Digital SLRs?

[soonami]

no AF motor drive, so that's a no for me.

The tilting screen is a nice feature, but other than that, I can't think of a reason not to get the D90 instead

[aizan]

ugliest dslr i've ever seen. olympus does a much better job with the whole movable screen thing.

[soonami]

Well idk abt ugliest ever, but it awfully bumpy, humpy and bludgy!

[revoked]

no AF motor drive, so that's a no for me.

The tilting screen is a nice feature, but other than that, I can't think of a reason not to get the D90 instead

might as well get a d40/d60 for no AF motor drive if you don't have the $$ for lenses + then move on to d90 or above once you can start affording to buy different lenses + flashes

[soonami]

I have a D40 now and the lack of AF motor on the body is irritating since I have a couple primes from film days like 50 1.8 and 35 2.0 and some zooms like 28-105 3.5-4.5 that I can use on the digital body but need to manual focus which sucks especially in low-light because it's often too dark.

I was thinking of upgrading soon depending on d5000 specs, but its pretty clear that for me, d90 is the way to go

[aizan]

sell your outdated screwdrive primes and buy the new af-s 35/1.8 dx...

[soonami]

I have a 35 1.8, but when I carry both bodies it's dumb to carry two lenses of the same focal length. And really, DX is bound to be outdated sooner than "full-frame" lenses once an FX DSLR becomes affordable. Maybe you will be happy shooting crop forever, but me I want super-wide perspectives, more detail, and better low light performance.

[aizan]

you have two d40s and the 35/1.8 dx? why didn't you say so!

anyhow, aps-c will be around for a long time. ff dslrs will either never make it into the entry level, or at least take another decade to do so. by then, all legacy ff primes will have been replaced with ones that have af-s and better optics. sensors will also improve, and it's not only possible, but likely that tomorrow's aps-c sensors will better today's ff sensors, and so on. the bottom will fall out of the used legacy lens market, so you might as well cash in your chips.

there are superwide zooms for aps-c, too. nikon just announced a new one along with the d5000: 10-24mm f3.5-4.5.

http://nikon.com/about/news/2009/0414_nikkor_02.htm

[soonami]

Right now, I have two bodies: D40, N80 and bunch of lenses: 18-55 DX, 55-200VR, 28-105 3.5-4.5D, 50/1.8D, and just got the 35 1.8 DX. I'm in the market for a more advanced body and a better normal zoom.

All of the DX sensors in entry level cameras are derived from their pro bodies before it (d60 sensor from D2). Now that the D3 and the D700 are 35mm equivalent, it will not be long until a top prosumer model (d400 model maybe) will have a full-frame sensor, which will then trickle down to amateur bodies.

The bottom will not fall out of "legacy" lenses because many of these lenses are made in Japan, out of solid metal and constructed to withstand the test of time. Why do you think that the D3X can meter an AI/pre-AF lens? These lenses are built well and have very good optics, but maybe not the bells and whistles of current lenses like AF-S or VR. However, for most people and situations, they are more than competent.

Yes, Nikon just released the 10-24mm, but other than the Tokina, not much else exists on DX that are that wide.

I think that over time sensors will become better and more efficient at collecting light allowing them to be bigger and better without much sacrifice in weight, cost, and efficiency. However, I think there will be a limit as to how big a sensor can get, but I think it is much more realistic that 35mm coverage will outlast cropped coverage as camera companies compete to sell more cameras.

[aizan]

the d700 already is the top prosumer ff model. if you're hoping that they will stick a ff sensor in a d90 level body, look at canon. three years after the 5d, we have the 5d mkii. same market position, $600 cheaper. maybe they'll put a ff sensor in a rebel chassis...three years from now? i think we're more likely to see an upgraded 5d mkiii that costs another $600 less. same story for nikon.

legacy primes will plummet in value, with a few exceptions, because sensors are much more demanding than film. they already reveal uncorrected aberrations that you never saw on film, and it'll only get worse.

look at prices of old slr lenses on ebay (i'm leaving out rangefinder lenses, whose values are inflated). only the most spectacular lenses hold or even gain value because of cult status, but even very good, solidly made lenses are available for a song. backwards compatibility is a value some manufacturers, such as nikon, pentax, and leica, maintain, but others, like canon, olympus, and hasselblad, are not afraid to abandon a lens mount for one reason or another. the fact that you can still mount old lenses on the newest body doesn't make much of a difference.

