joseph_albert

To minimize the compromise in image quality compared to your pro gear, why not consider either a Konica-Minolta A200 or Olympus C-8080. The K-M has a longer zoom range (equiv. to 28-200 in 35mm vs. equiv to 28-140) and anti-shake, whereas the C-8080 has a rugged all-metal body, more like a pro camera in build. I was able to buy a 1 month old C8080 and went that route, although the longer focal length capability and anti-shake of the KM was my preference. I'm not sorry I did as I carry the camera with me most days, and it gets knocked around, so the rugged build is worthwhile. My main complaints with the C8080 are the lack of an optical viewfinder (LCD or EVF are teh choices) and the lack of manual zoom capability. Otherwise, I've found it easy to use as the features not easy to control aren't ones I need generally. Changing resolution, ASA, white balance, exposure compensation, and changing modes between aperture/shutter priority or manual mode are all a snap. I generally leave it in aperture priority mode with exposure histogram displayed one the lower area of the LCD, and then set aperture, and adjust the exposure compensation wheel to center the histogram. shutter delay is not objectionable, but time to write a raw file to microdrive is a tad slow. SHQ JPEG is much faster though. my other cameras are all film cameras so I can't compare to other digital, but I get very clean, 8MP images at ISO 50. depth of field is so great with teh small sensor that I can generally handhold at ISO 50, occasionally using ISO 100. ISO 200 and 400 are noisy. I've used the noise reduction reature for long exposures and have done 15 and 30 second exposures to get images with no detectable noise.

If you like to use filters, forget all of the TLRs other than Mamiya. The Bay I filter mount means you can spend 100 euros on a filters pr two at the blink of an eye, and that's just getting started.

<p>

And koni-omegaflex TLRs, while quite cool cameras, are not a good choice since it will take you a very long time to get all the pieces you need (they are uncommon).

<p>

I don't think you can meet your requirements for 100 euros though-- most of these older TLRs require a CLA which will eat up over half of your budget right there. Probably a ricoh diacord for $50 + CLA and don't use filters.

There are some subjects for which a split-image screen is nice, and others for which it just doesn't work at all. For instance, the split-image focusing aid depends on finding a straight line or edge that you can position in the viewfinder so it runs across the split-image prism. It is not uncommon for this to be impossible. Try to focus on a leaf in dense foliage, or try to focus on a rough-textured plaster wall and you'll see that a split-image focusing aid is totally useless for this purpose. But it gets in the way of trying to focus by assessing the ground glass image.

<p>

And I've never figured out how microprism focusing aids are supposed to help. Half the time, the subject focal plane is in sharp focus in the ground glass when the microprism says it is out of focus. I don't know why this is, but I observe it regularly.

<p>

Split-image focusing aids are also questionable for macro work because lightl-loss to bellows factor often puts the effective aperture at a point where half of the split-image aid goes dark.

<p>

Try to focus on the eyes of a portrait subject and you need the camera to be vertical so the split-image assesses horizontal lines, but then if you plan to shoot a horizontal format image you have to re-rotate the camera and it may end up a different distance from the subject if you are shooting handheld, not to mention you'll miss candid shots rotating the camera back and forth.

<p>

a plain matte screen makes the most sense for an AF SLR because in the cases where split-image focusing aid is useful, AF works fine. In situations where AF is problematic, a split-image focusing screen often fails to work as well.

Don't fall into the trap of assuming you have to have a 1-size-fits-all camera for all of your photography. You could pick up a cheaper digital camera for, say, portraits, family snapshots etc. and still use your AE-1 for work that demands a higher quality image. I sold my Nikon film cameras and most of my lenses planning a switch to a D70, but the D70 just didn't cut it to replace a 35mm SLR for the applications for which I use a 35mm SLR. So I picked up a used C-8080 for digital applications, and found a Minolta SRT-101 and 4 lenses for about what I would pay for a single, lower end Nikon film camera to replace the ones I sold. I'll just wait for DSLRs to improve. I also shoot with a Mamiya twin lens outfit and I can't afford a digital camera that can match that quality.

