Any new digital body this year?

[jim_larson1]

Hasn't the 20D already been discontinued? I thought they were just selling out the existing inventory? (I think B&H still has a few *new* 10D's floating about).

 

A price war in the dSLR market would be nice. I look forward to it!

[andys]

Interesting times when it appears Canon are playing catchup for the first time in ages.

 

Is there any mileage in a 3D I wonder? Take a 1.3 sensor from a 1D mkii, with a EOS5D style body, add some weather sealing and functions from a 1dmkii and Bob's your Uncle (err not Mr Atkins....).

 

I'm fairly stunned that this hole has stayed in Canons lineup so long when the EOS 3 was recieved so well. A hole Nikon seemed to have aimed for with the D200 it seems.

 

Or is that just wishful thinking on my part, I suspect so..

[bobatkins]

Despite what some others think, my opinion is that the 1.3x format is dead. I know Canon have hinted that it's not, I just don't really believe them! I wouldn't buy one. My EF-S lenses won't fit, so that's a strike against it and if I'm going to dump my EF-S lenses, I think I'd rather wait for FF to become more affordable. My 20D is very good and the incremental increase in quality going to, say, 10MP 1.3x just wouldn't be worth it.

 

The 20D isn't discontinued as far as I know. It's still listed as a current model on the Canon website. Canon aren't pushing it and I doubt they're making many (if any), but as of now it's still a current model. It's not a bad buy either at the current price of just under $1000. If the Nikon D80 turns out to have higher noise and a slower frame rate and is priced at $900, the D80 wouldn't be a totally obvious better buy. Time will tell.

 

Canon could certainly hit the low end of the market with a $499.99 EOS 3000D, but that doesn't answer the question of what they are going to do with the 350D (XT). I suppose they could leave it as it is and not "upgrade" it. There's no law that says they have to do an upgrade after 18 months.

[andys]

One thing is known for sure, we will all know in about 4 weeks time....

[mark u]

There really are two possibilities that spring to mind. Canon announced a long while back that they were targetting producing a sub $500 DSLR (a "3000D"). Maybe its time has come. Another was the subject of an indirect hint from Canon's Chuck Westfall as recently as May 4th:

 

QUOTE

 

Canon's EOS 1D Mark II N uses yet a third standard, an intermediate-size sensor with a 1.3x conversion factor to balance processing speed with image quality. "We'd like to continue using that size as well," Westfall said.

 

UNQUOTE

 

Or even:

 

QUOTE

 

As buyers adjust to the conversion factors, new technology could lead to a resurgence of the older standards, Lee said. "I expect that in the future, full-frame sensors will become more affordable and therefore more popular and will make their way into less expensive consumer digital SLRs," he predicted.

 

Either way, Canon is unruffled by fluid digital SLR standards.

 

"The bottom line is what you see in sales," Westfall said. "It's the fastest-growing category in the entire digital-camera market."

 

UNQUOTE

 

Source:

 

http://news.com.com/Digital+SLRs+bring+lens+quandary+-+page+2/2100-1041_3-6068289-2.html?tag=st.num

 

Oh, and I doubt there will be a 40D, since the Japanese for four sounds the same as the Japanese for death.

 

http://en.wikipedia.org/wiki/Tetraphobia

[Landrum Kelly]

The biggest single WOW that Canon could produce would be a full-frame DSLR in the sub-$2,000 price range. That would be strong volley against the D200, especially if Nikon fails to come out with a full-frame DSLR. (An announcement of something from Nikon is imminent within about two weeks, but no one knows what it is.) Such a move would also go after all of the folks thinking about starting over with SONY.

 

A sub-$2,000 full-frame DSLR would also hurt the sales of the 5D, of course, but the 5D did not get the overwhelming reception that Canon must have expected. (I just bought one, for the record, but I simply got tired of waiting--therefore, when the price slipped into the $2500 range after rebate, I jumped for it, and I got it from B&H, not a discount house.)

 

The question is whether it is possible to make and market a sub-$2,000 full-frame DSLR, given the cost of the sensor. Since Canon makes its own sensors, it just might be possible. If they can merely break even on the body, they could win on the sale of glass.

 

When one starts talking about "legends," then one might have to deliver on something that is going to change the entire market, not just the high end. If full-frame DSLRs become standard, then the existing market will be blown sky-high for every manufacturer that didn't see it coming and prepare for it.

