Not upgrading to Weapon

[Steve Conry]

I'm having trouble wrapping my head around the math on this upgrade. I spoke with a RED rep at NAB about it too, and after going through it with him, he was a little stumped too. He ended up just kind of smiling and shrugging his shoulders and then said, "yeah it might not make sense to upgrade to be honest..."

Here's what I'm seeing:

I have an Epic MX body, originally purchased for -- ~20k.

In order to move to Weapon I must upgrade to Dragon -- $12,500

I also must buy a new Mini-Mag system, presumably w/ at least 2 cards, reader, side module -- $6,600

Then I must buy the Weapon -- $34,500 (lowest spec model, with limitations--fps, warranty, etc.)

So not including my ~20k original purchase, I'd have to spend about $53,600 to upgrade to Weapon.

As part of the upgrade program, RED will then offer me $13,000 off this price, bringing the total down to $40,000 for the Weapon body with basic accessories.

---note---

If I had uprated to Dragon prior to seeing Weapon specs, I would have saved $3,000 on Dragon, and gotten another $2,500 credit toward Weapon, bringing the total to $34,500. Which is ironically the exact cost of buying a brand new Weapon outright. If i had decided to put a deposit down before nab was over, I would also get a $5,000 accessory credit--however, this would quickly be spent on accessories I essentially already have for Epic, but no longer work with Weapon. So that is kind of a wash, especially since it's no longer an offer...

---

All this said, what I am having troubling working out is why it is in my best interest to "upgrade", when it is basically the same cost as buying a new one, only you don't get the second camera. It seems that the traditional model of camera companies obsoleting their own products every couple years would actually be better in this case. While I don't subscribe to the idea that my Epic is suddenly obsolete simply because there is a new one, I do think this is an odd pricing structure/business model given RED's motto of "Obsolescence Obsolete". It seems RED has "obsoleted" their camera (and most of its accessories) just as any other company would--the only difference is they are asking customers to trade in their camera as a down payment on the new model, at a very low valuation. If I upgraded to Weapon, before getting into accessories I'd be into RED for 60k, and only have 1 camera. With most companies, 60k buys you two or more high-end cameras.

Until the Weapon is available in the flesh and proves its worth 35k+ to me, I just don't see a reason to commit to it. And for the life of me, I can't see any reason to "upgrade" at all. Personally, I'm leaning toward doubling down on MX and picking up another body w/ some cards and accessories, at an all time low in the near future--or going a different direction on my next camera. The MX is an awesome camera and having another would be really valuable to my team.

I am really confused though... Am I missing something?

 

[Jeff Kilgroe]

or you could save $100k and just educate your clients.

The Microsoft HUB displays are not 8k. The 55" is only 1080p and the 84" is UHD, plus they are really designed for collaborative design work - video quality is not that great. You may be confusing the windows 10 announcement for 8k support, but that is just support for Metro apps to scale properly to 8k, currently they are limited to 2560x1600. The HUBs will be on display here in SF next week at the Microsoft build event if you want to check them out.

And yet MS did announce an 8K version. I even know which LG panel it's based on and I've already seen the panel in action. It [the MS version, anyway] is not releasing with the 4K models, release date TBD. Not to mention the other 8K displays coming out. These are clients who have multi-display walls made from UHD panels. 4K projectors in their conference rooms, just because they can. They don't want me shooting HD because "it's 2015, that's for phones." I've been delivering 4K to these guys since before I could buy a 4K display myself.

Not disputing that 4K is going to be with us a long while or that 8K is going to take some real time to reach any sort of mass proliferation in the market. And I'll take all the resolution I can get right now because I could use it. 50MP ain't enough on some of my architectural projects and we're stitching images to get what we want on a lot of these corporate gigs. That is an F'n pain in the ass, especially if we need the shot to move. Now we're talking expensive and finicky multi-camera rigs. I'd take a single camera any day that can provided the resolution. A $100K 16K camera would pay for itself before the check to pay for it even clears our bank account.

 

[Jeff Kilgroe]

It has to do with physics and facts, not semantics.

Yes, it does have to do with facts. And the interpretation of the points made above are semantical. I can assure you I'm right. I can furnish the math if you'd like, or the physical images to prove it. Or you could go test yourself, it's not hard.

Actually no. Not true. The size of hole at 200MM at f4 is much larger than the hole at 24mm otherwise you could not take light readings without inputting the lens, because a long lens, in a way, is a form of cropping out of light gathering, so the hole is bigger to give you more light, to compensate for the fact that you are seeing a narrower slice of the light emitting world outside the camera when your lens is longer.

