Savannah Miller

Firstly, the Alexa only does uncompressed raw.  Nothing else.
Secondly, there's no internal pixel binning, so you have to record full sensor resolution to get the full sensor size.  Not the same as 1080p prores.
Thirdly, and a big problem is RAW is not as simple as prores.  So at least at the VFX studio I work, we recieve the prores reels directly from editorial.  Then we get the EDL, and we generate the plates necessary for the shots, directly from the EDL and original camera prores with the necessary handles. This would likely be DPX format.  This is a 100% automated process.  We then composite with the necessary changes.  Any intermediate "precomps" would then be rendered as 32-bit linear EXR, and finals are exported back out as Arri-log DPX which are then converted back to Prores and delivered to the editorial/color.

With RAW it's not as nice because we can no longer deliver back in the same format that we recieve.  AND there are so many settings, especially with something like REDCODE that it's likely someone else now has to generate the dpx plates for us as we can no longer do that.  The ISSUE is they're not a VFX house and they likely do not have automated tools that do this, so that means extra time and money.  And then you might run into other problems slowdowns on the post side of things.  RAW is nice but it's not so simple just to use it as it causes extra hassle.  Prores 4:4:4 on the Arri Alexa makes everything work like a well-oiled machine.
Lastly, a lot of people believe that you have to have 4:4:4 in order to get accurate green screen keys, but what if I let you in on a secret.  I don't always KEY greenscreens.  You'll never hear this from anyone else, but if you take a greenscreen shot and despill it to where it's just a person over a now grey background, you can then difference the despill and the greenscreen to get a sort of fake alpha channel.  Multiply your background image over your despilled greenscreen plate and mask it by the result of that "fake" alpha channel and you get 100% perfect edges with no compositing.  And then you just adjust the white and black point of your bg image until it has the right level of contrast and color.

Firstly, the Alexa only does uncompressed raw.  Nothing else.
Secondly, there's no internal pixel binning, so you have to record full sensor resolution to get the full sensor size.  Not the same as 1080p prores.
Thirdly, and a big problem is RAW is not as simple as prores.  So at least at the VFX studio I work, we recieve the prores reels directly from editorial.  Then we get the EDL, and we generate the plates necessary for the shots, directly from the EDL and original camera prores with the necessary handles. This would likely be DPX format.  This is a 100% automated process.  We then composite with the necessary changes.  Any intermediate "precomps" would then be rendered as 32-bit linear EXR, and finals are exported back out as Arri-log DPX which are then converted back to Prores and delivered to the editorial/color.

With RAW it's not as nice because we can no longer deliver back in the same format that we recieve.  AND there are so many settings, especially with something like REDCODE that it's likely someone else now has to generate the dpx plates for us as we can no longer do that.  The ISSUE is they're not a VFX house and they likely do not have automated tools that do this, so that means extra time and money.  And then you might run into other problems slowdowns on the post side of things.  RAW is nice but it's not so simple just to use it as it causes extra hassle.  Prores 4:4:4 on the Arri Alexa makes everything work like a well-oiled machine.
Lastly, a lot of people believe that you have to have 4:4:4 in order to get accurate green screen keys, but what if I let you in on a secret.  I don't always KEY greenscreens.  You'll never hear this from anyone else, but if you take a greenscreen shot and despill it to where it's just a person over a now grey background, you can then difference the despill and the greenscreen to get a sort of fake alpha channel.  Multiply your background image over your despilled greenscreen plate and mask it by the result of that "fake" alpha channel and you get 100% perfect edges with no compositing.  And then you just adjust the white and black point of your bg image until it has the right level of contrast and color.

John Brawley said that and it's true. 90% of American TV shows seem to be shot on the Alexa in 1080p Prores 4:4:4.  Very few people do RAW or 4K, and if they do 4K it's usually Sony cameras shooting XAVC.  Only times you see raw are very high-end TV shows and RED shows for Netflix.  With now the Alexa LF, I'm sure a lot of shows will switch back to Arri.
It's not because of storage either because a lot of productions are fine with shooting even prores XQ, it's the extra costs of shooting RAW which people don't like.
I work in VFX where you occasionally will have shots where you push the Prores files harder than you ever would in grading and I generally see no issues.
Even when I do greenscreen stuff, the Alexa does 4:2:2 HQ during slow motion shots, and even then, those files are pretty robust too.

