MFT Sensor Reputation.

[Pannas]

I've recently read an article by a Rod Lawton, on his website "Life after Adobe", which is new to me but may be known to others on here.
He makes an interesting assertion that 'Lightroom is not very kind to MFT sensors, or small sensors in general'. He considers the Adobe Camera Raw Processing Engine is to blame, claiming that 'it produces quite a coarse noise pattern that's especially apparent with smaller sensor cameras'. He clearly suspects that this may explain much of the 'criticism of MFT sensors'. I guess because Lightroom is so widely used!
He adds that when 'opening up an ORF file in Capture One or DXO Photolab they look ten times better than in Lightroom'!
I use DXO products and have no experience of Lightroom since 2017. Is there any other evidence out there from other users to back up these claims.

[Ian]

I've recently read an article by a Rod Lawton, on his website "Life after Adobe", which is new to me but may be known to others on here.
He makes an interesting assertion that 'Lightroom is not very kind to MFT sensors, or small sensors in general'. He considers the Adobe Camera Raw Processing Engine is to blame, claiming that 'it produces quite a coarse noise pattern that's especially apparent with smaller sensor cameras'. He clearly suspects that this may explain much of the 'criticism of MFT sensors'. I guess because Lightroom is so widely used!
He adds that when 'opening up an ORF file in Capture One or DXO Photolab they look ten times better than in Lightroom'!
I use DXO products and have no experience of Lightroom since 2017. Is there any other evidence out there from other users to back up these claims.

I've known Rod for a very long time

I'm not really sure I know what he means. He may be criticising the default image quality once the RAW file has been imported to Lightroom. Lightroom doesn't have AI sharpening yet, but its AI noise reduction is now excellent.

My default settings are modified anyway.

Ian

[gazza95]

Ho old is the article you are referring to?

Certainly in the days of Em1x Capture One did process off files better than Lightroom. This was the reason I changed to Lightroom. The images a 3200 ISO were both cleaner and more detailed.

Since partially reverting to Lightroom for a new camera I'm finding the OM1 images in lightrooma nd Capture One are much closer than before.


Gary

[Pannas]

It's dated 3rd April 2025 and listed under tutorials, 'Three reasons why I edited this in DXO Photolab and not Lightroom'. With the associated edited photo.

[pdk42]

I've used LR for a long time, and also used it for years with my Olympus gear. I have also used Oly's Workspace (OW). You'd expect OW to be the reference for how to handle ORF files optimally.

Bottom line, I've no idea what he's talking about. I see no significant difference between LR and OW regarding noise. On top of that, use LR's latest AI NR and the results are outstanding.

As to the wider topic of "MFT sensor reputation", it seems to me that MFT sensor performance is exactly what you'd expect it to be given its size. It's no surprise that MFT performs less well than FF in terms of noise, and the gap is about 2 stops. It's always been that and it always will be, so long as the underlying sensor tech is consistent across the formats. That 2 stop difference is exactly what you'd expect of a sensor that is a quarter the area of FF.

Now whether than 2 stops matters is entirely down to your use case.

[gazza95]

pdk42

I regard OW as the reference with regards to Colour Edition when converting Raw files. With regards to noise and bayer processing I'm not convinced OW is the best of the raw converters.

With regards "That 2 stop difference is exactly what you'd expect of a sensor that is a quarter the area of FF." you seem to have missed the point that a 60mp Sony FF sensor is close to the 20MP Sony OM-1 M43 sensor pixel density yesterday has 2 stops more dynamic range, according to Phototophotos. Big difference is that the FF is 14bit data and M43 is 12bit data. This certainly shows up when you try to lift mid and dark toned areas.


As to which is the best, I've no idea these days.

Gary

[Ian]

pdk42

I regard OW as the reference with regards to Colour Edition when converting Raw files. With regards to noise and bayer processing I'm not convinced OW is the best of the raw converters.

With regards "That 2 stop difference is exactly what you'd expect of a sensor that is a quarter the area of FF." you seem to have missed the point that a 60mp Sony FF sensor is close to the 20MP Sony OM-1 M43 sensor pixel density yesterday has 2 stops more dynamic range, according to Phototophotos. Big difference is that the FF is 14bit data and M43 is 12bit data. This certainly shows up when you try to lift mid and dark toned areas.


As to which is the best, I've no idea these days.

Gary

Being 14-bit is no guarantee of significant additional quality. The two extra bits are containers. They need to be 'filled' with useful image data. Pushing the sensor will mean this headroom will eventually only contain noise.

Ian

[gazza95]

While I agree that 14bit is no guarantee of increased dynamic range. If both sensors use similar sensor technology with similar pixel pitch and the 14bit version has 2 stops advantage. It would be reasonable to expect M43 version to get more dynamic range if it used 14bits. Even if its not the full 2 stops.

The above only really applies at base or low ISO values.

[blackfox]

Hmmmm as some will have seen I’m currently playing with a Nikon D500 it’s roughly the same MP count as my OM1 and I PP in LR classic latest updated version , I have to admit that I like the look of the Nikon files as opposed to the Olympus ones but there’s honestly nothing that defines either as being superior. I have tested both across a range of light variations . and PP both exactly the same way .

