Lintelfilm

Exactly correct.
F-stop is a measurement of exposure - the amount of light hitting the sensor is the amount of light hitting the sensor. However aperture size as related to depth of field is a different thing entirely - not related to exposure but image character. Of course the two things are interrelated as one affects the other, but they should not be confused. 
There will always be real-world advantages to larger sensors, but all things being equal (on paper / mathematically) you can always re-create the "full frame" aesthetic on smaller sensors. 
THIS IS SERIOUS!!!! STOP BEING TRIVIAL!!!!

Exactly correct.
F-stop is a measurement of exposure - the amount of light hitting the sensor is the amount of light hitting the sensor. However aperture size as related to depth of field is a different thing entirely - not related to exposure but image character. Of course the two things are interrelated as one affects the other, but they should not be confused. 
There will always be real-world advantages to larger sensors, but all things being equal (on paper / mathematically) you can always re-create the "full frame" aesthetic on smaller sensors. 
THIS IS SERIOUS!!!! STOP BEING TRIVIAL!!!!

Idk, to me the full frame example clearly pushes the woman's hair off her right shoulder, whereas the m4/3 leaves it draped over her right shoulder. Damn distortion...

Ahhh Mattias I'm disappointed in you! I am quite a fan of your YT channel but you are WRONG here!
As @Timotheus says you are actually using an example that disproves your own point! Those two images ARE identical (within the realms of reasonable difference - of course you can never replicate the EXACT same look with two different lenses on different cameras). There is no significant difference between them. Background separation, compression, bokeh, etc are within the realms of "the same". 
This is a very confusing subject, because all the terminology gets messy when talking about equivalencies. I made the mistake in my earlier post of refering to "equivalent" F-stop, which you rightly took issue with, but I was only wrong in the terminology I used. What I should have said is "aperture size" (which as we know is a different thing). I absolutely do think there is a use for talking in terms of equivalents when discussing f-stop/aperture on different sensor sizes, but only when thinking about DOF. It makes sense to say F-stop should remain the same because that tells you about exposure/ the amount of light getting to the sensor - but it doesn't tell you about Depth of Field  image characteristics. Here you can talk in equivalencies.
Again, Timotheus is correct: 50mm 1.8 on FF = 35mm 1.2 S35 = 25mm 0.9 MFT
Of course in the real world you have to give a little room for error but in theoretical terms there is no difference in DOF characteristics. Forget about exposure - that is different. This is about depth of field and focal length.

Theoretically, yes. That's why I like the Sigma APSC zooms (yes I know I keep going on about them) on my C100 - because you can see the corners/edges. However different lenses have different corner characteristics. Some are not so nice.

Okay then, that brings up another question I was wondering about... Would it be better to use a DX lens with the D5500 to get the entire characteristics of the entire glass than a cropped portion of the full frame glass... Theoretically?
@Nikkor

Given equal quality of the footage (say, not a really soft 1080p output from one of the cameras): if you match exposure and colors of footage, very few will notice - even if given the task to spot the differences. Low-light footage will be an exception, where footage from smaller sensors will be noisier (if they're from the same generation of sensor technology).
One aspect I find important with full frame, is that I can easily find wide lenses with very little distortion for full frame. With the 28mm f/2.8 AI-S Nikkor, or with the 16-35mm f/4 I can shoot wide shots with very little distortion. The smaller the focal length, the harder it becomes to design lenses with little distortion.
 

@Mattias Burling we seem to disagree, so let's explore this al little bit, because I think understanding equivalence is useful for anyone, especially when juggling camera's with different sized sensors.
You didn't respond to what I said, i.e. you can get the same framing, same depth of field, shooting from the same spot...with different sensor-sized camera's. The key is using lenses that compensate for the differences in sensor size. The math concerns using crop factors for both focal length and f-stop to estimate the effects on framing and DOF. The physical f-stop obviously does not change. You show a screenshot from a Tony Northrup video that proves exactly these points! In the example using 100mm f5.6 on full frame yields the same framing and DOF as a 50mm f2.8 on MFT (2x crop). You can hear Northrup explain from 16:06...your example shows up right at 17:18 :-) Obviously there are limits as to what is currently possible. Getting the same framing and DOF as a fullframe 50mm F1.2 on a MFT camera would mean using a 25mm F0.6, which doesn't exist (yet!).