Bigger hole = thinner depth of field. This is why a 100mm lens at f2.8, when used with a doubler, effectively gives you a 200mm f4 lens, not a 200mm f2.8 lens.

If you think about it, otherwise your larger sensor claims would not make sense, because a smaller sensor is just a cropped sensor view, and if depth of field difference were equal, you could put a 2.8 16mm wide angle on lens a 16mm camera, and get as much depth of field differences as with a 50 mm 2.8 on an s35 sensor. This is not what happens.

Another example, on a EX1 or EX3, the lens goes to f1.8 - so how come I get almost no depth of field differentiation when I am zoomed all the way out? Because zoomed out, to give me the same field of view as 33 mm with s35 sensor size, I am actually at 6 mm on the lens, and the tiny 1/2 sensor is filming a crop of the center of the lens, and showing me a 35 mm field of view that way.

A 6mm lens has a tiny, tiny hole at f 1.8 - like the size of hole at f16 (I'm guessing, have not done exact math) one an actual, "normal" 35 mm.

This! ^^^^^^

This myth has come about because when you get farther from a subject, differences in depth of field require greater distance differences, and so it has confused people.

Exactly.

 

[Mike Nagel]

No, Jeff that is not correct. The focal length of the lens does not change the depth of field for any given field of view. For example a medium close up using a 50mm lens at F.4 will have the same depth of field as an MCU using a 200mm at F.4. The background will be more compressed but the depth of field is the same. The size of the sensor on the other hand does have an impact on the depth of field for a given field of view.

Kind regards,

Neal Norton

this is incorrect, focal length is absolutely a factor in DOF along side: COC, sensor size, subject distance and F-Stop...

here's an example for 3 focal lengths all with the same cam setup:

COC 0.3mm | sensor size w/ 1.17 crop factor to 35mm (Dragon 6K) | subject distance: 10m | (true) F-Stop: 1.8

DOF 25mm: 33.79m
DOF 50mm: 3.86m
DOF 100mm: 0.93m

 

[Anton Shavlik]

this is incorrect, focal length is absolutely a factor in DOF along side: COC, sensor size, subject distance and F-Stop...

here's an example for 3 focal lengths all with the same cam setup:

COC 0.3mm | sensor size w/ 1.17 crop factor to 35mm (Dragon 6K) | subject distance: 10m | (true) F-Stop: 1.8

DOF 25mm: 33.79m
DOF 50mm: 3.86m
DOF 100mm: 0.93m

Those aren't all medium close ups though. Depth of field changes with 4 variables, focal length, focal distance, aperture diameter, and circle of confusion.

When compensating for each other, focal distance and focal length changes will cancel out for the same subject being framed the same way.

Sensor sizes being larger or smaller will force you to compensate by either moving closer to your subject or using a longer focal length for the same framing of the subject (rather than both at the same time), both of which will decrease the depth of field.

Also, the 2k prores on the weapon will have a wider depth of field than the 6k raw, even though the focal length, focus distance, sensor size and aperture are all the same (obviously since it's exactly the same footage scaled to 2k).

 

[Mike Nagel]

Those aren't all medium close ups though. Depth of field changes with 4 variables, focal length, focal distance, aperture diameter, and circle of confusion.

Focal distance and focal length changes will cancel out for the same subject being framed the same way.

Sensor sizes being larger or smaller will force you to compensate by either moving closer to your subject or using a longer focal length for the same framing of the subject, both of which decrease the depth of field.

it doesn't matter if it is a WS, MCU, CU or ECU... DOF is DOF and it is calculated using the 5 parameters mentioned above.

if, like in ur statement, u are on purpose matching the FOV of different sensor sizes, ergo u are changing one parameter (in this case: the subject distance), then u are matching the DOF only because you adjusted parameters of the equation...

it is still 5 parameters in the equation, and focal length is one of them ;-)

 

[Anton Shavlik]

You can roughly assume that if you double your focal length you will need to double the distance to your subject for the same composition.

Actually, for example, a 200mm is 11x the magnification of a 20mm lens instead of 10x, so you can see the depth of field above isn't exact, but it's damn close.

 

[Jonas Klittmark]

Instead of engaging in this yelling contest I will simply quote from the Zeiss article, which is very interesting btw if anyone took the time to actually read it instead of asserting their misconceptions.

