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: 1217, showing: 1 – 20
« First‹ Previous12345Next ›Last »

The manual segmentation applications in their paper look interesting. This actually seems a lot like some of the old methods for colorizing B&W films....

Link | Posted on Mar 28, 2017 at 15:05 UTC as 8th comment

Back when I shot film, it was my Minolta MC W Rokkor Si 28mm f/2.5 -- and it is still a favorite. If it's one lens for everything, my Sigma 28-200mm f/3.5-f/5.6 Macro D Aspherical IF, a small and surprisingly sharp lens that cost me under $20, is my current FF answer. If the question is which lens do I actually use most, on APS-C the answer is my Sigma 8-16mm; a lens that is hard to use well, but regularly produces stunning images if used wisely.

However, I have a LOT of (mostly manual) lenses and don't really play favorites. I use what seems right for the circumstances, and deliberately rotate otherwise so I keep aware of my choices. It's really great that mirrorless cameras can use so many different lenses so well and so many old lenses now sell for tiny fractions of their utility-based value (most of mine were under $25).

Link | Posted on Mar 27, 2017 at 13:58 UTC as 128th comment

So, how do you know that there is actually a drone out there? It would be fairly easy to computationally render your flight from a terrain database. Think of it as a sort of "Touring" test. ;-)

Link | Posted on Mar 25, 2017 at 13:08 UTC as 26th comment | 3 replies
On article Re-make/Re-model: Leica Summaron 28mm F5.6 Samples (199 comments in total)

Well, that renders very much like a lens I just got: a 1990s Sigma 18-35mm f/3.5-4.5 zoom that cost me $32. Ok, the Sigma's actually better -- less vignetting and sharper corners -- but has similar colors and handles flare almost identically. I see "incredibly slim" motivation for this Summaron. Is this Leica's version of a lens in a body cap? :-(

Link | Posted on Mar 24, 2017 at 12:08 UTC as 71st comment
In reply to:

ProfHankD: The real advantage of Sony's rather-undersized "medium format" sensor should be that it could be used with FF lenses for alternative aspect ratios without needing a larger image circle. For example, a 30.6x30.6mm square format would fit the same FF-lens image circle while delivering 33MP. You could also do 2.35:1 cinemascope aspect ratio with 39.6x16.8mm at 24MP. It would be very easy for a camera to show these crops in the EVF and tag the EXIF data. However, nobody seems to be doing this. Why the heck not?

Karl Persson: that's a trick that I documented works very well for APS-C lenses on FF (in a paper I published in Electronic Imaging 2016). However, the aspect ratio is different here, and taking advantage of that does more than the teleconverter trick would do.

Link | Posted on Mar 22, 2017 at 11:05 UTC

The real advantage of Sony's rather-undersized "medium format" sensor should be that it could be used with FF lenses for alternative aspect ratios without needing a larger image circle. For example, a 30.6x30.6mm square format would fit the same FF-lens image circle while delivering 33MP. You could also do 2.35:1 cinemascope aspect ratio with 39.6x16.8mm at 24MP. It would be very easy for a camera to show these crops in the EVF and tag the EXIF data. However, nobody seems to be doing this. Why the heck not?

Link | Posted on Mar 22, 2017 at 02:16 UTC as 241st comment | 5 replies
In reply to:

ProfHankD: The phrase "effectively producing full-frame coverage on non-full-frame sensors" doesn't sound to me like a new claim for any of the focal reducers out there... why the "doesn’t work out exactly though" comment here and not on others? It is worth noting, BTW, that focal reducers do vary; of the five I've measured, the Speed Boosters provide the greatest reduction and the Lens Turbo II the least, with a spread from about 0.71x to 0.73x.

Let's put that in context. It turns out that APS-C isn't all the same size. Sensors in most brands actually have a crop factor of 1.52-1.54x while Canon's APS-C is 1.61x. BTW, actual APS film, C format, is 1.43x crop -- so they're all undersize. For that matter, 135 film wasn't always exactly 36x24, and standard slide mounts covered at least 2% of that area (which is part of why a lot of SLRs used to have 98% viewfinders).

