Panny Lenses with no OIS switch on E-M5

Started May 20, 2012 | Discussions
Anders W
Forum ProPosts: 15,444Gear list
Like?
Re: Panny Lenses with no OIS switch on E-M5
In reply to Detail Man, May 20, 2012

Detail Man wrote:

Anders W wrote:

Bryan Biggers wrote:

Anders W wrote:

According to the reports I have seen, there is a firmware bug that prevents OIS from working during the exposure itself (although it will work prior to the exposure).

Yes. It was discovered by forum user kenw when testing the IBIS on his E-M5 versus the OIS on the 14-45. Eventually, he discovered that the OIS was not turned on during the exposure itself when used on the E-M5. Here is the thread in case you are interested:

http://forums.dpreview.com/forums/read.asp?forum=1041&message=41297304

Has anyone considered that this may not be a firmware bug after all - but instead an intentionally engineered characteristic (due to technical limitations arising out camera-lens communications) ?

Good question. But yes I considered it. And judging by the E-M5 manual, it is a bug rather than a "feature". The manual (p. 49, under "cautions") provides the following advice:

When using a lens with an image stabilization function, turn off the image stabilizer function of either the lens or the camera.

That "caution" would hardly be needed if the OIS was meant to be disabled altogether. Nor does it seem likely that Oly would intentionally have allowed OIS to stabilize the view on the EVF/OLED prior to exposure but not the exposure itself, which is what the current E-M5 firmware manages to accomplish.

As to technical limitations, I can't see that there would be any aside from the firmware (which can fortunately be revised). Note that the OIS system always requires lens-body interaction. It is not regulated by the switch on the lens barrel alone, even on lenses having such a switch. As you know, there are multiple OIS modes (allow stabilization prior to capture or not, allow correction of horizontal movement or not) which are set via the menu system on Pany bodies. Consequently, the lens must ask the body for further instructions on when and how OIS should work once the switch on the lens barrel has been set to on.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH Panasonic Lumix G 20mm F1.7 ASPH +18 more
Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 14,837
Like?
Re: Panny Lenses (with or without OIS switch) on E-M5
In reply to Detail Man, May 20, 2012

One way that it appears that none of the Lumix G Series lenses are compatible with the E-M5 IBIS (when it is enabled) relates to the linear-motion correcting axes (making it a 3-axis system only):

Andy Westlake wrote:

Detail Man wrote:

Is the E-M5 linear motion compensation worse than that ? Do M43 lenses really accurately know and report the front nodal-point to plane-of-focus distance ? I have really been skeptical of how accurately all of that could occur - especially in the case of varifocal zoom lens-systems where internal elements move ...

Personally, I suspect that the E-M5 knows this information very well indeed for certain Olympus ZD lenses; most obviously the 12-50mm set to macro mode, the upcoming 60/2.8 Macro, and the Four Thirds 35/3.5 and 50/2 macros. But probably not through the lens reporting it to the body, instead by the IS system having the properties of these lenses already programmed-in. Obviously it also needs them to report the focus distance accurately, which means it could get less-accurate if you turn the lens reset function off in the menu.

For other lenses I'd expect the macro component of the IS system to be ineffective. There's no obvious way it could work with adapted manual focus macro lenses, for example.

http://forums.dpreview.com/forums/read.asp?forum=1041&message=41386159

Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 15,444Gear list
Like?
Re: Now you have me totally confused on this thread
In reply to TEBnewyork, May 20, 2012

TEBnewyork wrote:

Assume any OIS lens I use on Oly I want to use IBIS and not OIS.......forget about the viewfinder

You are saying the in all cases IBIS works switch or no switch.

Precisely.

The restrictions/problems all have to do with whether or not you can use OIS should you want to. But if you are fine with IBIS and don't want to use OIS, there is no problem at all.

Anders W wrote:

TEBnewyork wrote:

Anders W wrote:

TEBnewyork wrote:

So, my take is even if Panny has now sorted out the problems with the pancake zoom don't bother with it on the E-M5 until Oly sorts out the problems of IBIS on an OIS lens without a switch. So, glad it is one happy family of manufacturers. for m4/3

I can see what you mean by the last sentence and am prone to agree.

However, when it comes to what I think is your practical problem for the moment -- should you get the X 14-42 or not -- your only worry is if Panny has now sorted out at least part of the problem with that lens (blurry images at certain FLs and shutter speeds with OIS off). As long as that's the case, you can happily use it on the E-M5 with (the reportedly very good) IBIS.

Correct

The rest of the Panny problem (blurry images at certain FLs and shutter speeds with OIS on) and the whole Oly problem (how their firmware handles OIS on this or that body) should be of no concern to you. The OIS on the lens will be disabled when used on the E-M5, but why would you worry about that if you have IBIS?

Because read the above posts that apparently there is a bug that IBIS doesn't work with these lenses. Only the viewfinder is stabilized.

You misread my first post. What I said was that there is a bug in the current E-M5 firmware that prevents OIS (not IBIS) from working during the exposure itself (although it works for stabilizing the EVF/OLED view prior to exposure).

This bug is consequential only for lenses with a switch for turning OIS on/off on the lens barrel. On lenses like these, you should be able to use OIS even on the E-M5 but the firmware bug currently prevents proper operation.

For lenses like the X 14-42, which do not have an OIS switch on the lens barrel, OIS will be completely off on the E-M5 (and other Oly bodies) until Oly changes its policy on how to handle these lenses.

-- hide signature --
 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH Panasonic Lumix G 20mm F1.7 ASPH +18 more
Reply   Reply with quote   Complain
Bryan Biggers
Veteran MemberPosts: 3,510Gear list
Like?
Re: Panny Lenses with no OIS switch on E-M5
In reply to Anders W, May 20, 2012

Anders W wrote:

As to technical limitations, I can't see that there would be any aside from the firmware (which can fortunately be revised). Note that the OIS system always requires lens-body interaction. It is not regulated by the switch on the lens barrel alone, even on lenses having such a switch. As you know, there are multiple OIS modes (allow stabilization prior to capture or not, allow correction of horizontal movement or not) which are set via the menu system on Pany bodies. Consequently, the lens must ask the body for further instructions on when and how OIS should work once the switch on the lens barrel has been set to on.

Well, maybe not a technical limitation, but an intentional choice. The current behavior is actually how I would want the combination to work, even it if is a bug. With the current firmware, you can turn OIS on on the lens and IBIS on in the camera at the same time. You enjoy a stabilized preview, due to the OIS, and then use the possibly superior E-M5 IBIS during the exposure. What's not to like about that? It is the best of both worlds.

 Bryan Biggers's gear list:Bryan Biggers's gear list
Olympus Tough TG-1 iHS Panasonic Lumix DMC-LX7 Olympus PEN E-P3 Olympus OM-D E-M5 Olympus PEN E-PM2
Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 14,837
Like?
Re: Panny Lenses (with an OIS switch) on E-M5
In reply to Anders W, May 20, 2012

Anders W wrote:

Detail Man wrote:

Anders W wrote:

Bryan Biggers wrote:

Anders W wrote:

According to the reports I have seen, there is a firmware bug that prevents OIS from working during the exposure itself (although it will work prior to the exposure).

Yes. It was discovered by forum user kenw when testing the IBIS on his E-M5 versus the OIS on the 14-45. Eventually, he discovered that the OIS was not turned on during the exposure itself when used on the E-M5. Here is the thread in case you are interested:

http://forums.dpreview.com/forums/read.asp?forum=1041&message=41297304

Has anyone considered that this may not be a firmware bug after all - but instead an intentionally engineered characteristic (due to technical limitations arising out camera-lens communications) ?

