Yet another built-In stabilizer DSLR

[mars c]

"No, no, no you would not. The internal IS would probably be effective until about 100mm and then the displacements get too large for that soort of design. "

 

Heheheh.. thanks for the correction, I understood it, but forgot about it.

 

But Olympus has just introduced a 70-300 lens and that would become 140-600mm, I wonder if their IS would work in their own lens, only time will tell. but I hope so.

 

As for Canon, losing market shares on bodies, means losing more lens sales. Most people want the most feature they could get at the cheapest possible price, and the IS is a popular feature in a body, Undeniably worhty of consideration for a beginner and most people.

[jwhite3.0]

My understanding from all of your posts:

 

1. IS in the body would be less efficient with a mounted lens of >100mm

 

2. IS in the body would be less effecient as an IS system that has been tailored for a specific lens

 

3. We don't know if there would be an additive effect to having IS in the lens and body

 

4. The actual optics from Canon or Nikon are pretty good, we want more convenience and don't want to carry the traditional tripod

 

5. Already expensive Canon or Nikon dSLRs would be even more expensive regardless of one's own need for IS because they might not put out IS and non-IS dSLRs of the same model. Based on your posts I get the feeling in-body IS design would not be put on pro 1-series bodies due to the fact it is merely a marketing gimmick. Actually if they did I would appreciate that. Could result in a price drop in older non-IS pro dSLR models. I also recognize some of you would like even a less than 100% effictive IS in the body.

 

6. As someone using Canon products for almost a year I am amazed with the current technology. To be honest the optics of Canon products back to their first line of EF lenses is fantastic. Cool stuff.

[jim_larson1]

Honestly. . . .based upon the competition I think Canon WILL be forced to produce in body IS. You can't produce mid level cameras with lower pixel counts, without live preview, without anti-dust, without image stabilization and expect to retain market share.

 

This is mostly a slam at the 30D. Better image quality will only get you so far. The XTi addresses a few of these features. . .without affecting image quality. And did I not see some kind of anti-dust technology on the 1D-III?

 

Now. . .as to "which is better", in lens or in body. . . I think "it depends". I suspect if Canon was SMART they would produce in body IS, but CLAIM that the lens system was superior . . .and particularly superior for telephotos. They could claim that the in body would be specifically optimized for a certain focal range (say. . the 18-55 range. . .) and would not work well in the 70-300 range.

 

In that case. . .the only sales that would be affected would the prosumer IS lenses: the 17-85/EF-S, the 17-55/EF-S and 28-135/IS.

 

How would the pro market react? I dunno. If you had the 24/1.4L and 35/1.4L you probably would WELCOME in body image stabilization.

[j_smith6]

I've heard that the reason Canon is not making IS bodies is that it doesn't work as good as IS in lens + you can't use these 2 at once. If the results aren't good enough then why should they make it in pro bodies - people will choose IS lenses. On the other hand it's hard to make prime IS lens. Maybe they will make some IS bodies... who knows. It wouldn't really make any difference for me and I wouldn't choose a camera by checking if it has IS or not. I would take the one which has lowest noise, good AF, has enough megapixels, is comfortable then other things. 1-2 stops is almost nothing - usually.

[lester_wareham]

Eric Friedemann : "If Nikon or Canon have white papers, independent tests or even in-house tests proving the superiority of in-lens image stabilization..."

 

Eric, try a forum search. Someone did a few simple calculations recently to show what the limits of in-body IS are.

[lester_wareham]

Peter Quaedvlieg : "Another misunderstanding is the idea that the "gained" stops from an IS give other exposure specs."

 

In this context people are just talking about the gain in shutter speeds. The unit a Stop does not have to relate specifically to aperture ratios but includes shutter speed ratios or even just exposure value steps.

[lester_wareham]

mars c : "But Olympus has just introduced a 70-300 lens and that would become 140-600mm, I wonder if their IS would work in their own lens, only time will tell. but I hope so. "

 

I think the Oly format is quite small though, something like 2X as you say, that will make it more effective, even so they must be running into the end stops, bit like bottoming out your cars suspension. :)

[lester_wareham]

Also there may be engineering compromises for in-body IS.

 

Presumably in-body IS (or IBIS as I am getting tired of typing) seems to be best accomplished by moving the sensor around using transducers. This may have alignment issues affecting image quality, QA issues and service life issues.

[pmind]

Anyone here remember Konica/Minolta? I seem to recall that they also had anti-shake technology before going out of business, not sure how much importance I would place on it. Paul

[steve_dunn2]

<cite>Most likely the designers would turn off the internal IS if an IS lens was mounted.</cite>

 

<p>Since in-lens IS is more effective, this would make sense. But if they wanted to, they could do it the other way as well; the body can turn off in-lens IS when it wants to (several bodies do this in certain cases, such as self-timer or bulb).</p>

 

<cite>"I, personally, hope that they do not put it in the body."

<br><br>

Why hope against it?

