Tentacle wrote:
Forget liquid lenses for SLRs. And that's final. There is a simple reason for this. Liquid lenses work because of the surface tension that exist between two different fluids. The distance over which the suface tension is able to curve the transition between the two liquids is limited. Think droplet size. This means that there is a physical limitation that prevents you from getting beyond roughly a quarter inch diameter.
Now, such a small lens will work fine in, say, small-sensored mobile phone camera's, and maybe bigger sensors. It's simply not possible to upscale to large sizes. You can't make stable droplets of one inch big, gravity will pull them appart. ...Show more →
Some real futurologists are speculating about grouping liquid lenses to form images as an insect's eyes do - but I think we are safe in saying that this will not be in the next generation of Canon lenses!
Regards,
DaveMart
DaveMart wrote:
I reckon it is more likely to be data than power - does my idea of lens info like barrel distortion being passed to the camera for instant in-camra correction make any sense?
Could be awsome if Chuck gets his 'native 6400 ISO' going - stick any 5.6 coke botle on the camera and shoot the Olympics! ;-)
I somehow suspect that the new contacts would only go on pricey glass, so Canon would find a way of making us pay!
Could be barrel distortion, but I'm doubtful that this would require an additional data path. I don't see that communicating barrel distortion would necessitate such a large amount of data that it couldn't be carried over the existing path. (Along with barrel distortion, compensating for light fall-off could be done in camera; I'm tempted to guess that this might involve higher data volumes.)
Hmm, looking at the Birger link, it seems that all info back from the lens is pretty simple scalar values (focal length, aperture, etc.).* If barrel distortion couldn't be reduced to one or two simple values (ie. +1.5% or something), then perhaps an additional set of contacts would be needed for the increased data volume. Either that, or they want to add the ability to carry two different streams of data concurrently.
As for making us pay... I thought that was a given.
*Alternatively, Birger's only been able to reverse engineer only the simple responses; the lens is capable of more complex interaction, but he can't figure out how to do it.
Scott wrote: Great post Michael, but this bit of Canon marketing screams BS to me...
Scott,
I don't disagree. Sounds like a bit of PR hype to me, too. My point was simply that Canon knows that a lot of pro photographers use the 20D/30D bodies, and their efforts to beef the 30D up some recognized that. That, and the contention of some in this forum that the APS-C sensor is just for amateurs is unfounded. Many wildlife photographers have grown to love it. There's over a 100mm equivalent difference (based on a 400mm lens) between the 30D's 1.6x sensor and that of the Mark II N's 1.3x crop factor.
There is one thing I would like to see in a 30D replacement -- I'd like to see the same lens contacts/electronics of the "1" series bodies so that you don't lose autofocus when using a telextender on lenses slower than f/4. That would seem like a simple thing to do, though I admit, I don't understand how this works.
silverhalide wrote:
Could be barrel distortion, but I'm doubtful that this would require an additional data path. I don't see that communicating barrel distortion would necessitate such a large amount of data that it couldn't be carried over the existing path. (Along with barrel distortion, compensating for light fall-off could be done in camera; I'm tempted to guess that this might involve higher data volumes.)
Hmm, looking at the Birger link, it seems that all info back from the lens is pretty simple scalar values (focal length, aperture, etc.).* If barrel distortion couldn't be reduced to one or two simple values (ie. +1.5% or something), then perhaps an additional set of contacts would be needed for the increased data volume. Either that, or they want to add the ability to carry two different streams of data concurrently.
As for making us pay... I thought that was a given.
*Alternatively, Birger's only been able to reverse engineer only the simple responses; the lens is capable of more complex interaction, but he can't figure out how to do it.
E....Show more →
Another thing the lens could be communicating would be focus information so that the camera could allow for b/f, f/f issues - presumabbly you would run through a series where you focussed at different lengths, then afterwards it would correct - that would be wonderful, which is why I bring it up - but I really can't see why they would need anymore info from the lens than it already sends as distance info.
Regards,
DaveMart
Just another update on the GPS front, I met with some senior developers from SiRF, a company that designs and build miniturized integrated GPS receiver/processor chips for consumer electronics. They have a chip that does all the GPS signal processing and position resolving which is 17sqmm and it's used on units such as Garmins and GPS enabled watches and exercise monitors.
It turns out that they've been working with Canon and a whole bunch of other camera makers and the person mentioned to expect GPS enabled cameras and PDA devices within the next year or so. Also, I know for a fact that Research in Motion which designs and makes Blackberries are working to develop solutions for builtin GPS cability for their blackberry line. This is confirmed with the fact that they're currently hiring people with GPS experience for product development with their Blackberry line.
mfurtman wrote:
Scott wrote: Great post Michael, but this bit of Canon marketing screams BS to me...