Good question. But yes I considered it. And judging by the E-M5 manual, it is a bug rather than a "feature". The manual (p. 49, under "cautions") provides the following advice:

When using a lens with an image stabilization function, turn off the image stabilizer function of either the lens or the camera.

That "caution" would hardly be needed if the OIS was meant to be disabled altogether. Nor does it seem likely that Oly would intentionally have allowed OIS to stabilize the view on the EVF/OLED prior to exposure but not the exposure itself, which is what the current E-M5 firmware manages to accomplish.

OK. I figured that some kind of a statement of implied functionality in this situation existed - or else kenw and yourself likely would have taken that under advisement early on.

As to technical limitations, I can't see that there would be any aside from the firmware (which can fortunately be revised). Note that the OIS system always requires lens-body interaction. It is not regulated by the switch on the lens barrel alone, even on lenses having such a switch. As you know, there are multiple OIS modes (allow stabilization prior to capture or not, allow correction of horizontal movement or not) which are set via the menu system on Pany bodies. Consequently, the lens must ask the body for further instructions on when and how OIS should work once the switch on the lens barrel has been set to on.

It is precisely such things that got me wondering. Note that Olympus' statement that you have quoted above is a somewhat general one - one that makes no specific guarantees that the E-M5 will function with Lumix G-Series lenses which include on-lens OIS ...

One wonders why it would be that Lumix G-Series lenses which do not have a user-controllable OIS switch are not able to function (did I get that right ?) - and I wonder if perhaps such an (apparent) lack of camera-lens communication capabilities in those cases (for whatever technical reasons) may (also) extend to the case of the Lumix G-Series lenses which do have a user-controllable OIS switch.

The user-accessible ON/OFF switch function seems rather surpurflous as a difference here (considering all of the other camera-lens communications complexities involved), wouldn't you say ?

It might make sense that if such an incompatibility (specifically with Lumix G-Series user-switchable OIS lenses) exists, Olympus might not (for marketing and sales reasons) "shout it from the rooftops from the get-go" ... instead allowing it to be independently discovered in venues such as this ?

Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 15,444Gear list
Like?
Re: Panny Lenses with no OIS switch on E-M5
In reply to Bryan Biggers, May 20, 2012

Bryan Biggers wrote:

Anders W wrote:

As to technical limitations, I can't see that there would be any aside from the firmware (which can fortunately be revised). Note that the OIS system always requires lens-body interaction. It is not regulated by the switch on the lens barrel alone, even on lenses having such a switch. As you know, there are multiple OIS modes (allow stabilization prior to capture or not, allow correction of horizontal movement or not) which are set via the menu system on Pany bodies. Consequently, the lens must ask the body for further instructions on when and how OIS should work once the switch on the lens barrel has been set to on.

Well, maybe not a technical limitation, but an intentional choice. The current behavior is actually how I would want the combination to work, even it if is a bug. With the current firmware, you can turn OIS on on the lens and IBIS on in the camera at the same time. You enjoy a stabilized preview, due to the OIS, and then use the possibly superior E-M5 IBIS during the exposure. What's not to like about that? It is the best of both worlds.

But, as I asked already in my prior reply to you: Why would you want to do that?

Unlike prior versions of IBIS on Oly MFT bodies, the IBIS on the E-M5 can stabilize not only the exposure but also the preview. So what would be the advantage of having OIS stabilize the preview but let IBIS stabilize the exposure?

To me that seems to make things more complicated than they have to be (and increases the risk of unnecessary mistakes when turning this or that function on or off).

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH Panasonic Lumix G 20mm F1.7 ASPH +18 more
Reply   Reply with quote   Complain
Bryan Biggers
Veteran MemberPosts: 3,510Gear list
Like?
Re: Panny Lenses with no OIS switch on E-M5
In reply to Anders W, May 20, 2012

Anders W wrote:

Bryan Biggers wrote:

Anders W wrote:

As to technical limitations, I can't see that there would be any aside from the firmware (which can fortunately be revised). Note that the OIS system always requires lens-body interaction. It is not regulated by the switch on the lens barrel alone, even on lenses having such a switch. As you know, there are multiple OIS modes (allow stabilization prior to capture or not, allow correction of horizontal movement or not) which are set via the menu system on Pany bodies. Consequently, the lens must ask the body for further instructions on when and how OIS should work once the switch on the lens barrel has been set to on.

Well, maybe not a technical limitation, but an intentional choice. The current behavior is actually how I would want the combination to work, even it if is a bug. With the current firmware, you can turn OIS on on the lens and IBIS on in the camera at the same time. You enjoy a stabilized preview, due to the OIS, and then use the possibly superior E-M5 IBIS during the exposure. What's not to like about that? It is the best of both worlds.

But, as I asked already in my prior reply to you: Why would you want to do that?

Unlike prior versions of IBIS on Oly MFT bodies, the IBIS on the E-M5 can stabilize not only the exposure but also the preview. So what would be the advantage of having OIS stabilize the preview but let IBIS stabilize the exposure?

To me that seems to make things more complicated than they have to be (and increases the risk of unnecessary mistakes when turning this or that function on or off).

Well, you might want to do it because the E-M5 only stabilizes the view when you half press, and that of course usually locks the exposure and focus unless you have it set differently, so you can't get a stabilized "compose" view. By allowing the OIS to work during all of the preview, you get a more stable view while you compose; that's what you gain. I'm using it on my 45-200 right now, and I've been leaving both systems on, I guess because the "bug" allows me to. I' not sure why you would not want the stabilized preview, but you could always turn off the OIS if you have a lens with a switch.

 Bryan Biggers's gear list:Bryan Biggers's gear list
Olympus Tough TG-1 iHS Panasonic Lumix DMC-LX7 Olympus PEN E-P3 Olympus OM-D E-M5 Olympus PEN E-PM2
Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 15,444Gear list
Like?
Re: Panny Lenses (with an OIS switch) on E-M5
In reply to Detail Man, May 20, 2012

Detail Man wrote:

Anders W wrote:

Detail Man wrote:

Anders W wrote:

Bryan Biggers wrote:

Anders W wrote:

According to the reports I have seen, there is a firmware bug that prevents OIS from working during the exposure itself (although it will work prior to the exposure).

Yes. It was discovered by forum user kenw when testing the IBIS on his E-M5 versus the OIS on the 14-45. Eventually, he discovered that the OIS was not turned on during the exposure itself when used on the E-M5. Here is the thread in case you are interested:

http://forums.dpreview.com/forums/read.asp?forum=1041&message=41297304

Has anyone considered that this may not be a firmware bug after all - but instead an intentionally engineered characteristic (due to technical limitations arising out camera-lens communications) ?

Good question. But yes I considered it. And judging by the E-M5 manual, it is a bug rather than a "feature". The manual (p. 49, under "cautions") provides the following advice:

When using a lens with an image stabilization function, turn off the image stabilizer function of either the lens or the camera.

That "caution" would hardly be needed if the OIS was meant to be disabled altogether. Nor does it seem likely that Oly would intentionally have allowed OIS to stabilize the view on the EVF/OLED prior to exposure but not the exposure itself, which is what the current E-M5 firmware manages to accomplish.

OK. I figured that some kind of a statement of implied functionality in this situation existed - or else kenw and yourself likely would have taken that under advisement early on.

