FS7: New Sony PXW-FX9 with venice features and 6k full frame

Grug said:
A couple of points to consider. Both cameras have been exposed for 400ISO, with the grey card used to set a consistent exposure across the test. And the C500ii has been exposed half a stop darker than the FX9 to achieve that.
Click to expand...

Curious where the iso 400 info comes from(hopefully not that 400 on the chart, which is the light level in foot/candles). All the other tests (lighting) were conducted at 800 iso (as this is the native iso of all the cameras) , but with the exposure up 1stop from normal..as the test was looking at the response to various lights, and it’s common to expose these cameras in this fashion as a compromise between noise and DR.

The fx9/c500 test is an over/under, and you would assume the test would be conducted at native iso— 800. Relative to exposure, both cameras were shot identically, however the fx9s control patch read in resolve to be 1/2 stop slower, so the zero point on that camera was adjusted 1/2 stop brighter. They’ve been rolling out data regarding the tests over the last couple of days, but other than lens and lighting info, I haven’t seen anything regarding camera settings for this particular test.
 
Barry G said:
Curious where the iso 400 info comes from(hopefully not that 400 on the chart, which is the light level in foot/candles). All the other tests (lighting) were conducted at 800 iso (as this is the native iso of all the cameras) , but with the exposure up 1stop from normal..as the test was looking at the response to various lights, and it’s common to expose these cameras in this fashion as a compromise between noise and DR.

The fx9/c500 test is an over/under, and you would assume the test would be conducted at native iso— 800. Relative to exposure, both cameras were shot identically, however the fx9s control patch read in resolve to be 1/2 stop slower, so the zero point on that camera was adjusted 1/2 stop brighter. They’ve been rolling out data regarding the tests over the last couple of days, but other than lens and lighting info, I haven’t seen anything regarding camera settings for this particular test.
Click to expand...

From Geoff's notes:

"Light level will be set to 100FC measured in the middle of the Kodak Grey Scale card using the Sekonic C800 pointing directly at the camera.

The ISO of all cameras will be set to 800, this is the base setting of all of the cameras except one which is 500. The aperture will be set to T5.6, this will result in all cameras being exposed to a stop more than the manufacturer’s recommendation, this is normally the best balance between Dynamic Range and noise. The film will be exposed at 1/3rd stop over the rated ISO with daylight stock the lens will be set to T4 rather than T5.6 as with all other tests."
 
Grug said:
From Geoff's notes:

"Light level will be set to 100FC measured in the middle of the Kodak Grey Scale card using the Sekonic C800 pointing directly at the camera.

The ISO of all cameras will be set to 800, this is the base setting of all of the cameras except one which is 500. The aperture will be set to T5.6, this will result in all cameras being exposed to a stop more than the manufacturer’s recommendation, this is normally the best balance between Dynamic Range and noise. The film will be exposed at 1/3rd stop over the rated ISO with daylight stock the lens will be set to T4 rather than T5.6 as with all other tests."
Click to expand...

Yes.. those notes are for the lighting tests.. not the over/under comparison.. (regardless the cameras are rated at 800, not 400 In the lighting tests (then overexposed— which while it seems like the same thing, isn’t —exposure is the same, but gain isn’t).

The notes Adam Wilt provided for the fx9/c500 stress test (over/under), do not include the iso, but describe the rest of the process.

In addition we shot exposure ramps on the Canon C500 Mk II and the Sony FX9, starting from 7 stops over to 6 stops under in half-stop steps, using iris, shutter angle, and (on the Canon) an ND 0.6 filter.


As they’ve taken pains to use the manufacturers presets and native ISO’s throughout the tests, I would not assume they would rate the cameras 1 stop higher for an ISO/DR test.
 
Barry G said:
Yes.. those notes are for the lighting tests.. not the over/under comparison.. (regardless the cameras are rated at 800, not 400 In the lighting tests (then overexposed— which while it seems like the same thing, isn’t —exposure is the same, but gain isn’t).

The notes Adam Wilt provided for the fx9/c500 stress test (over/under), do not include the iso, but describe the rest of the process.

In addition we shot exposure ramps on the Canon C500 Mk II and the Sony FX9, starting from 7 stops over to 6 stops under in half-stop steps, using iris, shutter angle, and (on the Canon) an ND 0.6 filter.


As they’ve taken pains to use the manufacturers presets and native ISO’s throughout the tests, I would not assume they would rate the cameras 1 stop higher for an ISO/DR test.
Click to expand...

Could well be, I'll shoot an email to Adam for some clarification.