 

This is not to say that the era of the XT and the 10D, 20D, and 30D is dead, simply that they might increasingly look like lower-end consumer products by comparison--thereby inviting all of the owners of such cameras to upgrade as well.

 

The expiration of the summer rebates would also be consistent with the release of something that is going to turn the market upside down. I would even go so far as to speculate that Canon might accept a loss on a sub-$2,000 DSLR, if only because the profits on L glass have got to be very high. I know that I could only afford two good zooms to go along with my new 5D, and together they cost more than the camera.

 

It's ultimately about glass, even though we are talking about bodies, since, once you can get people to buy the body, they will buy the glass sooner or later, typically sooner.

 

The only other possibility that I see is an IS body from Canon that is 10 MP or higher.

 

--Lannie

[Landrum Kelly]

Sorry, Bob, I missed your link on the D80 above. So we do have a pretty good idea of what Nikon is up to, and it isn't FF.

 

--Lannie

[seragram]

I think more Mega pixels is a must for Rebel XT or they risk upseting followers further like after 30D release plus they must keep up with competition.

We may see a 30D upgrade soon after too, mybe a 1.3 sensor is in order.

[kevin conville]

I think those thinking of a 1.3x 30D replacement are missing a major point.

 

It wouln't be EF-S compatible.

 

Canon (I'm sure) likes selling EF-S lenses that cost as much, in some cases, as the cameras

they are attached to. One body can support a great number of lenses.

 

After developing these EF-S lenses, why would they want only one (entry level) body that

can use them? They wouldn't.

[robin_sibson1]

Kevin, that's a common assumption that may be worth some further thought. EF-S lenses with the projecting ring removed will fit on FF cameras but foul the mirror movement, and the 1D-series 1.3-factor cameras have, if I understand it correctly, a FF mirror and screen, so they don't work on those cameras. It may or may not be possible to design a 1.3-factor mirror movement that works with EF-S lenses - we just don't know. Also, the coverage of EF-S lenses almost certainly is not adequate for FF - but does anyone know how they perform over the 1.3-factor frame?

 

OK, OK, I am not actually expecting to see a 1.3-factor camera that takes EF-S lenses, but I'm also not jumping to conclusions!

[jim_larson1]

A beginning of a legend?

 

Hmmmm. While a sub $500 dSLR would be interesting, that would not be a legend in my mind.

 

Ya know. . . Is it time for a 1D-II replacement? Picture this:

 

Replace 1D-II with a full frame 12-MP version at the $3500 price point. That preserves the 5D (lower features, lower price), but finally merges the 1.3 and FF product lines at the pro level. The pro's would snap them up.

 

Of course, a full frame 5D at the $2000 price point would be nice: But I don't see it. (Note: I didn't see the 5D coming).

 

Want another legendary camera? How about a 12mp APS-C with 1 series AF (at the $2000 price point?). OK. I can dream. I don't expect that either.

 

My money is on a 1D-II replacement.

 

BTW: Any new camera MUST be greater than 10mp, or you can basically apply apply the term "legend" in the same sense the Edsel was a legend (as a buisness flop)

[thomas_sullivan]

well, now having read that other post's "...legend..." tease. None of the above is a beginning of a legend. It's just variations of past themes.

 

A true legend beginning would be a machine that combined the best of the DSLR world and the Video world.

[jim_larson1]

. . . or perhaps an image stabilized body. . .

[bobatkins]

EF-S lenses won't cover 1.3x, at least not over the whole zoom range and with acceptable edge/corner quality. Certainly not with high enough edge/corner quality to justify their price and use on a high pixel count $2000 camera body. You hear enough complaints about full frame lenses not being good enough for the 5D so EF-S lenses on 1.3x seems like a total non-starter.

 

The idea of a new 1.3x body is doubtful, but the idea of one that could take EF-S lenses is so unlikely as not really to be a possibility at all.

 

A full frame, high speed 1D is possible. Canon have pretty much said that that is the way they want to go. Whether there's the technology for a full frame 10-12MP camera that can do 8fps or more I don't know. Such a camera could hurt 1Ds sales if priced too low.

 

I don't see a significant 1Ds upgrade in the works. I don't hear current owners of the 1Ds complaining much about not having enough pixels and given it's the only game in town if you want a FF DSLR there's not really any competition to edge out. Even if there was it's doubtful that 1Ds owners would switch systems. I think most people would rather see a price drop than a pixel increase!