As to technical limitations, I can't see that there would be any aside from the firmware (which can fortunately be revised). Note that the OIS system always requires lens-body interaction. It is not regulated by the switch on the lens barrel alone, even on lenses having such a switch. As you know, there are multiple OIS modes (allow stabilization prior to capture or not, allow correction of horizontal movement or not) which are set via the menu system on Pany bodies. Consequently, the lens must ask the body for further instructions on when and how OIS should work once the switch on the lens barrel has been set to on.

It is precisely such things that got me wondering. Note that Olympus' statement that you have quoted above is a somewhat general one - one that makes no specific guarantees that the E-M5 will function with Lumix G-Series lenses which include on-lens OIS ...

One wonders why it would be that Lumix G-Series lenses which do not have a user-controllable OIS switch are not able to function (did I get that right ?) - and I wonder if perhaps such an (apparent) lack of camera-lens communication capabilities in those cases (for whatever technical reasons) may (also) extend to the case of the Lumix G-Series lenses which do have a user-controllable OIS switch.

Yes, you got it right that lenses without an OIS switch on the barrel can't use OIS on an Oly body. And the only reason they can't is that Oly has been too lazy to write the few lines of firmware required to include an OIS on/off option in their menu system similar to the one already available on Pany bodies.

Note that all Pany bodies didn't have this option from the beginning inasmuch as the first OIS lenses all had switches. It was added by means of firmware upgrades on the early bodies, such as my G1.

Note also that OIS, when it can be enabled, can operate in mode 1 only (both pitch and yaw corrected, stabilized preview) on an Oly body. Obviously, all three modes should be selectable, as should be the case on Pany bodies too of course, but isn't on some of them (e.g., the G3).

The user-accessible ON/OFF switch function seems rather surpurflous as a difference here (considering all of the other camera-lens communications complexities involved), wouldn't you say ?

I prefer to have the switch rather than visit the menus. But I certainly agree that whether a lens happens to have a switch or not shouldn't decide whether OIS can be enabled or not.

It might make sense that if such an incompatibility (specifically with Lumix G-Series user-switchable OIS lenses) exists, Olympus might not (for marketing and sales reasons) "shout it from the rooftops from the get-go" ... instead allowing it to be independently discovered in venues such as this ?

Possibly. But I am sure the "glitch" on the E-M5 is nothing more than that: a simple oversight. How Oly has generally chosen to handle OIS lenses with and without a switch, by contrast, is a conscious decision, but a pretty lousy one.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH Panasonic Lumix G 20mm F1.7 ASPH +18 more
Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 14,837
Like?
Re: Panny Lenses (with as well as without an OIS switch) on E-M5
In reply to Anders W, May 20, 2012

Anders W wrote:

Detail Man wrote:

As to technical limitations, I can't see that there would be any aside from the firmware (which can fortunately be revised). Note that the OIS system always requires lens-body interaction. It is not regulated by the switch on the lens barrel alone, even on lenses having such a switch. As you know, there are multiple OIS modes (allow stabilization prior to capture or not, allow correction of horizontal movement or not) which are set via the menu system on Pany bodies. Consequently, the lens must ask the body for further instructions on when and how OIS should work once the switch on the lens barrel has been set to on.

It is precisely such things that got me wondering. Note that Olympus' statement that you have quoted above is a somewhat general one - one that makes no specific guarantees that the E-M5 will function with Lumix G-Series lenses which include on-lens OIS ...

One wonders why it would be that Lumix G-Series lenses which do not have a user-controllable OIS switch are not able to function (did I get that right ?) - and I wonder if perhaps such an (apparent) lack of camera-lens communication capabilities in those cases (for whatever technical reasons) may (also) extend to the case of the Lumix G-Series lenses which do have a user-controllable OIS switch.

Yes, you got it right that lenses without an OIS switch on the barrel can't use OIS on an Oly body. And the only reason they can't is that Oly has been too lazy to write the few lines of firmware required to include an OIS on/off option in their menu system similar to the one already available on Pany bodies.

I find the "lazy goofs in engineering" hypothesis rather hard to swallow. Perhaps a genuine technical limitation due to Olympus' chosen design-course where it comes to E-M5 functionality exists, and/or the many millions of dollars involved in potential Olympus Zukio (as opposed to Lumix G-Series) lens-sales exists ? ...

Note that it appears (according to Andy Westlake 's speculations, anyway) that the linear-motion correcting 2 axes out of the 5 axes are quite likely non-functional - except for in the case of certain fairly recently released Olympus Zukio lenses:

See: http://forums.dpreview.com/forums/read.asp?forum=1041&message=41386159

(There) is a case where it may well be that specific technical barriers exist where it would come to attempting to implement the linear-motion correcting functionality with other lenses. However, such things do have way of "dovetailing rather nicely" with resulting volume of Olympus lens sales.

Note that all Pany bodies didn't have this option from the beginning inasmuch as the first OIS lenses all had switches. It was added by means of firmware upgrades on the early bodies, such as my G1.

Showing that now (in "modern times"), the "lazy goofs in engineering" at Olympus hypothesis may appear to be a weak argument ?

Note also that OIS, when it can be enabled, can operate in mode 1 only (both pitch and yaw corrected, stabilized preview) on an Oly body. Obviously, all three modes should be selectable, as should be the case on Pany bodies too of course, but isn't on some of them (e.g., the G3).

It is understandable that some barriers (in all cases discussed) may exist where it comes to the more technically complicated Mode 2, as well as the mode which enables angular-motion correction in the vertical axis only (so as not to mess up horizontal-axis "panoramaic" motions by the user).

But, Olympus has here designed a brand new camera system (certainly with a full knowledge of all those factors firmly in mind).

The user-accessible ON/OFF switch function seems rather surpurflous as a difference here (considering all of the other camera-lens communications complexities involved), wouldn't you say ?

I prefer to have the switch rather than visit the menus. But I certainly agree that whether a lens happens to have a switch or not shouldn't decide whether OIS can be enabled or not.

Once again, I wonder if this points to a deeper issue(s) surrounding the use of Lumix G-Series lenses with OIS on the E-M5 ?

It might make sense that if such an incompatibility (specifically with Lumix G-Series user-switchable OIS lenses) exists, Olympus might not (for marketing and sales reasons) "shout it from the rooftops from the get-go" ... instead allowing it to be independently discovered in venues such as this ?

Possibly. But I am sure the "glitch" on the E-M5 is nothing more than that: a simple oversight.

Hard (for me) to imagine such things being a mere "engineering oversight". I hope that you optimism in these regards will turn out to have been well-founded.

How Oly has generally chosen to handle OIS lenses with and without a switch, by contrast, is a conscious decision, but a pretty lousy one.

Well, there you go ... it (to me) may be an indication of (consciously arrived at) engineering and/or marketing reasons and/or motivations ...

Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 15,444Gear list
Like?
Re: Panny Lenses with no OIS switch on E-M5
In reply to Bryan Biggers, May 20, 2012

Bryan Biggers wrote:

Anders W wrote:

Bryan Biggers wrote:

Anders W wrote:

As to technical limitations, I can't see that there would be any aside from the firmware (which can fortunately be revised). Note that the OIS system always requires lens-body interaction. It is not regulated by the switch on the lens barrel alone, even on lenses having such a switch. As you know, there are multiple OIS modes (allow stabilization prior to capture or not, allow correction of horizontal movement or not) which are set via the menu system on Pany bodies. Consequently, the lens must ask the body for further instructions on when and how OIS should work once the switch on the lens barrel has been set to on.

