Luminous Landscape Forum
Equipment & Techniques => Mirrorless Cameras => Topic started by: mtomalty on October 16, 2015, 12:57:15 pm
-
Hi
Looking for some help sorting out a small issue with a newly
acquired used EM-1
Currently using the camera with Panasonic 15asph,25asph and 42.5mm
Getting wonderfully detailed images at exposure times shorter than 1/200
and longer than 1/60.
Using settings in the 1/60,1/125 and 1/160 (all with IS engaged) results in
files with a slight double exposure type blur as if the IS is moving the capture fractionally
which results in unusable files.
Any suggestions as to how to go about dealing with this in the menu?
Thanks
Mark
www.marktomalty.com
-
Hi
Looking for some help sorting out a small issue with a newly
acquired used EM-1
Currently using the camera with Panasonic 15asph,25asph and 42.5mm
Getting wonderfully detailed images at exposure times shorter than 1/200
and longer than 1/60.
Using settings in the 1/60,1/125 and 1/160 (all with IS engaged) results in
files with a slight double exposure type blur as if the IS is moving the capture fractionally
which results in unusable files.
Any suggestions as to how to go about dealing with this in the menu?
Thanks
Mark
www.marktomalty.com
do you have antishock = 0 to emulate EFCS ? if yes and you shoot hand held then may be defective (you got it used - so who knows what happened) camera... if no then set it because the range of exposure times is prime suspect for shutter shock effects...
-
Thanks
Definitely sounds like it would be the source of the problem.
I've now gone through the camera menu 3 times and can find no
item where this setting can be adjusted.
Was this a feature that was added after the cameras release?
The one I have is running on firmware 3.0
Mark
-
Thanks
Definitely sounds like it would be the source of the problem.
I've now gone through the camera menu 3 times and can find no
item where this setting can be adjusted.
Was this a feature that was added after the cameras release?
The one I have is running on firmware 3.0
Mark
(http://static1.squarespace.com/static/502f174524ac2660dce7d853/t/5161fdc4e4b0f29c92657409/1365376454339/Antishock+Menu.jpg)
-
Got it
Thanks for your patience.
Don't know why I didn't think to look in the ISO menu :)
Set to 0
Now to shoot the test
Mark
-
Got it
Thanks for your patience.
Don't know why I didn't think to look in the ISO menu :)
Set to 0
Now to shoot the test
Mark
note that antishock does not (may be, I sold E-M1 long time ago) work in some (all) burst modes... read the manual about that.
-
0 sec delay may not completely alleviate shutter shock. Firmware 4.0, scheduled for release next month, has an all electronic shutter option. This will do away with shutter shock, but introduces distortion for moving subjects.
-
0 sec delay may not completely alleviate shutter shock.
it is not a genuine EFCS, but 25ms delay - so may be, but as far as I remember it did for all practical purposes (at least unless you have some wobbly double barrel zoom mounted)
Firmware 4.0, scheduled for release next month, has an all electronic shutter option. This will do away with shutter shock, but introduces distortion for moving subjects.
rolling shutter is never a serious option I think it will resort to 10bit raw data just like Sony goes from 13bit to 12bit panasonic m43 sensor (if I recall spec correctly) will go from 12 bit to 10bit
-
Setting the delay to '0' certainly improved the results.
This evening was able to get some extremely sharp images in
The 1/4-1/8 range with the Pana 15 and 25 but earlier was still running into
some issues with the Pana 42.5 at 1/125
Images look decent at 50% but moving in to 100% there is still visible
doubling of subject edges as if things are ever so slightly out of register.
I had an EM-1 for a while last year with all Olympus lenses.
Going back to old files to compare shows no issues.
Is it possible that native lenses take better advantage of in camera stabilization
compared to non-native lenses?
Also,how important is it to set the focal length of the lens in use in the IS meno in the camera.
Next step is to wait for firmware update.