<p>

True, the C-8080 only produces really clean images at ISO 50, but with the tiny sensor, there is so much depth of field, that I rarely need to shoot at a higher ISO even handheld.

<p>

Film cameras can, with the help of a scanner, produce a digital image, but digital cameras can't produce a transparency, nor can they be configured to capture lots of tonal scale in the highlights (that is, you can't put negative film in a digicam, or otherwise do negative capture).

If this format dies, what I think will kill it is that nobody else is using it. Most likely, there won't be many (if any?) 3rd party lenses for these cameras, and photographers may fear investing in lenses for a format that may not keep up with the pack in image quality and whose future is uncertain.

<p>

The sensor size seems to have been chosen based on having reasonable size lenses that have no reduction in image quality moving from center to corner. This is because the lenes for this camera actually have a larger image circle, so that the sensor is capture what we normally think of as the center of the image. You also stop down 2 stops less than 35mm to get the same depth of field, so ISO 400 on 4/3 is like having ISO 1600 with 35mm if you use the same shutter speed. That is, if you hold depth of field and shutter speed fixed, then the 4/3 camera only needs to show less noise at ISO 400 than a "full frame" DSLR at ISO 1600. If course you only need to hold shutter speed fixed in the comparison if you are shooting handheld near the speed limits of handholdability.

<p>

Just like a film shooter might want to have a medium format outfit for highest quality or greatest enlargeability, and 35mm for ease of carrying the gear and convenience in use, a digital shooter might want a larger sensor camera to use in applications one might shoot in medium format film, and a smaller camera like a 4/3 sensor for handheld work-- photojournalists, sports and action photographers, travel and street shooters all could make good use of the format.

<p>

But that doesn't mean they will. Time will tell.

keep in mind that if you double the size of the image (and 35mm sized sensors are linearly 2x the size of 4/3) you need to stop down 2 stops to get the same depth of field. this in turns means you are 2 stops slower in shutter speed to keep exposure correct. The Olympus 4/3 may not be usable above ISO 400, but in terms of shutter speed, that's like using ISO 1600 shutter speeds with 35mm since you will have to stop the 35mm camera down 2 extra stops to get the same DOF. If you don't want to lose 2 stops of shutter speed, you'll need 2 stops faster film instead.

In response to Sean, the calculation you are doing assumes 1 pixel per sensor site even for cameras that only have 1 color per sensor. If one wanted to use jpeg-2000 for this there would be a pre-processing transformation that would pixelize the data generated by the sensor first, and the jpeg-2000 transform would apply to the pixelized data (3 colors per pixel).

<p>

In my original message I didn't mean to imply that JPEG-2000 should be the standard format for losslessly compressed data, just that it's been around for 5 years and the camera industry is still using proprietary compressed image formats.

<p>

Even if the standardized format were created on a computer after upload it would be better than today. If one wants to send a losslessy encoded image to a print business, what is the preferred format? Tiff? Seems pretty huge.

<p>

JPEG-2000 is like a discrete fourier transform over a wavelet basis, instead of the sine/cosine function basis normally used in the fourier transform. Because the wavelets have compact support, it is algorithmically cheaper than an FFT (fast fourier transform), so DSP chips to do it should not be expensive once the demand volume is in place.

Generally, Zeiss users will be more disturbed by having a single-coated Mamiya lens than a single-coated Rolleiflex TLR lens. The differences in image rendition between a Zeiss T* lens and a Mamiya TLR lens are subtle enough that you can hang images of the same genre made with both cameras in the same show without anyone knowing which camera made which.

<p>

the Mamiya TLR lenses with the late, purple coating are the lenses that have the later version shutter as well. This shutter has click stops both the aperture and shutter speed rings, whereas the earlier black Seiko shutter lenses with amber coating only have click stops on the shutter speed dial.