Well, maybe not a technical limitation, but an intentional choice. The current behavior is actually how I would want the combination to work, even it if is a bug. With the current firmware, you can turn OIS on on the lens and IBIS on in the camera at the same time. You enjoy a stabilized preview, due to the OIS, and then use the possibly superior E-M5 IBIS during the exposure. What's not to like about that? It is the best of both worlds.

But, as I asked already in my prior reply to you: Why would you want to do that?

Unlike prior versions of IBIS on Oly MFT bodies, the IBIS on the E-M5 can stabilize not only the exposure but also the preview. So what would be the advantage of having OIS stabilize the preview but let IBIS stabilize the exposure?

To me that seems to make things more complicated than they have to be (and increases the risk of unnecessary mistakes when turning this or that function on or off).

Well, you might want to do it because the E-M5 only stabilizes the view when you half press, and that of course usually locks the exposure and focus unless you have it set differently, so you can't get a stabilized "compose" view. By allowing the OIS to work during all of the preview, you get a more stable view while you compose; that's what you gain. I'm using it on my 45-200 right now, and I've been leaving both systems on, I guess because the "bug" allows me to. I' not sure why you would not want the stabilized preview, but you could always turn off the OIS if you have a lens with a switch.

OK. I see your point about having the preview always stabilized (as you can with OIS) versus having it stabilized only when you half-press (as you can with IBIS), although it wouldn't matter much to me personally. I begin by focusing (and move the AF frame prior to that if I need to) and then fine-tune framing (for which stabilization is useful). And I don't have exposure locked when half-pressing the shutter but lock it by other means (usually by shooting in M mode). But this is of course a matter of personal preference.

At any rate, I am afraid you will lose this rather special mode of operation unless you are ready to stick with the current E-M5 firmware release forever.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH Panasonic Lumix G 20mm F1.7 ASPH +18 more
Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 15,444Gear list
Like?
Re: Panny Lenses (with an OIS switch) on E-M5
In reply to Detail Man, May 20, 2012

Detail Man wrote:

I find the "lazy goofs in engineering" hypothesis rather hard to swallow. Perhaps a genuine technical limitation due to Olympus' chosen design-course where it comes to E-M5 functionality exists, and/or the many millions of dollars involved in potential Olympus Zukio (as opposed to Lumix G-Series) lens-sales exists.

Nah. If there would indeed be technical or business reasons, the logical thing to do would have been to disable OIS altogether, on lenses with as well as without a switch rather than leaving it in a semi-operational stage without properly documenting that fact.

Note that it appears (according to Andy Westlake 's speculations, anyway) that the linear-motion correcting 2 axes out of the 5 axes are quite likely non-functional - except for in the case of certain fairly recently released Olympus Zukio lenses. Here is a case where it may well be that specific technical barriers exist where it would come to attempting to implement the linear-motion correcting functionality with other lenses. However, such things do have way of "dovetailing rather nicely) with the resulting volume of Olympus Zukio lens sales ...

OK. I saw your post about that. Just didn't have the time to respond yet. But this is another issue so I'll deal with it in a reply to the post where you first brought it up.

Showing that now (in "modern times"), the "lazy goofs in engineering" at Olympus hypothesis may appear to be a weak argument ?

No. Whereas we can perhaps not expect Oly to immediately rush out new firmware for all their bodies just because Pany decides to skip the OIS switch on some lenses, I think we can legitimately expect that they fix the problem on newly released bodies right away and on older ones whenever it is time for a firmware update anyway. The MFT mount is supposed to be a standard that operates across brand borders, right?

It is understandable that some barriers (in all cases discussed) may exist where it comes to the more technically complicated Mode 2, as well as the mode which causes angular-motion correction in the vertical axis only (so as not to mess up horizontal-axis "panoramaic" motions on the part of the user).

No, I don't think so. It's just a matter of allowing the user (via the body firmware) to answer yes or no to some very simple questions asked by the OIS system. This is at most a couple of hours work for a skilled programmer. And none of the three OIS modes is any more technically complicated than the others. It's just a matter of allowing or not allowing the OIS system to do certain things that it is already able to do without any help by Oly.

Once again, I wonder if this points to a deeper issue(s) surrounding the use og Lumix G-Series lenses with OIS on the E-M5 ?

I find absolutely no reason to think so. On a lens with switch, the switch toggles the bit that means OIS on or off. A lens without the switch expects to get the same bit from the body instead of from the switch. That's all there is to it.

Hard (for me) to imagine such things being a mere "engineering oversight". I hope that you optimism in these regards will turn out to have been well-founded.

Well, for me it is very easy to imagine that it is a simple oversight, both on account of what I have seen others do and what I have done myself. I have a fair amount of programming experience, which means that I have full confidence in Murphy's law.

Well, there you go ... it (to me) points to (consciously arrived at) engineering and/or marketing reasons and/or motivations ...

When it comes to handling OIS more generally (lack of menu item for switchless lenses, operation in mode 1 only), yes. When it comes to the peculiarity of the E-M5 (OIS operative in preview but not during the exposure itself when OIS is set to on via the barrel switch), no.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH Panasonic Lumix G 20mm F1.7 ASPH +18 more
Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 15,444Gear list
Like?
Re: Panny Lenses (with or without OIS switch) on E-M5
In reply to Detail Man, May 20, 2012

Detail Man wrote:

One way that it appears that none of the Lumix G Series lenses are compatible with the E-M5 IBIS (when it is enabled) relates to the linear-motion correcting axes (making it a 3-axis system only):

Once a certain lens is mounted, the only free parameters required for all five axes (all five types of camera movement) to be properly corrected are the focal length (on zooms) and the focus distance. I am pretty sure that this information is already at the body's disposal on all electrically connected MFT lenses. Any other information required follows logically from these two once the body knows (as it does) what particular lens is mounted and the relevant specs of that lens. Information about these relevant specs can either be provided by the lens itself (which is the best solution) or stored as lookup table in the body firmware. So I doubt there's much of a real technical problem on this score either. As always, it still remains to be seen whether this simple job has been done and done right.

Andy Westlake wrote:

Detail Man wrote:

Is the E-M5 linear motion compensation worse than that ? Do M43 lenses really accurately know and report the front nodal-point to plane-of-focus distance ? I have really been skeptical of how accurately all of that could occur - especially in the case of varifocal zoom lens-systems where internal elements move ...

Personally, I suspect that the E-M5 knows this information very well indeed for certain Olympus ZD lenses; most obviously the 12-50mm set to macro mode, the upcoming 60/2.8 Macro, and the Four Thirds 35/3.5 and 50/2 macros. But probably not through the lens reporting it to the body, instead by the IS system having the properties of these lenses already programmed-in. Obviously it also needs them to report the focus distance accurately, which means it could get less-accurate if you turn the lens reset function off in the menu.

For other lenses I'd expect the macro component of the IS system to be ineffective. There's no obvious way it could work with adapted manual focus macro lenses, for example.

http://forums.dpreview.com/forums/read.asp?forum=1041&message=41386159

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH Panasonic Lumix G 20mm F1.7 ASPH +18 more
Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 14,837
Like?
Re: Panny Lenses (with an OIS switch) on E-M5
In reply to Anders W, May 21, 2012

Anders W wrote:

Detail Man wrote:

I find the "lazy goofs in engineering" hypothesis rather hard to swallow. Perhaps a genuine technical limitation due to Olympus' chosen design-course where it comes to E-M5 functionality exists, and/or the many millions of dollars involved in potential Olympus Zukio (as opposed to Lumix G-Series) lens-sales exists.

