fredmiranda.com Login
This topic is in the 2006 archive. You can read it and like its posts here. It is read only.

Archive 2006Interesting interview

  
 
rceres
Offline
Details
p.6 #1Interesting interview
rceres wrote: Camera shake does not really effect all lenses, or at least not equally. Why would anyone want IS on the 85 1.2L or 50 1.2L for example, never mind the 35 or 24 1.4 L's? IS is of little or no benefit there.
Then, MikeZ wrote: I hope you're kidding.
I can (unfortunately) show you plenty of slightly blurred shots from my 85L wide open at 1/80-1/100 where I couldn't handhold steady in low light.
There is a reason why a lot of people grab the 35L and marvel at the fact that they can shoot hand held at lower speeds in low light.
If I could get a blur-free images from 85L at 1/30 I would be in heaven.
You seem to have some nice primes in your arsenal, Let's see some of your shots from your EF100 f2 below 1/100 hand held, let's see how you can use this lens indoors without the flash hand held at low speeds.

Ha! Mike Z you are right. I have hardly any really slow shutter speed shots from my 100, because I almost always reach for the trusty 50mm 1.4. When I do use the 100mm at night I am usually leaning against a wall or some other odd pose. Using the 50 at slow shutter speeds I find I loose shots mostly from subject motion rather than camera shake. All that said, I'd still love to own the 50 or 85 1.2L, but sure won't be waiting for image stabilization either in camera or in len for either.
Sep 23, 2006 at 07:43 PM
RGS65
Offline
Details
p.6 #2Interesting interview
limitdown wrote:
Canon's decision is purely driven by the almighty dollar (or yen, or renminbi...)


This may be true, yet most of us will continue to buy Canon.
You go buy your Sony, I'll keep my Canon lenses thanks, with or without the IS.
Sep 23, 2006 at 08:43 PM
RGS65
Offline
Details
p.6 #3Interesting interview
Joe Tidwell wrote:
If canon would just lower the prices on entry level IS lenses they could compete.


Why would they do this? Compete? They are the leader, they don't need to lower prices when everyone is chasing them. They are a company. Their sole goal and mission is to make money. Period. They owe their shareholders that obligation.

Sep 23, 2006 at 08:48 PM
Joe Tidwell
Offline
Details
p.6 #4Interesting interview
RGS65 wrote:
Joe Tidwell wroteIf canon would just lower the prices on entry level IS lenses they could compete.
Why would they do this? Compete? They are the leader, they don't need to lower prices when everyone is chasing them. They are a company. Their sole goal and mission is to make money. Period. They owe their shareholders that obligation.



Don't get me wrong ill still buy canon but ultimately the power is in the consumers hands, if people just stopped buying canon then the prices would fall, but we know that isn't going to happen.

I think its good that the other companies have found a way to compete by offering more for less, competition is always good for the customer, hopefully it hurts canon enough so prices fall. But i guess I'm just dreaming.
Sep 23, 2006 at 10:01 PM
greeneggs
Offline
Details
p.6 #5Interesting interview
Me_XMan wrote:
I will never buy in-body IS. It's wasted money since depreciation is way faster with body than lens.


I don't get it:

1. Speed of depreciation. As long as you are buying a camera anyway, the in-body IS apparently adds very little to the cost. Or are you saying that you'd never buy a new digital camera? If you don't want your toys to depreciate, keep your money in the bank.

2. Comparison to lens depreciation rates. In-body IS makes IS on short lenses obsolete (or at least, it reduces their relative advantage significantly), and these lenses are all quite recent. If you buy a lens with IS built-in today, expecting it to hold its resale value, you may be very disappointed when in-body IS becomes standard across the line next year. Or in two years, when the next generation of in-body IS becomes available, being twice or four times as effective.
Sep 23, 2006 at 11:24 PM
ns66
Offline
Details
p.6 #6Interesting interview
thanks for the post, that's the last straw for me to switch, I was hoping 2007 canon will provide in-body IS but this ends that hope early, and thanks for that

IMHO in-body IS can and will work with lens IS and ADDs to stablization, they don't need any kind of coupling. here is why: when image pass through IS lens, to the sensor, it just appears to be a more stable image, just like someone with steadier hands, now why would in-body IS freak out? in-body IS will try to compensate the already-quite-steady image and make it MORE stablized, period. they don't need to be coupled or even aware of each other's existance, and they will make stablization only better.

so canon's argument of in-body IS agaist in-lens IS is nothing but excuse, and I am done with that company.