<p>

If you hold a multicoated lens under a light fixture and look at the color rendition of the sequence of reflections, you'll see blue or green in addition to purple and/or amber. Single coated optics sometimes has both purple and amber coatings on different and even on the same elements at times but this is not multicoating.

<p>

The 80/2.8S lens is purported by some to be multicoated. I think it is either single-coated, or has 1 multicoated element and the rest single-coated. When looking at its reflections, you don't see the tell-tale blue or green reflections, but there is a tiny blue dot in the center, which may well result from an internal element that is multicoated. But the outer element certainly is single-coated.

<p>

Technique has a bigger influence of color and contrast than brand of lens. The best 3 mamiya TLR lenses are the 55/4.5, 105/3.5 DS, and 180/4.5 Super, with 4th place for the 80/2.8 S. the 65mm and 250mm lenses are not as sharp as the 55mm and 180mm respectively, and the 135mm lens, while super sharp, is awkward to handle on the camera.

<p>

I personally find hasselblad cameras much less convenient to use. If you are basing a purchase decision solely on image quality, get a Mamiya 7 with 6x7 film area, or a Linhof 6x9 field camera with Rodenstock lenses that likely are the best medium format optics ever made.

To answer Bill's question... I use transparencies to archive my images. I've only acquired a digicam in the last few months (Olympus C-8080), and so far, my impression is that digital infrastructure is a huge mess. I haven't even gotten to the question of what format to use for archiving a digital image because I haven't really intended to archive them. But once I start using the digicam, darned if I don't suddenly have images needed archiving. I still shoot 35mm film for macro and flower photography, and medium format for other things I care about.

<p>

It will be interesting to see how many digitally archived images are no longer readable in 25-30 years, digital encoding is just one variable, media and media format another (8 inch floppies anyone?)

The price range pretty much rules out carbon fiber tripods, although after carrying a metal tripod around long enough you may break down and pony up for a carbon fiber one anyway.

<p>

Slik makes some tripods that add titanium and magnesium in with the aluminum to achieve a lighter weight/strength ratio than aluminum tripods, and these are certainly affordable.

<p>

Since carbon fiber is out of your price range, I'd recommend getting either a Bogen 3001BN or Slik Pro 700 DX (legset only) to keep the legset under $100 so you have enough $ left over for a decent quality head.

<p>

The value on a tripod legset today is probably the Slik Pro 700DX which can be purchased with a head for $85. It weighs 4.85 pounds and is as sturdy and tall as the old version Bogen 3021 (that weighed 5.75 lbs). The bogen 3001BN is $90 and weighs 3.7 pounds. It is shorter, and less sturdy than the 700DX, but still quite usable if you don't have heavy equipment, and it is easier to carry.

<p>

If you are on a really tight budget, just dispense with the idea of a ball head and get a Bogen 3030 pan head for $60. You can get RC2 style anti-twist plates for this head, and it isn't plagued with the lateral movement of the plates that plagues the separately sold Bogen RC2 quick release adapter-- plates really lock in place in the 3030.

I personally prefer panheads to ballheads in terms of usability, but they are awkward to pack and carry due to handles sticking out in all directions. As noted above, FLM, Giottos, Kaiser, and Velbon are the vendors to look to for moderately priced, serviceable ballheads.

Why are digicams still use proprietary image compression and raw image

formats? The JPEG-2000 standard has been out for over 5 years now and

is a standard (and superior) image compression that is lossless. It

would seem that having a standard so that all the digital

infrastructure could work together (software, printers, etc.) without

having to have proprietary software and plugins would be a win for all

concerned. With proprietary formats, owning more than one brand of

digital camera seems like it is an unbelievable disaster to manage the

environment.

