Is Olympus going in the wrong direction?

[Brownie]

Not sure I'd count dropping resolution to gain buffer speed an advantage, but I guess if that's something someone considers important. It wouldn't even be on my radar.

 

[Leolab]

I find the FF syndrome in M43 users comical. I left FF 3 years ago and when I settled on M43, I knew what I'd gain and lose. I made peace with that and have always understood that if I wanted that extra bit of quality, I would have to go with faster glass, which is small compared to FF for the equivalent fov, but I also had the option of tiny, excellent 1.8 primes and small and slower zooms if I was on a budget. My old 500 f4 next to my Panasonic Leica 200mm F2.8 with 1.4x attached gets me images as good but ,with far more ease than my old setup. M43 is a great system, on it's own and doesn't need to be constantly compared to other systems. View attachment 900840 View attachment 900841 View attachment 900842

I'm not sure where you got the impression that this thread was about comparing systems...this thread was about OMDS' future direction and as part of that discussion it is very relevant to understand what the competition is doing, the vast majority of which is in FF. The competition Olympus needs to fend off will come from FF and crop-sensors

You show a picture of a telephoto setup to prove the size advantage, I think we can all agree (as has been stated many times throughout this thread) that currently the main advantage of m43 is in the telephoto size/reach...we all get that...but the big size advantage has eroded in the other fields of view...and sensor technology in FF has advanced faster than in m43

I own both systems and there used to be more reasons for me to grab the m43, less so now...

Given Olympus and their strategy in m43 has lost money for years now...what do you suggest OMDS do about it to change their fortunes?

 

[Leolab]

I can see it if someone had a bunch of APS-C glass from another camera that they wanted to use on their new FF, but other than that there is no advantage.

For me its size.

IF they made a stellar long zoom for Sony E APS-C, i'd be all over it. Why? because it should be smaller for the reach, it would be native and I could AF with best-in-the world AF, and enjoy at least the same resolution as my current m43 cameras...It seems to me that this is exactly the same reason as to why we use m43 for telephoto in the first place

 

[greensteves]

For me, there are two scenarios where m4/3 shines, size and versatility. It's nice to have a compact carry that still puts out good image quality, like an EM10 II with a 14-42mm EZ lens. On the other hand, its also nice to have a not-light but still-comfortable carry that can cover a substantial focal range with excellent IQ, like an EM1 III with an 8-25mm or 40-150mm PRO lens.

For my wife, who wants both compactness and versatility at the same time, and with decent if not great IQ, a 1" sensor is the way to go, on a camera like the RX100 VI.

It would be great if computational photography can pump up m4/3 into an even higher level of IQ. In any case, I'm not likely to change to full-frame, though I acknowledge its advantages for some photographers.

How this plays out in the mass market is a whole other matter.

 

[BDR-529]

That's not actually true, there are FE to Z and m43 to E mount adapters in the wild already, the m43 to E adapters have been out for 8-9 years

I stand corrected. There are plenty of simple mechanical adapters for mount pairs with negative flange to focal differences (for example Konica F to Canon EF)

The drawback is that lens will not focus to infinity but I never realized that m4/3 to E-mount could actually work because there's 1,25mm positive difference and m4/3 bayonet is small enough to be machined inside E-mount Throat. The same is true for Z-mount which I sort of mentioned.

These adapters are still useless for your average m4/3 user because panny, leica and olympus lenses are so modern that few have even mechanical manual focus anymore. I can't immediately recall even one that has manual aperture control from these brands.

Also crop just 25% center of FF sensor area without AF, without aperture adjustment and without IS. In several cases there isn't even manual focus available

 

[fsi22]

I'm not sure where you got the impression that this thread was about comparing systems...this thread was about OMDS' future direction and as part of that discussion it is very relevant to understand what the competition is doing, the vast majority of which is in FF. The competition Olympus needs to fend off will come from FF and crop-sensors

You show a picture of a telephoto setup to prove the size advantage, I think we can all agree (as has been stated many times throughout this thread) that currently the main advantage of m43 is in the telephoto size/reach...we all get that...but the big size advantage has eroded in the other fields of view...and sensor technology in FF has advanced faster than in m43

I own both systems and there used to be more reasons for me to grab the m43, less so now...