the only significant market for prime lenses is pros. as far as consumers are concerned, they have since the late 90s (before digital came along) contented themselves with a zoom lens or two. a few may go on to purchase a small, fast standard prime, and/or a macro, but usually not even that. so we see that currently nikon makes a dx normal, but not a dx micro, and canon makes an ef-s macro, but not an ef-s normal (though olympus makes both). this is yet another hint that ff probably won't reach the entry-level. primes are and have been primarily used by pros, artists, and serious amateurs.

the nikkor 10-24mm is equivalent to a 15-36mm on ff. that's pretty much as wide as it gets on any format, except for the voigtlander 12mm ultra wide-heliar.

what "limits sensor size" is manufacturing yield. it has nothing to do with sensitivity. i don't see any reason why aps-c will be abandoned, unless there is a major, major breakthrough in sensor manufacturing technology, which we have not seen this whole decade. doesn't mean it can't happen, though. we may see canon abandon aps-h, but that's about it, since canon never made any lenses for that format. and a larger format will always incur a "penalty" in size and weight.

[soonami]

By paragraph:

1. You say that Canon has been making Full frame cameras for a while now and every three of years, they drop the price by $600. Extrapolating your approximation, in 10 years we should see the price of the FF drop by $1800, which means the $2700 MSRP Canon 5D mark II equivalent at that time should be $900. Also, with a $3000 MSRP, the D700 is not prosumer, it is professional. Pretty much only people that are paid to take pictures, or the retardly rich, can afford this piece of gear.

2. You state that legacy primes will drop in value because sensors are more demanding than film and as sensors get better then old lenses will not perform as well. But perhaps, you should consider that sensors are not as good as the lenses they are collecting images from. Primes are pretty much the least optically compromising of all lenses and there are few cases that I can think of, where a prime works significantly worse digitally than on film.

3. Old SLR lenses are cheap on ebay and some companies make lenses that are now backwards compatible. These two points are not really related, but yes some crappy lenses are cheap on ebay now that better lenses have replaced them. But anything that was worthy then is still now. Give examples of a functioning lenses, not a shitty 80s zoom, that is worth less than 10% (my definitions of a "song") of what is was, inflation corrected.

5.

the only significant market for prime lenses is pros. as far as consumers are concerned, they have since the late 90s (before digital came along) contented themselves with a zoom lens or two. a few may go on to purchase a small, fast standard prime, and/or a macro, but usually not even that. so we see that currently nikon makes a dx normal, but not a dx micro, and canon makes an ef-s macro, but not an ef-s normal (though olympus makes both). this is yet another hint that ff probably won't reach the entry-level. primes are and have been primarily used by pros, artists, and serious amateurs.

I have no idea what you are trying to say. That primes are mostly for pros? What does that have to do with whether or not full frame will be coming to lower level cameras? Many amateurs shot with 35mm SLR's set on full auto in the late 90s, not many moved to the APS size of the Pronea or w/e Canon had. If primes are so rarely used, then why are there so many 50mm lenses available in the second market. And wouldn't most SLR owners be considered "serious amateurs?"

[aizan]

if you wait long enough, they'll give you money to take cameras off their hands. ??? are you retracting your opinion that we'll be seeing ff cameras that don't cost as much as they do now? plenty of non-professionals are buying the d700, and most of them aren't rich, per se. want proof? look at flickr.

"sensors are not as good as the lenses they are collecting images from."

"there are few cases that I can think of, where a prime works significantly worse digitally than on film."

you've got it backwards. film obscures aberrations through grain and resolving power that decreases as film speed increases. there isn't a single general purpose film in 135 that can outresolve today's 10mp+ dslrs, though there are plenty iso 100 films that can break even. as the rule of thumb goes, if a lens tests well on digital, it will be fine on film. if you test a lens on film, you may be surprised what you see on digital.

btw, the best modern zoom lenses are as sharp as the best primes. that's not to say that i don't vastly prefer prime lenses, which i do, or that every zoom lens is high quality.

i haven't done the 10%, inflation adjusted math, but here are some cheap lenses off the top of my head:

almost every 50/1.4, 50/2, 50/3.5 macros, 58/1.4, 28/2.8, 100/2.8, 100/4 macros, 35/2, 35/2.8, 24/2.8, 85/1.8, 135/4, 200/4, 200/2.8 or thereabouts, for every slr lens mount.

exceptions include some high speed lenses, extreme focal lengths, pancakes, t/s or pc lenses, and assorted cult classics.

who's going to use a full range of prime lenses? who's going to spend the dough on ff cameras? entry-level users? anyone would buy anything if the price was right. if you like using primes, it's gonna present obstacles of some kind or another for a while. the perfect camera is not just around the corner.