I believe photodo only tested lenses while focused at infinity, and the data may not extrapolate well to other focusing distances. Moreover, MTF doesn't address the tradeoff between sharpest possible wide-open performance and smooth bokeh.

<p>

because there are so many dimensions to lens performance (and so many ratings of lenses try to reduce it all to a single, 1-dimensional scale) you really have to try the lens for the types of photography you plan to do with it. Sometimes even the most superlative lens optically handles very poorly for a certain type of photography, and if you miss the shot because you were fighting against the equipment, it doesn't matter how good the lens is.

<p>

Most lenses made by premium optical companies are marvelous. I find how convenient a lens is to use for a type of work is a much bigger predictor of whether I'll get interesting images with it, not whether it is more marvelous than some other marvelous lens.

<i>The Tokina has the speed of the Sigma but isn't quite there optically.</i>

<p>

Did you actually test them side-by-side, or is this just based on reputation, photodo, or other unreliable sources of information?

<p>

You might look at <a href="http://xoomer.virgilio.it/ripolini/100-300.htm">this review</a> of the Tokina lens by a professional nature photographer who rates this lens higher than the nikon 80-200/2.8 and 70-180/4-5.6 micro.

The portrait above shows why I don't like Nikon micro/macro lenses. To me, the bokeh (out of focus background area) is harsh enough to distract from the main subject. With macro shots, the situation is acute since DOF is in short supply-- harsh bokeh often becomes the thing about the photo that leaps out at you instead of the subject. Minolta, Olympus, and Tamron macro lenses have the smoothest bokeh among macro lenses I've used (which does not include Canon or Pentax or Leica or Zeiss).

<p>

Moreover, if you are considering the 55-60mm focal length because the longer lenses are out of your budget, it might be well to re-consider.

Working distance will be in such short supply that these shorter lenses are more awkward to use for nature, flowers, insects etc. Rather they are more suited to copy work. Get a 90mm or 105mm macro made by Vivitar, Tokina, Tamron, or Kiron (or Sigma if you don't mind sigma build quality). Heck, even the new Vivitar 100/3.5 that is made by Cosina and sells for around $125 is reported to be good (I haven't used it).

Two comments:

<p>

1. unless you have a particular application that is pushing some aspect of the performance of one of these lenses to or beyond its limit, it probably doesn't matter which you get-- they're all good. If you need to shoot handheld wide-open and need even illumination to the corners or sharpness to the corners on a 2-D subject, then you probably care which of these lensens you use. Otherwise, finding one in clean condition at a fair price should be your primary objective. Moreover, these lenses are not expensive, so it is by no means a once in a lifetime experience to buy one.

<p>

2. There is alot more to bokeh than number of aperture blades. Bokeh is one of the design tradeoffs in making a lens, that also includes contrast, and wide-open correction of the periphery of the image. Nikon generally designs prime lenses to be as sharp as possible to the corners wide open, and contrasty, at the expense of bokeh. Minolta seems to have made the right choices with the f/1.2 lenses-- you want as even illumination as possible wide open, a contrasty image, but you probably can sacrifice a little corner sharpness wide open, which means it won't be as good for 2-dimensional subjects wide open, but that's ok. DOF is so shallow at f/1.2, that poor bokeh would be a problem for far more images (3-d subjects) than loss of corner sharpness (2-d subjects). A near-circular aperture is more important in fast lenses since they have more shallow DOF apertures where the aperture is activated (wide open the aperture has no effect whether 8 blade or whatever).

Absolutely, you can just stop down more with a larger format to get the same DOF on a print. That's why I was suggesting it was a speed tradeoff not a DOF tradeoff. Normally with larger film formats, camera is on a tripod (for shots where you need DOF) and you just accept the slower shutter speeds. That's why the above ONLY applies to handheld shots (or other shots that require speeds fast enough to freeze action).

<p>

I'm not trying to claim that the smaller sensor digicams are as good as a DSLR across the board by any means, just that there do exist some applications where they may be preferable. When you are shooting handheld near the limits for shutter speed and DOF requirements, a larger sensor may not help much if at all.