I already have too many complains with canon, as a non-professional, and no interest in FF (1.6x just fine and in-body IS will work), all the FF and expensive L lenses and great line-ups mean nothing to me and in fact they are more negative since canon has to artifically dumb down lower end to protect their sales. I travel all over the world and for me, have a decent body with pentax K10D features (+D80's auto-iso), and a good all around lense of 18-200 are more important than anything else :P.

great EF lens line-up you say? tell me where is 18-135? 18-200 VR? the only available wide to normal/short tele lenese are 17-85IS and 17-55IS, either poor quality or too expensive/short, don't tell me canon can't make 18-135 or 18-200 IS, but they won't so we have to buy more lenses to cover the range, and why there's a great 24-105IS for FF and no counter-part for 1.6x? they want you to buy more FF bodies . thank goodness the world is not run by canon .

Edited by ns66 on Sep 24, 2006 at 08:53 PM GMT
Sep 24, 2006 at 02:32 PM
mh2000
Offline
Details
p.6 #7Interesting interview
Until someone presents a believable Matlab/Simulink model of an IS lens working with an in-camera IS system I will continue to believe that no one knows how the two systems could work together... you are all just guessing.
Sep 24, 2006 at 02:45 PM
Tentacle
Offline
Details
p.6 #8Interesting interview
ns66 wrote:
[...]

IMHO in-body IS can and will work with lens IS and ADDs to stablization, they don't need any kind of coupling. here is why: when image pass through IS lens, to the sensor, it just appears to be a more stable image, just like someone with steadier hands, now why would in-body IS freak out? in-body IS will try to compensate the already-quite-steady image and make it MORE stablized, period. they don't need to be coupled or even aware of each other's existance, and they will make stablization only better.

[...]


Ok, so the lens-IS detects a movement of, say, 5 degree per second and shifts the floating element at the appropriate speed. Now, the body-IS uses the very same type gyro sensors, and will also detect the very same 5 degree per second camera shake...

How will the body-IS electronics know how effective the lens-IS was at taking care of the shake? Remember, the current systems have no live preview, the shutter is closed, the mirror is down, the sensor CAN NOT do the detecting. (Even so, framerate of CCDs aren't that extreme to do high-speed shake detection.)

The point is: the Body-IS system of a hypothetical dual-IS system can not detect/measure how effective the lens-IS was. It uses the same kind gyro sensor input. There is no detecting how much shake there is left to compensate.

The only half-harted way to make these work together is to know EXACTLY how effective the lens-IS system is, and know EXACTLY how much left-over shake will remain... For any possible lens-IS, with every possible lens-IS revision, at any conceivable focal length, at all possible available battery power levels, at any possible combination of pitch acceleration and pitch speed and at any translation acceleration and translation speed... Only then the body-IS electronics will be able to GUESS what sensor shift needs to be applied.

If you don't see what kind engineering, controlling and signal processing nightmare this is, not to mention a debugging nightmare and a fine-tuning nightmare, then I don't know what will.
Sep 24, 2006 at 03:06 PM
ns66
Offline
Details
p.6 #9Interesting interview
I think you could be correct with the gyro sensors the only way to detect the "shake", so in the end, with current gyro implementation, the two IS need be somehow coupled to work together, I didn't really study how the IS works in detail but my guess is if the lens IS is done by calculation/motor (instead of based on inertia), that information can be relayed to body IS, and don't worry about complexity that kind of calculation is easier than convert raw to jpeg .

just a thought, wouldn't it be nice to have a CCD detect high-contrast area (similar to AF) and use that as shake detection, then we can do real IS (not realative to stable earth but relative to subject image), then we can also take care of "subject shake", and it will seeminglessly work with lens IS without any coupling. but of course that can only be done in body :P
Sep 24, 2006 at 04:04 PM
Richie S
Offline
Details
p.6 #10Interesting interview
I think someone mentioned before - that price is the main concern. It seems to me that on average, about $400 goes onto the price of a lens when IS is added - More so with the 70-200 f4L. Ok, you could argue there are other improvements, but in terms of the end user, that was about it (it was already a great optical performer).

Canon is riding the crest of the wave it began with the 300D, 10D and so on.It was best placed as the DSLR world expoded and has reaped the benefits as a result.
I only hope it's not assuming this will last forever.
I consider no price declines since IS was introduced, what 10 or more years ago, as a degree of exploitation - I'd also blame Nikon for this as they didn't put any price pressure on either.

I do assume that Canon will respond to the pressure now being applied - by essentially Samsung and Sony - the longer it fails to respond by keeping its prices up the more it will suffer in the long run.

Canon IS lenses are overpriced, as they've essentailly only had one fairly small competitor in the DSLR market for the past few years. This is changing. For exampe - do we honestly think that a stop or two is all that will EVER be achieved inbody?

