September 19, 2014, 08:16:30 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] 5 6 ... 66
46
EOS Bodies / Re: Do Sensors Make the Camera?
« on: August 28, 2014, 02:03:48 AM »
The 5D III's banding has actually proven to be pretty difficult to clean up with DeNoise...and it does not seem to be very consistent...the banding changes a bit from frame to frame...as if it is a mix of fixed and semi-random patterns.

When I was studying the nature of the FPN I could find on my 40D there was indeed a fixed pattern noise and another pattern noise structure that was not fixed but varied in location.
I shot some bursts at full speed and when I looked at the noise from those there was a pattern which was the same on all frames and another pattern which traveled vertically from frame to frame yet looked otherwise similar.
I'm guessing these traveling bands of noise patterns are due to high frequency interference from internal circuitry, probably some of the internal switch-mode power supplies.  It could also be ground loop type issues allowing the digital side to contribute noise to the analog side or a few other design/build compromises producing this problem.  Fortunately, my 40D's base noise levels are quite low, allowing me to push shadows quite hard w-o requiring NR but it's still not nearly as clean as a little Nikon D5100.

47
EOS Bodies / Re: Do Sensors sell the Camera?
« on: August 26, 2014, 01:40:27 AM »
Hi,
I know.... it's not like it's hard to do....
I can't think of anything at work that does not use at least 24 bit A/D and we have some test equipment that has 64 bit A/D and others that have 48 bit A/D running at 60Ghz sampling rates... I think that the last time I designed something with only 16 bit A/D was back in the 1980's....
Even though I don't know a lot about electronics I'll go ahead and wade in to water over my head.  While it may be easy to implement, it may have implications for the rest of the body.  Presumably, more precise sampling will draw more power proportional to the increase in precision: more bits of A/D will mean more components, all of which draw power.  Then the digital circuitry all the way from A/D to flash card has to be scaled-up to match which also draws more power.  All of this also generates heat which must be dealt with.  In my sophomoric opinion, this would result in slower framerate and heat issues for the sensor.  And that's not even considering the effect on battery life.  It's not impossible, but it's an extra set of engineering problems which incur greater cost, which affects retail cost and profit.

Also, why jump straight to 20 bit A/D when you can sell 14, then 16, then 17, then 18, then 19, then 20.   8)
    Hmm... Pentax use 22-bits ADC many years ago... on the Pentax K10D, but the raw file is only 12-bits... then Pentax K20D use back the 14-bits ADC... just wonder why they change back to 14-bits??

   Have a nice day.
I have a K-10D.  It's a nice tool but unlikely to make use of 22 bit ADC, the noise floor is too high to matter.
I'm wondering if there was a misread quote somewhere in there (dpreview article) where the ADC may have been a more moderate number of bits but 22 bits worth of processing precision were used?..
That said, I really like the color I can get out of that camera when it comes to subjects like deeply saturated flowers.  That and that I got a mint condition body for peanuts.  ;D

48
haven't seen a camera yet that didn't have, or eventually develop, some hot pixels.
That's what dark frame subtraction's for, even the in-camera process is usable for most shots.

At least SOME mfrs, NOT Canon, NOT Nikon, allow you to map out bad pixels right from the camera menu.

Not true, doing a 'sensor clean' with the cap on for a minute or so allows the camera to remap the sensor, and gone are all the hot pixels  8)

That is indeed a great and easy way to do it! Did that last winter - it was so easy I thought someone was just trying to pull our legs with that tip.
It's not as official as an actual menu-item so can someone describe the method in more detail and I'll test it out.
I have a Rebel that developed a bunch of hot pixels as I was shooting studio shots with it over a matter of minutes.  So I can test that out.

With the lenscap on the lens, go to the camera menu and select 'manual cleaning'. The mirror will flip up and the camera black out. Leave it for a minute, then switch off the camera to complete the 'cleaning'. You can do this with any lens on the camera, no need to use the body cap.

Thanks.
Will do this when I have time and post before and after pix... 
Leaving the lens cap on seems like a funny thing to do...  My other mfr's cameras simply do this by keeping the shutters closed.  Not that it matters...

49
EOS Bodies / Re: Are These The EOS 7D Mark II Specifications?
« on: August 25, 2014, 03:57:02 AM »
edit...
This has NOTHING to do with DXO here, BTW. Just to be very clear. This has everything to do with WHAT PHOTOGRAPHERS ARE ACHIEVING IN REAL LIFEa with the D800. I posted actual real world, artistic photographic examples, not some lab test of a step wedge or a bunch of numbers on paper (things you guys are often ragging on me about) and you guys are STILL denying it. Well...I guess what they say is true. Denial is the most predictable of human behaviors...