<br><br>

you can always turn it off when not needed and on when needed, IMO its better to have in body IS for my 50mm 1.8 rather than none.</cite>

 

<p>How about cost? If I already have IS in my lenses (which I do in my two longest lenses), why would I want to pay for it in the body as well? Given the amount of engineering that would have to go into designing an IS-capable body, I don't think it would be likely that Canon would sell IS and non-IS versions of the same body.</p>

[jim_greer]

Whoe! Back up! My whole basis of optical understanding here is being swamped. Way up top of this thread someone said "couple the new E510 with sigma 50-500, you'll have an image stabilized 100-1000mm lens." Then someone further down referred to the same concept with another manufacturer's zoom, and no one challenged either of them. I'm far from an expert in this stuff, just now switching from film SLR to DSLR, but done a lot of homework, and old film optical concepts seem still to apply. So regardless of IS in body or in lens or not at all, how in the world does mating the new E510 with the Sigma 50-500 produce a 100-1000mm lens? I assume these posters are multiplying by a 2.0 crop factor to get to the 100-1000, and if that's wrong and I'm missing the whole point just forget I wrote this, but if that is the point, it's all fooey, isn't it? If the lens angle of acceptance doesn't change, which it doesn't here, there's no change in magnification, pure physics, peripod, right? I can see where a cropped 50-500 might FEEL like a 100-1000, but isn't all that's happening is that the crop camera is cropping and enlarging the image, not in any way changing focal length? It all comes out in sharpness, right? Consider this: accepting these earlier posters' logic, you could let the camera do the first doubling through its inherent crop factor, and then double again in editing, and in effect conclude you've ended up with a 200-2000 lens, but really all that's happening is we've added more and more digital zoom atop a fixed amount of legitimate 50-500 optical zoom, correct? Hey, jump all over me if I'm wrong, or missing the whole point here, maybe IS somehow effects focal length and I'm totally ignorant to it, but please say something so I know my foundation of optical understanding either holds water or I've got to totally start over. Counting on these forums (here and elsewhere) to keep me grounded in the new digital world I'm entering, so thanks in advance for replies.

[bobatkins]

It doesn't matter which is more effective. Canon can't put IS in their wideangle primes for technical and optical reasons (or at least it would require a total redesign and possibly compromised optical performance).

 

Body IS is MUCH better than no IS at all.

 

Lens besed IS is certainly better for long telephoto lenses, it's also a much better user experience since you can see the stabilization taking place.

 

Body based IS could be difficult with full frame cameras because of the additional size and mass of the sensor, but body based IS works well with APS-C sized sensors and if you use a full frame lens there's no problem about the sensor moving outside the image circle of the lens.

 

With Olympus, Sony and Pentax now all selling stabilized bodies that only leaves Nikon and Canon left. If I was buying an entry level DSLR today, I'd seriously look at the Pentax K100D or K10D. The lack of body IS is going to start to hurt Canon at some point, and if Nikon ever go that way too, Canon's simoly not going to have a choice if they want to stay #1 in the consumer DSLR game. They won't like it, but they may just have to grit their teeth and do it.

 

Technically there's no problem with IS lenses. The camera knows when an IS lens is attached and could be set so that body IS turned off when an IS lens was used.

 

If Canon don't do it I may just be tempted to pick up a low end Pentax DSLR and shoot with adapted M42 mount wideangle and normal prime lenses on it!

 

My guess is that there little chance of seeing a Canon IS body for a year or two, but eventually they may have to do it due to market pressure.

[delwyn_ching]

I don't think Olympus' entrance into the IS DSLR market will affect Canon's stance on IS. Canon's IS is lens based and there are pros and cons to this method. Some schools of thought say this is more eficient than on the camera body itself plus what you see in the viewfinder is the stabilized image versus DSLR based IS. The pros and cons of this is the image in the viewfinder isn't stabilized but you don't have to buy IS lenses and it works with most or all existing lenses. either system works well.

[mtwhite]

Correct me if I'm wrong, but isn't the reason that Canon uses in-lens stabilization because IS was introduced when film SLRs were predominant? Moving a sensor is one thing, moving the entire film compartment is quite another ball of wax.

 

I agree on Bob Atkins' point on starting over. I like small bodies and small lenses and on-sensor IS, and Pentax's backwards compatibility with lenses is commendable. I'm been tempted to get a K100D with a pancake prime for street shooting for a long time.

[delwyn_ching]

Oh yeah, I forgot another point: film shooters. The IS lenses also works on film cameras, duh.

[phule]

<<My guess is that there little chance of seeing a Canon IS body for a year or two, but eventually they may have to do it due to market pressure.>>

 

Perhaps it will (or will also) force them to introduce in-lens IS 1) at lower cost and 2) in more products.

[bobatkins]

But I don't think Canon will ever make a 20/2.8IS or a 24/2.8IS or a $70 50/1.8 IS or a fishye IS or a T/S lens with IS.

 

Sure Canon might stuff IS into more zooms and they might drop the cost a little, but there are a whole bunch of lenses which will never be available in IS versions.