Anyway, focal reducers vary in coverage by less than 3% -- while Canon's APS-C is about 6% smaller than most digital APS-C. Meh.

badi: Missed again. The "Meh" is about this particular news item emphasizing a claim that the manufacturer didn't even really make. I.e., there isn't a significant magnification difference BETWEEN DIFFERENT FOCAL REDUCERS.

Link | Posted on Mar 17, 2017 at 15:41 UTC
In reply to:

ProfHankD: The phrase "effectively producing full-frame coverage on non-full-frame sensors" doesn't sound to me like a new claim for any of the focal reducers out there... why the "doesn’t work out exactly though" comment here and not on others? It is worth noting, BTW, that focal reducers do vary; of the five I've measured, the Speed Boosters provide the greatest reduction and the Lens Turbo II the least, with a spread from about 0.71x to 0.73x.

Let's put that in context. It turns out that APS-C isn't all the same size. Sensors in most brands actually have a crop factor of 1.52-1.54x while Canon's APS-C is 1.61x. BTW, actual APS film, C format, is 1.43x crop -- so they're all undersize. For that matter, 135 film wasn't always exactly 36x24, and standard slide mounts covered at least 2% of that area (which is part of why a lot of SLRs used to have 98% viewfinders).

Anyway, focal reducers vary in coverage by less than 3% -- while Canon's APS-C is about 6% smaller than most digital APS-C. Meh.

badi: I think you missed my point: these numbers are all much more approximate than you think; precision stated is much higher than accuracy. In fact, both the Lens Turbo and Lens Turbo II quote the same 0.726 magnification factor. Looking up my actual measurements, SB is actually 0.71x, LT is 0.73x, and LTII is 0.74x (even a little more variation than I remembered above).

The same inaccuracy applies for focal length of most lenses; several % error is normal. In fact, for some lenses the measured focal length is off by over 10% from the published length. Various zooms that supposedly go to 300mm really stop at more like 270mm, many fast "50mm" lenses are more like 53mm, and quite a few wide-angle lenses now are much wider than quoted (in order to compensate for loss of view angle when distortion corrections are applied). On the other hand, there are some "180-degree" fisheyes with more like 150-degree view angles.

So, don't worry about a few % unless you measure everything. ;-)

Link | Posted on Mar 17, 2017 at 15:07 UTC

The phrase "effectively producing full-frame coverage on non-full-frame sensors" doesn't sound to me like a new claim for any of the focal reducers out there... why the "doesn’t work out exactly though" comment here and not on others? It is worth noting, BTW, that focal reducers do vary; of the five I've measured, the Speed Boosters provide the greatest reduction and the Lens Turbo II the least, with a spread from about 0.71x to 0.73x.

Let's put that in context. It turns out that APS-C isn't all the same size. Sensors in most brands actually have a crop factor of 1.52-1.54x while Canon's APS-C is 1.61x. BTW, actual APS film, C format, is 1.43x crop -- so they're all undersize. For that matter, 135 film wasn't always exactly 36x24, and standard slide mounts covered at least 2% of that area (which is part of why a lot of SLRs used to have 98% viewfinders).

Anyway, focal reducers vary in coverage by less than 3% -- while Canon's APS-C is about 6% smaller than most digital APS-C. Meh.

Link | Posted on Mar 17, 2017 at 12:23 UTC as 23rd comment | 6 replies
On article Throwback Thursday: Our first cameras (386 comments in total)

Ok, I'll play too. Technically, my first couple of cameras were cheap plastic junk. The Konica C35V was my first serious camera... at least it was serious enough that I won a photo contest using it, shot my first published photos with it, and sold quite a few images shot with it. Always wished I had the rangefinder version, but zone focus worked with a (surprisingly good) 38mm that was f/2.8 wide open. After I moved on to Minolta SRT101 and XK, I still appreciated the little C35V's utter inobtrusiveness as a street shooter, but alas, it was stolen many years ago.

I used digitized video for a while (anyone remember the video snappy?), but my first digital was a Casio QV100. It had terrible IQ, but was really an impressively digital camera... not just in the lack of an optical viewfinder and the pivoting lens; I even gave a few lectures using slides uploaded to it! Actually, using the QV100 was a lot like using the C35V; it was also a very inobtrusive street shooter.

