Ad
Ad
Ad
Pages: « 1 ... 3 4 [5] 6 »   Bottom of Page
Print
Author Topic: The Missing Factor in Sony F828 Reviews  (Read 23344 times)
BJL
Sr. Member
****
Offline Offline

Posts: 5121


« Reply #80 on: February 09, 2004, 01:01:39 PM »
ReplyReply

Quote
So, ae you saying that smaller isn't better? Maybe bigger could be better?
Howard,

  it should be clear that my consistent opinion is not a mindless, partisan "smaller is better" or "bigger is better", but a weighing of advantages in each direction. Specifically, I have repeatedly argued that (i) larger sensors have some clear advantages in ultimate attainable image detail, when enough light is available (with digital as surely as with film), but that (ii) the advantages have often been misinterpreted and overstated, particularly by ignoring depth of field needs, increased thermal noise in the longer exposures typically needed with larger formats, etc., and  (iii) the cost and weight savings of smaller formats can in practice improve image quality for most photographers, by allowing them to buy and carry more and better quality lenses.

  When I weigh the factors, instead of looking at arguments on one side only, I come to the same conclusion as is indicated by the actions of every camera and sensor maker: (a) a sensor format of 2/3" or smaller is optimal for the great majority of photographers, including the moderately serious ones now being targeted by new 8MP, 2/3" models from no less than four major manufacturers, and ( even within the far smaller SLR market, the optimal sensor size will be in the smaller 4/3" to "1.5x" range. (I also expect that this severe numerical disadvantage in demand will therefore keep the economies of scale poor for larger formats like 35mm DSLR, so that they will probably never drop below the price range of professionals and very serious amateurs.)
Logged
BJL
Sr. Member
****
Offline Offline

Posts: 5121


« Reply #81 on: February 09, 2004, 03:08:59 PM »
ReplyReply

Quote
something I forgot... 18,000 photons resulting from a bright (isotrope) reflecting white surface.

Georg
Meaning 100% reflectivity? I hope so, because then our numbers are in agreement: for standard 18% reflectivity the number becomes about 3300 photons and so about 1000 electrons, or a bit over 100 electrons per square micron, matching close enough to the 92 eletrons per square micron from E-1 sensor data.

  This has a consequence for maximum attainable sensitivity (ISO speed) based on adequate shadow region quality: to get a decent minimal S/N ratio of 10:1 (a Kodak recomendation) in shadow areas needs at least 100 electrons, and for 3 square micron pixels at ISO 100, it seems that this would occur a bit more than three stops below "18% gray", about the low limit aimed at for decent shadow detail. So about ISO100 might be the upper speed limit for 3micron photosites, ISO400 for 6 square microns, ISO900 for the 9 microns of current high end sensors, and so on. (Higher ISO speeds on DSLR's must be "elecronic push processing", lower values are possibly "pull processing", with better shadow handling than ISO film speed standards require.)

P. S. and at most about ISO 80 for the 2.7 square micron photosites of Sony's 8MP 2/3" sensor, fairly close to its observed behaviour given how crude this reckoning is!
Logged
Ray
Guest
« Reply #82 on: February 09, 2004, 05:39:55 PM »
ReplyReply

Look, you guys! I sense a deep confusion amongst at least some of you (not naming anyone  ) and the seeds of this confusion have been sown by the Sony marketing department. The confusion lies in the nomenclature.

The Sony F828 is either a 7-50mm, F2, ISO 50 (64?) camera, or it's a 28-200mm, F8, ISO 800 camera, but not both. You can't have your cake and eat it. Unfortunately, that's what most of us like to do (have our cake and eat it) and Sony have capitalised on this, as do all other makers of small format cameras. (Can't blame them, I suppose.)

As a consequence of this mixing of terms, according to how attractive the numbers will appear to the consumer, there's a widespread impression that small format cameras are 'necessarily' low resolution and noisy.