Nah. If there would indeed be technical or business reasons, the logical thing to do would have been to disable OIS altogether, on lenses with as well as without a switch rather than leaving it in a semi-operational stage without properly documenting that fact.

Perhaps. But the "lazy goofs in engineering" hypothesis still to me seems rather an unlikely stretch.

From a marketing standpoint, if the E-M5 (for, let us say, purely technical reasons) happened to function successfully with other brands (other than Lumix G-Series) of OIS lenses - but not with Lumix G-Series OIS lenses - it would seem to be potentially quite damaging (from a public-relations standpoint) to count on prospective customers to be able to accurately differentiate between (just) Lumix G-Series OIS compatability, and the impression of a lack of a "global" OIS-lens compatibility (applying to all brands of OIS-lenses).

In such a case, it might seem better to allow such specifics to remain somewhat vague in the course of product promotion/documentation. Silence can surely be more "golden" than specific details which might serve to inordinately confuse and cause "dissonance" in the minds of customers

Showing that now (in "modern times"), the "lazy goofs in engineering" at Olympus hypothesis may appear to be a weak argument ?

No. Whereas we can perhaps not expect Oly to immediately rush out new firmware for all their bodies just because Pany decides to skip the OIS switch on some lenses, ..

Hold the phone ! All of these conditions have existed throughout the course of (specifically) the E-M5's engineering product development ...

... I think we can legitimately expect that they fix the problem on newly released bodies right away and on older ones whenever it is time for a firmware update anyway. The MFT mount is supposed to be a standard that operates across brand borders, right?

Well, not only is the E-M5 a new release, but the FW Level 1.1 Update (interestingly) did nothing whatsoever to address this matter. A seemingly strange and persistent "blind spot", then ? ...

It is understandable that some barriers (in all cases discussed) may exist where it comes to the more technically complicated Mode 2, as well as the mode which causes angular-motion correction in the vertical axis only (so as not to mess up horizontal-axis "panoramaic" motions on the part of the user).

No, I don't think so. It's just a matter of allowing the user (via the body firmware) to answer yes or no to some very simple questions asked by the OIS system. This is at most a couple of hours work for a skilled programmer. And none of the three OIS modes is any more technically complicated than the others. It's just a matter of allowing or not allowing the OIS system to do certain things that it is already able to do without any help by Oly.

I would "buy that" on a technical level. But, that might seem to only bolster the strength of a possible "marketing/monetary" related decision, then ? Of, course, I know that you would likely disagree with the above - so there is likely no need for you to re-iterate your viewpoint on that ...

Hard (for me) to imagine such things being a mere "engineering oversight". I hope that you optimism in these regards will turn out to have been well-founded.

Well, for me it is very easy to imagine that it is a simple oversight, both on account of what I have seen others do and what I have done myself. I have a fair amount of programming experience, which means that I have full confidence in Murphy's law.

When one reaches the point of placing their bets on "Murphy's Law" at play (which, from my own background in engineering design/development, I do indeed know well from direct experiences), there may also exist the possibility that a strong hope exists that (the rarely welcome) "Murphy's Law" will have turned out (in hindsight) to have eclipsed (the dreaded) "Marketing's Law" ...

Well, there you go ... it (to me) points to (consciously arrived at) engineering and/or marketing reasons and/or motivations ...

When it comes to handling OIS more generally (lack of menu item for switchless lenses, operation in mode 1 only), yes. When it comes to the peculiarity of the E-M5 (OIS operative in preview but not during the exposure itself when OIS is set to on via the barrel switch), no.

I hope (for the sake of all E-M5 customers/users) that you are correct. Just an "oversight" missed by the E-M5's design-group (and everybody else at Olympus) in all of the various Firmware Levels released throughout the E-M5 development-cycle (also includng publicly distributed FW Level 1.1).

Faith consists in believing not what seems true, but what seems false to our understanding .
- Voltaire

Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 15,444Gear list
Like?
Re: Panny Lenses (with an OIS switch) on E-M5
In reply to Detail Man, May 21, 2012

Detail Man wrote:

Perhaps. But the "lazy goofs in engineering" hypothesis still to me seems rather an unlikely stretch.

From a marketing standpoint, if the E-M5 (for, let us say, purely technical reasons) happened to function successfully with other brands (other than Lumix G-Series) of OIS lenses - but not with Lumix G-Series OIS lenses - it would seem to be potentially quite damaging (from a public-relations standpoint) to count on prospective customers to be able to accurately differentiate between (just) Lumix G-Series OIS compatability, and the impression of a lack of a "global" OIS-lens compatibility (applying to all brands of OIS-lenses).

In such a case, it might seem better to allow such specifics to remain somewhat vague in the course of product promotion/documentation. Silence can surely be more "golden" than specific details which might serve to inordinately confuse and cause "dissonance" in the minds of customers

Hmm. As far as I am aware there are no OIS lenses where the OIS could conceivably work on the E-M5 other than those marked Lumix.

Hold the phone ! All of these conditions have existed throughout the course of (specifically) the E-M5's engineering product development ...

In all likelihood yes. The first switchless OIS lens is the 14-42 introduced already with the G2 (if I recall correctly). But this doesn't matter much. The firmware, as we know, is about the last thing they finish on a new camera.

Well, not only is the E-M5 a new release, but the FW Level 1.1 Update (interestingly) did nothing whatsoever to address this matter. A seemingly strange and persistent "blind spot", then ? ...

In all likelihood, the bug hadn't even been discovered when FW 1.1 was released. That release came so early that it was probably based on what had already been discovered in house or by the selected few who had access to the camera before it started shipping.

I hope (for the sake of all E-M5 customers/users) that you are correct. Just an "oversight" missed by the E-M5's design-group (and everybody else at Olympus) in all of the various Firmware Levels released throughout the E-M5 development-cycle (also includng publicly distributed FW Level 1.1).

Yes, and if you have looked at Jonas B's thread about bugs in FW 1.1, it was hardly the only thing they overlooked. If many things can go wrong, quite a few will. Nothing surprising here. As far as I can tell, early firmware versions, including that released when a new camera first ships, like 1.1, usually contains quite a few bugs. The E-M5 doesn't seem to be an exception here although I don't have the impression that it is worse than average either. Besides, the particular bug at issue here is of really marginal importance. If the new IBIS works as well as it reportedly does, I don't see why people should be desperate to enable OIS instead.

Faith consists in believing not what seems true, but what seems false to our understanding .
- Voltaire

I put just as little stock in faith as Voltaire. Conjectures and refutations is more my kind of ball-game.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH Panasonic Lumix G 20mm F1.7 ASPH +18 more
Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 14,837
Like?
Re: All Lenses (with respect Linear-Motion Corrections) on the E-M5
In reply to Anders W, May 21, 2012

Anders W wrote:

Detail Man wrote:

One way that it appears that none of the Lumix G Series lenses are compatible with the E-M5 IBIS (when it is enabled) relates to the linear-motion correcting axes (making it a 3-axis system only).

Once a certain lens is mounted, the only free parameters required for all five axes (all five types of camera movement) to be properly corrected are the focal length (on zooms) and the focus distance.