Both my setups max out range wise at circa 840 mm so are evenly matched the only advantage of the Nikon is the zoom ability , both lenses are now used wide open as well .

As for using LR there are so many variations you can do to your raw file in PS as well that I can’t justify any criticism of it …. It is what you want it to be

[Ian]

I've used LR for a long time, and also used it for years with my Olympus gear. I have also used Oly's Workspace (OW). You'd expect OW to be the reference for how to handle ORF files optimally.

Bottom line, I've no idea what he's talking about. I see no significant difference between LR and OW regarding noise. On top of that, use LR's latest AI NR and the results are outstanding.

As to the wider topic of "MFT sensor reputation", it seems to me that MFT sensor performance is exactly what you'd expect it to be given its size. It's no surprise that MFT performs less well than FF in terms of noise, and the gap is about 2 stops. It's always been that and it always will be, so long as the underlying sensor tech is consistent across the formats. That 2 stop difference is exactly what you'd expect of a sensor that is a quarter the area of FF.

Now whether than 2 stops matters is entirely down to your use case.

gazza95 makes a good point regarding FF and resolution, or pixel pitch. Your Lumix S5 Has four times the sensor area accommodating about 25% more pixels - each pixel will be gathering at least three times more light. But not all FF sensors are made of the same stuff. The Canon EOS RP, I read, has significantly less DR and noise performance, despite being the same resolution as the S5. In fact, one review puts it firmly in the same league as the OM-1. If you look at the hugher 45 and 60 megapixel sensor mirrorless cameras, they do significantly sacrifice noise and DR performance, as well as continuous shooting rates.

Back to the software question originally highlighted - I think it's down to the default starting point settings, not the image potential.

Ian

[pdk42]

I agree, but I slipped in a constraint to my statement - so long as the underlying sensor tech is consistent across the formats.

[Harold Gough]

I never tried Lightroom. I open ORFs in DXO Optics Pro 9 and save them as TIFFs which I work on in PS CS6, using some Topaz plugins (versions no longer available).

Harold

[c12402]

I've been using DxO as the first step in my workflow since 2013 when I started with my first Olympus OM5. I use Lightroom or Photoshop as the second step in the workflow with very good results. I have not tested the latest versions of Adobe to see how they handle ORF's, maybe they are better now but for sure, some years ago the difference was very noticeable.

[griffljg]

I've also been using DxO PureRAW as the first step in my workflow since mid-2022. I use it primarily for noise reduction, but also for lens corrections, before exporting the image as a DNG file, which I then import into Capture One Pro for further processing. DxO PureRAW is so good at removing noise from OM-1 raw files, that I am usually able to get decent images out of raw files which have been taken at 25,600 ISO. - I kid you not!

With my E-M1 Mk III, I tend to keep to a maximum of 6,400 ISO.

Here is a link to an image in my Flickr page in which I used an ISO of 20,000. I have left all the EXIF intact:

https://www.flickr.com/photos/griffljg/52613714690/in/album-72157704292882111

Aah! I knew I had it somewhere. - Here is a bird photo taken at 25,600 ISO:

https://www.flickr.com/photos/griffljg/52837877692/in/album-72157704292882111/

[shakyHands]

griffljg On my recent trip to the Kimberly, I was using my OM-1 mk2 body with my big white lens for wildlife and took along my older OM-1 mk1 with a 12-45 for landscape/general shots. I seldom use the old body as it suffers from the dial issues that were common with the early original om-1 bodies. I use the front dial for exposure compensation. On a few occasions with the sticky dial, getting the front dial to move was difficult and I found that after the shots I must have pressed the ISO button with my thumb when I was attempting to shift the front dial. In consequence I ended up with a few shots at ISO 102400. I was surprised that DXO PR4 did a reasonable job of cleaning them up - here is a clipped part of the image with and without DXO

Without DXO -




With DXO -

[griffljg]

shakyHands, My OM-1, although one of the first sold in Australia, appears to have avoided all of the problems that seem to have afflicted some of the earlier cameras of that batch. Accordingly, I am still using it and am using the E-M1 Mk III with 12-45mm f/4 PRO as a backup / alternative in case I want to take scenic shots.

One thing that I have noticed is that the sensor of the OM-1 lends itself to noise reduction software a lot more than the earlier sensor in the E-M1 Mk III, which is the same sensor as in the E-M1X, E-M1 Mk II, E-M5 Mk III, OM-5 and OM-5 Mk II.

I was most impressed with what DxO PureRAW could do with high ISO images taken with the OM-1 and so am not at all surprised at what it managed to do with your image taken at 102,400 ISO, impressive that it is.

Like many others, I also hope for a new OMSystem camera with a higher resolution sensor, but not at the expense of losing the current high ISO flexibility of the OM-1, OM-1 Mk II and OM-3.

[shakyHands]

griffljg My om-1 mk1 dials were great for 18 months (I picked up mine in early march 2022) but then just started to get sticky and now vary between being reasonably smooth sometimes and difficult to turn at others.