly compensate for any bright regions in the circles of confusion the lens projects" sounds like a hot spot filter -- the Minolta STF does exactly the opposite. To improve bokeh, you want a filter that isn't compensating for anything but gets darker toward the edges like the STF one does.

Any significant gradient will confuse PDAF sensors,but should not impact CDAF too severely. It sound like Fuji is just taking what they can easily get from using a lens design that already had hardware to move the focus -- precisely what one would expect.

One last note: it is exceedingly difficult to make high-quality circular gradient filters. +$500 sounds reasonable.

Direct link | Posted on Sep 11, 2014 at 00:46 UTC as 6th comment
On Zeiss introduces 'no distortion' Otus 1.4/85mm article (337 comments in total)
In reply to:

ProfHankD: Zeiss specs look excellent, although it's around 0.8% distortion so saying how minimal distortion is makes little sense. All the optical qualities look very good, and the CA issues are much more important -- APOs do things postprocessing can't fix for other lenses. I'll also admit that my Samyang's "close focus" behavior drives me nuts, and extensions tubes cause a huge degradation in IQ for that IF lens design. Then again, for 1/18 the price....

You haven't removed all distortion if there is fairly significant geometrical distortion. The stuff they have fixed is largely just compliance with the definition of APO.

Direct link | Posted on Sep 9, 2014 at 11:35 UTC
On Zeiss introduces 'no distortion' Otus 1.4/85mm article (337 comments in total)

Zeiss specs look excellent, although it's around 0.8% distortion so saying how minimal distortion is makes little sense. All the optical qualities look very good, and the CA issues are much more important -- APOs do things postprocessing can't fix for other lenses. I'll also admit that my Samyang's "close focus" behavior drives me nuts, and extensions tubes cause a huge degradation in IQ for that IF lens design. Then again, for 1/18 the price....

Direct link | Posted on Sep 9, 2014 at 10:38 UTC as 59th comment | 3 replies
On Beginner's guide: shooting high-key at home article (69 comments in total)
In reply to:

Photomonkey: Pretty risky showing us how to get around Amazon's patent.

So you did! I missed that (traveling); sorry.

Direct link | Posted on Sep 9, 2014 at 01:35 UTC
Total: 473, showing: 21 – 40
« First‹ Previous12345Next ›Last »