Golly Jon, how'd you end up in this scrap?  ;)

here's one of my examples from 2 years ago that's a nice match for your sun-water example.
i could have pushed the foreground even lighter but then it looked phoney, not noisy.
Shots like this, and the extreme ease of post-processing, is why I dumped most of my Canon gear and went with Exmor-based goodies.  i like shooting into the sun!  I don't like spending a lot of time mucking around with NR software if I can easily avoid it.  ... Now I can, since 2012.

www.canonrumors.com/forum/index.php?topic=8105.msg161888#msg161888

and here's another, the shadowed bank and tree-trunks were far too dark in the as-shot image to make the shot look interesting.  Was also the 14mm end of 14-24, FWIW.

www.canonrumors.com/forum/index.php?topic=9082.msg172770#msg172770

to steal the plot line from a commercial...

money spent on changing over, a few $k difference.
effort in learning the new system, mild to moderate

time and hassle avoided, priceless

51
haven't seen a camera yet that didn't have, or eventually develop, some hot pixels.
That's what dark frame subtraction's for, even the in-camera process is usable for most shots.

At least SOME mfrs, NOT Canon, NOT Nikon, allow you to map out bad pixels right from the camera menu.

Not true, doing a 'sensor clean' with the cap on for a minute or so allows the camera to remap the sensor, and gone are all the hot pixels  8)

That is indeed a great and easy way to do it! Did that last winter - it was so easy I thought someone was just trying to pull our legs with that tip.
It's not as official as an actual menu-item so can someone describe the method in more detail and I'll test it out.
I have a Rebel that developed a bunch of hot pixels as I was shooting studio shots with it over a matter of minutes.  So I can test that out.

52
Technical Support / Re: Another my Stupid question = Sensor Sizes
« on: August 24, 2014, 01:51:39 AM »
Pixel size is irrelevant. SNR, and therefor dynamic range (assuming you have no other source of noise than what is inherent to the image signal itself) and noise are ultimately relative to total sensor area. That's it.

Uhmm... except when pixel size is not irrelevant.
I have to disaggree with you, somewhat, on one point; dynamic range will become limited when pixels become too small, and hence their full-well capacity decreases by more than just the ratio of their surface area.
I say this because, I suspect, the vertical dimension of the photodiode will have some aspect ratio limit with regards to the surface area.  When the surface area becomes too small, the other dimension will have to shrink also, and that will iimit the full-capacity/surface area, decreasing maximum DR.  You'll still be able to reduce noise levels quite effectively by binning/averaging, either hardware or software, but you'll reach a lower maximum when the pixel geometry gets too small.
I suspect something like 40MP smartphone camera may be an example.

EDIT:  Actually, we're already there in varying degrees.
Since many sensor systems are already counting individual electrons, smaller pixels are just gonna be DR-limited.  14bits at 1 bit per electron is only 16384 e-
Small pixels are useful even with full well counts well below that, like 2^10, but then that's already a 10-stop or less DR.  When you start averaging them, you're not gonna gain quite all of that DR back.  And then when you hit the aspect ratio limit for the photo-diode, the DR curve will really drop off.
Perhaps a resident math-whiz could graph that curve for a demo.... (nudge, hint-hint ;) )

53
haven't seen a camera yet that didn't have, or eventually develop, some hot pixels.
That's what dark frame subtraction's for, even the in-camera process is usable for most shots.

At least SOME mfrs, NOT Canon, NOT Nikon, allow you to map out bad pixels right from the camera menu.

54
Third Party Manufacturers / Re: D810 users are seeing spots
« on: August 23, 2014, 12:51:29 AM »
Weird. You must have had a dud. Couldn't possibly have been user error.

I'm sure I did get a dud.  Teach me to buy a first run of a new product! (Ooops, i did that again w d800s and XT1, no issues tho except the fuji's light leak)

The 5d2 banding flaw I first noticed in gray midtones and clear blue sky, then discovered I was not the only one complaining about it.
There were 2 sources of vertical banding. One was a FPN read noise with an 8 pixel cycle, like the 7D/60D/etc.

The other was a different issue, and actually the first flaw I'd noticed in the 5D2's smooth midtones.  After looking more closely I found the other noise issues.  Then I was just seriously annoyed!

But, back to the other banding issue.  I've also just discovered it in one of my recently acquired 2nd-hand Fuji XE1s.
About the central third of the sensor is affected.  When doing a dust-check at 200mm and f/22, I found this central region was also inhabited by wide bands of slightly lighter-darker areas, but not perfectly regular in alignment.  When I shot another test w-o any lens, I could no longer see this issue.  So it is some strange sort of interference effect?.. Anyone have a good guess?..  I thot it if was a simple interference it might make radial patterns but it created wider, relatively vertical bands.
When I still had the 5d2 I didn't try a no-lens shot to see if that effect was still there.
If I get some time, I'll post an example from the Fuji, it's very similar.

I suspect it might be minor misalignment of micro-lenses over their pixels producing subtle variations in effective QE.


55
Third Party Manufacturers / Re: D810 users are seeing spots
« on: August 21, 2014, 09:50:48 PM »
I also got my 5DII when it was first released and have never had an issue. Neither have I ever updated the firmware - 'cos I never do  ;)

The 5DII must be remembered as one of the best sorted cameras right from its inception; not a good example to use !

HEHE.. that makes me chuckle. :P

Whilst it may not have impacted you, I know people for whom it did delay their purchasing decision until after it was fixed/resolved.

yup, I remember seeing those xmas light black dot examples and I waited until after a FW update before getting one - but no matter what firmware version I ran on mine, it was a bandy bass terd of a camera.

