hjulenissen

Lives in Svalbard and Jan Mayen Islands Svalbard and Jan Mayen Islands
Joined on Aug 23, 2010

Comments

Total: 35, showing: 1 – 20
« First‹ Previous12Next ›Last »
In reply to:

ProfHankD: That "decoupling the shutter angle of capture from the shutter angle required for artistic effect" sounds like they've been reading my research papers on TDCI (time domain continuous imaging). :-)

Interesting that they are saying "755 RAW megapixels"... I guess they've given up on calling them megarays. Probably a good idea too.

Rishi:
If the 1/300 s exposures are shot back-to-back, then averaging them should in principle be equivalent to one single N * 1/300 s exposure. Only different read noise, saturation etc.

If one wants to emulate a particular shutter (e.g. rolling shutter), then surely that can be approximated by choosing what spatial-temporal pixels to include in the averaging, given that the sampling rate is high compared to the target rate.

Really, moving the mechanics of cameras into the digital domain opens up a large toolbox of quality and creativity possibilities. Why should a shutter-exposure be a temporal average, not e.g. a median filter? Why should all parts of the image have the same exposure length?

Link | Posted on Apr 12, 2016 at 07:38 UTC
In reply to:

BostonC: Is this for real?
755mp at 300fps for 14bit images, not including depth and direction of light ray info, will amount to half terabyte/sec.
This requires processor operating at 50GHZ or higher. I don't know.

Processor capabilities are not measured in GHz.

Link | Posted on Apr 12, 2016 at 07:31 UTC
In reply to:

ThatCamFan: The comments here about "nobody needs filters" really hurts my head, what will they say next? "Nobody needs a tripod" "nobody needs manual mode" "nobody needs a camera"

I dont understand why this hurts your head. People are argueing that the function of ND grad filters can often be substituted by higher DR cameras and/or HDR techniques.

This does not mean that nobody really needs filters, but I am guessing that many photographers (being conservative people) are using unnecessarily.

A tripod is needed for long exposures if you want maximum sharpness where you cannot rely on your hands or IS being steady enough. I don't think your analogy works.

Link | Posted on Apr 8, 2016 at 08:52 UTC
In reply to:

Franka T.L.: I wonder what kind of computing capacity is required for these , but judging from the current info it sure do look like its got a very very long way to go before even feasible for any form of application

The paper indicates processing time for 512x512 pixel frames of [75ms, 10s, 75s] for increasing order of algorithmic complexity.

The video seems to use the middle-complexity processing, while the paper contains still-images using all three. Frankly, I think that subjective quality scales poorly with computational complexity in their results.

Link | Posted on Feb 17, 2016 at 08:25 UTC
In reply to:

Henrik Herranen: This is easy:
If you thought a Bayer matrix did bad things to your quantum efficiency by throwing out over 50% of the light, how about leaving 95-99.9% of the light in your pinhole mask? While this system may have its applications, it surely isn't going to replace even the most humble camera phone module. This is a hard constraint no amount of signal processing or other cleverness can overcome.

Another thing is that as much as I appreciate the videos, and even though their nominal resolution were 512x512 pixels, they contain 100 pixels of real resolution at most in either direction, and even that is marred with noise.

All in all, very interesting, but at the moment this looks like a solution looking for its problem.

"the masks employed in some designs have transparent features only in a small central region whose area is invariably much smaller than the area of the sensor. In contrast, almost half of the features (spread across the entire surface) in our mask are transparent. As a consequence, the light throughput of our designs are many orders of magnitude larger as compared to previous designs. "
http://arxiv.org/pdf/1509.00116v2.pdf

Link | Posted on Feb 17, 2016 at 08:17 UTC
On article Opinion: Did Sony just do the impossible? (1075 comments in total)

I am very interested in this camera. I think that it offers (on paper at least) many of the things that I have lost hope that Canon will ever do.

It seems that Sony are at a very rapid release cycle. They launch cameras all over the place, and replace them fairly often. What is their track record on firmware upgrades? I am a bit surprised that they released a 36MP without (or so it seems) a robust solution for camera shake. It seems that the "lossy raw" came as a total surprise on them in the 7rII, even after the web had been filled with questions for years. While Sony seems to have the best sensors out there, they also seems to be a very different supplier than Canon/Nikon, meaning that while switching between C/N is predictable, switching to Sony has some potential pros and cons?

Link | Posted on Jun 23, 2015 at 05:12 UTC as 149th comment
On article CP+ 2015: Canon shows off new EOS 5DS and 5DS R (124 comments in total)
In reply to:

Just a Photographer: More resolution at the cost of image sharpness of everything over f8.0 due to diffraction. People will soon find out that the more megapixels doesn't automatically imply more sharpness and detail.

To start learning about physics and the laws of nature and why diffraction kicks in with this camera. Read this:

http://www.cambridgeincolour.com/tutorials/diffraction-photography.htm

FYI. The Canon 5Ds sensor pixel density is 4.1uM.