Holding cameras in the exact same position, you can take the same (framing, DOF) picture with:
a 50mm f1.8 on full frame a 35mm f1.2 on APS-C a 25mm f0.9 on MFT So there really is no specific aesthetic linked to a certain format, just math.

Not sure if you meant that reply for me or HelsinkiZim who asked about digital filters, but the ones I have -HDTVFX and UltraCon - are physical filters. 
I cant comment on Tiffens digital filters as I've never used them. I do always add some level of grain though, even if I want a clean digital look, just to dither a bit.
As an aside, I also built my own FCPX "lens character" plugin that is pretty cool if I do say so myself. FCPX users can download it free from my site. It's been very popular actually - Mike Matzdorf the guy who edited the Will Smith film Focus in FCPX has it and sent me a couple of tweets about it...
If you google Dawg Pu (yes it's a stupid name ripping off Dog Schidt Optics in a jokey way) you'll end up on my blog post about it. Richard Gale even gave his approval!
Of course I'd rather get Richard to build me a full customised set of DSO primes but getting effects optically is not always practical.
 

Not sure if you meant that reply for me or HelsinkiZim who asked about digital filters, but the ones I have -HDTVFX and UltraCon - are physical filters. 
I cant comment on Tiffens digital filters as I've never used them. I do always add some level of grain though, even if I want a clean digital look, just to dither a bit.
As an aside, I also built my own FCPX "lens character" plugin that is pretty cool if I do say so myself. FCPX users can download it free from my site. It's been very popular actually - Mike Matzdorf the guy who edited the Will Smith film Focus in FCPX has it and sent me a couple of tweets about it...
If you google Dawg Pu (yes it's a stupid name ripping off Dog Schidt Optics in a jokey way) you'll end up on my blog post about it. Richard Gale even gave his approval!
Of course I'd rather get Richard to build me a full customised set of DSO primes but getting effects optically is not always practical.
 

@Mattias Burling (and following on from my own answer) of course this is where super fast lenses for smaller sensors come come in. An f/0.95 on MFT will be ROUGHLY equivalent to a 1.8 on full frame. If we talk about zooms only, the Sigma 1.8 APSC zooms are roughly equivalent to 2.8 zooms on full frame. And seeing as you can't get 1.8 zooms for FF you could argue they negate the FF advantage. But then you get into advantages of specific lens features etc (such as IS, focal range, AF) and it comes down to which system suits your needs best. I moved from MFT to EF for various reasons but partly because building a lens collection around EF on a S35 sensor is as close as you get to a video/cinema standard now (below PL of course).

Speaking of the Sigma 18-35, it's the only "designed for APS-C" lens I am keeping in my kit; it does not vignette at all on the 1dx II in 1.4x 4k.  It's a stunner of a lens!

Baassiilllll!

Don't mention the war!

From Canon:
 

Thanks JP. I too got much less ghosting with everything off but it is there.
Personally I think the XC10 is still an awesome camera and although this ghosting issue should be acknowledged by Canon it's not keeping me awake at night. The camera uses a lot of digital fandangledness we really know little about to create a very sturdy and overall great image in most situations (esp. for the sensor size). I can live with this ghosting, because in motion it is only very noticeable in quite extreme circumstances. I'd guess that whatever is causing it is doing a job that is contributing a lot to the image quality so as with any camera I'll pay the price of compromise. 
The XC10 is really aimed at VJ's and other run+gun filmmakers who require versatility, speed and generally clean-out-of-camera images over "image perfection" (whereas a narrative filmmaker in contrast may prioritise the latter).
I'll contact CPS too but I'm not going to send the camera to them. Personally I'm more than happy using it as-is.
 