Correct. You stated this in any early thread relating to the E-M5 5-axis IBIS:

One implication of the above is that compensation for vertical/horizontal shift will not work unless the lens used is electrically connected to the body and can relay information about focus distance (unless there is an option for manually entering the focus distance on a shot-by-shot basis).

http://forums.dpreview.com/forums/read.asp?forum=1041&message=40558437

I am pretty sure that this information is already at the body's disposal on all electrically connected MFT lenses.

Well, [if] such is the case:

That each stand-alone M43 lens-system knows with some degree of precision what the front nodal-plane to plane-of-focus distance is - based upon mathematical calculations derived from from it's own internal mechanically-based measurements of the position of the movable lens-elements involved in focusing adjustements (to the extent that such measured data actually accurately reflects the lens-system's focal-length at infinity), the "effective" focal-length at close range (based upon the relevant magnification-factor), and the correspondingly (somehow) derived physical postion of the front nodal-point of the lens-system itself as a function of the physical position of the movable lens-element assembly - and is (furthermore) fully capable of communicating that data (in whole or in part) to any and all M43 bodies which that lens-sytem interfaces with ...

... then why do you think that Andy Westlake does not appear to agree with your viewpoint ?

It sounds like you may know something that he does not, and that perhaps you may want to contact our friend Andy Westlake in order to persuade him that his viewpoint is likely incorrect ?
Or, perhaps, the converse may be true ?

Any other information required follows logically from these two once the body knows (as it does) what particular lens is mounted and the relevant specs of that lens.

Yes, the (relatively) "easy" part.

Information about these relevant specs can either be provided by the lens itself (which is the best solution) or stored as lookup table in the body firmware.

In an academic "textbook" case, yes.

So I doubt there's much of a real technical problem on this score either.

In an academic "textbook" case, yes.

As always, it still remains to be seen whether this simple job has been done and done right.

Yes, this is how people often speak in the halls of academia at the University -
prior to then sauntering "out to lunch" relatively unencumbered by return-times.

Andy Westlake wrote:

Detail Man wrote:

Is the E-M5 linear motion compensation worse than that ? Do M43 lenses really accurately know and report the front nodal-point to plane-of-focus distance ? I have really been skeptical of how accurately all of that could occur - especially in the case of varifocal zoom lens-systems where internal elements move ...

Personally, I suspect that the E-M5 knows this information very well indeed for certain Olympus ZD lenses; most obviously the 12-50mm set to macro mode, the upcoming 60/2.8 Macro, and the Four Thirds 35/3.5 and 50/2 macros. But probably not through the lens reporting it to the body, instead by the IS system having the properties of these lenses already programmed-in. Obviously it also needs them to report the focus distance accurately, which means it could get less-accurate if you turn the lens reset function off in the menu.

For other lenses I'd expect the macro component of the IS system to be ineffective. There's no obvious way it could work with adapted manual focus macro lenses, for example.

http://forums.dpreview.com/forums/read.asp?forum=1041&message=41386159

Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 14,837
Like?
Re: Panny Lenses (with an OIS switch) on E-M5
In reply to Anders W, May 21, 2012

Anders W wrote:

Detail Man wrote:

Perhaps. But the "lazy goofs in engineering" hypothesis still to me seems rather an unlikely stretch.

From a marketing standpoint, if the E-M5 (for, let us say, purely technical reasons) happened to function successfully with other brands (other than Lumix G-Series) of OIS lenses - but not with Lumix G-Series OIS lenses - it would seem to be potentially quite damaging (from a public-relations standpoint) to count on prospective customers to be able to accurately differentiate between (just) Lumix G-Series OIS compatability, and the impression of a lack of a "global" OIS-lens compatibility (applying to all brands of OIS-lenses).

In such a case, it might seem better to allow such specifics to remain somewhat vague in the course of product promotion/documentation. Silence can surely be more "golden" than specific details which might serve to inordinately confuse and cause "dissonance" in the minds of customers

Hmm. As far as I am aware there are no OIS lenses where the OIS could conceivably work on the E-M5 other than those marked Lumix.

Really ? You would know far more about this than I. Was assuming that there may exist a Canon lens (or perhaps a Nikon lens) case. A lens-guru I am not. I will take your word for that ... at which point this rather glaring omission of functionality (might) seem even more suspect, then ...

Hold the phone ! All of these conditions have existed throughout the course of (specifically) the E-M5's engineering product development ...

In all likelihood yes. The first switchless OIS lens is the 14-42 introduced already with the G2 (if I recall correctly). But this doesn't matter much. The firmware, as we know, is about the last thing they finish on a new camera.

Have you (or do you) work(ed) in a private engineering/manufacturing environment, or in an academic/educational environment ? I am wondering if perhaps it might be the latter of the two ?

Well, not only is the E-M5 a new release, but the FW Level 1.1 Update (interestingly) did nothing whatsoever to address this matter. A seemingly strange and persistent "blind spot", then ? ...

In all likelihood, the bug hadn't even been discovered when FW 1.1 was released. That release came so early that it was probably based on what had already been discovered in house or by the selected few who had access to the camera before it started shipping.

That seems a far stretch, especially for a company with $1.7 Billion in hidden losses immersed in the scandal of their corporate lifetime, and counting on products such as the E-M5 to help to rescue them from their impending fiscal perils. Very sloppy "engineering form" such would seem to be ...

Why they would be such "perfectionists" in other respects of the design/development process, yet somehow fail to bother the notice the obvious (particulalry in light of your statement that Lumix G-Series lenses are the only non-Olympus Zukio lens-brand in existence), is beyond me (and, as well, might perhaps tend to cause one to look beyond a simple "Murphy's Law" scenario) ? ...

I hope (for the sake of all E-M5 customers/users) that you are correct. Just an "oversight" missed by the E-M5's design-group (and everybody else at Olympus) in all of the various Firmware Levels released throughout the E-M5 development-cycle (also includng publicly distributed FW Level 1.1).

Yes, and if you have looked at Jonas B's thread about bugs in FW 1.1, it was hardly the only thing they overlooked. If many things can go wrong, quite a few will. Nothing surprising here. As far as I can tell, early firmware versions, including that released when a new camera first ships, like 1.1, usually contains quite a few bugs. The E-M5 doesn't seem to be an exception here although I don't have the impression that it is worse than average either.

What particular knowledge-basis is your rating based upon ?

Besides, the particular bug at issue here is of really marginal importance. If the new IBIS works as well as it reportedly does, I don't see why people should be desperate to enable OIS instead.

Perhaps it may be better to rely upon personally conducted testing (as opposed to anecdotal testimonials) where it comes to these most complicated functions that are (understandably) rather hard (for any user) to evaluate in what truly constitutes a tangibly objective, repeatable manner ?

Faith consists in believing not what seems true, but what seems false to our understanding .
- Voltaire

I put just as little stock in faith as Voltaire. Conjectures and refutations is more my kind of ball-game.

I am aware that, in the spirit of true originality, you often do seem to prefer the wisdoms of "W".

DM ...

Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 15,444Gear list
Like?
Re: All Lenses (with respect Linear-Motion Corrections) on the E-M5
In reply to Detail Man, May 21, 2012

Detail Man wrote:

Well, [if] such is the case:

That each stand-alone M43 lens-system knows with some degree of precision what the front nodal-plane to plane-of-focus distance is - based upon mathematical calculations derived from from it's own internal mechanically-based measurements of the position of the movable lens-elements involved in focusing adjustements (to the extent that such measured data actually accurately reflects the lens-system's focal-length at infinity), the "effective" focal-length at close range (based upon the relevant magnification-factor), and the correspondingly (somehow) derived physical postion of the front nodal-point of the lens-system itself as a function of the physical position of the movable lens-element assembly - and is (furthermore) fully capable of communicating that data (in whole or in part) to any and all M43 bodies which that lens-sytem interfaces with ...

