August 27, 2014, 11:24:36 PM

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - Aglet

Pages: 1 [2] 3 4 ... 63
16
Canon General / Re: Canon lens comparible to a 150-500 or 150-600
« on: August 18, 2014, 12:00:57 AM »
I just got a tiny little Olympus 75-300mm for MFT, only weighs about a pound.  Yes, very slow with f/5.6-6.7 but it was only $450 new and, attached to a new EM10, it performs fairly well.  My intent was to use it as a smaller, lighter (and way cheaper) version of the 100-400 L + 60D when I need to travel with less.
So far, it's looking like a strong performer from 75-200mm (~150-400mm equiv), even wide open.  But I think my old L glass is pulling ahead at the long end AND, at least so far, I can get better, more consistent results from the Canon kit as far as sharpness.  I still need to learn how to optimize my use of the MFT system to squeeze the most performance from it as the IBIS may be less effective at that focal length.
As a mirrorless bonus tho, I can AF and then MF while the camera EVF instantly zooms my AF spot (to a level I define) so I can focus on a bird partly hidden by a branch.... Can't do that too readily with an SLR.

OTOH, a good friend is very satisfied with the performance from his 150-600m Tamron on his D7100; finds it quite sharp beyond 400mm but does get noticeably soft between 500 to 600mm.

So, even if CaNikon don't make such a lens at that price point, there are viable options.

17
Photography Technique / APOLLO missions - image inconsistencies
« on: August 17, 2014, 07:36:51 PM »
This seems about the most appropriate forum heading to post this so here goes.

I've often heard of controversy around various aspects of images from Apollo missions.  I'd never taken it too seriously but an interesting old video on youtube ..

www.youtube.com/watch?v=W79mIGx9Ib4

.. caught my attention the other night and I started watching it.
I've only gone thru the first 1.5 hours, it's ~3:40 total.

They present a variety of interesting discontinuities and other inconsistencies which could lend some credence to some of these images being produced in ways that are not congruent with the official story.
Whether differences in lighting or physical geometry, some things just don't look right.

None of this was obvious, when I was a wide-eyed kid watching these events unfold, back in the day on a small B&W TV.  Thinking about it now, that would have been an exceptional feat to accomplish with the technology of the time!  So there's room for doubt.. and deception.

Has there been a good discussion on this topic on this site before?

If not, with all the expertise available here, it should be possible to have a very interesting one.

For convenience' sake, if you comment on this, or any other video you may reference on the subject, please include the video time relevant to your reply.

18
Reviews / Re: DxO reviews Sony A7s: king of low light photography?
« on: August 06, 2014, 01:51:49 PM »
are you saying you disagree, that you can't squeeze 13 effective bits from properly processing a 12 bit data set?

Are you saying that information not captured initially (lost due to clipping at the bottom or top of the range) can be recovered by 'proper processing'?

If so, you may want to have a conversation with those fish you mentioned, which is about as logical as your suggestion.

so you now want to alter the parameters of the argument to include clipped data so that you're not wrong?

doesn't negate the fact that you can obtain somewhat more resolution from an ADC than the number of useful bits of that device.  That's just stat-math.
Sure, you can't recover data clipped at the high end, but it is possible to obtain more data at the low end.

19
Reviews / Re: DxO reviews Sony A7s: king of low light photography?
« on: August 06, 2014, 02:44:44 AM »
So, yes, it's possible to get more bits of data than you have available bits of an ADC.

Sure.  It's possible to upsample an image and get a print that looks good, too.  Doesn't mean the data are real. 

are you saying you disagree, that you can't squeeze 13 effective bits from properly processing a 12 bit data set?

if so, you might want to have a conversation with the engineers at NASA, Texas Instruments, HP/Agilent/name du-jour, Tektronix, National Semiconductor, etc.



Quote
Please go back to shooting with the lens cap on and pushing 4+ stops in post.  That seems to be something you enjoy...

you apparently still like trolling with the same old hook. ;)
your fishin' licence has expired

20
Third Party Manufacturers / Re: DXO uh-oh?
« on: August 06, 2014, 02:29:53 AM »
Quote
Why? Because when you integrate multiple frames, even if you don't even do any kind of dark or bias frame subtraction to remove read noise, your averaging those frames together. Averaging reduces noise. So, let's say you have the option of shooting one ISO 100 shot at 1/10th of a second, or four ISO 400 shots at 1/40th of a second. Integrate the ISO 400 shots, and you reduce noise by averaging. You reduce ALL noise, including deeper shadow read noise. In a Canon camera, ISO 400 has as much DR as ISO 100, so your did not lose anything by doing that, but because you could use a higher shutter speed, in the end, after integration, you gain something.