It's this misconception that I'm questioning in this thread. Small cameras are noisy, not because they are small but because they are effectively (in 35mm terms) very high ISO cameras.
Logged
BJL
Sr. Member
****
Offline Offline

Posts: 5121


« Reply #83 on: February 18, 2004, 12:29:00 PM »
ReplyReply

Quote
Ray, I didn't say Darwin became a Christian. He simply realized that his theory of evolution was fatllly flawed - was not correct. One does not have to be a Christain to beleve Darwin was wrong.
We are getting way off topic, but I think it is more to the point to say that one does not have to be an "evolution denier" to be a Christian (for example, the Pope isn't).
    Stories of Darwin deciding that his theory of evolution was fatally flawed are nonsense: he knew that there were details needing to be explained, some of which relied on the discrete nature of genetics and inheritence discovered later by Mendel, but "unfinished" is very different from "wrong".

   Anyway, attacking the credibility an idea on the basis of the alleged opinions of its original author is an "authoritarian" approach, maybe appropriate to religion but certainly not to science: in science, and all rational debate, the totality of informed assessment of the evidence and arguments over time is what counts, not the opinion of any one person.
Logged
Ray
Guest
« Reply #84 on: February 18, 2004, 06:54:33 PM »
ReplyReply

Quote
So what about the 1 in 1000? Well, suppose an organism evolved a complete eye on its forehead. A pretty neat trick in itself since the eye is a rather complex organ. But unless that eye also has an optical nerve and a site in its brain to process the information and understand what it is seeing, the new eye is worthless. The mutated eye will disappear in futre generations. The entire optical system would have to evolve in one step. Never happen. It is similar to a tornado going through a junk yard and leaving a fully assembled CAnon 1Ds with a 17-40mm zoom in its path.
Since no-one is listening except Digi-T, I suppose we can have a quiet conversation amongst ourselves, and after all, the eye is related to photography  Cheesy .

Howard, with all due respect, you're quoting the most obvious of religious propaganda. Any Jehovah Witness manual will use that argument to debunk Evolutionism.

The fact is, nobody believes that's how Evolution works. Mutations play a role, but complex arganisms like the eye, evolve very, very slowly over hundreds of millions of years and the stages of such evolution can be seen in the enormously wide variation in the sophistication of the eye in various creatures, from half blind worms and eyes so primitive that we'd hardly recognise them as eyes, to our own marvelous organs.
Logged
Howard Smith
Guest
« Reply #85 on: February 19, 2004, 10:44:05 PM »
ReplyReply

For information on the moth myth, see "Icons of Evolution" by Jonathan Wells, or maybe a newspaper article by Jonathan Wells.  I don't remember exactly which, but it was Jonathan Wells.

I got the idea you were an atheist from your post on Rejecting Yosemite, January 30, 2003.  You said, "I am an Atheist."  Atheism is quite foolish.  Agnostic is wimpish.

An agnostic believes there can be no proof of the existance of God (and I believe that too) but does not deny he possibility that God exists.  An atheist denys the existance of God.  I claim that the atheist cannot prove his position, but must take that position on faith - just like a believer in God.

Some Christians do believe in evolution.  While that belief is not supported by the Bible, I don't think it disqualifies one from being a Christian.  However, not taking the Bible seriously can be very dangerous.  It is easy to be deceived when you seek explainations from the secular.  Ray, keep an open mind to the exisance of God.  Just look around you next time you are out photographing God's creations.  It is truely wonderful and could not have happened by chance.
Logged
Ray
Guest
« Reply #86 on: February 20, 2004, 08:12:22 AM »
ReplyReply

Quote
Instead of using the 256 micro sensors, which are still analog by design, just use one and use it to count photon strikes via a digital counter capable of counting up to the 15000/3 layer = 5000 photons expected for a bright day or more to add extra dynamic range. This would be a 13 bit counter giving 8192 levels ov brightness. Then you would have a 100% digital system with unlimited dynamic range, depending on the max value possible in the counter. The digital value in the counter would then be read out after the exposure. Speed of operation of the counter would be key to its possibility and also squeezing the small detector and counter registers into a 4u square area. This is unlikely at the pressent but who knows what the nano technologists may be able to do on silicon in the not to distant future!
Sounds a lot less cumbersome than my idea  Cheesy . But I really haven't much idea how practical such an arrangement would be. How big are digital photon counters and how fast can they count? I tend to think they're rather large instruments and/or would be easily overwhelmed by large amounts of light. I mean, those photons are not arriving in a nice orderly manner, one after the other  Smiley .
Logged
Lin Evans
Sr. Member
****
Offline Offline