Also, because of Canon's delberate downspeccing of cameras in an attempt to force consumers to spend more, Canon is now playing catchup. If the Pentax K10 has even moderate ISO performance then it's the pick of the entry+ level.

And all these people who rubbish Sony should maybe take a look at why Canon is so successful. They have this great share price because of the huge margins on their products. They've essentially been able to get you (and me for that matter) to pay far far more for their Cameras and lenses than it costs to manufacture them.
Sep 24, 2006 at 04:30 PM
Tentacle
Offline
Details
p.6 #11Interesting interview
ns66 wrote:
I think you could be correct with the gyro sensors the only way to detect the "shake", so in the end, with current gyro implementation, the two IS need be somehow coupled to work together, I didn't really study how the IS works in detail but my guess is if the lens IS is done by calculation/motor (instead of based on inertia), that information can be relayed to body IS, and don't worry about complexity that kind of calculation is easier than convert raw to jpeg .


Doing the actual calculation itself will indeed be easier than RAW to JPG. But that's beside the point. If you want to estimate the left-over shake that needs to be corrected, then you need a ton of data. You'll need to have the effectiveness of all popular IS lenses, with all their revisions, with all known extenders, in all conceivable motion scenarios. And if you're unlucky, the effectiveness also depends on ambient temperature and battery levels. That is a very tall order.

Only once you've "mapped" each lens-IS system, you have the information (look-up table with all the data, a calculation algorithm and the actual shake) to calculate the compensation.

just a thought, wouldn't it be nice to have a CCD detect high-contrast area (similar to AF) and use that as shake detection, then we can do real IS (not realative to stable earth but relative to subject image), then we can also take care of "subject shake", and it will seeminglessly work with lens IS without any coupling. but of course that can only be done in body :P

Okay, that idea might work. But it will open another can of worms: A high speed CCD which can compare the image several thousand times per second will need a pretty hefty processing chip to overlay each new image onto the old one to compare and calculate the shift.
Sep 24, 2006 at 04:30 PM
Richie S
Offline
Details
p.6 #12Interesting interview
It just seems to me that Canon's current business plan needs them to be a good distance ahead of competitors in order to get people to pay more for their products.

It'll be interesting to see what they have to offer, as they'll be loath to compete on price.
Sep 24, 2006 at 04:45 PM
Tentacle
Offline
Details
p.6 #13Interesting interview
Richie S wrote:
[...]

Canon IS lenses are overpriced, as they've essentailly only had one fairly small competitor in the DSLR market for the past few years. This is changing. For exampe - do we honestly think that a stop or two is all that will EVER be achieved inbody?

[...]


There are a few limits which are inherent to in-body IS. We've been through those before: APS-C works, but there is a limit to how much it conceivably can shift, so it will not work at super-tele range. APS-H is a small maybe, FF is a no-no.

Expect lens-based IS to survive simply because it can do more than in-body IS. Then again, it will be pushed into the niche market of longer lenses. In-body IS will take over the lower end and become mainstream in the APS-C market.
Sep 24, 2006 at 04:47 PM
Mr Grim
Offline
Details
p.6 #14Interesting interview
10DFT is right on the money. Ok, my expertise in optical physics is mostly focused on spectrophotometers and telescopes, but the physics of the situation is unchanged. Controling image shake at the node point (the place where the light is concentrated into the smallest area) requires: 1) less movement distance, 2) less power, 3) works regardless of the size of your sensor -- no special consideration for large sensor sizes, etc.

Also, lens-based IS also improves autofocus, although this advantage becomes aparent only as f/l becomes longer. Your camera autofocuses when the mirror is down. In lens autofocus allows a more stable image on the AF mechanism, improving its performance and increasing the probability that a long tele shot of a small target stays on the AF sensor point longer and more consistently. Body IS cannot do these things.

All that being said (and optical physics aside) marketing/economics is something else again. Pros and advanced amateurs will understand the benefits of in-lens IS and probably be willing to pay for it. Newcomers probably won't be willing to pony up the cash for the improved performance - especially if they don't understand or appreciate the difference. The good part of this is that competition in the market drives both prices down and innovation forward. Having in-body IS spreads your technically smaller advantage over a wider range of lenses, mile wide, inch deep so to speak. The in-lens IS concentrates your advantage on those few lenses you purchase with the IS built in; inch-wide, mile deep.

This kind of smacks of the full-frame vs. crop sensor debate. Decide what works best for you -- and buy it! (Canon, Nikon, Sony, whatever) Your dollars will help drive the market in the direction of your personal choice and convince manufacturers to invest more in your chosen baby. In a free market, we all win. In physics, this ain't the case.