I'm gonna poke a small hole in this argument, even if i've made some math errors.
Otherwise, I know the basic premise is correct and I agree with you.

Averaging 4 at iso 400 vs 1 at iso 100 will not net as much of a return because the SNR of any sensor also gets worse at higher ISO by a ratio that's mathematically pretty close to the same as the benefit of the stacking, at the same ratio of iso to n images stacked.
Accordian tu Ducks-o-mark.  5.1 dB worth (39.7 -  34.6) on the 5d3, for instance, at 18% gray (screen) (similar for d800 as well)
stacking 4 gives about 6dB of benefit??  then, net 0.9dB improvement, not likely noticeable.
You'd need to stack more images at the same iso to get better results.

However, where it's needed most, in the deeper shadow areas, the SNR difference between 100 iso and 400 iso is much smaller and averaging 4 iso 400s will certainly make an improvement over 1 at iso 100.  This is more a characteristic of Canon's sensors that can be exploited more effectively with this method.

Also, this is only applicable to random noise.
if theres FPN, then averaging can make it worse by reinforcing it unless you also apply a spacial shift to the images you're stacking and then realigning them later.

21
Third Party Manufacturers / Re: DXO uh-oh?
« on: August 05, 2014, 08:41:12 PM »
So now <the D810> certainly IS competition for the 5d3 in more types of shooting.

Yet Nikon IS still predicting greater sales losses than Canon.  Some competition...   ::)

as long as SoNikon sell enough to stay in production and a step ahead I'm OK with that.
I don't own stock in any of them. :P

Edit: typo

22
Reviews / Re: DxO reviews Sony A7s: king of low light photography?
« on: August 05, 2014, 08:38:18 PM »
Nope it's an oversimplified example to illustrate a simple point: When you downscale an image you gain additional information per pixel in the downscaled image (assuming you use a sensible downscaling algorithm). Each pixel in the downscaled image will use information from multiple pixels in the original.  In practice it is really trivial to observe, take a somewhat noisy image and downscale, which is more noisy and thus has less dynamic range: a pixel in the original image or a pixel in the downscaled image? Obviously you don't gain editing latitude for the image as a whole (you loose it), but each pixel individually gain DR.

Gains DR to a point.  There's a ceiling, and that ceiling is the maximum DR at capture.  Sorry, but your suggestion that combining four pixels with 14-bits at capture and getting 16-bits of real data is ludicrous.  But you're probably making dilbert happy as you waft the stench of DxO's BS (aka Biased Scores) around the forums.  Nothing to be proud of, IMO.

It sounds like you're dismissing it out of hand when you really should be saying something else.

I'm sure you can remember a day, back when (are you old enough?), early digital scientific instrumentation often made use of (noise) dithering to improve dynamic range and reduce noise.
I don't care to get into all the math, so I don't know just HOW much effect it can have, but it does work.
In the case of a camera sensor, averaging 4 pixels is gonna buy some extra DR at the expense of spacial resolution.

Read about it here, for those interested:

www.analog.com/library/analogDialogue/archives/40-02/adc_noise.html

EDIT:  adding quote from above referenced article to keep things interesting:

Digital Averaging Increases Resolution and Reduces Noise
The effects of input-referred noise can be reduced by digital averaging. Consider a 16-bit ADC which has 15 noise-free bits at a sampling rate of 100 kSPS. Averaging two measurements of an unchanging signal for each output sample reduces the effective sampling rate to 50 kSPS—and increases the SNR by 3 dB and the number of noise-free bits to 15.5. Averaging four measurements per output sample reduces the sampling rate to 25 kSPS—and increases the SNR by 6 dB and the number of noise-free bits to 16.


Yes, I've played with digital averaging and dithering (both before and – with much less consternation – after MATLAB).  My $300 EOS M does multishot noise reduction.  Woot.

Dismissing out of hand?  No.  Simply defining boundaries beyond which the logic breaks down.  I notice your pasted example indicates signal processing to achieve 16-bits of real data from a 16-bit ADC.  Would you care to provide an example demonstrating signal processing which delivers more than the bit depth of real data initially acquired (as in msm's suggestion of combining 14-bit pixels to achieve data with true 16-bit depth)?

canon's only got 12 bits of data on a 14 bit conversion.
So you're telling me those last 2 bits of noise are a FEATURE by implementing noise dithering? ;)

If only the noise were purely random, it could qualify.  :D  They are getting closer to that.

In fact, it's entirely likely that Canon did go to 14b ADC back then so that noise dither would improve tonal transitions; something I remember being stated in a promo for the 40D in its day.  Altho, "noise" was not mentioned, don't want people to get any negative impressions or customer confusion from that word!