Posts: 269


WWW
« Reply #87 on: February 03, 2004, 06:45:54 AM »
ReplyReply

Quote
Thanks for posting that vibrant and spectacular shot of the painting. All we need now is a comparison between the F828 at f8 and the 10D at f13, of a highly detailed subject with the exposure the same in both cases.

I'll see if I can set one up, but I can assure you that there will be no comparison in depth of field. I frequently shoot art of this type with my 1DS, and sometimes with my 1D and 10D and I can't get the DOF with any of my dSLR's that I get with the F828. Typically I shoot at between F11 and F16, but I would need to go to F32 to get equivalent DOF and the sharpness just isn't there at the super small apertures.
Logged

Lin
georgj
Newbie
*
Offline Offline

Posts: 6


WWW
« Reply #88 on: February 10, 2004, 03:06:50 PM »
ReplyReply

Hi BJL,

you're right with the intervals mentioned based on a very high accuracy for finding the respective intensity class; I've allowed a certain overlap resulting in ~460 steps on the basis of a better than 2/3 chance for identifying the correct part of the interval (the mean error equals sqr(n) but is not equally distributed).

Georg
Logged
Howard Smith
Guest
« Reply #89 on: February 19, 2004, 05:40:01 PM »
ReplyReply

Ray, I certainly don't know you are right.

For natural selection to work, the mutation must be beneficial or attractive for the organism - give it an edge.  Given the time available to "test" the mutation and the fact that most mutations are fatal or non-beneficial, seems impossibe for evolution to work.  You know that.
Logged
Guest
« Reply #90 on: February 20, 2004, 09:23:24 AM »
ReplyReply

This thread has really strayed from its origins. And since we are now firmly treading in the area of religion, and this is a photography-related site, I'd like to suggest that this discuss either be continued in private or dropped.

Like at the dinner table, religion, sex and politics are best avoided here.

Michael
Logged
Laco
Guest
« Reply #91 on: January 23, 2004, 02:56:10 AM »
ReplyReply

Lenses have normally their best performance at 2-3 f-stops from being wide open. After that their performance slightly decreases.
Logged
Howard Smith
Guest
« Reply #92 on: January 26, 2004, 05:06:58 PM »
ReplyReply

Depth of field is a function of f/stop, focus distance, and the acceptable circle of confusion size.  Period.  It has nothing to do with film/digital/format.  Again I suggest you read Michael's tutorial on Understanding Depth of Field.
Logged
Howard Smith
Guest
« Reply #93 on: January 26, 2004, 10:40:12 PM »
ReplyReply

Ray, I re-read your last post.  But don't forget, if you reduce the focl length of the ense to get great depth of field, youalso reude the image size.  The image will require greater elargement to et back to the same size as before you chose a new lens.  That greater enlargement will increase the size of the circle of confusion on the print.  So what you gained in decreasing the focal lngth is lost in restoing the image size.

The depth of field is doubled if the f/number is doubled.  Going from 2.8 to 5.6 doubles the depth of field.

Depth of field changes with the square of the focal distance.  Change the focal distance by a factor 2 and depth of field changed by a factor of 4.  But keep in mind the image size also changes.

Depth of field also changes inversely to the squaure of the focal length.  Ah ha,you say.  Depth of field is a function of focal length.  But the missing factor is the "negative" must be enlarged more times to produce an equal image size on the print.

When all things are  taken togeter, the focal length doesn't matter when the image size is constant.  (Don't forget that a circle of confusion is an image too.)  To get the same image size on the film or sensor, when you reduce the focal length by a factor of 2, you must move closer by a factor of 2 to get the same size image.  Or enlarge the shorter focal length image more to get the same image size.