 

A sub $2000 full frame DSLR is a possibility I think. I'm just not sure it can be done right now, but if anyone can do it, Canon can. Yes, it would hurt 5D sales but it's quite possible Canon could live with that. They've been heavily discounting the 5D ($300 rebate) so sales are likely soft and maybe they've scaled back production. If they replaced it with a $2000 model they might actually make more money in the long run from increased body and lens sales. The 5D has been out for a year and Canon have been known to replace a model after 1 year in the past. A $2000 FF DSLR would certainly make quite a splash, if they can do it.

[bennyboy]

I'd be surprised if 5D sales are as soft as people say, my local store can't keep them on the shelves very long, and most of the UK photographers that I talk to on PN have gone and bought a 5D already - I know we're not a representative bunch here but it's got to go for something.

 

My bet is on a same pixel count 400D with spot and other goodies at the same price point as the 350D, a 10mp full frame 40D, and a new 1D meld of the IDMKIIN and 1DSMKII, with inbetween pixel count and 8fps. But what the hell do I know :D

[robert_potts1]

Canon has two things to contend with. The first is the new A100/D80. The second is the new dSLR that Sony is supposed to announce this year which will probably be shadowed by a Nikon clone. Logically the new cameras will either be pro models or scaled down dirt cheap models. I expect the former more than the latter.

 

The Canon folks have definitely painted their way into a corner. If they want to create maximum havoc for the competition, a consumer grade 1.3x with 11mp at $1,000.00 would do it. It would also shoot Canon in the foot.

 

Another approach would be to introduce a truly novel camera of some sort. They certainly have more ideas than I do to work with, but a 4:3 ratio camera with 11mp that falls between 1.6x and full frame at a competitive price would be interesting.

[joeb]

The 4:3 system has a 2x conversion factor. The quality of images would be suspect. I don't see canon signing on to a generic lens system like the 4:3. Money is made selling lenses. They would never sign on to an open system like the 4:3. Thats for companies trying to break into the DSLR with out the lens base Canon brings to the table.

[splat]

Since we are all just guessing I through out my crazy idea. How about an EF-S compatible body that has a square sensor. Same width as the 1.6 crop sensors but with additional height. Heck, while speculating, why not a cirular sensor that covers the full image circle of an EF-S lens. Then the orientation of the camera when you take the shot means nothing. Just crop it the way you want it in post.

 

I don't really think any of this will happen but it's full to speculate. Afterall we are talking about digital cameras. They aren't limited to doing things the way film did.

[bobatkins]

Nobody wants a square sensor, or at least nobody that matters and is going to buy enough cameras to make it worthwhile doing.

 

Circular sensors don't make sense either from a production standpoint. Too much wasted wafer space when fabricating the devices. That translates to higher sensor costs.

 

4:3 is an aspect ration as well as a system. However it's seen as an "amateur" format in digital. I don't think it would be a "feature" that most users would want as a format ratio and I'm 100% sure Canon won't buy into the olympus 4/3 sensor size.

 

Canon could drop the 30D and come out with a 40D with a 10 MP sensor, but there would be a lot of very pissed off 30D owners if they did.

 

Canon could also stick to their "We're right and our 8.2MP images are better than the competition's 10.2 MP images" line, and they may well be correct. How it would affect sales is anyone's guess, but I'd bet Canon have spent a LOT of money on market reseach to try to answer that question.

[kevin conville]

Circular sensors? How many ways wrong can that be?

 

As Bob mentioned, a waste of valuable space. Would you also include a circular viewfinder?

 

Just how would one compose with this view? I envision lots of unwanted ommisions that

"just" missed a rectangular crop. Or, would we start printing on round paper?

 

Clearly a miss, as I suspect square and (to a somewhat lesser extent) 4/3 is as well.

 

Certainly not the stuff of "legends" anyway.

[kevin conville]

Oh, and one other point about changing the aspect ratio.

 

If one were to make a square (for example) sensor that an EF-S lens would cover, roughly

24 x 24mm, wouldn't that require the same size mirror (on that axis) as a FF sensor which

is 24 x 36mm?

 

Same problem applies to 4/3 in that the "height" of the sensor can really be no more than

about 16-17mm. So what you "gain" is just a loss of area and an increase in the dreaded

multiplication factor.