50mm lenses are flooding the second hand market because they were sold with every slr for several decades. now that people aren't shooting film much, they're clearing out their closets.

i don't think buying a dslr means someone is a "serious amateur." as far as the market is concerned, it means "more than $1000, less than $3000."

[Osmosis Jones]

^ those were the most well constructed posts about SLRs. ever.

[joelz]

Yea that was mad informative.

My current question is do I invest in non-DX lenses only (I shoot with a d90) in the interest of eventually upgrading to a FF camera (I realize, should I switch to canon I'm screwed) or do I still get DX lenses which will work perfectly well on my APS-C/DX sensor?

[ericdw]

well since the d90 is relatively new, presumably you bought the camera relatively recently. why are you already thinking about upgrading? and if you're just thinking about it and don't plan on it for a while, just get the DX, 5 years from now when you upgrade, there will be better lenses

[soonami]

My opinion, stated from the beginning was that 35mm equivalent-sized sensor DSLRs will be available soon for an affordable price ~$1000. Maybe not this year, but probably within 5 years.

You state the resolution of 35mm is less than the image of a 10+mp dslr, but you are incorrect. http://www.l-camera-forum.com/leica-forum/customer-forum/80607-resolution-digital-versus-35mm-film.html#post844368 Various estimates for resolution of 35mm would require resolution of about 20mp for digital to be equivalent to slow, high-content film. However, at high ISO, I concede that digital vastly out-performs film

Digital sensors still do not have the dynamic range of sensitivities that film does. The highlights in many digital images are easily over-exposed and over-blown, with the information is not recovered by dodging or lowering brightness/lowering contrast that can be done with film negatives.

Sure the best pro modern zooms are as sharp at many of the focal lengths that primes normally occupy, but how about distortion? Or max aperture or weight or size or price? When it comes to photography, sharpness isn't everything. A lens can be razor sharp, resolve up to 50 MP but if has weird distortions that cannot be corrected, or weighs 5 pounds, or costs $3000, then where's the advantage?

No one carries 10 primes anymore because zooms are so good that you don't need a prime to get a good image. However, even in these modern days of image stablization, and high sensitivity sensors, primes still have a place when it comes to low-light photography and capturing motion.

You ask:

who's going to spend the dough on ff cameras?

but you basically say that many non-professionals are using the d700 which is FF, and falls under your serious amateur definition of $1000-3000.

Speaking of primes:

-First example of the 50mm f/1.4 as a cheap prime that is flooding the used market. But these lenses are still selling for a good price. You can get a 50mm f/1.8 AF-D new for about $120 shipped, but people are still spending ~$100 for non-AF 50mm F/1.4 on ebay:

here

here

A AI-S 50 1.4 in near mint condition sells for almost $300, about the same price as a brand new 50 1.4 Af-d and more than half as much as a 1.4 AF-S:

here

here

85mm f/1.4:

These sell for about $1000 new and they really hold their values, used ones sell for about how much a new one costs because so view are made and sold in the first place. Even non-autofocusing AI-S lenses still sell for around $600

here

here

here

35mm f/2:

here

here

here

here

The point is that primes will hold their place, even old ones, because they possess benefits: speed, size, weight, depth of field, etc that cannot be replaced. And if no one shoots film, then why are there people buying the 35 f/2 AIS when the 35 f/1.8 is cheaper, has a larger aperture, and silently auto-focuses? You are wrong about prime lenses, Nikon brand lenses are cannot be bought for "a song" nor are they flooding the secondary market. You forget that although the demand for some of these old is very low, the supply in many cases is even lower.

_______________

Back to the original point you made:

anyhow, aps-c will be around for a long time. ff dslrs will either never make it into the entry level, or at least take another decade to do so. by then, all legacy ff primes will have been replaced with ones that have af-s and better optics. sensors will also improve, and it's not only possible, but likely that tomorrow's aps-c sensors will better today's ff sensors, and so on. the bottom will fall out of the used legacy lens market, so you might as well cash in your chips.

You state that FF DSLRs will either never make it to entry levels or need at least a decade to do so, but by your own estimates, in less than 10 years we should get a Canon FF DSLR with a MSRP of less then $1000.

You say that all legay FF primes will be replaced by AF-S with better optics, but the fact is that Nikon primes hold their value well. Even 20 y/o lenses are comparable in price to modern lenses of perceived higher functionality.