But instead of decreasing the focal length, I can pick up my long lens and move way back.  If I increase the focus distance by a factor of 2, I will increase the depth on field by a factor of 4 because I have made the image and the circle of confusion smaller.  Same thing happens if I switc to wide angle lense and then move up to make the image bigger.
Logged
Howard Smith
Guest
« Reply #94 on: January 28, 2004, 07:33:43 PM »
ReplyReply

I suppose the photodo experiment could have "leveled the field" by keeping the shutter speed about constant and using T-Max 100 for the 35mm and Tri-X for the 4x5.  I don't might leveling the field as long as it makes it level.

"Luminous Landscape" sot of implies the subjet and envionment we may be interested in.  Normally, cameras will be tripod mounted and shutter spedd will not be much of an issue.  Yes, I know there may exceptions, but let's think "normally."

If smaller is better, it strikes me as odd that folks much smarter than I (the Canon and Nikon pros for starters) are staying up nights trying to make bigger sensors.  What are they thinking?

Call me old fashiond (I am), but when I think 30x40 print, I think 4x5 (or bigger if I had it).  I don't think 2/3" digital or even full frame 35mm digital.  If I think small enough, then my Sony 717 is plenty good enough.
Logged
Howard Smith
Guest
« Reply #95 on: January 29, 2004, 10:56:12 AM »
ReplyReply

BJL, I would point out that except for Pentax and Fuji, I don't think any camera maker you mentioned is a player in the medium format market.  Nikon does make large format lenses, but no cameras.  So it may be not so surprising that the makers you named are working in the "35mm" market place.

I am old fashion and still prefer film to digital.  So what?  I was maybe the last person to buy a CD player.  However, I do recognize that digital is the "wave of the future," for now.  I'm not so narrow minded to think the current digital technology is the ultimate.  Someday, and maybe not too long from now, there will be a camera that will make you want to throw you digital stuff away (same way many have done with film).  Film has been with us longer than the airplane.  Within my life time, man has gone to the moon with a medium format film camera.  Digital is a baby, and a remarkable baby at that.  But just as film was not the ultimate, I doubt digital is either.  It may not have the life expectancy of film.  When man makes that next giant leap foreward and steps onto Mars, I doubt he will be armed with a 6x6 film Hassy.  And maybe not even 100 megapixal Minox.
Logged
Ray
Guest
« Reply #96 on: January 30, 2004, 06:35:50 PM »
ReplyReply

Quote
Even if such a sensor were perfect, the resulting images would be no better than the lens that formed it.

It does not make economical sense for one technology to progress much beyond the other.
Howard,
Can I refer you to the following quote from Norman Koren's site?

Quote
The total detail a lens can resolve at large apertures, where performance is aberration-limited, is relatively independent of format. It is a function of lens quality and design. A good lens can resolve about the same detail at f/5.6 for 35mm as for 4x5, where the image is much larger, but 4x5 images will have more detail because 35mm images are limited by film resolution.

The crucial point here is, it's the lack of quality of film that gives the larger format its advantage. The closer you get to creating a 'perfect' film, the smaller the advantage of the larger format, at least in terms of noise and resolution. Even without any further improvement in lens design and construction, the 'perfect' film would be a worthwhile thing to strive for.

Of course, we'll never get there and we'll probably never need to get there. The questions is, how close to the 'perfect' film do we need to get, to make large format obsolete for all purposes other than creating shallow DoF?
Logged
Ray
Guest
« Reply #97 on: February 01, 2004, 06:50:24 AM »
ReplyReply

To further illustrate this point, the Olympus E-1 would be an excellent example because the Zuiko lenses appear to be so much better than the rather noisy and low resolution 5MP sensor.

DESIGN BRIEF: A sensor for the Olympus E-1 capable of delivering image quality as good or better than that from any camera in existence, including the 22MP Leaf Valeo, 8x10 field camera or 6x17cm panorama camera.