Link | Posted on Mar 16, 2017 at 11:46 UTC as 277th comment
In reply to:

sleibson: Sentence the van driver to community service... on the salt-flat repair team.

The $5000 fine and 6 months in jail combo does seem less appropriate than having to pay for the towing and restoration effort and being assigned significant community service. Perhaps this is something to prosecute for civil damages, not just as breaking the law?

Link | Posted on Mar 15, 2017 at 13:06 UTC
In reply to:

AlexanderHorl: Does the app save pictures as raw or is just jpeg possible?

Actually, much as I like raw too, the fact that the RESULT is A raw means it can't be doing all that wonderful computational alignment Sony does in most multi-shot capture modes. This is why you need a tripod. Ideal would be a JPEG of the final image and/or the set of raws from which it was constructed along with a specification of how to merge them.

Link | Posted on Mar 15, 2017 at 12:40 UTC
In reply to:

ProfHankD: $30 for something that could easily be a free app or even a firmware update. This is unfortunately copying an old Minolta thing (program cards) too closely. :-(

It shouldn't be too long now before the Open Memories interface starts producing more interesting and useful apps for free....

_Frederico_: I'm a lot more revolutionary than that, and implement lots of stuff inside commodity cameras. ;-) For example, my current work has largely centered on being able to capture and process image data so that you can adjust the time interval represented by an exposure after the fact (TDCI -- time domain continuous imaging) and we've actually implemented this inside Canon PowerShots using CHDK, as reported in a paper at Electronic Imaging 2017.

I'm sure this is a nice app; I just don't think Sony should restrict app development to themselves nor do they need to charge $30 for one.

Link | Posted on Mar 15, 2017 at 12:21 UTC

$30 for something that could easily be a free app or even a firmware update. This is unfortunately copying an old Minolta thing (program cards) too closely. :-(

It shouldn't be too long now before the Open Memories interface starts producing more interesting and useful apps for free....

Link | Posted on Mar 15, 2017 at 04:20 UTC as 23rd comment | 2 replies
On article Ask the staff: electronic or optical viewfinder? (889 comments in total)
In reply to:

ProfHankD: The whole point of SLRs was to get you as close as possible to seeing how the camera will capture the scene, and EVFs do that better. Beyond that, they free you from seeing your composition only by the (dim) light that comes through the lens, especially stopped down. They also free the camera design from the awkward mirror/pentaprisim optics and mechanics, simultaneously allowing shorter/better-utilized space between the lens rear and focal plane... and an EVF live view stream can be electronically routed anywhere. As for timing, early initiation of captures allows true zero-lag capture -- which might not be widely implemented yet, but is impossible with a moving-mirror SLR.

In sum, stop thinking about digital cameras as emulating film. They are an entirely different medium, and EVFs / live view are key components of that medium.

keepreal: the EVF images do indeed map the full captured sensor range into a tonal range a human can see, whereas you can't see some of the tones that the sensor captures looking through an OVF. Why is it so hard for people to understand that the sensor itself is the best way to determine what the sensor will record?

Link | Posted on Mar 12, 2017 at 21:49 UTC
On article Ask the staff: electronic or optical viewfinder? (889 comments in total)
In reply to:

ProfHankD: The whole point of SLRs was to get you as close as possible to seeing how the camera will capture the scene, and EVFs do that better. Beyond that, they free you from seeing your composition only by the (dim) light that comes through the lens, especially stopped down. They also free the camera design from the awkward mirror/pentaprisim optics and mechanics, simultaneously allowing shorter/better-utilized space between the lens rear and focal plane... and an EVF live view stream can be electronically routed anywhere. As for timing, early initiation of captures allows true zero-lag capture -- which might not be widely implemented yet, but is impossible with a moving-mirror SLR.

In sum, stop thinking about digital cameras as emulating film. They are an entirely different medium, and EVFs / live view are key components of that medium.

Huh? Most TLRs and many rangefinders compensate for parallax using a moving mask. The big thing for SLRs was looking through the taking lens so you could see details like lens flare, focus and DoF, etc. As for dynamic range, surely you don't think an OVF view is a more accurate representation of captured dynamic range than an EVF is? The EVF display might map the DR into a smaller range of brightness, but even that's a huge plus given that many sensors now exceed the instantaneous DR of human sight (e.g., all the Sony A7 series does).

Link | Posted on Mar 12, 2017 at 18:21 UTC
On article Ask the staff: electronic or optical viewfinder? (889 comments in total)

The whole point of SLRs was to get you as close as possible to seeing how the camera will capture the scene, and EVFs do that better. Beyond that, they free you from seeing your composition only by the (dim) light that comes through the lens, especially stopped down. They also free the camera design from the awkward mirror/pentaprisim optics and mechanics, simultaneously allowing shorter/better-utilized space between the lens rear and focal plane... and an EVF live view stream can be electronically routed anywhere. As for timing, early initiation of captures allows true zero-lag capture -- which might not be widely implemented yet, but is impossible with a moving-mirror SLR.

In sum, stop thinking about digital cameras as emulating film. They are an entirely different medium, and EVFs / live view are key components of that medium.

Link | Posted on Mar 12, 2017 at 17:06 UTC as 285th comment | 5 replies

Great argument for why cameras should be supporting user apps. If you plot non-smart phone (non-app capable) sales I don't think you'll see anywhere near as happy numbers. Sony really needs to open PlayMemories to 3rd-party authors.

Link | Posted on Mar 7, 2017 at 01:22 UTC as 97th comment | 2 replies
In reply to:

Hachu21: This graph explain something important :
I see frequently -and I tend to agree- that even high end dslr are missing important features that you get with you sub 1000$ smartphone (high dpi screen, fast and easy UI, straightforward possibilities to share images, quite good 4k video (I'm looking at you Canon), some powerfull integrated feature and easy to use (the panoramic mode on the iphones come to mind).
On this graph, you can see the abysmal difference between smartphones sales numbers and dedicated camera sales numbers. It means that the funds, people, technical innovation and concurrence pressure involved in the smartphone market is just in a different galaxy compared to the "old and small" camera world.

It also means that R&D costs are divided on a much higher volume that allow high tech to almost affordable prices.

Hachu21: you're not wrong. CHDK and ML incidentally document what type of processor is in Canon PowerShots and EOS (respectively). They're not very beefy.

Link | Posted on Mar 7, 2017 at 01:18 UTC
In reply to:

ProfHankD: Well, that explains the $2600 price!

As an engineer, I can't help but be impressed that such a complex electro-mechanical-optical system can be commercially viable... and the complexity certainly explains the use of no-user-serviceable-parts-inside silicone glue.

However, something about this also feels very wrong. In computer systems, as system complexity has become incomprehensibly high, the physical component count hasn't. Materials with internally-complex structures (VLSI chips) basically absorb that complexity, and connections between things get simplified from point-to-point wiring to standardized interfaces/networks. We do the same in computer software, hiding complexity in reusable functions. It's the whole industrial revolution idea that keeps systems manageable: standardized interfaces between interchangeable (and customizable) parts.

In contrast, lenses seem chock full of CUSTOM structures and interconnections. Isn't it time that started to change?

The lens/body interface is pretty well standardized (per manufacturer)....

However, let me give you a really simple example of non-standardization that seems unwise. For under $10, you can buy passive extension tubes for just about every camera mount. The extension segments on those tubes all are screw threaded, so you can insert as many or as few threaded extensions as desired between the end pieces that mate to your lens and body. The wacky thing is that each such tube set I've gotten has a different diameter for the middle sections, so you can't use a middle piece from an A-mount tube set with an EF-mount tube set, etc. The diameters are not usefully different -- usually just a mm or two -- so what's the benefit to the users or to the manufacturer?

It would seem one could have various standardized lens barrel and electronics components to select from in building lenses, rather than full custom parts....

Link | Posted on Mar 4, 2017 at 17:55 UTC
Total: 1217, showing: 1 – 20
« First‹ Previous12345Next ›Last »