56
Photography Technique / Re: APOLLO missions - image inconsistencies
« on: August 19, 2014, 03:46:02 PM »
...
And the incredible Eidophor Projectors they used in mission control!!
http://en.wikipedia.org/wiki/Eidophor[/b]

Eidophors used an optical system somewhat similar to a conventional movie projector but substituted a slowly-rotating mirrored disk or dish for the film. The disk was covered with a thin film of high-viscosity transparent oil and through the use of a scanned electron beam, electrostatic charges could be deposited onto the oil, causing the surface of the oil to deform. Light was shone on the disc via a striped mirror consisting of strips of reflective material alternated with transparent non-reflective areas. Areas of the oil unaffected by the electron beam would allow the light to be reflected directly back to the mirror and towards the light source, whereas light passing through deformed areas would be displaced and would pass through the adjacent transparent areas and onwards through the projection system. As the disk rotated, a doctor blade discharged and smoothed the ripples in the oil, readying it for re-use on another television frame.

That is really cool!  I'd never heard of this thing before.

57
The idea of being able to do this is absolutely drool-worthy.
Will be interesting to see how well it performs and how much movement range it will have on a FF.

www.imaging-resource.com/news/2014/08/14/hartblei-introduces-hcam-master-ts-14-24mm-tilt-shift-optic-for-sony-e-moun

Dang!  Sony's the only system I don't carry... yet.

58
EOS Bodies - For Stills / Re: Canon mirrorless: Status?
« on: August 18, 2014, 05:27:27 PM »
I'm intrigued why more fixed-mount lenses with a simple, high quality zoom aren't offered more often.  Right now, the best bet for fixed-mount lens with a small, high quality zoom are some "cheaper" APS-C Leicas or the high-end point and shoots like the Sony RX100 series or the G1X II.  Fuji has the X10, X20 bodies that do this as well, I think...

I think the reason why is manufacturers want lens pullthrough dollars, so the added cost / hassle of making it modular in as many body designs as possible is more profitable in the longer term.  Just guessing, though.

- A

...and there's that great new Panasonic too.
But I think you've nailed it, it's more about profit than making the best possible all-in-one that would meet the needs of 95% of people, 95% of the time.
Build such an ideal camera and you'll only sell a lot of them until you saturate the market, then you'll be out of customers and unable to sustain a robust operation.

59
Photography Technique / Re: APOLLO missions - image inconsistencies
« on: August 18, 2014, 05:21:15 PM »
I'll answer just this one.  The lander is up on legs and next to a slight hill on camera right.  This makes it's shadow appear like it's going a different direction when it's not.  Look carefully at the lander and its legs and you'll see that its shadows are consistent with those in the rest of the image.

http://www.lpi.usra.edu/resources/apollo/images/print/AS14/68/9487.jpg

By the way, this one and many more were debunked by the Mythbusters.  Go watch the episode.

Thanks for the link, certainly changes the appearance and perspective of the image when the R side of it isn't cropped away!  Lots of other great images there to explore too.

This is a gag post, right?  I went to CR and the Onion site breaks out.

HAHA!  Not completely gag. :)
Just looking at poor quality, compromised images and explanations for them leave too much wiggle room.  Combine that with the pressures such a mission would have, from so many angles, that it's conceivable to have also spent a lot of effort to generate some contingency images....  Then again, this was pre-Nixon era... but not by much. ;)

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.

Bull. The 747 was designed from scratch and brought into production in 28 months during that period, without computers. Remember the Concorde?

It was an exceptional feat. That's the kind of thing you get when thousands of dedicated individuals, with extraordinary resources, work for a long time on a common goal.

I agree.  Feats like this are possible when great leaders inspire great minds and many talented/smart individuals to work together to achieve a common goal.  Those 1960's folks did all that with slide rules and an enormous amount of guts, sweat and sacrifice.  (Not to mention copious amounts of coffee and cigarettes!)

Unfortunately, I'm afraid we have mostly forgotten that spirit of self sacrifice and achievement in this country.  Our leaders are largely selfish, corrupt or just too weak to make a difference.   The majority of our youth are now poorly educated, apathetic and self centered.  (And those are the ones that aren't on public assistance programs.)  Our smart and motivated youth are in short supply these days.  It's a terrible shame.  Now more than half of our populace fears the future is less hopeful than the past.   :'(

That's another good point.
Cold war era saw tremendous innovation and expansion of technical capability, really pushing the limits of the technology of the day.
Despite much great technological information and abilities in the small-scale ways we have now, I don't think we've seen anything, in a big-picture technology push, that's comparable to what was done back then.  Lots of great unmanned probe stuff has been accomplished but manned exploration, unfortunately, seems to have faltered due to budget cuts and possibly also a different mindset in general.
The muddling about with the space station and shuttles was a good example of how that became inefficient.

If the moon landing program had not been done back then, and were, for whatever reason, a project of this time, I wonder how smoothly it would progress, how much it would cost, and how long it would take.
Perhaps I need to look at some news from China and India to get an idea.

Pages: 1 2 3 [4] 5 6 ... 66