John Brawley said that and it's true. 90% of American TV shows seem to be shot on the Alexa in 1080p Prores 4:4:4.  Very few people do RAW or 4K, and if they do 4K it's usually Sony cameras shooting XAVC.  Only times you see raw are very high-end TV shows and RED shows for Netflix.  With now the Alexa LF, I'm sure a lot of shows will switch back to Arri.
It's not because of storage either because a lot of productions are fine with shooting even prores XQ, it's the extra costs of shooting RAW which people don't like.
I work in VFX where you occasionally will have shots where you push the Prores files harder than you ever would in grading and I generally see no issues.
Even when I do greenscreen stuff, the Alexa does 4:2:2 HQ during slow motion shots, and even then, those files are pretty robust too.

Rx100 doesn't have more tech, but it has a lot packed into a tiny body.  Look at the zoom range of that lens.  The internal ND filters, that pop up viewfinder.  Crazy stuff for a tiny camera.
There's a few things Blackmagic does not have that the Panny cameras do have, but largely I feel if Blackmagic had more experience building a smaller dslr/mirrorless form factor camera, then the Pocket would cost exactly the same AND have more features.  Blackmagic didn't add a waveform monitor because they said it's very computationally expensive, and probably not worth the tradeoff when they can add other features.  They probably also felt that anamorphic is seems pretty niche and if you really need it, they have other cams that can do it.
Rotating screen they do not have, but not sure how easy it is to build a 5" rotating screen.  They did take the easy route and use a display they know that people are happy with.
Autofocus is the main thing that Blackmagic doesn't have.  If they were able to add phase detect autofocus and used some of the newer Sony sensors, allowing higher ISO performance, they would steal more business from other manufacturers.  In many ways this cam at least fixes some of the gripes that people have about blackmagic's being either too big and heavy or poor ISO range.  The aggressive pricing of this camera and better ISO performance is a huge step in converting a lot of new people to Blackmagic.

Does panasonic sell higher volumes than Blackmagic?
The Rx100 series of cameras has tons of features packed into it, so I can see how development costs are high, but I do feel that the Panasonic cameras are just priced like everyone else, with higher pricing given to their better models.  Blackmagic just prices things however they want, so you really see how cheap it is to build cameras.  They priced this one at $1295, but if they could get it down to $1000, I'm almost certain they would.  It does sort of put into perspective what stuff really costs, because this camera is basically an entirely new body design with a new type of sensor as well, with tons of PRO features as well.  Blackmagic is the only company building cameras less than $2000 that license Prores inside of them.

Blackmagic always puts unusual lenses on their cameras.  The first pocket cameras at NAB had outrageously huge zoom lenses on it.
The veydras look very nice on the pocket camera.  A nice size relative to the camera.

You don't need shallow mount to use a focal reducer.
You just need a mount that provides enough room for the lenses necessary to perform focal reduction.  In the case of the Lucadapter, the lenses are in the tube that is inside the 4.6K EF mount, so you just screw a new one in.
In the case of most speedboosters it's just easier to start with a shorter mount and expand to a longer one, such as EF or PL

Also did you watch some of those CookeTV videos with Geoff Boyle?  About 6 months ago they posted a video where Geoff talks about using a certain companies camera that lacks an OLPF and testing it for a few months.  He doesn't mention said company by name but we all know who it is.  He says if he shoots certain patterns he can produce moire, just like with the herrinbone suit that was seen during the HBO camera test.
 
BUT after all of that he says it doesn't matter when you actually get the camera and shoot REAL jobs.  Sure you can take a Blackmagic camera, underexpose it 2 stops and find more FPN than a certain competitors camera, and yes, you can maybe find moire if you shoot certain patterns, but in the small chance that it does happen, i can easily be fixed by moving the camera a little bit.
 
The magenta issue was one of the only problems that was very "real" and I would define as a serious issue, but they fixed that almost completely with firmware updates.

Also did you watch some of those CookeTV videos with Geoff Boyle?  About 6 months ago they posted a video where Geoff talks about using a certain companies camera that lacks an OLPF and testing it for a few months.  He doesn't mention said company by name but we all know who it is.  He says if he shoots certain patterns he can produce moire, just like with the herrinbone suit that was seen during the HBO camera test.
 
BUT after all of that he says it doesn't matter when you actually get the camera and shoot REAL jobs.  Sure you can take a Blackmagic camera, underexpose it 2 stops and find more FPN than a certain competitors camera, and yes, you can maybe find moire if you shoot certain patterns, but in the small chance that it does happen, i can easily be fixed by moving the camera a little bit.
 
The magenta issue was one of the only problems that was very "real" and I would define as a serious issue, but they fixed that almost completely with firmware updates.

