428 replies to this topic

[gkarris]

Decided to start a thread on Micro Four-Thirds...

 

There's lots of discussion on You Tube saying MFT is dead

 

https://www.youtube....h?v=qjXSnNMZ0PU

 

plus lots of rebuttals like it won't and MFT doesn't really have much noise and such (from Olympus Ambassadors nonetheless):

 

https://www.youtube....h?v=uXrEKHwX_BU

 

(plus a lot of the MFT vids on YT seem to be about how good a video camera it is, what about just regular pics?)

 

My first dSLR was an regular 4/3 Olympus E-500 then I got a much smaller E-420, but with the low light pics and cropping the noise was horrendous, plus Olympus promptly abandoned that mount for MFT. I switched to Canon APS-C then to Nikon (8 years ago)...

 

I've been using a D3300 and only a Nikon ED VR 18-200 as my walk around for over 4 years, but of course it is kind bulky....

 

I decided to give MFT a try and got a deal on a cheap Lumix G3 and 45-200 OIS lens to see how it goes with walking around and low light performance.

 

Is MFT just getting a bad rap or are the rumors true? Can you get in on digital photography on the cheap with people getting rid of this format? Are the mirrorless APS-C's a better deal for about the same size?

 

Please post your experiences here. Thanks as always...

[Grimstod]

Its just getting a bad rap. In astrophotography Quantum efficiency is what matters. Smaller sensors are more quantum efficient. Take for example ZWO 1600. It is a 4/3rds sensor and is in high demand. ZWO some time ago made some 35mm sensors but they were just not efficient enough and are not discontinued. 

 

Most of the anti 43 hype comes from a hugely inaccurate article written by DPR Review. Its their equivalency article. It really shows just how dumb and ignorant people have become about photography now in this age. I blame our educational system for this. 

 

Tony especially is not to mathematically astute. He has been called out multiple times on making inaccurate statements. I know I notice one just about every video he makes. 

Edited by Grimstod, 03 August 2019 - 12:42 PM.

[Alen K]

Its just getting a bad rap. In astrophotography Quantum efficiency is what matters. Smaller sensors are more quantum efficient. 

Smaller sensors are more quantum efficient? Please explain. 

[Grimstod]

Scroll left to right. You can see that the QE gradually goes up with smaller sensors.
https://compare.astr...ing-camera.com/

[Alen K]

Scroll left to right. You can see that the QE gradually goes up with smaller sensors.
https://compare.astr...ing-camera.com/

I don't think you can come to that general conclusion from that information. Those smaller sensors also have smaller pixels and sensors with smaller pixels tend to be back-side illuminated, which does generally lead to higher QE (with some exceptions). Older small sensors without BSI don't have such high QEs. I asked for an explanation because I had never heard that relationship stated. 

 

That said, the issue here was MFT vs something larger like APS-C or FF. When comparing two sensors you might find an MFT with higher QE than a larger sensor but the increase would have to be larger than the ratio of the areas to make the MFT better in terms of SNR. If the QEs are similar, with any given lens or telescope the MFT will merely have a smaller field of view, as long as the pixel sizes are the same. But if the MFT sensor has smaller pixels that would result in lower per-pixel SNR.

 

The noise issue for general photography is usually from the optics. MFT-camera manufacturers like to show you how their tiny 150mm f/4 lens for their camera is just like the 300mm f/4 lens for an FF camera. But the aperture diameter of the 150mm f/4 lens is half that of the other lens so lets in 1/4 the light. The 150mm lens would have to be f/2 to get a similar SNR (and the same depth of field, but that is a non-issue for AP).

Edited by Alen K, 04 August 2019 - 11:34 AM.

[gkarris]

The noise issue for general photography is usually from the optics. MFT-camera manufacturers like to show you how their tiny 150mm f/4 lens for their camera is just like the 300mm f/4 lens for an FF camera. But the aperture diameter of the 150mm f/4 lens is half that of the other lens so lets in 1/4 the light. The 150mm lens would have to be f/2 to get a similar SNR (and the same depth of field, but that is a non-issue for AP).

 

Isn't this part of the MFT debate? You have to apply the crop factor to the f-stop as well, so a 150mm f/2 in FF is really a 300mm f/4 in MFT?

 

Whenever this issue is brought up elswhere, a flame-war ensues...

 

It's clearer to me now being into Astronomy - it's all about "that light bucket"....

[gkarris]

Went Planespotting with the Lumix G3/45-200

 

Pros: Small size easy to just carry around.

 

Con: Smaller size = smaller battery + always liveview w/ mirrorless = not as long a battery life.

 

The only thing is that this new camera and format and resolution means I have to learn how to do stuff in post (compared to Nikon APS-C).

 

Very long telephoto (400mm equivalent) now there's much more heat distortion (compared to my Nikon 300mm equivalent).