Note that I suggested two solutions: That the lens communicates the information or that it is stored in a lockup table in the body firmware. Note also that it suffices that the information is reasonably accurate for the purpose at hand. Note finally that it is sufficient that the information is provided for circumstances in which correction for vertical and horizontal shift actually makes a difference. At longer focus distances, correction for these types of camera movements makes no practical difference.

... then why do you think that Andy Westlake does not appear to agree with your viewpoint ?

It sounds like you may know something that he does not, and that perhaps you may want to contact our friend Andy Westlake in order to persuade him that his viewpoint is likely incorrect ?
Or, perhaps, the converse may be true ?

I am not sure Andy Westlake and I are in any serious disagreement here and I doubt that he has more inside information about how it actually works than I do.

Any other information required follows logically from these two once the body knows (as it does) what particular lens is mounted and the relevant specs of that lens.

Yes, the (relatively) "easy" part.

Information about these relevant specs can either be provided by the lens itself (which is the best solution) or stored as lookup table in the body firmware.

In an academic "textbook" case, yes.

So I doubt there's much of a real technical problem on this score either.

In an academic "textbook" case, yes.

As always, it still remains to be seen whether this simple job has been done and done right.

Yes, this is how people often speak in the halls of academia at the University -
prior to then sauntering "out to lunch" relatively unencumbered by return-times.

No need to distinguish between academia and other circumstances here, I think. To determine the effective FL and the location of the front nodal point at different focus distances with sufficient precision for the purpose at hand would hardly be "rocket science" for an optical engineer. In all likelihood, they can get it straight out of the software by means of which they originally created the lens design.

Andy Westlake wrote:

Detail Man wrote:

Is the E-M5 linear motion compensation worse than that ? Do M43 lenses really accurately know and report the front nodal-point to plane-of-focus distance ? I have really been skeptical of how accurately all of that could occur - especially in the case of varifocal zoom lens-systems where internal elements move ...

Personally, I suspect that the E-M5 knows this information very well indeed for certain Olympus ZD lenses; most obviously the 12-50mm set to macro mode, the upcoming 60/2.8 Macro, and the Four Thirds 35/3.5 and 50/2 macros. But probably not through the lens reporting it to the body, instead by the IS system having the properties of these lenses already programmed-in. Obviously it also needs them to report the focus distance accurately, which means it could get less-accurate if you turn the lens reset function off in the menu.

For other lenses I'd expect the macro component of the IS system to be ineffective. There's no obvious way it could work with adapted manual focus macro lenses, for example.

http://forums.dpreview.com/forums/read.asp?forum=1041&message=41386159

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH Panasonic Lumix G 20mm F1.7 ASPH +18 more
Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 15,444Gear list
Like?
Re: Panny Lenses (with an OIS switch) on E-M5
In reply to Detail Man, May 21, 2012

Detail Man wrote:

Really ? You would know far more about this than I. Was assuming that there may exist a Canon lens (or perhaps a Nikon lens) case. A lens-guru I am not. I will take your word for that ... at which point this rather glaring omission of functionality (might) seem even more suspect, then ...

Well, to my knowledge, the OIS system on any lens requires electric power to function. Furthermore, this power is expected to be supplied by the body. Do you know of any OIS lens other than those marked Lumix for which an MFT body can meet that requirement?

Have you (or do you) work(ed) in a private engineering/manufacturing environment, or in an academic/educational environment ? I am wondering if perhaps it might be the latter of the two ?

In what way would that be relevant for the issue at hand (time of firmware finalization)? To my knowledge there is tangible evidence that firmware is the component finalized last. In fact, I am sure that the firmware on the E-M5 isn't final quite yet.

That seems a far stretch, especially for a company with $1.7 Billion in hidden losses immersed in the scandal of their corporate lifetime, and counting on products such as the E-M5 to help to rescue them from their impending fiscal perils. Very sloppy "engineering form" such would seem to be ...

I don't think Oly is sloppier than anyone else. Things like these happen all the time. And in some cases, the problem is far more serious and far more costly to fix than in the present case. Think of software-related recalls in the automotive industry for example, especially in the domain of safety systems. Certainly not meant to happen. Still does happen.

Why they would be such "perfectionists" in other respects of the design/development process, yet somehow fail to bother the notice the obvious (particulalry in light of your statement that Lumix G-Series lenses are the only non-Olympus Zukio lens-brand in existence), is beyond me (and, as well, might perhaps tend to cause one to look beyond a simple "Murphy's Law" scenario) ? ...

In my experience, even the best fail to deliver bug-free software on the first try. It would surprise me if your experiences are any different.

What particular knowledge-basis is your rating based upon ?

No rating at all. I am not sure there is one for the type of product we are talking about. But none of the bugs so far reported seem to be a kind that seriously limits the functionality of the product. That's what I meant when I said that I don't think the E-M5 is worse than average. The only thing that currently worries me are the reports about banding when used with certain lenses, in particular the 20. But (regrettably) I am far from convinced that this is merely a firmware issue.

Perhaps it may be better to rely upon personally conducted testing (as opposed to anecdotal testimonials) where it comes to these most complicated functions that are (understandably) rather hard (for any user) to evaluate in what truly constitutes a tangibly objective, repeatable manner ?

I was not relying merely on anecdotal evidence. There are at least a three pretty systematic tests of the new IBIS out there (optyczne.pl, Pekka Potka, and kenw). And I am not sure that my own tests (when I am in a position to perform them, which I am not in this case) are always superior to those performed by others. That said, it would obviously be preferable if everyone could decide for themselves whether they should rely on OIS or IBIS with a certain lens on the E-M5. And to reach a conclusive verdict on that issue, OIS must of course first be made fully operative on the camera in question.

I am aware that, in the spirit of true originality, you often do seem to prefer the wisdoms of "W".

Sure. Just as you prefer the wisdom of "DM". But I try my very best to have the wisdom of "W" subjected to the wisdom of "P".

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH Panasonic Lumix G 20mm F1.7 ASPH +18 more
Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 14,837
Like?
Re: All Lenses (with respect Linear-Motion Corrections) on the E-M5
In reply to Anders W, May 21, 2012

Anders W wrote:

Detail Man wrote:

Well, [if] such is the case:

That each stand-alone M43 lens-system knows with some degree of precision what the front nodal-plane to plane-of-focus distance is - based upon mathematical calculations derived from from it's own internal mechanically-based measurements of the position of the movable lens-elements involved in focusing adjustements (to the extent that such measured data actually accurately reflects the lens-system's focal-length at infinity), the "effective" focal-length at close range (based upon the relevant magnification-factor), and the correspondingly (somehow) derived physical postion of the front nodal-point of the lens-system itself as a function of the physical position of the movable lens-element assembly - and is (furthermore) fully capable of communicating that data (in whole or in part) to any and all M43 bodies which that lens-sytem interfaces with ...

Note that I suggested two solutions: That the lens communicates the information or that it is stored in a lockup table in the body firmware.

Note that "(in whole or in part)" above addressed your (already understood by me) alternatives.

Note also that it suffices that the information is reasonably accurate for the purpose at hand.

That's easy for us to say. It has a sweeping and sufficient sounding ring to it.