From page 5:
"You may sometimes come across those who hold the viewpoint that a longer focal length has a larger image-side depth of focus. That is not true, however, because the image-side depth of focus is only dependent on the f- number. This misconception comes from confusing the image-side depth of focus with the depth of the three-dimensional image."

There is one exception however, which becomes relevant when the focus approaches infinity:
"The fact that the depth of field is only dependent on the imaging scale regardless of the focal length no longer applies with very large distances. Even at closer taking distance, two photographs of an object will not be identical if they are taken with two different focal lengths, even if the depth of focus is practically identical. Besides the perspectives, the maximum blurriness of the distant background differs. It is lower for shorter focal lengths than for longer ones."

 

[Jonas Klittmark]

it is still 5 parameters in the equation, and focal length is one of them

Fair enough. But the point some of us are making is that given the same magnification grade of the object in focus, focal length does not matter. And yes, this obviously means moving the camera if you are changing focal lengths.

The original statement that started all this was "DOF is still primarily a function of focal length." This is clearly not a good summary of what determines DOF.

Practical scenario: say you need to get a 2-shot with both actors in focus and the frame should be filled with the two actors. Focal length is NOT a way for you to adjust DOF in this situation. It does not matter whether you shoot close with an 18mm or far away with a 180mm. If the actors are the same size in the frame, the DOF will be the same (although the perspective will not). The only way to change the DOF is to change aperture, or if possible change sensor size/COC such as on a RED where we can crop the sensor and shoot at 4K instead of 6K.

This is not to be confused with bokeh. The bokeh is affected by the focal length and a longer lens "magnifies the blurriness" of the background, which many confuse with having more shallow DOF. But bokeh is not the same as DOF, which is objectively measurable and not open to interpretation. Other than COC, which is actually sort of arbitrary and... oh well. At least the COC will only be decided once per camera and display format and not on a per shot basis.

 

[Mike Nagel]

Fair enough. But the point some of us are making is that given the same magnification grade of the object in focus, focal length does not matter. And yes, this obviously means moving the camera if you are changing focal lengths.

The original statement that started all this was "DOF is still primarily a function of focal length." This is clearly not a good summary of what determines DOF.

Practical scenario: say you need to get a 2-shot with both actors in focus and the frame should be filled with the two actors. Focal length is NOT a way for you to adjust DOF in this situation. It does not matter whether you shoot close with an 18mm or far away with a 180mm. If the actors are the same size in the frame, the DOF will be the same (although the perspective will not). The only way to change the DOF is to change aperture, or if possible change sensor size/COC such as on a RED where we can crop the sensor and shoot at 4K instead of 6K.

This is not to be confused with bokeh. The bokeh is affected by the focal length and a longer lens "magnifies the blurriness" of the background, which many confuse with having more shallow DOF. But bokeh is not the same as DOF, which is objectively measurable and not open to interpretation. Other than COC, which is actually sort of arbitrary and... oh well. At least the COC will only be decided once per camera and display format and not on a per shot basis.

Jonas,

I agree that is one possible scenario, but for many reasons a lot of times on set u can not freely move the camera...

here's an example - shoot with ONE camera:

COC and sensor szie are fixed as we're using one cam. u still have aperture, focal length and subject distance. Director or DP wants a razor thin DOF let's say (to prove a point) 10cm (0.1m). u are WFO @ 1.8 with all available lenses. U only got subject distance and focal length left as ur adjustable parameters.

now, most likely u can NOT use ur wide angle lenses because u would need to get mega close to the actor (to create 10cm of DOF) and u might then block light or cast shadows and/or because some lenses have a minimum focus distance so u won't get close enough to the subject to create that DOF with a wide angle lens (besides the extra lens distortion of wides vs. long lenses), so u are forced to use a higher focal length to create that DOF...

another scenario that is very common:

u just shot a MS of an actor using a 50mm, camera stays in exact same position as we have no time to move the camera, because now we quickly shoot a CU of the exact same actor and scene slapping on a 85/100/135mm... the longer focal length will give you the close up and also create the small DOF that you want to separate the subject from the background...

this scenario happens all the time and focal length is the only deciding factor here for the change in DOF.

 

[Mike Nagel]

I'm having trouble wrapping my head around the math on this upgrade. I spoke with a RED rep at NAB about it too, and after going through it with him, he was a little stumped too. He ended up just kind of smiling and shrugging his shoulders and then said, "yeah it might not make sense to upgrade to be honest..."

Here's what I'm seeing:

I have an Epic MX body, originally purchased for -- ~20k.