You commented that "tomorrows APS-C sensors will better today's FF sensors". I, of course, agree, but tomorrow's FF sensors will be better than today's FF sensors as well, so the comparison across time and market segment is not fair.

You claim that the "bottom will fall out of the used legacy lens market." I can speak only of what I know, Nikon, but I would love to see the day that I can get a 85mm AF-D for like $100, or the 50mm 1.4 AF-D for $50, when the bottom falls out of the market, but that has never happened and by considering 40 years of Nikon history seems unlikely to happen. Now if Nikon changed the lens mount from F-mount to something else, then maybe this would happen, but also seems unlikely.

In summation, I pretty much disagree with everything you stated, think that you are flat wrong, and have not been the least bit successful at convincing me otherwise.

[soonami]

My current question is do I invest in non-DX lenses only (I shoot with a d90) in the interest of eventually upgrading to a FF camera (I realize, should I switch to canon I'm screwed) or do I still get DX lenses which will work perfectly well on my APS-C/DX sensor?

If you don't have any non-dx lenses, I would just plan to buy what works best for your D90, i.e. not get the 35mm f/2, when the 35mm f/1.8 is cheaper and faster. If you do decide to switch to FX later, then you should be able to sell anything DX for not too much of a loss because someone else will probably still want it for a DX body. I just wouldn't drop 1+k on the 17-55, 2.8 though...

[aizan]

two quotes from the post you linked:

"...and you see why, in the huge majority of cases, any double-digit sensor will usually do at least as well, and usually better, than a 35mm film image at equivalent ISOs - in terms of resolution."

"But obviously, the range would be 54 Mpixels to 10 Mpixels (or less) - depending on what assumptions you make about the conditions, enlargement, etc."

i dunno about you, but i don't bother shooting spur orthopan in my leica. i just switch to my 6x9 fujis. and i like the look of tmax 3200, while i still don't like the look of digital noise, even after noise reduction.

in practice, dslrs only have a little more dynamic range than slide film, but there is much more exposure latitude. if you shoot raw, you can recover a lot of the highlights and shadows. it's ludicrous to say otherwise, because it just isn't true.

i agree with all of your points made regarding primes vs. zooms, which is why i hardly ever use zooms. waveform distortion creeps the hell out of me, but it doesn't matter for most photography, and it's trivial to correction barrel and pincushion. let's say you're an architectural photographer, and distortion really does matter. what lens are you going to buy? the one that's a hassle to use, or the one that gets the job done? the options are out there. dslrs aren't the preferred choice of architectural photographers anyway. it's shifted over to mfdb mounted on technical or view cameras, no pun intended.

the 85/1.4 ai-s was not produced in large numbers, it's extra fast, and a cult classic. of course it's not cheap. i already listed the exceptional characteristics that preserve value on the used market. most lenses are not like that. the 85/1.4 ai-s is still a bargain compared to the $1200 85/1.4D IF. i'd like to see how much the 35/2 af goes for after it's replaced (the 35/1.8 dx doesn't count, obviously). same goes for the 85/1.8 af-d that you want.

$100 is pretty inexpensive, imo. if i didn't have plans and strict guidelines for buying camera gear, i'd be dropping way too much money on this, that, and the other thing i can finally get without eating ramen for a month. it's a film shooter's paradise these days. i did a quick search, and there were 240 nikkors that sold for under $100 in the last month. i also found a near mint 85/1.8 non-ai converted to ai by john white for $250. that's a screaming good deal. i also found a near mint 28/3.5 non-ai with original caps for $60. that's the lens sam abell has used for his personal photography since the early 70s.

a more corrected, better coated (especially important for digital) lens with a new af mechanism presents benefits of its own. on the other hand, i'm fully aware of the reasons why an older lens can sometimes hold its value beyond normal expectations. the 35mm pre-asph summilux is a good example. this lens is revered in japan for its optical signature, especially at f1.4 where residual aberrations and low flare resistance create a delicate luminosity. they now sell for about $1500-3000, the same range as mint secondhand 35mm summicron and summilux asphs, which are far more perfected optics. mr. kobayashi of cosina voigtlander even payed homage to this great walter mandler design by creating the 35mm nokton classic, which mimicks its optical performance and size. it has much better flare resistance and costs a lot less, too.

i was joking about prices forever dropping at the current rate. it's not going to. if we look at the price of the canon eos 3 (the last film era counterpart to the canon 5dmkii), it was ¥185,000 in 1998. adjusting for inflation using the average monthly wage, and converting to dollars with today's exchange rate, that's $2,378.96USD. that number looks familiar.

and where did i say that nobody shoots film? i mostly shoot film. there are plenty of film dinosaurs and photo students. it sounds like you do too. you're just a little out of touch, is all.