I agree that it totally makes sense to run the under/over tests from the manufacturers' default settings. At the same time, it's odd to set a standard for a test (in this case exposing 1-stop over) and then discard it for other parts of the test.
 
Base ISO doesnt really mean anything in todays world of post.

The experienced operator only knows the camera.

They might say, this camera holds highlights well at 1600 but is noisy or this camera is clean but insensitive at 400 either could be called base.

Tests should be run from clipping and closing down iris (or increasing shutter)
 
Grug said:
...it's odd to set a standard for a test (in this case exposing 1-stop over) and then discard it for other parts of the test.
Click to expand...

It was a completely different test :-). (Different charts, different light levels, different lenses, different goals— one was testing lighting and color response across a range of lighting fixtures and cameras / the other was an exposure ramp / stress test of two cameras. (Or.. thinking about it another way, how exactly do you do a 13 stop exposure ramp “one stop over”?). Regardless, the question is whether they rated the cameras at 400. That would be inconsistent with the lighting test, as all those cameras were rated at 800, not 400.
 
New firmware announced for FX9 Version 2.0 features include:

180 fps full-frame HD recording
4K (4096×2160) DCI recording
Ability to load user 3D LUTs
HDR shooting function recorded in Hybrid Log Gamma
Unlocked touch screen
Eye AF

Still no job shooting but I'm tempted to go for it. Sony will probably continue to expand possibilities with future firmware releases.
 
Coming October:

- 16-bit RAW output (with XDCA-FX9)
- Full-frame 4K DCI RAW 60p/50p recording with oversampling (Image sensor scan size is around 5K)
- FHD 180fps shooting
- DCI 4K recording
- Touch Screen Focus control
- Eye Auto Focus
- Focus transition speed can be set to an Assignable Button
- Touch Screen control for Menu Status Screen
- User 3D Luts
- 6G-SDI support
- XLR-K3M digital support
- Higher selectable ISO (up to 102,400 ISO)
- Support for DWX digital wireless audio receiver (with XDCA-FX9)

https://www.newsshooter.com/2020/04/30/sony-fx9-version-2-0-firmware/

A lot of great stuff...I think adding touch screen AF is really nice.
 
NorBro said:
- 16-bit RAW output (with XDCA-FX9)
- Full-frame 4K DCI RAW 60p/50p recording with oversampling (Image sensor scan size is around 5K)
Click to expand...

F5/F55/F65 do their raw debayer in post. Does the FX9 do that after having downsampled the sensor to 4K? How is that possible? Is this linear raw? Will you need AXS R7 recorder ($6,980)? I asked too many questions, so edited it down.

I guess we'll find out in October.
 
Last edited:
I think providing full-frame 6K makes most sense, and I'm not sure why they are limiting it. Or what they exactly are doing for that matter as I don't have any information/knowledge about that.

I guess if the image looks really nice (and accomplished without external hardware) then most won't be too bothered or interested in how they are doing it.
 
RAW from FX-9 will be via SDI spigot on XDCA-FX9 to Atomos Shogun 7. The Shogun 7 will require firmware update, available simultaneously to v2.0 firmware. Also note it will be ProRes RAW "for the moment." Don't know what that means-- only repeating what Atomos told me. But until Adobe supports ProRes RAW, post is limited to FCP X.
 
These sound like great additions but is anything coming now or all due in October? (not that it matters alot for the time being.)
 
The initial messaging out of Sony and Atomos was that there would need to be a new product developed to handle the raw out of the FX9 (and extension unit, which is the only way to get raw out of this camera.) Atomos said it would be the Neon line because existing hardware wasn't capable of handling the data.

So Atomos seems to have found a way to manage the data with the Shogun 7 but it is limited. ProRes Raw is limited to 12 bit so the Shogun 7 is taking in a 16 bit signal from the extension unit and converting it into 12-bit ProRes Raw. The Shogun 7 is also limited to 4K 60p (not 120p). So I imagine Atomos will still be developing a Neon-line unit to pair with the FX9 and extension unit, but that will likely be released closer to October when the FX9 firmware is released.
 
It's such a shame Sony designed the FX9 well and then still persist on adding that ugly bulky piece of extension back. Yikes.
 
I don't own the extension for my FS7 but I've worked with fully loaded FS7 with the piggy back brick several times and it just takes away the compact design. If Sony can do raw out from the FS5 without adding bulk why o why not on the FX9?
 