Given Olympus and their strategy in m43 has lost money for years now...what do you suggest OMDS do about it to change their fortunes?

At every equivalent focal length, you have a smaller lens, not just telephoto .

OMDS have the smaller lenses, they also now have the faster glass. Smaller and cheaper glass didn't disappear with the release of their faster glass. M43 is the most complete lens system, between Panasonic and OMDS, no focal length for the budget or size conscious aren't filled.

 

[fsi22]

And a rookie shall lead us...

Extremely well said @fsi22 I've tried to make this point many times, but you've done a much more eloquent job here than I ever have. I bolded what I think are the most important points. Many m4/3s users have this format envy problem, and have to get over it. We can ignore the trolls, enjoy our very, very good gear, and stop the constant gnashing of teeth. We really can.

Thank you. When I moved to M43. I was amazed at the creative features like Custom focus limiter, Pro Capture, Live composite, Live ND on the newer models and off course the ibis. I found those so useful that they became settings that I now use all the time and have helped me get many of my favourite images.

 

[demiro]

I'm not sure where you got the impression that this thread was about comparing systems...this thread was about OMDS' future direction and as part of that discussion it is very relevant to understand what the competition is doing, the vast majority of which is in FF. The competition Olympus needs to fend off will come from FF and crop-sensors
...
Given Olympus and their strategy in m43 has lost money for years now...what do you suggest OMDS do about it to change their fortunes?

I think the point is that OMDS should be marketing this system the way fsi22 teed it up - that m4/3s is a great system and doesn't need to be constantly compared to FF.

They shouldn't be 'fending off' anyone. They should be defining and communicating the value the format provides, which to me is small, highly capable and stylish on one end, and small[er] and less expensive on the other end, with great performance for wildlife.

If you nail that value proposition the comparisons look better. FF doesn't really do 'small and stylish', and while it can nail wildlife you're going to pay more money and carry more weight on a mm of reach basis. You do have to allow FF to win in sports and event-type stuff. Not that m4/3s can't do that well, but FFs advantages shine brighter in those spaces.

APS-C does small and stylish. Well, Fuji does, at least. Call that a draw vs OMDS. But they don't do tele nearly as well. CanNikSony will compete on tele side. Only OMDS brings that versatility.

The failure of Olympus, to me, was fighting too hard for the middle ground. The f/1.2 lenses and PRO standard zooms. I know they are great kit, but were clearly bad strategy. Fight where you can win. If that piece of the pie isn't enough to run your business then pack it in.

 

[PakkyT]

This is because the differences are based on total illumination (ie the absolute amount of photons received by the sensor during an exposure which is the major determinant for noise performance and a major aspect for DR as well, as opposed to illumination per unit area which is how we determine exposure)

No, I believe you are claiming something stated two different ways are different. 6 of one vs. half dozen of the other type thing. The FF photosite size advantage is the the amount of photons collected by a single FF pixel is more than a single m43 pixel for a given exposure (same f-stop, same shutter time) because the m43 pixel is smaller. The illumination per unit area determines how much each pixel site sees in photons. If the m43 pixel is one-fourth the size of the FF pixel, then yes, the single m43 pixel will receive only one fourth the photons as the single FF pixel. But once the FF and m43 pixels are the same sizes, then there is no FF vs. m43 pixel. They are identical so call them, say, "Moe pixels". As they are all Moe Pixels each will gather an identical number of photons assuming the same sensor technology.

To your point four times larger FF sensor will gather more absolute photons than the m43 sensor because it covers 4 times the area. But if they are all Moe sized pixels then both sensors gather the same number of photons per Moe. They should share identical image quality characteristics.

I can see it if someone had a bunch of APS-C glass from another camera that they wanted to use on their new FF, but other than that there is no advantage.