[splat]

I didn't think a circular sensor was a real possibility. Just posting crazy ideas. Still I don't buy ALL the arguments against it. Couldn't the chip designers be a little smarter and still make the chip square but use the corners for other things? There is no law that the whole surface of the chip needs to be used for light collection. Sure it would have to be a departure from the way things are done now but that's what I was getting at. Not the specific feature of a circular sensor. My point was just that the next camera doesn't have to be "the next logical step" as we see it. It could be a complete departure from anything that we can think of. Regarding the size of the mirror I don't believe that would preclude a circular sensor either. I'm sure the is a engineering solution to get the mirror out of the way without hitting the rear element of the lens. The viewfinder configuration could have various guides then too. You select the aspect ratio and the viewfinder could project a circle with the diameter of the short side of your crop. Inside the circle would indicate "definately in the picture" and the outer extreme of the viewfinder could indicate "possibly in the picture". Again I don't think this will happen. Just trying to think up crazy ideas. Like Nikon with their adjustable crop camera. That was a great idea and I wish Canon would impliment it on their full frame cameras. Sometimes you don't need all that resolution but you do want to capture a RAW format image rather then a jpeg. A grid in the viewfinder that you could turn on and off would be nice too. And guides for 5x7,8x10,11x14, etc. that you cound turn on and off would be nice to have as well.

 

I'm don't know the science behind sensor technology but what if they could record not just how much light hits a particular area of the sensor but could determine the spectrum of light that hit a site on the sensor. Or if they could somehow capture the directional quality of the light hitting the sensor rather then just the quantity. To me these type things would be ledgendary (though I doubt the next camera we see from Canon will truly be ledgendary)

 

Please don't take this post to mean that I think that any of these thing are likely or even possible. I'm just tossing out some more "out there" ideas. Not necessarily good ideas. Just trying to point out that it's possible that Canon has in store for us something that none of us have even dreamed of (though I doubt it)

 

Anyway happy speculating to everyone. i really enjoy this time of year and look forward to reading all the continued speculation for the next 4 weeks or so.

[bobatkins]

"I don't know the science behind sensor technology but what if they could record not just how much light hits a particular area of the sensor but could determine the spectrum of light that hit a site on the sensor."

 

Yes, it's called the Foveon sensor design. Sigma used it in their DSLRs. It has some technical problems as well as some theoretical technical advantages. It has yet to be proven better then the current Bayer designs (in a practical application).

 

I do not expect any new sensor designs from Canon right now. Their in-house CMOS technology is already better than most of their competitors.

[robin_sibson1]

Well, Bob, thank you for answering my question about 1.3-factor coverage of EF-S lenses. It is no surprise at all, as my post indicated, to learn that the designers have taken full advantage of the 1.6-factor image circle in their EF-S designs, but I would be interested to know if there are any published tests to demonstrate what the coverage is. Presumably a combination of lens surgery and MLU would make such tests possible.

[bobatkins]

It's been done. The EF-S 18-55 is quite easy to convert to a standard EF mount and since it's a cheap lens you don't risk much (except for your camera...)

 

See http://www.bobatkins.com/photography/tutorials/efs-10d.html

 

Others have reported converting the 17-85IS (for use on a 10D).

 

As I recall just looking through the viewfinder, you got what looked like full frame illumination at the long end, but not at the short.

 

BTW mirror lock up wouldn't help in full frame testing of EF-S lenses, since the mirror comes back down as soon as you take a shot, so even if you can lock it up before mounting the lens, it will hit the lens on its way back down! Normally there's only a problem at the short (wide) end of zooms though. At longer focal lengths the rear elelemnt of the lens moves away from the sensor.

 

Since many people (not me, but others!) claim the 18-55 is pretty poor even on an APS-C format camera, you can't expect much on full frame even if it does actually illuminate the whole frame. Basically a waste of time.

 

So no, Canon will NEVER, EVER make a 1.3x EF-S camera. NEVER.

 

I suppose if they wanted to go to 1.5x like Nikon then the EF-S lenses would probably have enough coverage, Could happen. Seems pretty unlikely. I'm surpeised that the 1.5x APS-C DSLR makers don't make a bigger deal out of the fact that they are 1.5x and Canon is "only" 1.6x. After all a bigger sensor is a bigger sensor, even if it's only slightly bigger. They've made big deal out of lesser differences in the past.