MT
-
This evening was able to get some extremely sharp images in
The 1/4-1/8 range
this range is outside shutter shock affected range though
-
was still running into
Is it possible that native lenses take better advantage of in camera stabilization
compared to non-native lenses?
Panasonic lens has OIS and E-M1 has IBIS... I 'd assume E-M1 body shall disable OIS on Panasonic lens, but if not - did you switch it off on the lens just in case ? Power OIS switch to off.
-
The IS mode has some effect on SS. I find IS Auto to work best.
Some Panasonic lenses have a OIS ON/OFF switch. For those that don't, go into MENU, Gears, C. Release, and set Lens I.S. Priority to Off for IBIS and On for lens OIS.
-
Thanks for that additional info
The three Panasonic lenses I have do not have the
OIS switch
Will set it up that way tomorrow and see if that further improves things
Still seeing a bit of shake at 1/60,1/125,and 1/200 with the 42.5 mm
Mark
www.marktomalty.com
-
You are having SS issues over the usual range of shutter speeds. One thing I've noticed with the three different OM-D model cameras I have is that I have very good IQ at 1/15 sec. for almost all focal length lenses.
-
Just ran another series of tests.
Since shutter shock is the culprit I removed myself from the equation and
worked on a tripod shooting from f2-f11 with both the Oly 45 1.8 and the
Pana 42.5 1.7
Results were same for both so no difference for on and off brand lenses in this case.
Apertures used allowed me to shoot from about 1/500 through to 1/15
Shot the series with IS on and off.
Every frame was critically sharp with no blur or faint secondary image whatsoever with or without IS
Thinking then that since the camera was 100% stationary the problem might be associated with
how IS is working,or not, at critical shutter speeds.
I exposed 5 frames shot at 1/125 (where I am seeing most problems) with IS off and then
5 frames at 1/125 with IS on.
All were handheld
Of the 5 frames with IS off I had 4 critically sharp and 1 with visible camera movement
With IS engaged I ended up with 4 unusable images with detectable movement and
1 frame that was just useable.
Would these results then suggest that shutter shock is not the culprit but rather
how IS is performing in the sensitive shutter speeds
Like you, I get excellent results, handheld, with all lenses with shutter speeds slower
than 1/60
Anyways.......
Mark
-
Just ran another series of tests.
Since shutter shock is the culprit I removed myself from the equation and
worked on a tripod shooting from f2-f11 with both the Oly 45 1.8 and the
Pana 42.5 1.7
Results were same for both so no difference for on and off brand lenses in this case.
Apertures used allowed me to shoot from about 1/500 through to 1/15
Shot the series with IS on and off.
Every frame was critically sharp with no blur or faint secondary image whatsoever with or without IS
Thinking then that since the camera was 100% stationary the problem might be associated with
how IS is working,or not, at critical shutter speeds.
I exposed 5 frames shot at 1/125 (where I am seeing most problems) with IS off and then
5 frames at 1/125 with IS on.
All were handheld
Of the 5 frames with IS off I had 4 critically sharp and 1 with visible camera movement
With IS engaged I ended up with 4 unusable images with detectable movement and
1 frame that was just useable.
Would these results then suggest that shutter shock is not the culprit but rather
how IS is performing in the sensitive shutter speeds
Like you, I get excellent results, handheld, with all lenses with shutter speeds slower
than 1/60
Anyways.......
Mark
when I had E-M1 and then the firmware version with antishock = 0 was in place already I absolutely did not have any issues with Olympus 45/1.8 or other lenses around the shutter shock prone range of exposure times and I always kept IS on (yes, auto or whatever that parameter was)... I do suggest to visit for example dpreview forums and ask there - much more users who can share their experience... but there is always a possibility that something is wrong with the camera
-
The EM-1 has a dual shutter design that has caused its owners problems going back to 2012. Mine had the shutter lock up in cold weather, i.e. around 32F degrees and no such problems with the E-M5 which has a more straightforward shutter mechanism.
-
Rumor has it the E-M1 will get Anti-shock 0 and L shutter with the net firmware. E-M5mkII does have it. I hope that's true!
-
it is not a genuine EFCS, but 25ms delay - so may be, but as far as I remember it did for all practical purposes (at least unless you have some wobbly double barrel zoom mounted)
rolling shutter is never a serious option I think it will resort to 10bit raw data just like Sony goes from 13bit to 12bit panasonic m43 sensor (if I recall spec correctly) will go from 12 bit to 10bit
I do not believe the above is correct. The 0 sec antishock is indeed engaging an EFCS mode.
Also, a drop to 10bit Raw is not a certainty. Some Panasonic cameras do it. Some do not. The Oly EM5II has full electronic shuttering and that does NOT drop to 10bit. The Panasonic GX7 likely uses the same Panasonic sensor as the EM1, and that also does NOT drop to 10bit mode when in e-shutter.
-
I do not believe the above is correct. The 0 sec antishock is indeed engaging an EFCS mode.
I do not suggest you to believe - I suggest you first to read the relevant topics in dpreview m43 forum (those will be close to 2 years old by now) where this was properly investigated along with recorded timed (in ms) sonograms of how the shutter blades are working.
Also, a drop to 10bit Raw is not a certainty.
for Panasonic sensor used in E-M1 ( http://chipworks.force.com/catalog/ProductDetails?sku=OLY-E-M1_Pri-Camera ) = http://www.semicon.panasonic.co.jp/ds4/MN34230PL_E.pdf = "Electronic shutter (Full scan) : Full scan 10bit mode (22.5 fps) 1 / 22.5s 1 / 81300 s (1/81300 s step)"
this is an official sensor documentation from Panasonic
now camera's firmware might try to mask that when writing the raw, but 10bit is what sensor output - at least what Panasonic itself is saying
as for E-M5 mk II or GX7 - do they use Panasonic 16mp or Sony 16mp sensor ? both exist ...
-
I do not suggest you to believe - I suggest you first to read the relevant topics in dpreview m43 forum
I suggest you read more yourself. Since you like DPR, have a look at this:
http://www.dpreview.com/articles/4134393686/olympus-updates-om-d-e-m1-with-electronic-first-curtain-anti-shock
The bit in the URL that says "olympus-updates-om-d-e-m1-with-electronic-first-curtain-anti-shock" might be a bit of a giveaway that you are wrong.
for Panasonic sensor used in E-M1 ( http://chipworks.force.com/catalog/ProductDetails?sku=OLY-E-M1_Pri-Camera ) = http://www.semicon.panasonic.co.jp/ds4/MN34230PL_E.pdf = "Electronic shutter (Full scan) : Full scan 10bit mode (22.5 fps) 1 / 22.5s 1 / 81300 s (1/81300 s step)"
this is an official sensor documentation from Panasonic
now camera's firmware might try to mask that when writing the raw, but 10bit is what sensor output - at least what Panasonic itself is saying
as for E-M5 mk II or GX7 - do they use Panasonic 16mp or Sony 16mp sensor ? both exist ...
Most people believe the GX7 uses the MN34230 sensor. The same as used in the GM1 and EM1 and GH4.
The GM1 & GH4 drop to 10bit mode in e-shutter. The GX7 remains at 12bit.
I've had this argument with someone else here many months ago, because if you look at the datasheet you linked to it shows that the sensor has both a 10 and a 12 bit output mode, and I argued that just because the e-shutter field only quotes the (fast) 10 bit readout speed, this doesn't necessarily mean that you cannot configure the chip to ALSO do e-shutter in 12 bit mode (albeit with a slower readout speed). Whoever I was arguing with tried to use the conclusive data that the GH4 was 10bit in e-shutter, as conclusive proof that the MN34230 chip can ONLY do 10 bit. This is flawed logic.
If the GX7 does indeed use the same chip, then the fact that its been shown that it does actually operate in a slower 12bit e-shutter mode means I was correct, an that person as well was wrong.
But we do need to be 100% sure of what sensor is in the GX7, and I'm not aware of any teardown that has been done on that camera.
As for the EM1, well we'll have to wait and see. My point was that its not a dead certainty it will be a 10bit e-shutter mode. There is enough evidence to cast doubt on that assumption. Maybe it will be, maybe it won't.
-
I suggest you read more yourself. Since you like DPR, have a look at this:
http://www.dpreview.com/articles/4134393686/olympus-updates-om-d-e-m1-with-electronic-first-curtain-anti-shock
The bit in the URL that says "olympus-updates-om-d-e-m1-with-electronic-first-curtain-anti-shock" might be a bit of a giveaway that you are wrong.
and I really suggest you to read the reasech done by users who tested what is really going on with shutter blades after that firmware update was released, instead of PR releases.
-
Most people believe the GX7 uses the MN34230 sensor. The same as used in the GM1 and EM1 and GH4...GM1 & GH4 drop to 10bit mode in e-shutter. The GX7 remains at 12bit.
it may be so, but you "believe" too much in various things ... with E-M1 we know (instead of "believing") for sure what was the sensor (thanks to ChipWorks) and the topic is about E-M1... GX7 might as well be using a different revision of the sensor or Sony sensor, etc - we don't know for sure, unless you produce something other than "beliefs"... and one might logically ask a question as to why a flagship GH4 has 10bit when lowish GH7 maintain 12bit for still imaging with electronic rolling shutter with the same sensor and GH4 having no less powerful everything else and GH4 released _after_ GX7 on top of that ? why step back, why cripple GH4 - unless GX7 has a different sensor with different electronic rolling shutter output ? DxOMark shows quite different results for GH4 vs GX7 too
I've had this argument with someone else here many months ago, because if you look at the datasheet you linked to it shows that the sensor has both a 10 and a 12 bit output mode, and I argued that just because the e-shutter field only quotes the (fast) 10 bit readout speed, this doesn't necessarily mean that you cannot configure the chip to ALSO do e-shutter in 12 bit mode (albeit with a slower readout speed).
and why 'd Panasonic not list 12bit electronic rolling shutter mode in its documentation - not logical to omit a feature that is present to be used
Whoever I was arguing with tried to use the conclusive data that the GH4 was 10bit in e-shutter, as conclusive proof that the MN34230 chip can ONLY do 10 bit. This is flawed logic.
like yours based on the "belief" that GX7 has the same sensor and that datasheet omitted an electronic rolling shutter mode (not to mix with non electronic rolling shutter modes) present in sensor
If the GX7 does indeed use the same chip, then the fact that its been shown that it does actually operate in a slower 12bit e-shutter mode means I was correct, an that person as well was wrong.
if my grandmother had balls... or what is the proper wording in English ?
-
and I really suggest you to read the reasech done by users who tested what is really going on with shutter blades after that firmware update was released, instead of PR releases.
Lets see. We'll do a quick google on "EM-1" and "Electronic first curtain" shall we.
Oh look. Its not Just DPR mentioning it, its also:
Imaging resource, DigLloyd, Ming Thein, etc etc etc. In fact, EVERYBODY. Every single web site. Except you.
And oh look. Also, Robin Wong, who says "The 0 second anti shock basically used First Curtain Electronic Shutter". And he's an Olympus Employee. Right.
So either they are all Aholes who have been taken in by erroneous PR releases .... or the Ahole is you.
I know which one my money would be on. The onus is now on you to prove otherwise and that the EM1 does not in fact use EFCS.
-
Got it
Thanks for your patience.
Don't know why I didn't think to look in the ISO menu :)
Set to 0
Now to shoot the test
Mark
In addition to choosing Anti-Shock 0sec in the "E" menu, you also have to go into the drive mode menu (the front button on the top left of the camera, near the power switch) and choose the single-frame shooting icon with the diamond next to it. If you choose one of the modes that doesn't have a diamond, the anti-shock feature won't be activated.
-
Also, a drop to 10bit Raw is not a certainty
Just to dig up this thread from a couple of months ago. To recap, AlterEgo stated quite categorically that the EM1 does not have an Electronic front curtain shutter mode, and also that when the new firmware was released to give a full e-shutter mode, this would drop the sensor output to 10 bit mode.
He was wrong about the EFCS, and now we know that he was also wrong about the drop to 10bit. It remains as a 12 bit file.
I confess there is a certain satisfaction in being able to say, "I told you so" ;)
Perhaps I should also trawl back through some even older threads, since I recall having an equally unpleasant argument with other people here, who were trying to tell me point blank that the quoted panasonic MN34230 data sheet for the sensor used in the GH4/EM1 showed categorically that e-shutter was ONLY possible in 10bit mode - and I argued back then that it was probably a programmable chip, and that the (very limited) data sheet was only stating the headline figure of 10bit as it gave the fastest readout.
Its all looking very likely that Panasonic COULD provide the GH4 with a 12 bit e-shutter mode if they wanted.
-
Just to dig up this thread from a couple of months ago. To recap, AlterEgo stated quite categorically that the EM1 does not have an Electronic front curtain shutter mode, and also that when the new firmware was released to give a full e-shutter mode, this would drop the sensor output to 10 bit mode.
He was wrong about the EFCS, and now we know that he was also wrong about the drop to 10bit. It remains as a 12 bit file.
I was not wrong about EFCS, EM-1 does not have true EFCS - operation like in Sony cameras for exаmple (which do not employ tricks like 25ms delay to fight shutter shock)... also as noted the sensor used in E-M1 is 10bit in electronic shutter (Full scan) - which is a __rolling shutter__ (as in video, just with full sensor readout) mode when mechanical shutter is not used at all... just to give you an idea - using EFCS does not mean that sensor is operating in electronic shutter mode... take Sony sensors - when I use EFCS, single shot I have "13bit" raws, when I use silent shutter = electronic shutter (no EFCS naturally) mode I am dropping one bit.
-
I was not wrong about EFCS, EM-1 does not have true EFCS - operation like in Sony cameras for exаmple (which do not employ tricks like 25ms delay to fight shutter shock)... also as noted the sensor used in E-M1 is 10bit in electronic shutter (Full scan) - which is a __rolling shutter__ (as in video, just with full sensor readout) mode when mechanical shutter is not used at all... just to give you an idea - using EFCS does not mean that sensor is operating in electronic shutter mode... take Sony sensors - when I use EFCS, single shot I have "13bit" raws, when I use silent shutter = electronic shutter (no EFCS naturally) mode I am dropping one bit.
There are two seperate issues. One is the EFCS, the other is e-shutter.
You are still wrong about the EFCS not being a 'true' EFCS. Its just that because the shutter mechanism was not designed in the first place to properly accommodate the ideal mechanical operation, Olympus have ALSO put in a delay. But putting in this additional delay does NOT mean that the EFCS is somehow false as you implied. Its just not as optimal an overall implementation.
The second point was you insisted the e-shutter mode (the new one that Olympus calls 'silent') would drop to 10bit. I said this wasn't a certainty. I was right. You were wrong. The e-shutter mode is (thankfully) still 12 bit.
-
You are still wrong about the EFCS not being a 'true' EFCS. Its just that because the shutter mechanism was not designed in the first place to properly accommodate the ideal mechanical operation, Olympus have ALSO put in a delay. But putting in this additional delay does NOT mean that the EFCS is somehow false as you implied. Its just not as optimal an overall implementation.
EFCS is a combined operation of sensor and mechanical shutter... so once one thing is not operated as it should you do not have it
The second point was you insisted the e-shutter mode (the new one that Olympus calls 'silent') would drop to 10bit. I said this wasn't a certainty. I was right. You were wrong. The e-shutter mode is (thankfully) still 12 bit.
and proof is ?