[freecat]

man i want a 5D

[jcab.]

man i want a 5D

I'd have one by now if it wasn't for market... T_T

[guywithpants]

i'm thinking of getting a pentax k10d

what are some good starter lenses for it?

[aizan]

pentax is the only brand that has a decent selection of aps-c primes. i'd start with the 35mm f2.8 macro limited.

[Lloyd]

Enjoying this discussion.

in practice, dslrs only have a little more dynamic range than slide film, but there is much more exposure latitude. if you shoot raw, you can recover a lot of the highlights and shadows. it's ludicrous to say otherwise, because it just isn't true.

What determines the dynamic range of a digital camera? I'm confused, because it seems that digital cameras should be able to capture a higher dynamic range than they do in practice.

Errors introduced during image capture should play an important role in determining the dynamic range possible.. The only three errors I can think of are read noise (caused by errors in the camera's sensor, such as extra electrons knocking around), shot noise (caused by the fact that photons arrive at random times on the sensor and don't interact with each other) and quantization noise (caused when you convert the signal from a camera's sensor into an n-bit number, corresponding to the camera's bit depth). These sources of noise result in a minimum number of photons that can be absorbed by each pixel in a camera sensor, in order to generate a usable signal.

Another important factor is the amount of light, by pixel, that a camera's sensor can absorb, before it can no longer respond to incident light. This property will produce a maximum number of photons that can be absorbed by each camera pixel.

The dynamic range of a camera should then be determined by the ratio of max number of photons to min number of photons. In terms of stops, the dynamic range is log base 2 of this ratio.

If you work out the dynamic range for a typical digital camera using the above properties, typically you get a large dynamic range, perhaps 10 stops plus. However in practice this sort of range isn't observed (seems to be more like 6 stops). Is there something I'm missing? Are there other sources of noise than the above? Or is it something more subtle? The post above mentions that shadows and highlights can be recovered from RAW files. Is the smaller dynamic range then introduced by some process further down the the image processing line from actual image capture?

[aizan]

nope. you're not missing a thing. the limiting factor in practice is exposure and scene contrast. spend enough time in an amenable situation, you can eke out every last stop of dynamic range. things rarely turn out that good. raising the iso, for starters, will lower the dynamic range.

btw, if you do shoot raw, it doesn't mean you can ignore white balance when you're taking pics. the histogram is generated from the the white balance setting, so you might end up clipping a channel.

[guywithpants]

where do all of you guys get your gear from

certain online stores

or local camera shops

[soonami]

Online vendors I like:

adorama.com

bhphoto.com

jr.com

amazon.com (super cheap accessories like filters)

and of course ebay, when live search cashback and ebay discount coupons are in effect

I like Calumet stores to check stuff out since they have everything and you can even rent pro lenses for special events.

Locally shops are good sometimes but more often hit or miss. They will often have older lenses and film bodies if you need something for a class. But selection is much more limited, but service is always nice and they will order anything you need even though you will pay a premium

[itzkoolkidkam]

what do you guys think about Leica P&S? the gf wants a new camera but doesnt want to lug a dslr around. Any recs?

[the Repeater]

what do you guys think about Leica P&S? the gf wants a new camera but doesnt want to lug a dslr around. Any recs?

The DLUX-4? Same as the Panasonic LX3. Though prices have fallen on both so maybe the styling/vanity of the Leica is worth it...

I think it's a great camera for those with small hands. Takes fantastic pictures, but I have large hands and the controls/grip is small, so it wasn't something I wanted to use myself.

I wish there was a camera just like it without the conservatively compact body.

[guywithpants]

the canon g10 is bigger than the panasonic lx3

and almost the same price g10 being slightly more expensive

[the Repeater]

the canon g10 is bigger than the panasonic lx3

and almost the same price g10 being slightly more expensive

I've seen and tried both. The G10 is a fantastic body with a drastically undersized sensor. The LX3 takes far better pictures. Though I didn't get down to the pixel-level, the reviews and pixel density of the G10 demonstrate it is not the camera it could be.

Shame, really. But the G10 is obviously just the top end of Canon's compacts, no more. Certainly designed to not interfere with the SLR range.