No that is not what I meant. Ignore the using cropped lens idea (it is a silly idea really). I just mentioned that the discussion of using those lenses made me think that if m43 stayed at 20MP, once FF sensors reached 80MP then the "photosite collection area" advantage goes away. FF sensors would have no image quality advantages over m43 sensors if you made the m43 sensors with the same generation of same sensor technology.

 

[Leolab]

I think the point is that OMDS should be marketing this system the way fsi22 teed it up - that m4/3s is a great system and doesn't need to be constantly compared to FF.

They shouldn't be 'fending off' anyone. They should be defining and communicating the value the format provides, which to me is small, highly capable and stylish on one end, and small[er] and less expensive on the other end, with great performance for wildlife.

If you nail that value proposition the comparisons look better. FF doesn't really do 'small and stylish', and while it can nail wildlife you're going to pay more money and carry more weight on a mm of reach basis. You do have to allow FF to win in sports and event-type stuff. Not that m4/3s can't do that well, but FFs advantages shine brighter in those spaces.

APS-C does small and stylish. Well, Fuji does, at least. Call that a draw vs OMDS. But they don't do tele nearly as well. CanNikSony will compete on tele side. Only OMDS brings that versatility.

The failure of Olympus, to me, was fighting too hard for the middle ground. The f/1.2 lenses and PRO standard zooms. I know they are great kit, but were clearly bad strategy. Fight where you can win. If that piece of the pie isn't enough to run your business then pack it in.

I agree with this...I think the resources tied up 'chasing' FF with the 1.2 PRO lenses would have been better spent filling out the small, high quality, prime lineup...updating those perhaps with nicer build and aperture rings for those that like the manual controls...adding a nice small 10.5mm f2 AF prime, and maybe a nice small 8mm f2 prime

 

[Leolab]

At every equivalent focal length, you have a smaller lens, not just telephoto .

OMDS have the smaller lenses, they also now have the faster glass. Smaller and cheaper glass didn't disappear with the release of their faster glass. M43 is the most complete lens system, between Panasonic and OMDS, no focal length for the budget or size conscious aren't filled.

You should check out the Samyang and new Sony small primes for their FE system, these are roughly similar in size to the Oly 1.8 primes, the Samyang primes being cheaper. The Sony bodies are also roughly similar in size to the Oly/Panny bodies

As far as all focal lengths being 'covered' by Panny or Oly, I still don't have a 21mm equivalent AF small prime, nor do I have access to an 18mm equivalent AF small prime, i don't think we have a fantastic choice of small lens at 35mmEquiv (the Oly 1.8 being sub-par in my experience, the 1.2PRO being huge), the 28mm equivalent is missing a nice 1.4-2 prime (15mm Panny is closer but noticeably tighter, particularly wrt horizontal field of view), no small longer-normal (58-75 equivalent)...As is mentioned in earlier post, many of the lenses from Oly lack common ergonomics (snap-focus on some, metal bodies on some, plastic bodied others, the silver on some is actually champagne), panasonic chose to go after some weird focal lengths (15mm) with some having aperture rings, others not, some being metal others plastic, some gray and black, others silver or black...)...they have the opportunity of creating some small common-design-language primes that are f1.4-f2 with great mechanics...its a shame they are not doing this

 

[mawz]

No, I believe you are claiming something stated two different ways are different. 6 of one vs. half dozen of the other type thing. The FF photosite size advantage is the the amount of photons collected by a single FF pixel is more than a single m43 pixel for a given exposure (same f-stop, same shutter time) because the m43 pixel is smaller. The illumination per unit area determines how much each pixel site sees in photons. If the m43 pixel is one-fourth the size of the FF pixel, then yes, the single m43 pixel will receive only one fourth the photons as the single FF pixel. But once the FF and m43 pixels are the same sizes, then there is no FF vs. m43 pixel. They are identical so call them, say, "Moe pixels". As they are all Moe Pixels each will gather an identical number of photons assuming the same sensor technology.

To your point four times larger FF sensor will gather more absolute photons than the m43 sensor because it covers 4 times the area. But if they are all Moe sized pixels then both sensors gather the same number of photons per Moe. They should share identical image quality characteristics.

There are two aspects to the IQ equation.