Note finally that it is sufficient that the information is provided for circumstances in which correction for vertical and horizontal shift actually makes a difference. At longer focus distances, correction for these types of camera movements makes no practical difference.

Seems like marginalizing issues implicitly involved in the actualization of much celebrated "advances".

then why do you think that Andy Westlake does not appear to agree with your viewpoint ?

It sounds like you may know something that he does not, and that perhaps you may want to contact our friend Andy Westlake in order to persuade him that his viewpoint is likely incorrect ?
Or, perhaps, the converse may be true ?

I am not sure Andy Westlake and I are in any serious disagreement here and I doubt that he has more inside information about how it actually works than I do.

Well, he (too) opts for the "in-camera look-up table" hypothesis based on some possibly crude lens-provided mere estimates of lens-system focal-length. But we strongly suspect that he has actually held an E-M5 in his hands - and further perhaps (himself, or his associate Richard Butler) even performed some actual evaluations of the effectiveness of the linear-motion correcting functionality demonstrated by the E-M5 with the particular Olympus Zukio lens that he cites. It would be interesting if Andy might explain to us how his knowledge may differ from the entirely hands-off approach from people like us who have never even used or tested the E-M5 whatsoever

I wonder why Andy (by all appearances) has not included a single Lumix lens in his speculation ???

Any other information required follows logically from these two once the body knows (as it does) what particular lens is mounted and the relevant specs of that lens.

Yes, the (relatively) "easy" part.

Information about these relevant specs can either be provided by the lens itself (which is the best solution) or stored as lookup table in the body firmware.

In an academic "textbook" case, yes.

So I doubt there's much of a real technical problem on this score either.

In an academic "textbook" case, yes.

As always, it still remains to be seen whether this simple job has been done and done right.

Yes, this is how people often speak in the halls of academia at the University -
prior to then sauntering "out to lunch" relatively unencumbered by return-times.

No need to distinguish between academia and other circumstances here, I think.

As that is in itself an "academic" statement, I will assume the former (unless informed of the latter)

To determine the effective FL and the location of the front nodal point at different focus distances with sufficient precision for the purpose at hand would hardly be "rocket science" for an optical engineer. In all likelihood, they can get it straight out of the software by means of which they originally created the lens design.

That's easy for us to say. It has a sweeping and sufficient sounding ring to it.

Andy Westlake wrote:

Detail Man wrote:

Is the E-M5 linear motion compensation worse than that ? Do M43 lenses really accurately know and report the front nodal-point to plane-of-focus distance ? I have really been skeptical of how accurately all of that could occur - especially in the case of varifocal zoom lens-systems where internal elements move ...

Personally, I suspect that the E-M5 knows this information very well indeed for certain Olympus ZD lenses; most obviously the 12-50mm set to macro mode, the upcoming 60/2.8 Macro, and the Four Thirds 35/3.5 and 50/2 macros. But probably not through the lens reporting it to the body, instead by the IS system having the properties of these lenses already programmed-in. Obviously it also needs them to report the focus distance accurately, which means it could get less-accurate if you turn the lens reset function off in the menu.

For other lenses I'd expect the macro component of the IS system to be ineffective. There's no obvious way it could work with adapted manual focus macro lenses, for example.

http://forums.dpreview.com/forums/read.asp?forum=1041&message=41386159

Reply   Reply with quote   Complain
Detail Man
Forum ProPosts: 14,837
Like?
Re: Panny Lenses (with an OIS switch) on E-M5
In reply to Anders W, May 21, 2012

Anders W wrote:

Detail Man wrote:

Really ? You would know far more about this than I. Was assuming that there may exist a Canon lens (or perhaps a Nikon lens) case. A lens-guru I am not. I will take your word for that ... at which point this rather glaring omission of functionality (might) seem even more suspect, then ...

Well, to my knowledge, the OIS system on any lens requires electric power to function. Furthermore, this power is expected to be supplied by the body. Do you know of any OIS lens other than those marked Lumix for which an MFT body can meet that requirement?

A lens-guru I am not. I will take your word for that. Panasonic must be "it" ... so they FUBARed it.

Have you (or do you) work(ed) in a private engineering/manufacturing environment, or in an academic/educational environment ? I am wondering if perhaps it might be the latter of the two ?

In what way would that be relevant for the issue at hand (time of firmware finalization)?

One would likely need to have worked within a private engineering/manufacturing environment to understand that relevancy (in ways that constitute more than academic speculations from afar) ...

To my knowledge there is tangible evidence that firmware is the component finalized last.

"Last", but certainly not "arbitrary" or "least" ...

In fact, I am sure that the firmware on the E-M5 isn't final quite yet.

He he. They should rightly be paying customers for "beta-testing services rendered", then ? The money truly seems to be flowing in the wrong direction here. Hold on to your wallet for now ...

That seems a far stretch, especially for a company with $1.7 Billion in hidden losses immersed in the scandal of their corporate lifetime, and counting on products such as the E-M5 to help to rescue them from their impending fiscal perils. Very sloppy "engineering form" such would seem to be ...

I don't think Oly is sloppier than anyone else. Things like these happen all the time. And in some cases, the problem is far more serious and far more costly to fix than in the present case. Think of software-related recalls in the automotive industry for example, especially in the domain of safety systems. Certainly not meant to happen. Still does happen.

Something probably widely taught in "Apologetics 101" at the industrial "School of Hard Knocks".

Why they would be such "perfectionists" in other respects of the design/development process, yet somehow fail to bother the notice the obvious (particulalry in light of your statement that Lumix G-Series lenses are the only non-Olympus Zukio lens-brand in existence), is beyond me (and, as well, might perhaps tend to cause one to look beyond a simple "Murphy's Law" scenario) ? ...

In my experience, even the best fail to deliver bug-free software on the first try. It would surprise me if your experiences are any different.

I guess that they are not the mythical gods of "Mount Olympus", after all, then ? Sounds industrial.

What particular knowledge-basis is your rating based upon ?

No rating at all. I am not sure there is one for the type of product we are talking about. But none of the bugs so far reported seem to be a kind that seriously limits the functionality of the product. That's what I meant when I said that I don't think the E-M5 is worse than average. The only thing that currently worries me are the reports about banding when used with certain lenses, in particular the 20. But (regrettably) I am far from convinced that this is merely a firmware issue.

How could mere firmware code address something perhaps likely arising out electrical interference ?

Perhaps it may be better to rely upon personally conducted testing (as opposed to anecdotal testimonials) where it comes to these most complicated functions that are (understandably) rather hard (for any user) to evaluate in what truly constitutes a tangibly objective, repeatable manner ?

I was not relying merely on anecdotal evidence. There are at least a three pretty systematic tests of the new IBIS out there (optyczne.pl, Pekka Potka, and kenw). And I am not sure that my own tests (when I am in a position to perform them, which I am not in this case) are always superior to those performed by others. That said, it would obviously be preferable if everyone could decide for themselves whether they should rely on OIS or IBIS with a certain lens on the E-M5. And to reach a conclusive verdict on that issue, OIS must of course first be made fully operative on the camera in question.

Fair enough.

I am aware that, in the spirit of true originality, you often do seem to prefer the wisdoms of "W".

Sure. Just as you prefer the wisdom of "DM". But I try my very best to have the wisdom of "W" subjected to the wisdom of "P".

"Plato" ? "Pascal" ? "Panasonic" ? Mythical Norse god "Pedantor" ? Please don't keep us in suspense

DM ...

Reply   Reply with quote   Complain
Keyboard shortcuts:
FForum MMy threads