 

 

 

 

I did take test photos with my 3 dslr's in my living room with just a lamp on at night at a dark corner at ISO 6400. Best is Nikon D3300, second Canon T3i, somewhat distant 3rd Lumix G3.

 

Of course, that Sony 24MP APS-C Sensor in the Nikon is gonna be tough to beat...

Edited by gkarris, 04 August 2019 - 01:14 PM.

[outlierrn]

People have been saying that micro 4/3rds was doomed since it first came out.  Here's a couple I took with a Panasonic G9, flame nebula with a 50-200mm lens, and M8-20 with the Redcat 51.  

Attached Thumbnails

• 
• 

[gkarris]

People have been saying that micro 4/3rds was doomed since it first came out.  Here's a couple I took with a Panasonic G9, flame nebula with a 50-200mm lens, and M8-20 with the Redcat 51.  

Those pics are awesome!!!

 

I just saw some YouTube vids on that Redcat - very nice...

 

Is that Leica 50-200 f/2.8 you used?

 

Thanks

[outlierrn]

Those pics are awesome!!!

 

I just saw some YouTube vids on that Redcat - very nice...

 

Is that Leica 50-200 f/2.8 you used?

 

Thanks

Thank you, yeah that's the lens, although it's a F2.8-4, so I was shooting at F4.  Everything's a tradeoff, of course, and you can see some distortion with a zoom lens.  OTOH, I don't think I'm quite nailing focus with the Redcat yet, but neither of those issues are camera related.  

[Grimstod]

Isn't this part of the MFT debate? You have to apply the crop factor to the f-stop as well, so a 150mm f/2 in FF is really a 300mm f/4 in MFT?

 

Whenever this issue is brought up elswhere, a flame-war ensues...

 

It's clearer to me now being into Astronomy - it's all about "that light bucket"....

 

This is where the war starts.    So when you crop to a smaller sensor it does not change the F stop. This is something that was universally scepter for over a 150 years. Then DPR wrote a horrible article saying it is real. It isn't, never was, never will be. A F4 lens on a APSC sensor is an F4 lens on a 35mm and on a 43rds. If the so-called equivalency thing was true then my Mamyia 6x7 camera would have a 4.3 stop advantage over my 35mm. That means a 2.8f lens I have on the 6x7 would actually be F 0.5, does that sound ridiculous? Yes it does and is. What you loose in going to a smaller sensor is resolution. That is it. With the crazy amount of resolution offered by todays cameras. I choose lower resolution and faster lighter lenses. 

[Grimstod]

I don't think you can come to that general conclusion from that information. Those smaller sensors also have smaller pixels and sensors with smaller pixels tend to be back-side illuminated, which does generally lead to higher QE (with some exceptions). Older small sensors without BSI don't have such high QEs. I asked for an explanation because I had never heard that relationship stated. 

 

That said, the issue here was MFT vs something larger like APS-C or FF. When comparing two sensors you might find an MFT with higher QE than a larger sensor but the increase would have to be larger than the ratio of the areas to make the MFT better in terms of SNR. If the QEs are similar, with any given lens or telescope the MFT will merely have a smaller field of view, as long as the pixel sizes are the same. But if the MFT sensor has smaller pixels that would result in lower per-pixel SNR.

 

The noise issue for general photography is usually from the optics. MFT-camera manufacturers like to show you how their tiny 150mm f/4 lens for their camera is just like the 300mm f/4 lens for an FF camera. But the aperture diameter of the 150mm f/4 lens is half that of the other lens so lets in 1/4 the light. The 150mm lens would have to be f/2 to get a similar SNR (and the same depth of field, but that is a non-issue for AP).

As I explained in my last post you'r talking about equivalency. Something DPReview invented. It is actually total hogwash. A F4 lens is an F4 lens no mater what format it is on. What changes is the resolution. 

Also smaller sensors can be cooled to much lower levels then can larger ones. This is another thing that should be taken into account considering that noise from heat Doubles with every 1 degree of temperature increase if I am reading my astrophotography book correctly. 
My statement stands that smaller sensors or generally more Q Efficient. This is because smaller sensors or easier to make. And it is a variable that you will see across every Astrophotography camera line. 

[Grimstod]

An Astrophoto I took of Orion using a 600mm f7.5 using an Olympus OMD Em1 When I put a 43rd camera on my 600mm f7.5 it does not suddenly become a f15, it is still a f7.5

[gkarris]

This is where the war starts.    So when you crop to a smaller sensor it does not change the F stop. This is something that was universally scepter for over a 150 years. Then DPR wrote a horrible article saying it is real. It isn't, never was, never will be. A F4 lens on a APSC sensor is an F4 lens on a 35mm and on a 43rds. If the so-called equivalency thing was true then my Mamyia 6x7 camera would have a 4.3 stop advantage over my 35mm. That means a 2.8f lens I have on the 6x7 would actually be F 0.5, does that sound ridiculous? Yes it does and is. What you loose in going to a smaller sensor is resolution. That is it. With the crazy amount of resolution offered by todays cameras. I choose lower resolution and faster lighter lenses.