 

Even if Canon do bring out more IS lenses, they won't make my current non-IS versions stabilized! For the additional cost of the IS in the 70-200/4L, I could buy a Pentax K100D...but unfortunately my Canon lenses won't fit on it!

[phule]

All of which is why I included the "will also" in my response.

 

I guess I should be more clear in my replies.

 

And I think you're wrong on one point: not only will Canon not make those lenses you seek, I don't they'll be making new ones at all. It would seem that consumer primes are a thing of the past to Canon. (The exception being the 60mm EF-S).

[phule]

Though, I hope they prove me wrong.

[kin_lau]

What are the chances that Sigma might take a big risk and release a EOS mount DSLR with body IS _and_ a Foveon chip?

 

Sigma already has technology to put IS/OS in their lenses, and the Kodak DCS Pro SLR/c was (according to DPreview anyhow) a Sigma made EF mount body.

[tom_matus1]

Interesting conversation, but, it's all moot when it really boils down to the most basic components.... PROFIT MARGIN AND GAIN for CANON and they most likely won't do that until forced to by competitors or they are bored.... funny how some threads take off and some sit stagnant.... This has been fun to read...

[bobatkins]

I think this thread shows that people WANT body IS. Whether they will switch brands to get it I don't know, but Canon must be listening. I think Canon have essentially said "nothing's impossible" when asked about body based IS.

 

I don't think Canon have necessarily given up on prime lenses, though they are pretty slow to update anything but the cream of the crop (50/1.2, 85/1.2).

 

The chances that Sigma will come up with a compatible mount body must be close to zero. If they could do it, you'd think they'd have done it by now. Technically it's possible I guess, but there may be patent issues on either the hardware or software and/or the market projections say it's a bad move. Given Sigma's problems in the past with simply getting the lens interface right, I'd be a bit skeptical that they could make a body compatible with all past and future EOS lenses.

 

I agree it all comes down to $$$. If Canon think their bottom line will be better if they don't introduce a body with built in IS, they won't introduce one. However the longer they leave it and the more of their competitors start to offer it, the greater the likelyhood that the economics will shift in favor of them doing it.

 

It's not like they would lose sales of their 300, 400, 500 and 600mm lenses, since they are only offered in IS versions. It's not likely they'll make a FF cameras with body stabilization, so they won't lose their pro zoom IS lens sales.

 

It pretty much all hinges on the consumer end of the market. If the Rebel XTi starts loosing significant sales to Pentax/Sony/Olympus, they'll have to start thinking about it.

[boblester]

I was so impressed with the idea of in body stabilization that I sold my Canon stuff and bought a K100D and two of the pancakes. I was also looking for a decent 70-300ish zoom to tide me over until Pentax released something and bought the 80-320. I was not impressed. The in-body stabilization just didn't do much at 200, never mind 320. Not when compared to Canon's 70-300IS (non-DO). You also have to crank up the ISO which in turn introduces quite a bit of noise. And the pancakes did not compare to the 17-55IS. They were sort of hit and miss in low light. You really had to know where the limits were. Sometimes they were sharp and sometimes they weren't. The K10D might have better stabilization than the K100D but I don't think it is a consistent 3 stops like the 17-55IS and I'll bet stabilizing a long zoom like that "100-1000" would be quite a stuggle above 200mm. Anyway, I am back to my 17-55IS and 70-300IS for now.

[mars c]

"How about cost? If I already have IS in my lenses (which I do in my two longest lenses), why would I want to pay for it in the body as well? Given the amount of engineering that would have to go into designing an IS-capable body, I don't think it would be likely that Canon would sell IS and non-IS versions of the same body."

 

The competetors offer bodies with IS at same price or cheaper than the xt or xti, I dont see why canon cant do that, considering that the rebel was the first well featured, high performance dslr that cost much less than the competition.

 

Abouy the 100-1000mm stabilized lens, I admit that was a mistake. But even if the In body Is is not very effective, squeezing 1 or 1.5 stop of hand holdability out of my 50mm 1.8 would be a great help, IMO.

[jim_larson1]

Impossible doesn't mean what it used to mean.

 

Consider: 2 days prior to the 5D being introduced, who would have thought a $3000 full frame dSLR was possible? There was no shortage of people claiming the absudity of the leaked (but 100% accurate) spec sheet.

 

Consider: Who would have thought that the XTi would sport the 30D af system, with a sensor the had BOTH a higher pixel count AND antidust tech?

 

Consider: 100 years ago, indoor plumbing was a rarity. Commercial radio unheard of. TV unimaginable.

 

So. . . .will canon produce an image stabilized body? I submit market pressure will dictate. Would I pay a $200 premium for an image stabilized 40D? HECK NO!

 

But. . .I bet that a 40D with image stabilization would cost the SAME as the 30D when first introduced ($1399). If Canon waits too long. . .an image stabilized 40D with 12mp would probably cost what a 30D costs today: About $1100.