1. How many photons each pixel gathers.
2. How many pixels in the image.

The two are directly related and both impact IQ. More pixels improves IQ so long as the pixels stay the same size, bigger pixels improve IQ so long as the overall pixel count stays the same. And per-pixel performance is largely irrelevant today except when comparing similar size/similar pixel density sensors because the pixel density/overal image IQ tradeoff is pretty much even as long as the sensor fill factor is high enough (ie so you aren't losing photons to the circuitry between the pixel sites)

The reality is that noise performance very closely approximates total illumination with current technology sensors, and pixel size really only matters if the pixels are either unusually large (really low-MP FF sensors like the A7S 12MP sensor) or small (a notional 44MP m43 sensor) and even then not as much as most thing. Yeah, you still see some differences between big pixel sensors and little pixel sensors, but far less than in the past, because we're hitting the point where the primary determinant of noise is shot noise (aka quantum effects) and those differences also only show up in really low light situations (ISO6400+)

This is why the noise performance difference between m43 and FF trends at 2 stops for similar tech sensors. You're getting 2 stops more total illumination.

As an aside, this is also why checking noise performance at 100% magnification is no longer useful, as it tells you nothing about how the image will look.

This is easy to test with any camera with a crop mode. A D850 in DX mode generates files essentially identical to a D500. Same basic sensor tech, almost identical pixel density and cropped to the same size and approximate pixel count. But compare an FX format image from the D850 to the same image from the D500 and the D850 image has better DR and lower noise across the image. Same size pixels with the same per-pixel performance, but 25 Million more of them giving data.

Equalizing for pixel pitch only equalizes per-pixel performance (which is really not relevant with modern cameras), not overall image performance and that FF sensor still gets 4x as many photons per image as the m43 sensor.

 

[mawz]

I stand corrected. There are plenty of simple mechanical adapters for mount pairs with negative flange to focal differences (for example Konica F to Canon EF)

The drawback is that lens will not focus to infinity but I never realized that m4/3 to E-mount could actually work because there's 1,25mm positive difference and m4/3 bayonet is small enough to be machined inside E-mount Throat. The same is true for Z-mount which I sort of mentioned.

These adapters are still useless for your average m4/3 user because panny, leica and olympus lenses are so modern that few have even mechanical manual focus anymore. I can't immediately recall even one that has manual aperture control from these brands.

Also crop just 25% center of FF sensor area without AF, without aperture adjustment and without IS. In several cases there isn't even manual focus available

Z mount is interesting because it offers electronically coupled adapters for FE lenses and could also do so for m43. RF mount is probably capable as well, FE wouldn't be due to space limitations (there's just barely enough room for a mechanical adapter).

AF coupled Z adapters are possible for m43, it would just require development. Techart certainly could do so, it's not like the m43 AF protocols aren't well documented.

Remember, there's a LOT of mechanically coupled m43 lenses. The AF ones are useless because of focus by wire and no aperture ring, but CV, Laowa, 7Artisans, TTArtisan, Pergear and more make mechanical m43 lenses.

 

[Brownie]

No that is not what I meant. Ignore the using cropped lens idea (it is a silly idea really).

Possibly, but manufacturers are doing it. Someone must think it works.

 

[doxa750]

I think the point is that OMDS should be marketing this system the way fsi22 teed it up - that m4/3s is a great system and doesn't need to be constantly compared to FF.

They shouldn't be 'fending off' anyone. They should be defining and communicating the value the format provides, which to me is small, highly capable and stylish on one end, and small[er] and less expensive on the other end, with great performance for wildlife.

If you nail that value proposition the comparisons look better. FF doesn't really do 'small and stylish', and while it can nail wildlife you're going to pay more money and carry more weight on a mm of reach basis. You do have to allow FF to win in sports and event-type stuff. Not that m4/3s can't do that well, but FFs advantages shine brighter in those spaces.

APS-C does small and stylish. Well, Fuji does, at least. Call that a draw vs OMDS. But they don't do tele nearly as well. CanNikSony will compete on tele side. Only OMDS brings that versatility.