I have had the extension back on both my FS7 and FS7 II. I found it helps balance the camera when attaching one of my big Anton cine batteries. I have played with RAW from the XDCA FS7 to the Odyssey when I owned it and honestly found the sheer bulk of CinemaDNG files did not make it worthwhile. I am resisting ProRes RAW to my Shogun Inferno (do not yet have the Inferno 7 and no point in spending money now) but -- flame me if you like-- I just intensely dislike FCP X and see it as an NLE of last resort when needing to put together something quick. I just never got into it and still cannot wrap my head around it. ProRes RAW will be more useful to me when Adobe supports ProRes RAW on the Mac. PC users can try the beta ProRes RAW plug in (amazing that support for an Apple CODEC is more quickly ported to Windows rather than MacOS). The XDCA FX9 is a bit more useful because of the wireless mic slot, which will be enabled with firmware 2.0. As far as appearance of camera, I actually like the appearance with the back a lot better than that "Aaton look" of the bare camera. Guess it's a matter of taste. All I want to do now, though, is to shoot something other than the squirrels digging up the few remaining bulbs that they didn't eat last fall. And as far as FX9 vs FS5, I think these are positioned for two different shooting situations. And the FS5 is later than the FS7, thus allowing more internal engineering to accomodate RAW without the back. FX9 development is based on the FS7 so I suspect to keep costs lower, Sony chose not to develop a new i/o circuitry. This is only a guess and I could be wrong, so don't quote me.
 
I just think Sony sees the FS7/FX9 more as broadcast/documentary/actuality cameras than they do the FS5 line and as such they require you to purchase the extension unit to unlock raw. I think in their minds they assume most people who purchase an FX9 aren't looking for raw whereas the market for the FS5 is indie shooters who want a small budget cinema camera that can punch above its weight class fairly easily.

The thing is that most people don't see it the way Sony sees it. Sure, most of us FS7/FX9 owner ops would use the XAVC-I codec way more than we would raw as the majority of us are shooting projects that don't really need raw. But what about those projects where we do? We are paying a premium price over the FS5 line, so why shouldn't we have access to raw as easily as that camera does? Just because we wouldn't use raw primarily doesn't mean we wouldn't ever use it.

Canon followed a similar approach for a while. They aimed the lower priced C200 at the indie cine crowd which is why they didn't give that camera a 10-bit file and instead only gave it raw internal along with a tiny .MP4 file. They figured the C300II was aimed more toward a similar market as the FS7 crowd, so that camera had the 10-bit file, but not raw internal. Very similar thinking.

The thing is, though, is that Canon has moved past that way of thinking in their latest generation of cameras. They finally accepted that today's shooters want cameras that are flexible and that can easily be used for broadcast/corporate/documentary work one day and cinema/drama/commercial work the next. They smartly gave their cameras the ability to play in both worlds by giving them a great internal raw option that's affordable to use as well as an easy to work with 10-bit file for daily work. This is what the market expects in 2020.

Sony, for whatever reason, just doesn't seem to want to go there. They've created a machine (FX9) that is a world class camera in many ways, but its one flaw is that its recording options haven't grown with the times. It's stuck at 4K 60, it doesn't offer raw internal and the XAVC-I files and bitrates are the same as they were when the FS7 came out in 2014. To me, Sony is really missing the boat here. If they were to have offered this camera with 6K 30p, 4K up to 120p and an internal raw option, it would be dang near perfect and would position itself better against the C500II and C300III.

I'm betting we will eventually see some changes in Sony's way of thinking as the market will force their hand. It's just how long will that take? I do think the FX6 will punch above its weight class again as I'm betting Sony will give it either internal raw or once again raw with just the addition of a recorder. This camera will have to have better internal options than the FS5 and for that to happen, it's going to step on the FX9's toes. The question is, just how much?
 
Last edited:
I predict that the FX6 will not have better internal codecs than the FS5. Sony doesn't want to step on the FX9's toes, and also a lot of those buyers may not care too much about the codec but be overjoyed by the AF and low-light performance.
 
Joshua Milligan said:
The thing is, though, is that Canon has moved past that way of thinking in their latest generation of cameras. They finally accepted that today's shooters want cameras that are flexible and that can easily be used for broadcast/corporate/documentary work one day and cinema/drama/commercial work the next. They smartly gave their cameras the ability to play in both worlds by giving them a great internal raw option that's affordable to use as well as an easy to work with 10-bit file for daily work. This is what the market expects in 2020.
Click to expand...

Canon has a big loss in market share. It seems they learned that their feature crippling wasn't helpful for the business. It was probably a learning on a hard way but who knows what's behind their change of strategy. The C300iii is the first Canon cinema camera I ordered.
 
You must log in or register to reply here.
Back
Top