Scott,
I don't disagree. Sounds like a bit of PR hype to me, too. My point was simply that Canon knows that a lot of pro photographers use the 20D/30D bodies, and their efforts to beef the 30D up some recognized that. That, and the contention of some in this forum that the APS-C sensor is just for amateurs is unfounded. Many wildlife photographers have grown to love it. There's over a 100mm equivalent difference (based on a 400mm lens) between the 30D's 1.6x sensor and that of the Mark II N's 1.3x crop factor.
There is one thing I would like to see in a 30D replacement -- I'd like to see the same lens contacts/electronics of the "1" series bodies so that you don't lose autofocus when using a telextender on lenses slower than f/4. That would seem like a simple thing to do, though I admit, I don't understand how this works. ...Show more →
The one thing I would love to see in the 30D replacemnt would be the ability to AF at an effective f/8. This would definitely get me to consider buying one over another 1 series.
DaveMart wrote:
[
I don't know what the answer to Monito's points about shutter lag and dust are though - but presumably there is more than one way to skin a cat.
Regards,
DaveMart
Possible solution to dust issue:
'Actually the problem of sensor dust would be improved. You can have a fixed clear filter where the mirror would be. The area behind the clear filter and the sensor would be sealed in a clean room at the factory. Dust falling on the clear filter is much further away from the sensor, so is much more out of focus and much less of a problem.
This is how the Genesis is. The problem of dust is not 100% totally eliminated, but it is radically reduced.'
Funnily enough, this was posted over at dpreview - some there do sometimes come up with worthwhile information.
Regards,
DaveMart
DaveMart wrote:
Possible solution to dust issue:
'Actually the problem of sensor dust would be improved. You can have a fixed clear filter where the mirror would be. The area behind the clear filter and the sensor would be sealed in a clean room at the factory. Dust falling on the clear filter is much further away from the sensor, so is much more out of focus and much less of a problem.
This is how the Genesis is. The problem of dust is not 100% totally eliminated, but it is radically reduced.'
Funnily enough, this was posted over at dpreview - some there do sometimes come up with worthwhile information.
Regards,
DaveMart ...Show more →
Wonder what's to stop Canon from doing this now (ie, in a camera with a mirror)? Place the filter in front of the arc of the mirror, and behind the furthest distance a lens can extend into the box. I wonder if there's enough clearance for a 2mm thick piece of glass to squeeze into? would this get in the way of the AF sensors?
DaveMart wrote:
Funnily enough, this was posted over at dpreview - some there do sometimes come up with worthwhile information.
Well, thank you. Please continue your labors over there sifting through the postings for the occasional bit of worthwhile information to communicate here so that we may be saved the effort.
silverhalide wrote:
Could be barrel distortion, but I'm doubtful that this would require an additional data path. I don't see that communicating barrel distortion would necessitate such a large amount of data that it couldn't be carried over the existing path. (Along with barrel distortion, compensating for light fall-off could be done in camera; I'm tempted to guess that this might involve higher data volumes.)
I think that simply identifying the lens, focal length, and aperture would be all that's needed. The camera could have a database of all the necessary information in firmware, updatable when new lenses are introduced. In terms of barrel distortion and light-falloff, a Canon 16-35/2.8 is a Canon16-35/2.8 is a Canon 16-35/2.8 - all copies possess roughly the same characteristics.
Monito wrote:
Well, thank you. Please continue your labors over there sifting through the postings for the occasional bit of worthwhile information to communicate here so that we may be saved the effort.
Yeah, and I will make sure I maintain an 'engineering attitude' too- just think, all these years I imagined that qualifications and study were needed to evaluate issues authoritatively- if only I had known that I could evaluate solely by having an 'attitude'.
This could revolutionise camera design and electronic engineering.
Regards,
DaveMart
10DFT wrote:
I think that simply identifying the lens, focal length, and aperture would be all that's needed. The camera could have a database of all the necessary information in firmware, updatable when new lenses are introduced. In terms of barrel distortion and light-falloff, a Canon 16-35/2.8 is a Canon16-35/2.8 is a Canon 16-35/2.8 - all copies possess roughly the same characteristics.
Yes, but a Canon 16-35/2.8 @ 16/2.8 is going to have different characteristics than @35/11. It may make more sense for the lens to keep this info rather than the camera to have a huge database of all lenses at all focal lengths at all apertures. (Although putting it into the camera rather than the lens would be a great way to mess with 3rd party lens makers.)
silverhalide wrote:
Wonder what's to stop Canon from doing this now (ie, in a camera with a mirror)? Place the filter in front of the arc of the mirror, and behind the furthest distance a lens can extend into the box. I wonder if there's enough clearance for a 2mm thick piece of glass to squeeze into? would this get in the way of the AF sensors?