In order to move to Weapon I must upgrade to Dragon -- $12,500

I also must buy a new Mini-Mag system, presumably w/ at least 2 cards, reader, side module -- $6,600

Then I must buy the Weapon -- $34,500 (lowest spec model, with limitations--fps, warranty, etc.)

So not including my ~20k original purchase, I'd have to spend about $53,600 to upgrade to Weapon.

As part of the upgrade program, RED will then offer me $13,000 off this price, bringing the total down to $40,000 for the Weapon body with basic accessories.

---note---

If I had uprated to Dragon prior to seeing Weapon specs, I would have saved $3,000 on Dragon, and gotten another $2,500 credit toward Weapon, bringing the total to $34,500. Which is ironically the exact cost of buying a brand new Weapon outright. If i had decided to put a deposit down before nab was over, I would also get a $5,000 accessory credit--however, this would quickly be spent on accessories I essentially already have for Epic, but no longer work with Weapon. So that is kind of a wash, especially since it's no longer an offer...

---

All this said, what I am having troubling working out is why it is in my best interest to "upgrade", when it is basically the same cost as buying a new one, only you don't get the second camera. It seems that the traditional model of camera companies obsoleting their own products every couple years would actually be better in this case. While I don't subscribe to the idea that my Epic is suddenly obsolete simply because there is a new one, I do think this is an odd pricing structure/business model given RED's motto of "Obsolescence Obsolete". It seems RED has "obsoleted" their camera (and most of its accessories) just as any other company would--the only difference is they are asking customers to trade in their camera as a down payment on the new model, at a very low valuation. If I upgraded to Weapon, before getting into accessories I'd be into RED for 60k, and only have 1 camera. With most companies, 60k buys you two or more high-end cameras.

Until the Weapon is available in the flesh and proves its worth 35k+ to me, I just don't see a reason to commit to it. And for the life of me, I can't see any reason to "upgrade" at all. Personally, I'm leaning toward doubling down on MX and picking up another body w/ some cards and accessories, at an all time low in the near future--or going a different direction on my next camera. The MX is an awesome camera and having another would be really valuable to my team.

I am really confused though... Am I missing something?

the stunning part to me is that if somebody would have told me one year ago that RED will do a body upgrade for 2-3 times the usual cost, but the major feature of this upgrade is a format output and it's Prores, I would have not believed it... ;-)))) hahahaaaaaa

so we have a body upgrade, but the body on it's own is now less functional, there's no power-in, there's no audio in... but Prores, the feature that 99% of RED owners and shooters couldn't give a crap about, is NOT an optional module. It's a fixed feature. I don't know anybody that owns RED gear that wants to pay for Prores crap. All have been doing RAW, whether they liked it or not, for many years and for many of us R3dcode was a main reason to buy RED...

Modularity gone the wrong way. IMO.

Would have loved to see them improve the body so that it has basic, important connectors and functionality so that I can actually ditch some of the many accessories we need to use. And THEN obviously tackle image quality even more.

I could not care less about the additional weight reduction or the bit smaller form factor... Alexa is a fat pig and it gets used A LOT.

As an upgrade, it is really not attractive to me, but if u buy new - looking at the pricing - W is the obvious choice.

 

[Jonas Klittmark]

Jonas,

I agree that is one possible scenario, but for many reasons a lot of times on set u can not freely move the camera...

here's an example - shoot with ONE camera:

COC and sensor szie are fixed as we're using one cam. u still have aperture, focal length and subject distance. Director or DP wants a razor thin DOF let's say (to prove a point) 10cm (0.1m). u are WFO @ 1.8 with all available lenses. U only got subject distance and focal length left as ur adjustable parameters.

now, most likely u can NOT use ur wide angle lenses because u would need to get mega close to the actor (to create 10cm of DOF) and u might then block light or cast shadows and/or because some lenses have a minimum focus distance so u won't get close enough to the subject to create that DOF with a wide angle lens (besides the extra lens distortion of wides vs. long lenses), so u are forced to use a higher focal length to create that DOF...

another scenario that is very common:

u just shot a MS of an actor using a 50mm, camera stays in exact same position as we have no time to move the camera, because now we quickly shoot a CU of the exact same actor and scene slapping on a 85/100/135mm... the longer focal length will give you the close up and also create the small DOF that you want to separate the subject from the background...

this scenario happens all the time and focal length is the only deciding factor here for the change in DOF.