(1) Foveon type sensor consisting of 3 layers, one layer for each primary colour.

(2) Photodetector size - 4 micron, resulting in a total of 15 true megapixels with no interpolation (sensor size 18mmx13.5mm - hope my maths is correct).

(3) Each 4 micron detector divided into 16x16 sub-detectors (total 256).

(4) Each sub-detector of 0.25 micron diameter to have a different sensitivity so that each main detector will have a discrete numerical value ranging from 0-256 depending on the intensity of light passing through the microlens.

(5) Each sub-detector will be either 'on' or 'off' and the most sensitive of the sub-detectors will be switched on by the minimum number of photons that are necessary to rise above the system noise.

(6) The minimum number of photons required to switch on the most sensitive of the sub-detectors will be directly related to the camera's optimum ISO setting.

(7) Higher ISO settings will result in less than 24 bit colour, but no noise.

Okay! I've designed it. You engineers now build it!  Cheesy .
Logged
Howard Smith
Guest
« Reply #98 on: February 03, 2004, 08:37:54 AM »
ReplyReply

Ray, even Darwin recognized that his theory of evolution was fatally flawed.  It's those that don't look deepr that continue to hang on to such notions, usually to prove one of their own view points.

You seem to think the cost of medium and large format is too high.  For whom?  You perhaps, but there does seem to a niche market out there.  In fact, I ind large format film to be not that expensive, especially for the results.  The cost of a large format lens is right in there with Canon 35mm lenses, or less.  A camera body is a few hundred dollars to a few thousand for top of the ine - much like 35mm.

Yes, I have a lot of money tied up in medium format.  But I've had the same system for almost 20 years, still works as good as ever.  I watch the turn over in 35mm and digital, and maybe a top quality medium format system isn't all that much.

Now, if you happen to be a photographer who doesn't require the latest in electronic do-everything for you but remove the lens cap and aim, even 35mm can be inexpensive.
Logged
BJL
Sr. Member
****
Offline Offline

Posts: 5121


« Reply #99 on: February 04, 2004, 05:12:09 PM »
ReplyReply

Quote
As I understand it, there are three broad types of noise; photon noise; dark noise and read noise. It's the additional read noise (the process of quantifying the electronic signal) which might be unavoidable if one substitutes fewer bigger pixels for a lot of smaller ones.
Ray,
  you make an important point: perhaps you read the Roper site more thoroughly that I did! I only considered the main sources of noise that affect the electron counts recorded in the wells of the sensor, photon noise and dark current noise, but not read noise from the subsequent analogue amplification before A/D conversion.
  Perhaps it is time to add one mathematical fact; for constant exposure time, aperture ratio, and sensor size, photon noise and dark current noise scale with the square root of photosite area, and so does discretization error in the A/D converter. Any noise source that scales with this square root behaviour is cancelled by binning or downsampling, because the S/N of combined data decreases in proportion to the square root of the number of values combined.

  But it is possible that as photosite size increases, the read noise increases more slowly than this square root pattern, and if so, increasing photosite size can give a visible benefit. In this situation, increasing site size enough would lead to read noise becoming distinctly smaller than the other sources, so noise would then be back to roughly "square root" scaling, at which point further site size increases would have no further benefit for image quality.

  So considerations of read noise (and that "well fraction" issue) will indicate a natural site size, below which noise problems get noticably worse; above which there is no significant benefit. Clearly this size limit decreases with time, as chip feature size continues to shrink and amplifiers improve, and that trend probably underlies the steady shrinkage of site size across the whole spectrum of digital sensors, with 9microns the current upper limit even in US$30,000+ digital backs.

  I speculate that Sony might have pushed a bit beyond this limit with the 2.7micron pixel pitch of its 8MP 2/3" format sensor, on the slender evicence that, according to DPReview, the 5MP downsampled output option of the Sony 828 has higher noise than the native 5MP output of the Sony 717.
Logged
Pages: « 1 ... 3 4 [5] 6 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad