Industry News: OM System launches the OM-1

AlanF

Stay at home
CR Pro
Aug 16, 2012
9,724
14,940
I haven't at any time said that shooting with only 20MP is impossible for BIF, only that it is a great deal easier if you have more MP to allow for framing errors with fast moving birds.

You have greater ability at targeting subjects than me and you clearly devote a great deal of time to BIF. You will also have more success simply because you shoot a far greater number of images than most people - if I recollect correctly, in a recent post you said you'd taken 60,000 shots on your R5. Your shots of mallard and wigeon in flight are excellent, as I would expect from anyone as devoted to BIF as you clearly are. I've only been shooting BIF for about a year, and it only accounts for about 5% of my photography. I need the safety margin that the 45MP of the R5 provides, to achieve similar BIF shots.

I'm not going to post my own images, because I don't want to turn this into a silly competition to prove who is "best" at anything. I prefer instead to just discuss equipment, and its suitability for its intended tasks, which is why I have posted about the OM1, and why I consider it to be a less than ideal choice.
If you have been taking BIF for only about a year, and only 5% of your shots are of BIF, do you think you have the knowledge and experience to make repeated confident and dogmatic disparaging comments about 20 Mpx sensors and write off the Olympus?
 
  • Haha
  • Like
Reactions: 1 users

entoman

wildlife photography
May 8, 2015
969
1,150
UK
Nowhere did I say it was impossible, just that your previous claim that a global shutter alone would make it possible is incorrect.
Er, I didn't claim that a "global shutter alone would make it possible". I appreciate that there are many posts here and that it's easy to miss things that other posters have said, but I also mentioned that a faster processor and faster cards would be needed to achieve high burst speeds with pixel-shift technology. I also mentioned that I believe that mechanical IBIS will fairly soon be ousted by improved digital stabilisation, which will make for faster pixel-shift speed. Olympus/OM System have more experience than other brands with AI tech too, so if anyone can do it, they can.
 

entoman

wildlife photography
May 8, 2015
969
1,150
UK
If you have been taking BIF for only about a year, and only 5% of your shots are of BIF, do you think you have the knowledge and experience to make repeated confident and dogmatic disparaging comments about 20 Mpx sensors and write off the Olympus?
Nowhere have I "written off" Olympus so do not try to twist my words. I'm brand-neutral, format-neutral and I have heaped praise on other aspects of the camera.

What I've said is that I believe a camera using pixel-shift to achieve hi-res is far less suitable for BIF than a camera with native hi-res. I've given my reasons for that, and I stick firmly with that view.

I've also been a photographer for over 50 years, including 30 years as a professional industrial photographer, and I've spent the last 10 years as a hobbyist with over 2000 published wildlife images, so I have a full appreciation of the relative merits of different sensor sizes, different pixel densities and different technologies.

Sadly, I feel that continuing this subject with you is degenerating and heading down a path that I have no wish to follow, so I'll wish you goodnight.
 

AlanF

Stay at home
CR Pro
Aug 16, 2012
9,724
14,940
Nowhere have I "written off" Olympus so do not try to twist my words. I'm brand-neutral, format-neutral and I have heaped praise on other aspects of the camera.

What I've said is that I believe a camera using pixel-shift to achieve hi-res is far less suitable for BIF than a camera with native hi-res. I've given my reasons for that, and I stick firmly with that view.

I've also been a photographer for over 50 years, including 30 years as a professional industrial photographer, and I've spent the last 10 years as a hobbyist with over 2000 published wildlife images, so I have a full appreciation of the relative merits of different sensor sizes, different pixel densities and different technologies.

Sadly, I feel that continuing this subject with you is degenerating and heading down a path that I have no wish to follow, so I'll wish you goodnight.
I have no doubt you can offer excellent advice on industrial photography and, I assume from posts about your travels, butterfly photography.
 

tron

EOS-1D X Mark III
CR Pro
Nov 8, 2011
5,124
1,503
1. If you have such appreciation of different sensor sizes why you continue with the thought that OM-1 is a low resolution camera? (or at least a lower resolution camera than other ones?)

In the part of the R5's, Z9's and D850's sensor in the center in an area equal to Olympus sensor size (2X crop) it has about 11.25 Mpixels.
Sony's A1 corresponding sensor part has 12.5 Mpixels

A bird that covers all the frame of Olympus would have 20Mpixels.
If shot with Canon, Nikon or Sony at the same distance with the same lens Focal Length it would contain almost half the megapixels (11 to 12 as stated above).

So OM-1 as all other 4/3rds cameras have higher pixel density than almost all DSLRs and mirrorless cameras (one exception I know is Canon's 90D). It just doesn't use a FF or a APS-C sensor but a smaller one with its advantages and disadvantages.

2. Also who mentioned in Olympus that pixel-shift is meant to be used for moving subjects ?

Sure if a bird is ... sleeping you could use it.... :D
 
Last edited:

raptor3x

EOS RP
Jan 26, 2012
631
114
State College, PA
whumber.com
Er, I didn't claim that a "global shutter alone would make it possible". I appreciate that there are many posts here and that it's easy to miss things that other posters have said, but I also mentioned that a faster processor and faster cards would be needed to achieve high burst speeds with pixel-shift technology.
Here's what you said
Many people here including myself hoped that the camera would have a global shutter, and if that had been the case (as I pointed out in another post) it WOULD have been possible to use hand-held hi-res for BIF and action, and would have put the OM on par with our FF Nikon, Canon and Sony cameras. Virtually instantaneous readout would have meant the period between the successive shots was so brief that they could be composited quickly enough to virtually eliminate motion blur and the weird artefacts that occur with moving water and foliage in current pixel-shift cameras.
The bolded statement is simply incorrect.
I also mentioned that I believe that mechanical IBIS will fairly soon be ousted by improved digital stabilisation, which will make for faster pixel-shift speed. Olympus/OM System have more experience than other brands with AI tech too, so if anyone can do it, they can.
Digital stabilization is fundamentally incompatible with pixel shift. There's a disconnect in your understanding of how either digital stabilization work, how pixel-shift methods work, or both.
 

entoman

wildlife photography
May 8, 2015
969
1,150
UK
Here's what you said

The bolded statement is simply incorrect.

Digital stabilization is fundamentally incompatible with pixel shift. There's a disconnect in your understanding of how either digital stabilization work, how pixel-shift methods work, or both.
Please clarify *why* you believe the bolded statement is incorrect. It represents what I believe to be possible. Are you saying that a global shutter would not have "virtually instantaneous readout"? Do you think that pixel-shift is and always will be impossible for HHHR, and if so, why? Are you saying that if by its use, HHHR would still not be fast enough to be comparable with FF cameras that have high native resolution?

I understand how pixel-shift works - the sensor is moved in tiny increments for successive shots, which are then merged in-camera to produce a single image with higher resolution, typically twice the native linear resolution of the sensor. In what way is it "fundamentally incompatible" with digital stabilisation? I'm very happy to withdraw that part of my statement if you can explain why digital stabilisation can't be used for pixel-shift (and that would also negate my comment about DS being capable of reducing the overall time to take and composite the images).
 

entoman

wildlife photography
May 8, 2015
969
1,150
UK
1. If you have such appreciation of different sensor sizes why you continue the idiotic thought that OM-1 is a low resolution camera? (or at least a lower resolution camera than other ones?)

In the part of the R5's, Z9's and D850's sensor in the center in an area equal to Olympus sensor size (2X crop) it has about 11.25 Mpixels.
Sony's A1 corresponding sensor part has 12.5 Mpixels

A bird that covers all the frame of Olympus would have 20Mpixels.
If shot with Canon, Nikon or Sony at the same distance with the same lens Focal Length it would contain almost half the megapixels (11 to 12 as stated above).

So OM-1 as all other 4/3rds cameras have higher pixel density than almost all DSLRs and mirrorless cameras (one exception I know is Canon's 90D). It just doesn't use a FF or a APS-C sensor but a smaller one with its advantages and disadvantages.

2. Also who mentioned in Olympus that pixel-shift is meant to be used for moving subjects ?

Sure if a bird is ... sleeping you could use it.... :D
1 - It seems to me that anyone buying a lens would choose it primarily for its angle of view! Wide angles for wide views, long focal lengths for narrow angles of view to pick out smaller or more distant objects.

The whole point of M43 for many people is the fact that you can obtain the same angle of view on M43 with a nice lightweight 300mm lens that you would need a heavy and vastly more expensive 600mm lens if shooting on FF!

To make a comparison between formats, as you are doing, based on focal length, is a bit odd, to say the least...

2 - Pixel-shift on early cameras was slow and resulted in ghastly artefacts on moving water or foliage. It rendered moving objects, even slow-walking humans occupying a small part of the frame, as blurs. Processors are faster now, BSI sensors have faster readouts, and M43 has the advantage that the smaller sensor has to be moved a shorter distance to produce pixel-shift, so is faster, and more usable for hand-held hi-res, and for slow moving subjects.

Ultimately the aim must be for M43 to be able to compete directly with FF by offering similar (or better) performance, including hand-held hi-res bursts. The OM1 can't do that, which is why I'm disappointed. But, I believe if it had a global shutter, a faster processor and used CFE-B instead of slower SD cards, it would come much closer to matching the performance of A1, Z9 and R5. And guess what - if it did, I'd sell my FF gear and switch to Olympus!

I'll certainly be very interested in following further developments with the OM system, and hopefully a global shutter, faster processor and CFE-B slot will appear within a couple of years.

3 - Accusing a fellow poster of having "idiotic thoughts" is provoking escalation and usually results in retaliation or rudeness. I have no intention of descending to that level, but it's something you should consider when replying to people.
 
Last edited:

raptor3x

EOS RP
Jan 26, 2012
631
114
State College, PA
whumber.com
Please clarify *why* you believe the bolded statement is incorrect. It represents what I believe to be possible. Are you saying that a global shutter would not have "virtually instantaneous readout"? Do you think that pixel-shift is and always will be impossible for HHHR, and if so, why? Are you saying that if by its use, HHHR would still not be fast enough to be comparable with FF cameras that have high native resolution?
As I pointed out to you earlier, global shutters read out all of the pixels on a sensor simultaneously but that doesn't mean you get higher framerates. That is largely controlled by how quickly you can move data off the sensor. In global shutter sensors you will tend to have more electronics built into the sensor devoted to the global shutter which tends to result in reduced ability to move data off the sensor quickly. Again, the point is that global shutter is great but you need global shutter and a sensor with a massive framerate, a global shutter doesn't automatically mean you can shoot at a high framerate.
I understand how pixel-shift works - the sensor is moved in tiny increments for successive shots, which are then merged in-camera to produce a single image with higher resolution, typically twice the native linear resolution of the sensor. In what way is it "fundamentally incompatible" with digital stabilisation? I'm very happy to withdraw that part of my statement if you can explain why digital stabilisation can't be used for pixel-shift (and that would also negate my comment about DS being capable of reducing the overall time to take and composite the images).
It's because digital stabilization only works for video to keep the viewpoint from moving around but has no ability to stabilize a still image (or even a single video frame) to keep the image sharp or make sub pixel shifts required by a fixed pattern hires mode. The Olympus cameras in handheld hires modes rely on a combination of turning off the IBIS to allow random hand tremors to reposition the sensor and then quickly turning the IBIS back on to stabilize each individual sub-frame.
 
  • Like
Reactions: 1 user

tron

EOS-1D X Mark III
CR Pro
Nov 8, 2011
5,124
1,503
1 - It seems to me that anyone buying a lens would choose it primarily for its angle of view! Wide angles for wide views, long focal lengths for narrow angles of view to pick out smaller or more distant objects.

The whole point of M43 for many people is the fact that you can obtain the same angle of view on M43 with a nice lightweight 300mm lens that you would need a heavy and vastly more expensive 600mm lens if shooting on FF!

To make a comparison between formats, as you are doing, based on focal length, is a bit odd, to say the least...

2 - Pixel-shift on early cameras was slow and resulted in ghastly artefacts on moving water or foliage. It rendered moving objects, even slow-walking humans occupying a small part of the frame, as blurs. Processors are faster now, BSI sensors have faster readouts, and M43 has the advantage that the smaller sensor has to be moved a shorter distance to produce pixel-shift, so is faster, and more usable for hand-held hi-res, and for slow moving subjects.

Ultimately the aim must be for M43 to be able to compete directly with FF by offering similar (or better) performance, including hand-held hi-res bursts. The OM1 can't do that, which is why I'm disappointed. But, I believe if it had a global shutter, a faster processor and used CFE-B instead of slower SD cards, it would come much closer to matching the performance of A1, Z9 and R5. And guess what - if it did, I'd sell my FF gear and switch to Olympus!

I'll certainly be very interested in following further developments with the OM system, and hopefully a global shutter, faster processor and CFE-B slot will appear within a couple of years.

3 - Accusing a fellow poster of having "idiotic thoughts" is provoking escalation and usually results in retaliation or rudeness. I have no intention of descending to that level, but it's something you should consider when replying to people.
You keep repeating the same thing over and over again without reading the arguments against it. I was harsh so I apologize. I will delete that comment from the post. I made the mistake of wasting a lot of my time trying to respond to you - and I am still doing it but finally I marked this thread as "ignore" - so that's it.

My mistake of course as I said.
 
Last edited:

Czardoom

EOS RP
Jan 27, 2020
500
1,100
I think that there are trade-offs with different systems and the choices that people make will depend on what is more important to them. If a wider angle of view is what you are looking for so that you can better track and capture fast moving birds, then you might not choose the
MFT camera and will choose the FF. If higher resolution is more important, then you might want to choose the MFT camera, where, if you crop the R5's FF image down to MFT size, you would get 11.25 MPs instead of 20 MP for the Olympus. Different strokes for different folks, so to speak.
 
  • Like
Reactions: 1 user

entoman

wildlife photography
May 8, 2015
969
1,150
UK
As I pointed out to you earlier, global shutters read out all of the pixels on a sensor simultaneously but that doesn't mean you get higher framerates. That is largely controlled by how quickly you can move data off the sensor. In global shutter sensors you will tend to have more electronics built into the sensor devoted to the global shutter which tends to result in reduced ability to move data off the sensor quickly. Again, the point is that global shutter is great but you need global shutter and a sensor with a massive framerate, a global shutter doesn't automatically mean you can shoot at a high framerate.

It's because digital stabilization only works for video to keep the viewpoint from moving around but has no ability to stabilize a still image (or even a single video frame) to keep the image sharp or make sub pixel shifts required by a fixed pattern hires mode. The Olympus cameras in handheld hires modes rely on a combination of turning off the IBIS to allow random hand tremors to reposition the sensor and then quickly turning the IBIS back on to stabilize each individual sub-frame.
Thank you for the explanation, which is appreciated.
 

entoman

wildlife photography
May 8, 2015
969
1,150
UK
Panasonic GH6 newly leaked specs indicate 25MP, which is an insignificant jump from the Olympus 20MP, but may swing it for some people.
Unfortunately it still has contrast detect DFD and will (probably) still suffer from "DFD wobble".
I know several people who use GH5 and say they don't notice the jittering at all, but others find it irritating and distracting, especially for macro or video.

Given the choice between the two, I'd go for the Olympus, due to the legendary durability and weather-sealing, and (probably) faster and more efficient AF and tracking.
 

SteB1

EOS M6 Mark II
Feb 22, 2019
55
93
I'm certainly interested in this camera/system. My particular interest is as a replacement for my 7D mkII, 100-400mm f4.5-5.6 mkII combination, which is my standard walk around combination (I have a shedload of macro lenses, other lenses, bodies), but this is my do all combo for nature shooting. I have got a FF camera, a 5Ds at the moment, but FF is a disadvantage in reach terms, both for long lens and smaller macro stuff. In practice if I use FF, I'm just cropping to smaller than APS-C for most opportunistic photos, so FF is wasted. Don't get me wrong, FF is great for my landscape, large butterflies, large dragonflies etc, but birds, smaller insects etc, getting close enough to fill the frame is rare and difficult. I do have to make a mirrorless choice sometime. I'll certainly wait to see if an R7 type camera comes out. If it's 7D level, I'm interested, but if it's cheap plasticky body with M50 type specs, I'm not so interested.
 

SteB1

EOS M6 Mark II
Feb 22, 2019
55
93
Panasonic GH6 newly leaked specs indicate 25MP, which is an insignificant jump from the Olympus 20MP, but may swing it for some people.
Unfortunately it still has contrast detect DFD and will (probably) still suffer from "DFD wobble".
I know several people who use GH5 and say they don't notice the jittering at all, but others find it irritating and distracting, especially for macro or video.

Given the choice between the two, I'd go for the Olympus, due to the legendary durability and weather-sealing, and (probably) faster and more efficient AF and tracking.
Personally, I haven't found the difference between my 20mp Canon 7D mkII (+70D) files or those from the 24mp 80D that great in practise. Yes, sure I noticed at first the 24mp had slightly better detail at 100%. But in the real world, for all intents and purposes, they're the same. I think it needs a jump to 30+mp before you start seeing a real world difference.
 

entoman

wildlife photography
May 8, 2015
969
1,150
UK
Personally, I haven't found the difference between my 20mp Canon 7D mkII (+70D) files or those from the 24mp 80D that great in practise. Yes, sure I noticed at first the 24mp had slightly better detail at 100%. But in the real world, for all intents and purposes, they're the same. I think it needs a jump to 30+mp before you start seeing a real world difference.
Absolutely, assuming a 3:2 format, here are the relative frame widths for various resolutions:

20MP = 5472 pixels
24MP = 6000 pixels
30MP = 6720 pixels
45MP = 8192 pixels
50MP = 8640 pixels
61MP = 9504 pixels

Doubling the megapixel count only adds 50% to the frame width.

It's very difficult to tell any difference between a 20MP and 24MP sensor (assuming same format, generation and degree of enlargement), and a latest generation 20MP sensor will often produce more *apparent* resolution in a print than a previous generation 24MP sensor.

Pixel-shift hi-res of course quadruples the MP, resulting in doubling the frame width, at the expense of introducing motion blur and strange artefacts if there is any subject movement.
 
Last edited:
  • Like
Reactions: 1 user

AlanF

Stay at home
CR Pro
Aug 16, 2012
9,724
14,940
There’s a simple way of working out the increase in resolution on increasing the number of pixels - the number of pixels wide varies as the square root of the total, so doubling the total increases the number in the width by 41.4%. Increasing from 20 to 24 increases it by 9.5%. A 500mm lens on a 24 Mpx sensor will give you the same resolution as a 548mm lens on a 20 Mpx sensor. Similarly, an increase from 20 to 30 Mpx gives a 22.5% increase in resolution, so your 500mm on the 30 Mpx would be equivalent to having a 610mm lens on the 20 Mpx.
 
  • Like
Reactions: 1 user

EOS 4 Life

EOS R
Sep 20, 2020
1,300
1,035
500 fps could easily be achieved with global shutter
A global shutter just means the exposure happens all at once.
It says nothing about the time between exposures.
Most global shutter CCD cameras have much lower FPS than rolling shutter CMOS cameras.
 

EOS 4 Life

EOS R
Sep 20, 2020
1,300
1,035
As I pointed out to you earlier, global shutters read out all of the pixels on a sensor simultaneously
Global shutters do not read out from the sensor simultaneously.
The entire sensor in on at the same time so the exposure is simultaneous.
The image processor can still read the serialized image in chunks
 
  • Like
Reactions: 1 user

EOS 4 Life

EOS R
Sep 20, 2020
1,300
1,035
I think that there are trade-offs with different systems and the choices that people make will depend on what is more important to them. If a wider angle of view is what you are looking for so that you can better track and capture fast moving birds, then you might not choose the
MFT camera and will choose the FF. If higher resolution is more important, then you might want to choose the MFT camera, where, if you crop the R5's FF image down to MFT size, you would get 11.25 MPs instead of 20 MP for the Olympus. Different strokes for different folks, so to speak.
The way I see it is that an MFT camera can use a focal reducer with every EF lens as a full-frame camera while gaining a stop of light while a full-frame camera can use a 2x teleconverter with a limited number telephoto lenses while losing two stops of light.
While an MFT camera with a focal reducer can't beat a full-frame camera it is better at being a full-frame camera than a full-frame camera with a 2x teleconverter is at being an MFT camera.
Of course, we can split the difference by going APS-C
#R7