This is all true but it's sort of trivial in the sense that we all know an ECU will have more shallow DOF than a WS. It's not always actionable either because if we want to shoot an entire film with shallow DOF, it would be pretty lame to have it consist of all ECUs. Sensor size and aperture on the other hand can be used to manipulate DOF without relying on constricted framing and perspective. Which is why a lot of us are excited about the VV size RED sensor!

In either case, the main point that a lot of folks seem confused about is that it is not the focal length in and of itself that gives a shallow DOF, it's the increased magnification that gives the shallow DOF. A wide lens close up with the face the same size as on a long lens at "normal" distance would give the exact same DOF. (But obviously a very different perspective and bokeh.)

 

[Corey Robson]

I work in network series and am tied to delivering Prores masters so I was initially really looking forward to this upgrade for my Dragons. I was hoping to have Weapons to replace at least a couple of our Alexas for next season on "Arrow". It will not be happening though for a few reasons.

Every episode I shoot a lot of 90 degree shutter action at night, usually with dark wardrobe in dark locations. D.E.B. has been essential for cleaning up my shadows at 1600ISO or higher. I was disappointed to hear at NAB that D.E.B. will not likely ever be an option in camera on Weapon or Dragon. I also heard from the RedcineX engineer on the NAB floor that D.E.B. may be implemented for Red Rocket-X transcodes eventually but for the foreseeable future it's only an option for GPU/CPU processing. If I want to deliver the cleanest Prores shadows possible I'd still have to transcode with multiple GPUs only even if I did get Weapons. On a CW budget, we simply can't afford the time to transcode full debayer PR444 near set or afford the expense of having our dailies facility to do it for us.

Also I need 96-120FPS framerates from time to time so the magnesium Weapon is not on the menu as a solution for bypassing Prores transcoding. For me CF Weapons would be the only option worth getting and would make for a very hefty Visa bill. This is ultimately the upgrade killer for me.

That being said I do trust that the upgrade prices are what they are in order to keep the Red business model in the black. I for one haven't tried to bring my own 8K camera to market lately so who am I to say it's overpriced? I do love where Weapon is going in terms of features and ergonomics but for now I'll have to stick to my Dragons. Maybe next round.

(Phil, it was great to meet you finally!)

 

[David Litchfield]

I don't know if its been covered elsewhere, but are people not having to pay for the same sensor twice? You have to do the $9500 Dragon upgrade before you go to Weapon. But Red are not swapping the sensor over (as far as I'm aware). So part of the Weapon upgrade has to include paying for a second Dragon sensor?

 

[Marcello Altieri]

The excitement lost their vision and mine too lol

 

[luigivaltulini]

for now remain stationary in place.
I spent $ 50,000 for carbon about 1 year ago ... I have lost 60% in a short time ..
I want to update a Weapon, with all my heart, but $ 30,000 to upgrade now, they are still so many in such a short time! and 40000 for 8k even more

Technology advances, and everything has its price ....ll wait the next sensor!
maybe later this year to upgrade ...

 

[Guidofilippi]

I don't know if its been covered elsewhere, but are people not having to pay for the same sensor twice? You have to do the $9500 Dragon upgrade before you go to Weapon. But Red are not swapping the sensor over (as far as I'm aware). So part of the Weapon upgrade has to include paying for a second Dragon sensor?

Same happened with Red One MX to Epic MX...

 

[Guidofilippi]

Maybe a more wise move may have been to make a Red Epic 2, or Evo, or XT, or watever...
Same back plate (reinforced to eliminate adaptor plate), capable to use same accessories, plus all the new accessories, plus the cableless contacts for LCD and EVF, plus internal ProRes recording capability...
To make a new, very similar design, 5% smaller and 5% lighter, with all-new, increasingly expensive accessories, is a risky move... and looks like a change of strategies too.
You keep your Epic 1, or you buy Epic 2, you move to the new accessories little by little...
Arri makes several Alexas... probably helps to keep value of used gear too.
Just saying...

 

[Meryem Ersoz]

I guess I'm getting off at this stop, too. Hopping off the bus with a gorgeous pair of profitable cameras, one MX and one Dragon.

Thanks, RED! what a ride!

all that's left to do is to make a few more films.

 

[Elsie N]

I guess I'm getting off at this stop, too. Hopping off the bus with a gorgeous pair of profitable cameras, one MX and one Dragon.

Thanks, RED! what a ride!

all that's left to do is to make a few more films.

As an observer of REDuser and its members, I think it's a great loss if you do not continue along the upgrade path.

As the most engaged female on REDuser, you and the 8k V,V sensor would be a leader for your gender.