From flicking back and forth between ISO and (fine) gain in a non-Log picture style, ISO 400 = 7.5db  and ISO 500 = 10.5db.
Going down from 10.5db to 8.5 using fine gain controls is -2db, so that's a 1/3 of a stop. It seems to only make a difference in WideDR and CLog. Could be useful for another little bit of exposure latitude when shooting outdoors.

I think this is a very sensible point of view - you can great footage from the cam as is. Nevertheless, if it is a NR issue I'd prefer to have the option of using Canon's NR processing in camera, or using my own in post if I could avoid ghosting. Or having it fixed somehow.
Sometimes people talk about motion cadence and its importance and even if you can't point out ghosting with normally exposed footage at base ISO without looking at individual frame grabs, it nevertheless is there and is performing a streaking of motion that affects our perception of movement. And, for me, the premise of this camera is that it is a camera that moves - small, light, handheld, image stabilisation, on the run, on a gimbal. So its rendering of movement should be good. And that is compromised at the moment.

How would it be normal when they never done that before?
The XA20/25, XA30/35, XF100/105 and XF200/205 have all been available together as the exact same models with only one connection separating them.
For Canon to all of a sudden abandon this practice and starting to promote an XX5-Sibling over another would be a first.

Thanks JP. I too got much less ghosting with everything off but it is there.
Personally I think the XC10 is still an awesome camera and although this ghosting issue should be acknowledged by Canon it's not keeping me awake at night. The camera uses a lot of digital fandangledness we really know little about to create a very sturdy and overall great image in most situations (esp. for the sensor size). I can live with this ghosting, because in motion it is only very noticeable in quite extreme circumstances. I'd guess that whatever is causing it is doing a job that is contributing a lot to the image quality so as with any camera I'll pay the price of compromise. 
The XC10 is really aimed at VJ's and other run+gun filmmakers who require versatility, speed and generally clean-out-of-camera images over "image perfection" (whereas a narrative filmmaker in contrast may prioritise the latter).
I'll contact CPS too but I'm not going to send the camera to them. Personally I'm more than happy using it as-is.
 

I'd wait to sell it until we confirm that Tom is having the same results with ALL of the same parameters.
Honestly, I'm not too concerned about it because I will rarely shoot over ISO 500, I know it can still rear it's ugly head, but I never noticed it any of my footage until I bumped it up to ISO 20,000. If Tom's XC15 footage showed it, I would have chalked it to... Don't use any camera at 20,000 ISO. Lol.

I retested Pikachu with less light and was able to get alittle more ghosting to appear with Clog.  It happens with horizontal movement with hard pans.  Vertical only doesn't seem produce it but horizontal + vertical movement does. EOS Standard and Cinema EOS Standard have the least ghosting artifacts.  

*ISO is set to 6400 in these test.
Clog Ghosting Horizontal


Clog Ghosting Vertical


Cinema EOS Standard Ghosting Horizontal

Cinema EOS Standard Ghosting Vertical


With this level of lighting I would not be shooting with the XC10 anyway....but if i did i would not use clog.  Normal panning with Cinema EOS Standard was usable with very little ghosting.  Wide Dynamic range also had better performance than Clog.
 

On another note, I have been watching some XC10 videos on YouTube and there is some great footage out there. I am amazed how many people are using all auto and getting some incredible results out of this camera. I may be on a a mission to shoot a completely auto, run and gun feature with the XC10.

Personally I think this is an issue you will see in any XC10. We don't have anyone here really looking for the ghosting and saying they can't find any.
I think you'd be wasting your time to send your unit in. I'd press them to try it on a random unit (or ten) to confirm/underline that it is a universal issue.
 
Posted these on the other thread ages ago. Granted they're out of focus but I think they show the real world effects of the issue pretty well. Look at the weird banding on the faces, and in the final image the ghost of his profile. It's really very weird.