The failure of Olympus, to me, was fighting too hard for the middle ground. The f/1.2 lenses and PRO standard zooms. I know they are great kit, but were clearly bad strategy. Fight where you can win. If that piece of the pie isn't enough to run your business then pack it in.

I agree that F1.2 Pro strategy didn't really work out for them. However, I am very glad they exist as it prevents me from needing to add FF. I am chasing my little ones in-house a house and it is priceless to be able to capture those precious moments while she is still willing . It is not going to equal FF but it is way more than good enough for me personally.

I think their F1.4 Pro Lens line up should do well judging how small PenLeica is. i.e. 25mm F1.4. I would be pre-ordering 10mm F1.4 Pro lens and/or 100mm F2.8 Macro Lens if OMDS ever decides to make them.

Cheers

 

[Aristophanes]

I think the point is that OMDS should be marketing this system the way fsi22 teed it up - that m4/3s is a great system and doesn't need to be constantly compared to FF.

They shouldn't be 'fending off' anyone. They should be defining and communicating the value the format provides, which to me is small, highly capable and stylish on one end, and small[er] and less expensive on the other end, with great performance for wildlife.

If you nail that value proposition the comparisons look better. FF doesn't really do 'small and stylish', and while it can nail wildlife you're going to pay more money and carry more weight on a mm of reach basis. You do have to allow FF to win in sports and event-type stuff. Not that m4/3s can't do that well, but FFs advantages shine brighter in those spaces.

APS-C does small and stylish. Well, Fuji does, at least. Call that a draw vs OMDS. But they don't do tele nearly as well. CanNikSony will compete on tele side. Only OMDS brings that versatility.

The failure of Olympus, to me, was fighting too hard for the middle ground. The f/1.2 lenses and PRO standard zooms. I know they are great kit, but were clearly bad strategy. Fight where you can win. If that piece of the pie isn't enough to run your business then pack it in.

The history of photography is the 135 format eats up all smaller formats.

Economy of scale and defined standards.

FF manufacturers don’t have to make the smallest kits. They just have to get near enough to take away the margin market share.

 

[PakkyT]

There are two aspects to the IQ equation.

1. How many photons each pixel gathers.
2. How many pixels in the image.

Hmmm, a lot to digest there. Acknowledged and will have to look into it further,

 

[Aristophanes]

There are two aspects to the IQ equation.

1. How many photons each pixel gathers.
2. How many pixels in the image.

The two are directly related and both impact IQ. More pixels improves IQ so long as the pixels stay the same size, bigger pixels improve IQ so long as the overall pixel count stays the same. And per-pixel performance is largely irrelevant today except when comparing similar size/similar pixel density sensors because the pixel density/overal image IQ tradeoff is pretty much even as long as the sensor fill factor is high enough (ie so you aren't losing photons to the circuitry between the pixel sites)

The reality is that noise performance very closely approximates total illumination with current technology sensors, and pixel size really only matters if the pixels are either unusually large (really low-MP FF sensors like the A7S 12MP sensor) or small (a notional 44MP m43 sensor) and even then not as much as most thing. Yeah, you still see some differences between big pixel sensors and little pixel sensors, but far less than in the past, because we're hitting the point where the primary determinant of noise is shot noise (aka quantum effects) and those differences also only show up in really low light situations (ISO6400+)

This is why the noise performance difference between m43 and FF trends at 2 stops for similar tech sensors. You're getting 2 stops more total illumination.

As an aside, this is also why checking noise performance at 100% magnification is no longer useful, as it tells you nothing about how the image will look.

This is easy to test with any camera with a crop mode. A D850 in DX mode generates files essentially identical to a D500. Same basic sensor tech, almost identical pixel density and cropped to the same size and approximate pixel count. But compare an FX format image from the D850 to the same image from the D500 and the D850 image has better DR and lower noise across the image. Same size pixels with the same per-pixel performance, but 25 Million more of them giving data.

Equalizing for pixel pitch only equalizes per-pixel performance (which is really not relevant with modern cameras), not overall image performance and that FF sensor still gets 4x as many photons per image as the m43 sensor.