I think this was exactly the kind of users I was referring to in my earlier post.  There are different sensor designs, and each design has drawbacks and positives.  The Ursa Mini 4K was the CMOSIS CMV12000 and that sensor sucked.  This pocket 4K uses a back illuminated Starvis design so I doubt it will have issues with FPN like other Blackmagic sensors do.  And as far as "unusable" in anything but bright sunlight.  Are you sure you weren't just underexposing your images?
First gen blackmagic cameras had a bit more aliasing/moire then previous cameras and they did have slightly more fpn than newer models, but they didn't have huge issues.  And likewise, the 4.6K sensor as a whole is very good and doesn't have many issues either.
Testing is mostly  useful to calibrate and spot bad sensors, as not every sensor is perfectly made, and I assume you can streamline the process once you get better at making cameras.  The "issues" you claim plauged first generation Blackmagic cameras have nothing to do with testing but rather their choice of sensor, which you seem to not like.

I would guess only a little, but Blackmagic probably figured out how to streamline testing as they know how to make cameras by now.
I think whatever Blackmagic does to deliver their cameras at a low cost is fine by me because they at least seem to be very open about replacing defective cameras with a new one. 
And it's the same with computer hardware. If you look on newegg or similar sites all you see are bad reviews because no one who recieves hardware in good condition actually leaves a review.  If it works as expected, everyone knows from benchmarks and specs how a particular piece of computer hardware performs.  Likewise, no one is out making threads about how good their camera works, or how it arrived with no defects.
Blackmagic seems no worse than other brands it's just due to the amateur pricing of their products, you get a lot of idiots who don't know how to use the cameras properly and claim defects because they don't know how to expose.  There was a guy who compared a bunch of different cameras including Alexa Mini and a few others and he declared the 4.6K "unusable" due to magenta all over the images and the manufacturer Rec 709 lut looking bad.  Maybe if he had used the same kind of ND that is internal with the Alexa Mini and learned how to grade he wouldn't be saying these ridiculous things.

I would guess only a little, but Blackmagic probably figured out how to streamline testing as they know how to make cameras by now.
I think whatever Blackmagic does to deliver their cameras at a low cost is fine by me because they at least seem to be very open about replacing defective cameras with a new one. 
And it's the same with computer hardware. If you look on newegg or similar sites all you see are bad reviews because no one who recieves hardware in good condition actually leaves a review.  If it works as expected, everyone knows from benchmarks and specs how a particular piece of computer hardware performs.  Likewise, no one is out making threads about how good their camera works, or how it arrived with no defects.
Blackmagic seems no worse than other brands it's just due to the amateur pricing of their products, you get a lot of idiots who don't know how to use the cameras properly and claim defects because they don't know how to expose.  There was a guy who compared a bunch of different cameras including Alexa Mini and a few others and he declared the 4.6K "unusable" due to magenta all over the images and the manufacturer Rec 709 lut looking bad.  Maybe if he had used the same kind of ND that is internal with the Alexa Mini and learned how to grade he wouldn't be saying these ridiculous things.

Here's a Black Sun from an Alexa shoot of mine, and a nice OLPF flare from a DXL / RED Monstro on the latest Mission Impossible trailer.
JB

Out of likes. Agree on that. But the MōVI® M10 came out in 2013. So they had gimbals soon after the Canon C series stuff came out, and the Red Scarlet.

This is known as “the pattern”
Resident eps are are shot in 8 working days plus one day of second unit. The second unit happens concurrently with the first day of the second episode shoot. 
6 days are shot in studio and 2 day are out on location.
We average about 35 setups or “slates” per day. Have done as high as 60 on Resident. 
Many setups are three cameras. So that’s about 70-90 “shots” per day. 
Each camera shoots about 60-90 mins of footage per day but it’s not unusual to have 2 hours per camera (for action or slow motion heavy days)
i day we have three cameras but I really mean we have theee camera operators.
We have many more cameras that are pre-build for different roles. A, B and C are Alexa Mini. D camera is a full time Steadicam camera (Alexa XT for weight and mass).  It goes on down to letter q or something silly. 
There are three sets of Primo Primes and each camera has its own 11-1 as well. We also have a couple of other specialty lenses like a CP 50 macro, and I also use a lot of SLR Magic primes on the micro and APO Hyper primes on the Ursa Mini.
We shoot about 6-10 script pages per day.
I always say you can’t really go much faster than a setup every 15 mins.  That’s 4 per hour.  On a 12 hour workday it means we theoretically can do 48 setups.  It’s very hard to maintain that kind of pace in a day (and have it look good)
Sometimes you can get the shot turnover down by leapfrogging cameras. Have the A cameras do the first shot, prep the B camera to come right in after that setup is finished and the A camera pulls out.
There are 12 in the camera department not including me.  Ops, firsts and seconds x 3 plus two utilities and a loader.
Time is the thing we all struggle most with. Time to light, time to shoot, time to tweak.  
The pace of TV has a way of “honing” your choices and teaching you to react and trust your instincts. 
JB

The blackmagic is not THAT ugly, and likely when you hold it, all your opinions will change.  The Pocket 4K feels good in your hands.