First, there is very little clearance between some Canon lenses and the mirror, as it is.
Second, a thin bit of glass so far forward would easily be broken accidentally and would be a nightmare to support.
DaveMart wrote:
Yeah, and I will make sure I maintain an 'engineering attitude' too- just think, all these years I imagined that qualifications and study were needed to evaluate issues authoritatively- if only I had known that I could evaluate solely by having an 'attitude'. This could revolutionise camera design and electronic engineering.
Now I think you are veering into the kind of sneering and derision that you were scolding for earlier. Further, you twist my meaning when I referred to an "engineering attitude" and you have quoted it out of context.
silverhalide wrote:
Yes, but a Canon 16-35/2.8 @ 16/2.8 is going to have different characteristics than @35/11. It may make more sense for the lens to keep this info rather than the camera to have a huge database of all lenses at all focal lengths at all apertures. (Although putting it into the camera rather than the lens would be a great way to mess with 3rd party lens makers.)
E.
Yes, but the focal length data as shot is already transmitted to the camera, as is data telling the camera what lens was used (you can see the data in EXIF when you open the file in DPP). All the camera needs to do is take that data and apply it to known quantities. PTLens software does this in post-processing for a variety of lenses, including Canon and others. All Canon needs is the data for its own lenses. I could be wrong, but I don't think the amount of data needed is excessive.
An alternative would be to provide the data in post-processing software like DPP.
Shutter lag on EVF cameras appears to have been essentially eliminated. http://www.dpreview.com/reviews/panasonicfz8/page5.asp
and is of the order of 0.005 of a second http://www.popphoto.com/photonews/3722/panasonic-unveils-evf-camera-with-12x-zoom.html
There of course remain two components to delay - AF speed and video lag.
AF lag is greatly mitigated by the possiblity of pre-focussing - there was a Oly camera I remember which had the rather spooky ability to give you shots taken BEFORE you had fully depressed the shutter - it shot a sequence.
This is not to say of course that there are not weaknessess in phase detection AF, and is one reason why it seems to me far more likely that EVF will be used in DSLRs first at the low end, not the high end.
The second component, video lag, would likely be eliminated by better EVFs - I haven't manged to track down the exact figures yet, but am told that it is basically nil in the high-end EVF's used in the Genesis, for instance.
Of course, another approach is to use two sensors, one for imaging and the other for AF and so on.
That may solve a lot of issues.
So all in all I very much doubt that we will see EVFs iun this coming generation, but have little doubt that it will be here, starting at the low end, within the next few years.
Regards,
DaveMart
Monito wrote:
Now I think you are veering into the kind of sneering and derision that you were scolding for earlier. Further, you twist my meaning when I referred to an "engineering attitude" and you have quoted it out of context.
I agree that no futher purpose is served by continuing as we have done, and suggest that we should both continue dealing with the actual issues and try to refrain from value judgements, such as the merits of other sites.
I have just posted a comment on shutter lag issues in EVFs', which, whilst it does not answer fully your objections, or indeed show why they do not apply, seems to indicated that perhaps the restrictions that you stipulate may in fact be possible to circumvent somehow - as in actualf EVF cameras now I couldn't find them in the figures.
Not totally sure of what is going on though, and your objection wass good one.
Regards,
DaveMart
Patrick Cox wrote:
Well, I can't resist "stirring the pot" a bit. I am on an email list of a wildlife photo workshop operator that uses Canon gear and in his most recent email update that I just received today, he had the following comment...
"...The other news is from the camera manufactures......... Officially it will out in the next few week and I think you will be pleased ..."
Since he is a wildlife shooter, I am assuming that we will see some announcements that will please wildlife shooters. I hope a new long lens is part of the announcement!
Feb 20, rather than March 1, is the embargo date - this means Canon is very likely to have a last-minute presence at UK's Focus In Imaging 2007 trade show in addition to PMA 2007.
There has been a lot of hype about Canon releasing a 100-400 /f4L IS. I have used the f5.6 version of this lens but it seems hard to obtain consistent sharp results under a range of lighting conditions. I have been holding off buying one but need something longer for a shoot next weekend. While asking about what they had on hand I asked my CPS man if he had heard anything about a f4 version coming out - sadly he said no. One of my favourite lenses was the FD 150-600 f5.6 lens which had a little gearbox on the side that you pulled back that had so much more controlability than the current 100-400 which can be super stiff or super loose. There has to be something long and exciting on the horizon.
..."