<p>

This seems to be in contrast to the situation with film, at least C41 where there are in the last 5-10 years, the appearance of very good 400 and 800 speed C-41 films. You can shoot NHG 400 or 800 in medium format and get better quality than shooting Reala in 35mm, as an example. I suspect that DSLRs are getting close to their MP resolution limits and future enhancements will be in reducing noise, particularly at faster speeds.

<i>

With the D-SLR I can get higher resolution with less DOF or lower resolution with close-to-equal DOF.

</i>

<p>

I assume you mean just when shooting handheld, or maybe I'm missing something and you find this to be the case even with a tripod available?

Thanks, that is helpful. It seems that the optimal point will vary with the DOF requirements so there won't be 1 minimax point for every image.

The content I posted originally was something I noted over the weekend when I was using an Olympus C8080 to shoot a miniature golf birthday party in a photojournalistic style and I concluded that the niche for these 2/3" sensor cameras is shooting handheld with high DOF requirements. With a DSLR with larger sensor, shooting handheld and needing DOF means a faster ISO setting, and the image quality difference is at least narrowed if not nullified.

No, actually, it is the sensor size and not the lens. At fixed magnification, DOF is independent of focal length. With a smaller sensor, you are working at a lower magnification to get a given composition, and hence DOF is greater.

<p>

I do understand that a limitation of small sensors is their ability to get shallow DOF (unless you could build an f/0.1 lens or somesuch, which may even be technically impossible, aside from impractical). I shoot medium format when I want the freedom of shallow DOF.

<p>

However, the idea that a larger format requires stopping down more and hence gets better image quality from the lens is not clear. It is easier to build a lens with more modest coverage, which could translate into a sharper lens for the smaller format. It is MUCH easier to make an f/2.8 lens for a smaller format than a larger format (that's one reason lenses for larger formats are generally slower). Also, stopping down increases diffraction effects that limit resolution, so it is by no means clearcut that a lens for larger format is sharper because it is stopped down more. In fact, I don't know if I've ever taken a photo I considered flawed because of lack of resolution, despite the intense scrutiny photographers often place on this aspect of a photosystem's imaging capabilities. With film, grain, and not resolution is usually the limiting factor in enlargeability. When digital images are not acceptable from a quality point of view, it is color fringing, moire patterns and noise that is the problem.

<p>

Thus, my question really just concerns relative noise of, say, a 2/3" sensor at, say, ISO 50, and a full-frame sensor at, say, ISO 800. Anyone have any actual data or subjective observations concerning such a comparison?

<i>

No one has EVER built a true interchangeable lens leaf-shutter system for 35mm.

</i><p>

<a href = "http://www.photoethnography.com/ClassicCameras/index-frameset.html?ZeissContaflexIII.html~mainFrame">

Zeiss-Ikon Contaflex III and Contaflex IV</a> were 35mm SLRs that took lenses in leaf-shutters.

It has been articulated here that sensor size places a lower bound on

the noise in a digitally captured image. My understanding of the

problem is that smaller sensors have smaller pixel sensors and smaller

pixel sensors produce a weaker photoelectric effect, and thus need

greater amplification to get a read of the image. The greater

amplification also amplifies dark noise more than less amplification

would. (If that is an imcomplete or incorrect understanding, I'm sure

someone will jump in to clarify :-)

<p>

However, smaller sensors also have more depth of field, which means

that you can get more DOF with a wider aperture, which in turn means

that at a given shutter speed you can shoot at a slower ASA setting.

This tradeoff is often well understood by medium and large format

users who are used to thinking in terms of losing shutter speed in

comparison to a smaller format after stopping down more to maintain

desired depth of field.