@Just a Photographer:
At what point will we hit the theoretical "diffraction wall"? There are estimates of 288 MP - for APS-C (!):
http://www.dpreview.com/forums/post/37510786

"don't start complaining on the web if you find your pictures not being that sharp and detailed after f8.0 as you thought they would be at 50MP!"
Luckily, most of us not that ignorant. We investigate theory and practice before shelling out 4 grand. As my Canon 7D has 4.3 micron sensels compared to the 4.1 micron sensels of the 5Ds, I believe that sharpness loss in the centre of the frame due to diffraction at the same aperture-setting will be comparable on a pixel-to-pixel basis.

"What is the point of owning this camera?"
I'd say that anyone wanting to own digital MF but could not justify the expense (or lack of flexibility), might be interested in this camera. Landscape, portrait, product,... It is no doubt a "specialized" camera, unlike the 5Dmk3 that is more of a swiss army knife.

Link | Posted on Feb 14, 2015 at 15:54 UTC
On article CP+ 2015: Canon shows off new EOS 5DS and 5DS R (124 comments in total)
In reply to:

Horshack: I hope this thread doesn't descend into the typical Canon chatter about dynamic range, with everyone asking dpreview staff about how the camera performs in that regard. To the staff: Would you say there was a wide range of people at the show, with long saturated queues of photographers waiting to raise the 5Ds in their hands to see if there are any surprises lurking in the dark areas of the Canon booth?

@Just a Photographer:
Yes, really. You will find out if you try to read up on the subject.

I probably won't buy the 5Ds, but not due to some home-made faulty theory on resolution that you offer, but because it does not seem to offer the DR that Sony will probably offer in an A9. That camera may even be priced lower.

Link | Posted on Feb 14, 2015 at 13:34 UTC
On article CP+ 2015: Canon shows off new EOS 5DS and 5DS R (124 comments in total)
In reply to:

Just a Photographer: More resolution at the cost of image sharpness of everything over f8.0 due to diffraction. People will soon find out that the more megapixels doesn't automatically imply more sharpness and detail.

To start learning about physics and the laws of nature and why diffraction kicks in with this camera. Read this:

http://www.cambridgeincolour.com/tutorials/diffraction-photography.htm

FYI. The Canon 5Ds sensor pixel density is 4.1uM.

@Just a Photographer:
"Blame yourself for being ignorant and stupid to fall for Canons' marketing"
I am not falling for anything. I am using my engineering education and good references in order to try to understand what is going on. Perhaps you should try something similar? Get an education, or at the very least a physics text book?

"Ignore the 'internet fool' who told you so! "

This is actually the best advice that you have given so far, and that is exactly what I will do. Being able to do a google search does not make you an expert on camera technology, and by now you have been given ample opportunity to explain your misleading statements.

-h

Link | Posted on Feb 14, 2015 at 13:28 UTC
On article CP+ 2015: Canon shows off new EOS 5DS and 5DS R (124 comments in total)
In reply to:

Horshack: I hope this thread doesn't descend into the typical Canon chatter about dynamic range, with everyone asking dpreview staff about how the camera performs in that regard. To the staff: Would you say there was a wide range of people at the show, with long saturated queues of photographers waiting to raise the 5Ds in their hands to see if there are any surprises lurking in the dark areas of the Canon booth?

@Just a photographer:
In fact, every photography, shot at every aperture using any sensel density will have decreased sharpness due to diffraction.

The 5Ds/r should have better resolution at any aperture than the 5Dmk3, everything being equal (i.e. similar AA-filtering etc).

Link | Posted on Feb 14, 2015 at 11:51 UTC
On article CP+ 2015: Canon shows off new EOS 5DS and 5DS R (124 comments in total)
In reply to:

Just a Photographer: More resolution at the cost of image sharpness of everything over f8.0 due to diffraction. People will soon find out that the more megapixels doesn't automatically imply more sharpness and detail.

To start learning about physics and the laws of nature and why diffraction kicks in with this camera. Read this:

http://www.cambridgeincolour.com/tutorials/diffraction-photography.htm

FYI. The Canon 5Ds sensor pixel density is 4.1uM.

@Just a Photographer:
I suggest that you either cut down on the condescending tone or just accept that people over here generally will classify you as just another internet fool.

If you truly are interested in the topic, I suggest that you search the forum for knowledgeable people who have contributed considerably more than "google this and you'll learn". Once you show that you comprehend what has allready been said, you are in a better position to be taken seriously.

Link | Posted on Feb 14, 2015 at 11:49 UTC
In reply to:

hjulenissen: I have the RX100M2. Here is my wish list:
-A simple popup EVF
-The built in flash can be pulled to give indirect flash. Great. Now make it stick so I don't have to keep my finger on it.
-Loose the tiltable LCD screen, as it adds bulk without adding value.
-Loose the NFC and WiFi gimmicks
-I don't need continous zoom. Give me 24 or 28mm equivalent at the current f/4.9 equivalent _and_ 100mm at a slightly larger aperture than the current f/13 equivalent. I don't need the intermediate steps, and if this lets you optimize weight/quality, go for it.
-A touch screen seems natural in a product like this, and the interaction with the GUI screams for one. E.g. touch to focus.
-Use gorilla glass in the LCD
-Please add certain "enthusiast" features to the firmware, such as usable exposure bracketing and flexible limiting of exposure parameters in (semi) automatic exposure modes
-Please let someone skilled in ergonomy have a look at the control ring

All in all, though, I am quite happy with the images that this camera has enabled me to take. Chances are slim that I will update a 1 year old expensive camera for anything new. The only possibility that I can dream of would be a water proof version.