Regards,

Gene Grim
Sep 24, 2006 at 04:52 PM
Richie S
Offline
Details
p.6 #15Interesting interview
Tentacle - I hope in-lens IS survives, and it appears to be, from people who understand it's workings far better than me, the best current solution. I'm just hoping that Canon finally feels some pressure.
Sep 24, 2006 at 05:06 PM
Pixel Perfect
Offline
Details
p.6 #16Interesting interview
rceres wrote:
Camera shake does not really effect all lenses, or at least not equally. Why would anyone want IS on the 85 1.2L or 50 1.2L for example, never mind the 35 or 24 1.4 L's? IS is of little or no benefit there. Now for 200mm and up, IS is probably a miracle, but IS on a 300mm F2.8 lens is hardly a significant portion of the price. And, if you are spending that much on the lens, wouldn't you want the best possible (ie. in the lens) vibration mitigation? I sure would.


Why. Have you been in an art gallery museum or aquarium where no tripod are allowed. Even at ISO 1600 f/2 and you can get 1/25s or slower. I would love to shoot a 50/85 at say f/5.6, ISO 200-400 sometimes (DOF is often needed) and IS is the only hope I have.

A super fast lens can still use the benefits of IS. It's all about choices. I shouldn't have to always revert to super high ISO and wide open, or flash (often not allowed either)
Sep 24, 2006 at 07:37 PM
Pixel Perfect
Offline
Details
p.6 #17Interesting interview
chez wrote:
Alright, all you people bashing the in-body IS...would you be singing a different tune if the 400D had it. I'm sure you would. Who would not want in-body IS. You can turn it off if you don't want it. I just don't understand why anyone would defend Canon for not putting it into their bodies. Especially since it does not really seem to cost very much for this technology...unlike the lens based IS which costs you a miniumu of $500 per lens.



of course they would. Well known phenomenon, the competitions technology sucks until we use it, but then we implement much better and it rocks. Doesn't matter what sort of forum it's always the same. If Canon had introduced this 2 years ago with the 300D say, the fan boys would be shouting from the top of the Rockies telling how freaking amazing it is and that Nikon are losers becasue they don't have it.
Sep 24, 2006 at 07:41 PM
JohnJ80
Offline
Details
p.6 #18Interesting interview
fanboy, hmm. I guess maybe for some.

I've always been a skeptic when it comes to IS. It works ok but it is no panacea. If it is true that the in body Sony IS is only about 1 stop, then it is a big don't care and much ado about nothing., IMHO

Currently, IS (3G, IIRC correct my generation numbers if I have them wrong, please) in the 24-105 is the first that I thought was meaningful. The 1G IS (28-135 etc...) is pretty much a don't care. 2nd generation, so-so. G3 ok. Remains to see what 4G looks like (ala new 70-200 f/4L IS USM).

So, if this is about the capability of 1G - then it is no big deal. Probably is no big deal anyhow.

J.
Sep 24, 2006 at 08:21 PM
ns66
Offline
Details
p.6 #19Interesting interview
Tentacle wrote:
ns66 wrotejust a thought, wouldn't it be nice to have a CCD detect high-contrast area (similar to AF) and use that as shake detection, then we can do real IS (not realative to stable earth but relative to subject image), then we can also take care of "subject shake", and it will seeminglessly work with lens IS without any coupling. but of course that can only be done in body :P
Okay, that idea might work. But it will open another can of worms: A high speed CCD which can compare the image several thousand times per second will need a pretty hefty processing chip to overlay each new image onto the old one to compare and calculate the shift.


I think AF sensors are good candidates to do this additional "shake detection" job...

as about speed, I kinda remember sony P&S or video camera use CCD image check to do their IS... I think it's doable

I dislike canon more and more, they are now on the opposite side of users, trying all kinds of tricks to milk our last $, while other companies trying their best to make users happy, they might not have the resource to do better, but I appreciate the fact they are "on our side"
Sep 24, 2006 at 11:06 PM
Roy NN7DX
Offline
Details
p.6 #20Interesting interview
Mr Grim wrote (excerpts):
Ok, my expertise in optical physics is mostly focused on spectrophotometers and telescopes, but... (snip)
Also, lens-based IS also improves autofocus, although this advantage becomes aparent only as f/l becomes longer.... (snip)


Thank you Gene... This helps me a lot... I have been looking for a qualified opinion...
Sep 25, 2006 at 12:36 AM
FM ForumsCanon ForumJoin Upload & Sell
  
 

You are not logged inLog inRegister