The blackmagic is not THAT ugly, and likely when you hold it, all your opinions will change.  The Pocket 4K feels good in your hands.

The Red Raven is a micro 4/3 camera in 16x9 mode so it would be so much easier if it was a true Micro 4/3 mount, but that's likely not possible because the sensor is too far into the body.  Sticking with the DSMC2 design limits them with certain features they can/cannot add as well as cooling.  Blackmagic is much more flexible in building cameras that match the sensors.
RED probably gave the full 4.5K due to competition from the ursa mini 4.6K.

No.  The sensors in red cameras are sunk too far into the bodies for certain mounts (e-mount being one of them.)  And because of the distance the sensor is from the body opening, it's very hard to build a speedbooster.  The currently developed abt speedbooster for the RED cameras is ridiculously expensive because every speedbooster is built by hand from one of red's mounts (which are already expensive.)  And on top of that, the ABT speedbooster is not a huge sensor size increase like Metabones is, it's much smaller.  Imagine a speedbooster on the Raven, which is basically a micro 4/3 EF mount camera.  Not possible.
 
RED has promised to stick with the DSMC2 body design for another few years so that will also limit what they can do with their cameras.

What's interesting is Kinefinity fixed a lot of issues that RED has with their cameras.  From the beginning they always designed a shallow sensor so that you can mount many different lens options.  On top of that, their sensors can be speedboosted easily.
 
I'm 100% for a collaboration with Olympus to build a model with IBIS and Autofocus.  They are ahead of the game in those areas for micro 4/3.
If blackmagic built a camera with a sony sensor (like mavo 6k, DJI, etc.), shrunk the size down,  kept the same color science, and added autofocus, it would be a slam dunk against cameras like the c200 because of no market segmentation and professional features like prores.  I mention a sony sensor because it will have higher iso performance and lesser cooling requirements, leading to a smaller camera.
Blackmagic is the only brand building cameras without market segmentation and including every feature they can.  If they just add a few things to their cameras and shrink the size a bit, they can cause a huge shift in the industry as a lot of other camera manufacturers will lose sales.

The Red Raven is a micro 4/3 camera in 16x9 mode so it would be so much easier if it was a true Micro 4/3 mount, but that's likely not possible because the sensor is too far into the body.  Sticking with the DSMC2 design limits them with certain features they can/cannot add as well as cooling.  Blackmagic is much more flexible in building cameras that match the sensors.
RED probably gave the full 4.5K due to competition from the ursa mini 4.6K.

No.  The sensors in red cameras are sunk too far into the bodies for certain mounts (e-mount being one of them.)  And because of the distance the sensor is from the body opening, it's very hard to build a speedbooster.  The currently developed abt speedbooster for the RED cameras is ridiculously expensive because every speedbooster is built by hand from one of red's mounts (which are already expensive.)  And on top of that, the ABT speedbooster is not a huge sensor size increase like Metabones is, it's much smaller.  Imagine a speedbooster on the Raven, which is basically a micro 4/3 EF mount camera.  Not possible.
 
RED has promised to stick with the DSMC2 body design for another few years so that will also limit what they can do with their cameras.

What's interesting is Kinefinity fixed a lot of issues that RED has with their cameras.  From the beginning they always designed a shallow sensor so that you can mount many different lens options.  On top of that, their sensors can be speedboosted easily.
 
I'm 100% for a collaboration with Olympus to build a model with IBIS and Autofocus.  They are ahead of the game in those areas for micro 4/3.
If blackmagic built a camera with a sony sensor (like mavo 6k, DJI, etc.), shrunk the size down,  kept the same color science, and added autofocus, it would be a slam dunk against cameras like the c200 because of no market segmentation and professional features like prores.  I mention a sony sensor because it will have higher iso performance and lesser cooling requirements, leading to a smaller camera.
Blackmagic is the only brand building cameras without market segmentation and including every feature they can.  If they just add a few things to their cameras and shrink the size a bit, they can cause a huge shift in the industry as a lot of other camera manufacturers will lose sales.

No.  The sensors in red cameras are sunk too far into the bodies for certain mounts (e-mount being one of them.)  And because of the distance the sensor is from the body opening, it's very hard to build a speedbooster.  The currently developed abt speedbooster for the RED cameras is ridiculously expensive because every speedbooster is built by hand from one of red's mounts (which are already expensive.)  And on top of that, the ABT speedbooster is not a huge sensor size increase like Metabones is, it's much smaller.  Imagine a speedbooster on the Raven, which is basically a micro 4/3 EF mount camera.  Not possible.
 
RED has promised to stick with the DSMC2 body design for another few years so that will also limit what they can do with their cameras.