 
They physical F-stop doesn't change, obviously, the light entering an F4 lens is the same no matter what camera it's on.
 
It is about equivalency (my understanding) 35mm being the "standard".
 
An F/2.8 lens on a 6x7 is like having an F/0.5 lens on a 35mm...
 
So F4 equivalency on MFT is F8 light on the surface of the sensor compared to 35mm...
 
Some people talk "equivalency" some talk the physical lens.
 
In the end, it's all about the pics and no matter what you use, a great pic is what matters.
 
Like ^^^Grimstod's pic...

An Astrophoto I took of Orion using a 600mm f7.5 using an Olympus OMD Em1 When I put a 43rd camera on my 600mm f7.5 it does not suddenly become a f15, it is still a f7.5
FullSizeRender.jpg

Incredible!!

Edited by gkarris, 06 August 2019 - 08:37 AM.

[Grimstod]

Never be afraid to use 43rds for astrophotography. Many dedicated astro cameras use the sensor. 

[Jeff Lee]

I've strongly been considering a ZWO294 based on my use of the 224. But this thread has convinced me to try out my G9 and Gx8 (I have tether programs for both). SharpCap will work with a capture folder, and I tried it once but got involved in making the 224/ES102 like each other. Next clear night going to put the M43's cameras to the test. Thanks Grimstod.

[Grimstod]

If you do eventually spring for a cooled camera from ZWO give the 1600 a serious look. Its a 4/3rds sensor and with their lens adapter you can use your current lenses on it. 

[Jeff Lee]

I considered that model but it will be used mainly with SharpCap at home on the ES102 and the 224 on the C8 when my herniated disc is healed.

[outlierrn]

One thing to know about equivalency, is that terrestrial photographers use F-numbers to judge depth of field, something astro types don't.  F4 is F4 in terms of how long it takes to collect a million photons, but the depth of field of F4 on a M4/3s and a full frame are different.

[Lumix.guy]

I switched from Canon DSLR to M43 years ago and I've owned Panasonic GF1, G1, G2, G6 and now the G9.  I've done casual astrophotography with all of them (and my old Canon 30D) and I can tell you the G9 is a huge step forward compared to the G6 as far as thermal noise.

 

I also own the ZWO ASI1600MC-Cool and it uses a Panasonic M43 sensor and is a wonderful astronomy camera.  I have used the ZWO M43 lens adapter and it works well.  Lenses to consider: Olympus 8mm f/1.8 fisheye, Panasonic 25mm f/1.7 and Panasonic 42mm f/1.7.

 

The G9 also has a night mode that turns the viewfinders red. 

 

John

[17.5Dob]

Try and find a 12mm f1.4 for a 4/3 camera (24mm f1.4 on ff) and lets talk equivalency/ Etendue. 8.6mm aperture with the supposed 12mm f1.4 and 17mm with the 24mm. F ratio is the same. Which lens is going to deliver more light......?

Edited by 17.5Dob, 02 September 2019 - 08:05 PM.

[Lumix.guy]

For point sources (stars) the 24mm wins, but for extended objects (nebulae) they are much more similar than you might think.  For extended objects you also need to look at sensor pixel size, quantum efficiency, vignetting and other details.

[17.5Dob]

For point sources (stars) the 24mm wins, but for extended objects (nebulae) they are much more similar than you might think.  For extended objects you also need to look at sensor pixel size, quantum efficiency, vignetting and other details.

And the discussion is not about a 1600 cam or any other astrocam using a 4/3 sensor. We are talking about using an actual 4/3 sensor camera, for "wide angle" MW shots,  (itself an oxymoron, because you'd need a 9-10mm lens to get the same wide angle FOV as an 18-20mm lens on FF).

For the same pixel size, same QE, same f-ratio, a larger aperture lens will always beat a smaller one, to get the same FOV, particularly for extended objects.

You are trying to side step the issue of using an actual 4/3 camera.

There is no debating the success of the 4/3 chipped astocams, but in real life usage, a 4/3 camera on a tripod trying to shoot the MW..............................? Unless you shoot and stack a few dozen frames, you'll be forced to use ISO 12,000 plus to compensate for the pinhole lens required with a 4/3 camera.

[outlierrn]

Maybe I'm missing something, but this is a single frame 60 sec, iso 1600, 12mm FL shot on a G9 and a Skyquider pro.

 

eta f3.2

Edited by outlierrn, 03 September 2019 - 12:07 AM.

[outlierrn]

Hmm, I don't know why CN reduced my 421kb file to 184kb, but it's much crisper than that.