Back in the 80s, I used to get annoyed by articles from math geeks telling us designers how to get an extra half or full bit (or more?) effective resolution from a 12 bit ADC by dithering and processing.  This was the beginning times of digital processing.
I never paid much attention because, for the instruments I was building, I could get ADCs with enough precision and resolution without having to resort to any DSP for a result.
If I need 16 bits, I'll buy 18 or 20 bit converters.  I don't have to worry about cost cutting for consumer scale production.
If noise was an issue, filtering and averaging was easily done with simple software routines the end users preferred to have control of anyway.

see

DITHER

http://en.wikipedia.org/wiki/Analog-to-digital_converter#Dither


and

OVERSAMPLING

http://en.wikipedia.org/wiki/Analog-to-digital_converter#Oversampling


So, yes, it's possible to get more bits of data than you have available bits of an ADC.
The process is similar whether you have a 16b ADC with 2LSB worth of noise or a 14b ADC you dither and oversample.

So, I think there are already some examples that define that the boundary of a 14b ADC is potentially greater than 14 bits.

Again, plenty of astute signal processing math geeks here can explain how that works.
Jristas astrophotography examples, pulling nebula detail from stacks of dozens or hundreds of images, are a good example of how to extract more bits of info from limited ADCs and noise which equates to a terrific amount of DR and effective bits of conversion.

23
Third Party Manufacturers / Re: DXO uh-oh?
« on: August 05, 2014, 06:33:46 PM »
ya, the d810's many little improvements make it a much better all-around camera than the previous 800s.
So now it certainly IS competition for the 5d3 in more types of shooting.

As for Tony's videos... I'm not a fan.

24
Reviews / Re: DxO reviews Sony A7s: king of low light photography?
« on: August 05, 2014, 06:28:01 PM »
Nope it's an oversimplified example to illustrate a simple point: When you downscale an image you gain additional information per pixel in the downscaled image (assuming you use a sensible downscaling algorithm). Each pixel in the downscaled image will use information from multiple pixels in the original.  In practice it is really trivial to observe, take a somewhat noisy image and downscale, which is more noisy and thus has less dynamic range: a pixel in the original image or a pixel in the downscaled image? Obviously you don't gain editing latitude for the image as a whole (you loose it), but each pixel individually gain DR.

Gains DR to a point.  There's a ceiling, and that ceiling is the maximum DR at capture.  Sorry, but your suggestion that combining four pixels with 14-bits at capture and getting 16-bits of real data is ludicrous.  But you're probably making dilbert happy as you waft the stench of DxO's BS (aka Biased Scores) around the forums.  Nothing to be proud of, IMO.

It sounds like you're dismissing it out of hand when you really should be saying something else.

I'm sure you can remember a day, back when (are you old enough?), early digital scientific instrumentation often made use of (noise) dithering to improve dynamic range and reduce noise.
I don't care to get into all the math, so I don't know just HOW much effect it can have, but it does work.
In the case of a camera sensor, averaging 4 pixels is gonna buy some extra DR at the expense of spacial resolution.

Read about it here, for those interested:

www.analog.com/library/analogDialogue/archives/40-02/adc_noise.html

EDIT:  adding quote from above referenced article to keep things interesting:

Digital Averaging Increases Resolution and Reduces Noise
The effects of input-referred noise can be reduced by digital averaging. Consider a 16-bit ADC which has 15 noise-free bits at a sampling rate of 100 kSPS. Averaging two measurements of an unchanging signal for each output sample reduces the effective sampling rate to 50 kSPS—and increases the SNR by 3 dB and the number of noise-free bits to 15.5. Averaging four measurements per output sample reduces the sampling rate to 25 kSPS—and increases the SNR by 6 dB and the number of noise-free bits to 16.


25
Reviews / Re: DxO reviews Sony A7s: king of low light photography?
« on: August 03, 2014, 09:38:55 PM »
I'd be willing to bet every print you've ever had professionally printed came off of an Epson, Canon, or possibly HP printer.

Then there's all the Fuji and Noritsu (and other) printers used by most low-end commercial printers these days, department store type stuff.
i haven't used them since Noritsu wet-labs went by the wayside.  Too bad, I liked the (laser-based?) wet prints i used to get from them once I had the lab dialed in and had them keep the equipment zero'd for my print jobs.

When they went to "dry-lab" operation it was just a commercial sized CMYK printer and the results looked horrible to me.  Low ppi meant things like fine detail and gradations in high key shots were just stochastic dots instead of continuous tone.  Went with a pro-level printing service after that but wasn't convenient for small prints under 12x18.  The big Prographs can do an awesome job when properly set up tho. 
My Epson R3000 outputs satisfactory prints with little effort at home, usually stands up well to loupe inspection.