All sensor images are composites of pixels. The more pixels with greater bit depth, the more accurate the image. Translates to larger sensors with greater DR and lower noise. Makes that 2-stops baked in between m43 and FF. Just physics.

 

[demiro]

The history of photography is the 135 format eats up all smaller formats.

Economy of scale and defined standards.

FF manufacturers don’t have to make the smallest kits. They just have to get near enough to take away the margin market share.

I don't think the past always predicts the future, especially in the digital camera space of 2021. Miscellaneous business terms notwithstanding.

 

[swifty]

Skimming through this thread it's obvious there are a wide variety of views on what directions OMDS should take. I may as well add my thoughts into the mix.

Coming as a dual system user, I just can't see any advantages of OMDS going into FF other than being a premium third party lens supplier. I just find that this segment is already the most competitive with respect to bleeding edge technology and being fought out by the three largest camera manufacturers with a fourth alliance trying to vie for the small remaining marketshare.
But OMDS will have retained many of the lens design expertise for large sensor ILCs so they can either design and manufacture for other mounts eg. L-mount if allowed to join that alliance or just farm out design work, which happens a lot in the industry anyways.

If (and this is a massive if) they are to try another format, I agree with one of the previous posters in suggesting that they should try medium format. It's a much smaller market that is vastly different to m43, with high margins and without the expectation to create a really big ecosystem as it's not expected to address every niche like the mainstream FF market.
Here I think OMDS can really shake things up by creating a very compact MF system which Fujifilm and Hasselblad have already entered into but I think OMDS can do it better with their history of compact systems. But the resources required remain very large and I'm just not sure JIP will inject the type of capital required for a committed crack at it.

So I think we're 'stuck' with m43 mount. Which IMHO really isn't a bad thing as the rest of the internet seem to want to insinuate. Of course ppl on this board already know this.
We have a well considered modern, high bandwidth mount that is appropriately large for an excellent IBIS system, supported by a large number of third party lens and accessory manufacturers in addition to OMDS and Panasonic.

Where I think m43 and OMDS can go is at least maximise the image circle that is guaranteed available on all m43 lenses, which is 21.6mm in diameter. Panasonic already does this with their mutiaspect ratio sensors for some GH products but that's only maximising things for video in the wider aspect ratios.
I'd like to see OMDS take this approach further doing multiapsect ratio all the way from 16:9 to 1:1.
That means using a sensor sized at least18.8 X 15.3mm instead of 17.3 X 13mm so alternatively an APS-C sensor @24X16mm would also work but with an active area of 18.8X15.3mm so there are some wasted real estate going this route.

For the benefits, you'd get:
APS-C vs oversized m43 vs normal m43:
3:2 = 384mm² vs 216mm² vs 199mm²
4:3 = 341mm² vs 225mm² vs 225mm²
16:9 = 324mm² vs 199mm² vs 168mm²
1:1 = 256mm² vs 234mm² vs 169mm²
Depending on the aspect ratio, it ranges from a small improvement (3:2) to almost equaling APS-C (1:1) from an initial gap that was less than a stop to begin with.
The additional engineering required would be for the IBIS unit to support the larger mass of a larger sensor.
Incidentally a 43MP Sony Semi APS-C stacked sensor is rumoured to be floating around, which also has a higher pixel density than 20MP m43 sensors, so that can be a candidate and it'd remove m43's dependency on a special sensor designed just for m43. They can basics utilise any APS-C sensor which covers a much larger volume market but of course we’re interested in the high pixel density stacked ones.
As an added bonus, third party designs eg. some from Sigma are used for both APS-C and m43 such as the 56mm f/1.4. These lenses with large enough image circles can also be used in both m43 and APS-C mode.

Beyond that, for the lens system I think OMDS can focus on three areas:
- Macro
- Telephotos in a moderate size
- Fast but not uber fast small primes eg. f/1.4 series that’s been rumoured but keep it on the small side and WR.

As for features, make weather resistance common for all their products.
Continue to lead in ibis and experiment more with computational features like live composite and pro capture.