<p>

It seems to be universal that digital cameras have noisier images as

ISO speeds get faster. So there are tradeoffs here-- larger sensors

produce less noise at a given ISO speed, but smaller sensors can shoot

at slower ISO speeds while maintaining the same shutter speed and

depth of field as a larger sensor, and slower ISO speeds means less

noise.

<p>

This seems to suggest an interesting minimax problem to find the

optimal sensor size for handheld shooting. Of course, if you have a

tripod, you have the luxury of shutter speeds as slow as desired in

most cases (subjects that won't hold still being an exception in the

absence of flash to freeze the action).

<p>

To illuminate the problem further, let's compare a full-frame (ie 35mm

sized) DSLR with one of the non-interchangeable lens cameras with

sensors 1/4 as large in linear dimension as the full-frame 35mm

camera. Examples of the latter camera today would be Konica-Minolta

A200, Olympus C-8080, Nikon Coolpix 8700 etc., ie cameras with the

so-called 2/3" sensor. These are 8MP cameras, but I don't know if

there are any 8MP full-frame DSLRs around.

<p>

Now, if you increase format size by 4x in linear dimension, then to

frame the same shot you need a lens that is 4x longer in focal length,

and you need top stop down the aperture an additional 4 stops to get

the same depth of field. To compensate for the 4 stop closing of the

aperture, you need to regain 4 stops via shutter speed or ISO setting.

Let's suppose we are shooting handheld and shutter speed has to

remain fixed. If you are shooting at ISO 50 on the all-in-one

digicam, you are now shooting at ISO 800 on the full-frame DSLR.

<p>

This raises some interesting questions: is ISO 50 on a 2/3" sensor

noisier, less noisy of similar to ISO 800 on a full-frame DSLR? for

handheld shooting, is larger always better in terms of image quality

or is there a minimax point where the two tradeoffs give most

favorable results? if so, where is this minimax point?

<p>

Thoughts?

I noted that the stop-down button on the lens doesn't work either-- doesn't seem to have the internal linkage to the disphragm lever. I suspect this linkage was taken out at some point and that is why the lens has the very slight play in it at full extension. The price was US$ 99, so I may decide to keep it anyway if test images are ok.

The Kiron-made wide angle lenses are of high optical quality, have rugged build quality, and generally are priced lower than similar lenses bearing the same brand as your camera.

<p>

However, there are a couple of things you have to live with. Generally, they are a little less resistant to flare than, say, a Minolta wide angle prime lens. They also generally produce images that are noticeably cooler in color balance than a Nikon or Minolta lens as examples. If these limitations aren't a problem for you, then these lenses represent excellent value. The 24/2 and 28/2 lenses are very sharp by f/4, and usable at f/2.8. The 24/2 and 28/2 also have 55m filter threads, which is convenient for vintage Minolta users. I once owned an M42 screw mount Vivitar 28/2.5, most likely the same lens as this as noted above, and while image quality was good, it is heavy and required 67mm filters if I remember correctly, making it cumbersome to use.

Contrast, snappy colors, and flare control are the things you want to be looking for in a 200/3.5 unless you shoot subjects that are unusual for a 200/3.5 such as 2-dimensional subjects or subjects where the point of focus is on material in the corner of the image. This is because if you focus on the central 2/3 of the image, then corner sharpness is totally irrelevant unless your subjects is 2-dimensional-- the depth of field isn't even within the range of acceptable focus.

 

Also, keep in mind that a an f/3.5 lens is stopped down a little at f/4 so it may have a little better corners already.

 

Lastly, these evaluations are being done with slow sharp slide film, right? if you are comparing negatives, then by the time the defects caused by the enlarging lens and print paper or scanning is entered into the equation, resolution performance gaps between different lenses is narrowed considerably.

Thanks for the reply. It is very slight-- less than a nikon autofocus lens, which have more play than I've ever seen in a manual focus lens. I guess I was hoping someone who actually owns the 100/3.5 minolta macro might comment about whether there is this very very slight play normally with this lens or not.