26
Reviews / Re: DxO reviews Sony A7s: king of low light photography?
« on: August 03, 2014, 04:30:00 PM »
Oh yes. Don't even get me started on printing for different lighting conditions. I use a high CRI 150W CFL bulb to preview my prints with, as it is a neutral sunlight white and very bright, but if the print is going to normally be viewed under say tungsten light, you have to factor that into the preparation and printing.

As for lifting, I guess it depends on how you do it. I tend to just change the black point, which really does tend to "compress" rather than "stretch". When you compress the histogram, you don't usually run into problems with read noise. Your basically just reducing the contrast of the image (on screen...in print it looks different), then shifting the tones around. If you do a normal lightroom-style shadow pull, that is stretching information, rather than compressing it, in which case Sony sensors certainly have an advantage. If all you ever do is print from Lightroom, rather than use Photoshop to prep your images for print, then a better sensor would be useful.
I'll compare those methods next time I have to do some extra tweaking for print.
I used to just create a file that looked good on my display, send it to my printing guru and he returned a piece of media that looked the way I expected it to.
Now that I can make some smaller prints on my own, I'm finding how much of a hassle it can be to obtain that level of perfection!

27
Reviews / Re: DxO reviews Sony A7s: king of low light photography?
« on: August 03, 2014, 04:25:34 PM »
I print hundreds of large canvases and art prints that go all over the world. Never once have I been asked for a print to be 'optimised for viewing in dimmer lighting conditions'.

Perhaps you don't offer the service or your clients are not aware that it's even a possibility
OTOH, your images are more likely to be located as more of an exhibition piece, so lighting is probably adequate and this issue may not come up.

Quote
A picture is there to be viewed - you need to be able to see it - therefore you need to have a reasonable amount of light falling on it.

I've occasionally tailored prints to be viewed under particular lighting conditions.  While i'm no expert in this area, the adjustments I made resulted in a better appearance than the usual mill-run type image when viewed side-by-side in said conditions.

Quote
However aren't you the guy who wants to lift heavily underexposed shadow area by four, five stops or more ? In that case I suggested the dimmer the viewing light the better. Preferably no light at all.

I'm the guy who does what's necessary to create the image I intend.

EDIT - no idea why this reply nested under Chuck's post.

28
Reviews / Re: Pentax 645z Review
« on: July 31, 2014, 08:22:49 PM »
I've never hand-on mf body before. To me, this body looks  :-\
It's a bit heavy but it's actually quite nice handling without a tripod

29
Reviews / Re: Pentax 645z Review
« on: July 31, 2014, 08:21:29 PM »

30
Third Party Manufacturers / Re: DXO uh-oh?
« on: July 29, 2014, 02:49:49 PM »
@zigzagzoe
Thanks for your posts.

There's a few of us here who will heartily agree with most of what you said.  I was a Canon fan until the superior raw files from a $400 Nikon consumer body blew my 5D2 into the weeds!  Well, at that time I was already a Canon fan in decline and searching for better options as I'd realized the 5D2, so hyped by fanboys, was a bit of a turd with very noisy low ISO raw files and a tendency to underexpose; or at least the one I had was.

Your example of the model, whose arm was in deep shadow, and looking like it has a bad rash because of the read noise, is a good one.
When pushing the limits, sure, there are plenty of "work-arounds" to make Canon cameras able to produce the desired shot.  People doing that seem to be forgetting they're compensating for under-performing hardware!  Poor hardware IS a good way to improve your skills as a photographer, as you have to be extra creative to overcome the limitations of your tools.  Reduce the limitations placed on your shooting by those under-performing (Canon) tools and you're now free to benefit from a range of other advantages like less setup time, less time in post, more freedom and flexibility in lighting and exposure, etc.

Ardent fanboys will continue to argue that technique matters more.  It sure does, especially when you chose to use gear with more limitations for certain kinds of shots.

As for the lens argument some make, sure, Canon has some excellent lenses.  But there's also plenty of good glass for Nikon and no matter how bad the Nikon lens may be, it still doesn't change the read noise issue.

My D800/e are my favorite bodies for my kind of shooting yet and I only wish I would have not been so prejudiced in the past so I could have benefitted from using the advantages of other camera brands much earlier.  I haven't purchased a new Canon camera for years now.  But I have purchased Nikon, Pentax, Fuji and Olympus and have enjoyed the benefits they all bring while none of them are as limited by FPN and read-noise issues as Canon. 
Despite the sales and marketing success, Canon is now the underdog, and I actually hope they can perform some sort of a comeback miracle; it might give me one more tool to use.

EDIT:  BTW, I would not be surprised if Nero isn't already pondering and composing an acrid response to your posts.  ;)

Pages: 1 [2] 3 4 ... 63