-h

Link | Posted on Apr 29, 2014 at 10:12 UTC

I have the RX100M2. Here is my wish list:
-A simple popup EVF
-The built in flash can be pulled to give indirect flash. Great. Now make it stick so I don't have to keep my finger on it.
-Loose the tiltable LCD screen, as it adds bulk without adding value.
-Loose the NFC and WiFi gimmicks
-I don't need continous zoom. Give me 24 or 28mm equivalent at the current f/4.9 equivalent _and_ 100mm at a slightly larger aperture than the current f/13 equivalent. I don't need the intermediate steps, and if this lets you optimize weight/quality, go for it.
-A touch screen seems natural in a product like this, and the interaction with the GUI screams for one. E.g. touch to focus.
-Use gorilla glass in the LCD
-Please add certain "enthusiast" features to the firmware, such as usable exposure bracketing and flexible limiting of exposure parameters in (semi) automatic exposure modes
-Please let someone skilled in ergonomy have a look at the control ring

Link | Posted on Apr 29, 2014 at 10:11 UTC as 12th comment | 3 replies
In reply to:

forpetessake: Before people start cheering and celebrating, they need to understand that Sony (just like other manufacturers) mislead the ignorant public providing equivalent FLs and conveniently forgetting to convert the f-stops to FF as well. So they would advertise RX-10 lens as 24-200mm f/2.8 equivalent, while it's 24-200mm f/7.6 equivalent.
Same with these lens designs, they provide an equivalent FL and 'forget' to provide an equivalent f-stop.
If you look at the actual equivalent numbers, like the first design is 26-112mm f/5.1-9.0 you'll be a lot less jubilant. The lenses are quite dim, no faster than they were in the compact 35mm film cameras of yesteryear.

@Tapper123

In terms of total amount of light on the sensor for equivalent pictures, you need to scale the aperture.

My primary interest as a photographer is the images, not some parameter crusade. The truth is that my RX100M2@ 10mm f/1.8 compares most sensibly to my Canon 7D@18mm f/2.8.

-h

Link | Posted on Apr 29, 2014 at 09:58 UTC
On article Apple applies for dual-sensor camera patent (71 comments in total)
In reply to:

agentul: Panasonic has been making video cameras with three sensors for years.

But they have shared a common lens, thus are quite different from the illustration

Link | Posted on Mar 26, 2014 at 13:18 UTC
On article Can computer corrections make simple lenses look good? (160 comments in total)
In reply to:

danijel973: This is not really impressive as I duplicated this result with a simple "sharpen" command in Gimp. Also, you can't get more information than you put in, meaning that you can't create detail from blur. You can clarify detail that's already there, but I would always prefer to do it optically to the maximum possible extent, and only then use software to try to go even further. Intentionally designing bad lenses and relying on software to make them mediocre is not a good idea.

>>Also, you can't get more information than you put in,
right
>>meaning that you can't create detail from blur.
wrong

If you encrypt your harddrive, the bits will look like a blurry mess. Given the right algorithm and key, you can have all of the information back, though.

Link | Posted on Oct 1, 2013 at 08:15 UTC
On article Can computer corrections make simple lenses look good? (160 comments in total)
In reply to:

ProfHankD: I've been studying PSFs for several years now. The biggest problem with deconvolution is that the PSFs are not really convolved in the first place -- especially for out-of-focus regions of the image. Still, there's lots one can do with better computational methods; I use genetic algorithms for this sort of thing.

What do you mean by the PSFs "not really convolved"? Does it mean that the idealized model of a (slowly varying) linear convolution does not describe the errors contributed by the lens? If not, what kind of physical process is it?

If you had access to highly detailed info about the lens (e.g. sweep monochromatic light from 400-800nm on a target print of impulses (or wavelets) distributed across the frame and sweep this target from close focus limit towards infinity), how much better could things be? Is it fundamentally a problem of gathering enough data, or is it about finding the right algorithms to apply?

Link | Posted on Oct 1, 2013 at 08:13 UTC
On article Can computer corrections make simple lenses look good? (160 comments in total)

If the lens designers knows that a given lens correction is available, they might be able to "tailormake" a PSF that is easy to correct (no deep zeros, gaussian-like?), rather than a PSF that is as small as possible.

Perhaps that would allow better system-performance for a given cost/size/Weight?

Link | Posted on Oct 1, 2013 at 05:39 UTC as 36th comment | 1 reply
On article Can computer corrections make simple lenses look good? (160 comments in total)
In reply to:

123Mike: I think the example is fake because there details in the "improved" version that do not exist in the "original".

Visual inspection is not sufficient to determine that such examples are fake.

Link | Posted on Oct 1, 2013 at 05:35 UTC
Total: 35, showing: 1 – 20
« First‹ Previous12Next ›Last »