FS7: FS7 Firmware v3 Format Tests

I did some tests with my FS7 and the version 3 firmware today. I recorded in every format the camera can output, including those with the XDCA unit (when paired with an Odyssey 7Q+). I recorded three files for each format: reflective white at 61%, +2 stops, and -2 stops. This was done by setting correct exposure with the internal ND wheel at 1/4, then recording with the ND wheel clear, at 1/4, and at 1/16.


There seems to be a number of people asking questions about one format vs another. Hopefully this will provide a useful tool to answering those questions.


The total file size is 2.4 GB. In an effort not to shut down my website’s server, I’ll try to distribute this with bittorrent. I’ve not used bittorrent before, but this seems to work within my office. However, it did take about 15 minutes to make a connection and begin downloading.

https://www.dropbox.com/s/lu0hplq56218dyf/Sony FS7 Firmware v3 Camera Tests.torrent?dl=1

My conclusion wasn't that there was no need for ProRes, just that in most situations you wouldn't see the difference between ProRes and XAVC-I. If you are doing closeup or tabletop work, ProRes gives you a ton more detail. Take this shot of a bouquet of flowers I shot today. When I should it in XAVC-I, most of the fine detail on the petals was lost. Only in ProRes am I able to get the high-detail look I'm trying to achieve. Note that this shot has a LUT and a little color shifting in Resolve, but no noise reduction or sharpening.

As someone who does not own the extension unit and does not have the opportunity to rent one easily, is there any comparison photos between the two?

Im rather content with XAVC I, but would love to see the exact difference so that i know what i can expect from certain situations.
 
As someone who does not own the extension unit and does not have the opportunity to rent one easily, is there any comparison photos between the two?

Im rather content with XAVC I, but would love to see the exact difference so that i know what i can expect from certain situations.

Hi Martin,

Just so you know, recording 4K ProRes doesn't require the extension back. Just an external recorder that can take a 4K HDMI feed and record it to ProRes, like the Odyssey 7Q+ (the 7Q+ can, the 7Q cannot).

I'd be happy to post some comparison images, but I'll be out of the office until Wednesday of next week. If no one else has posted some by then, I'll do so at that point.
 
Hi Martin,

Just so you know, recording 4K ProRes doesn't require the extension back. Just an external recorder that can take a 4K HDMI feed and record it to ProRes, like the Odyssey 7Q+ (the 7Q+ can, the 7Q cannot).

I'd be happy to post some comparison images, but I'll be out of the office until Wednesday of next week. If no one else has posted some by then, I'll do so at that point.

Aha, so thats enough to see an increase in sharpness in some images?

I just got myself an a7S. Might revisit the Convergent Design in the future then.
 
Hi Chance,

It's hard to tell based on the small images that you posted, but it looks like you haven't brought the levels up enough.
 
ChristopherA,

this thread is brilliantly timely for me. I have a shoot coming up which will include FS7 4k material; the client's specifications require ProRes 422HQ capture. Your HDMI to ProRes findings have greatly reassured me. Shame it has to be an HDMI connection - it'll be a week out in the middle of nowhere so I'll take plenty of spare leads :-), but great that it works, and avoids the 12-bit linear shenanigans of FS7 raw.

I can't load from torrents on the setup I'm at now, so haven't downloaded your images; can you confirm that the 4k HDMI signal is coming out at a full 10-bit depth? I'm planning to shoot log and hear suggestions of 4k HDMI being 8-bit, but noone directly refuting that (people answering on the Sony forum have tended to be working RAW to ProRes). If you are seeing 4k HDMI as the best picture, I'm hoping your answer will relieve me of my one remaining pang! Thanks,

Ben
 
Hi Cameramanben,

I'm not sure if there is a conclusive test to determine 8-bit vs 10-bit, but it does appear to be 10-bit. I'm certainly able to pull on the footage quite a bit with no visible banding. In case you have a better way, I've put just the 4K HDMI > ProRes clip on dropbox and a link is below. If you do have a way to determine 8-bit vs 10-bit video, I'd love to learn it.

https://www.dropbox.com/s/9ty92zdjq8fa3c8/4k HDMI Correct Exposure.mov?dl=1
 
I'm sure I don't need to tell you this, Ben, but it's worth pointing out that with Levels in Clip Attributes in Resolve set to Auto, Resolve will choose Data for S-Log3 XAVC-I and Video for ProRes. So you need to either use a different LUT depending on which format you have, or manually override the interpretation. That is why the Odyssey preset LC709A LUT is modified compared to the one which ships with Resolve. The Odyssey LUT is scaled so it works correctly with the way ProRes is interpreted by default, and the Resolve (Sony) LUT works with the way XAVC-I is interpreted by default.
 
Nick,

as a part-time dabbler and not a regular ProRes user, I din't know that specifically about Resolve. I do generally specify the Clip Attributes - it is what Sony recommends somewhere in their suggested workflows and, as I think we've discussed, the variability of levels in different bits of software and hardware gives me more headaches than anything else with LUTCalc! Good to know, so thanks for that. Does ProRes auto to Video levels in all circumstances?

My upcoming shoot is for a company which is Adobe-centric, which is a bit different again LUT-wise, and as there is some complicated post-production effects involved at 4k I think I'll steer clear of suggesting any looks. LUTs in the monitors for myself and the director, but don't add to post's pain once the rushes have gone off! ;-)

Ben
 
Does ProRes auto to Video levels in all circumstances?
Every ProRes Clip I've tested does, but I don't claim to have tested ProRes from every camera or recorder that can record it. That list has got a lot longer recently.

Recent updates of Resove have got better at setting the range based on metadata, rather that a single assumption about a given format. For example Sony XAVC-I with a LUT baked in should be interpreted as Video, and S-Log XAVC-I should generally be interpreted as Data. But the problem with ProRes is that it often comes from an external recorder and therefore does not usually include any metadata about the "gamma" of the recording. So in the absence of other specific information, assuming Video for ProRes is the safest bet. But that assumption is wrong if the ProRes is S-Log, and you want to apply a Sony LUT.
 
I know this topic can and will get convoluted and messy quickly...so to try and clarify - Nick are you talking specifically about shooting Sony Slog (without burned in LUT) as ProRes file to an Ody 7Q+ via HDMI? And then using Resolve as your specific software to apply a LUT to that ProRes footage?
 
I know this topic can and will get convoluted and messy quickly...so to try and clarify - Nick are you talking specifically about shooting Sony Slog (without burned in LUT) as ProRes file to an Ody 7Q+ via HDMI? And then using Resolve as your specific software to apply a LUT to that ProRes footage?
Kind of! It's not specific to HDMI and the 7Q+. If you recorded the HD SDI S-Log3 output of an FS7 as ProRes the situation would be the same. Other workflows may require other approaches.

I have not tested whether FS7 internal ProRes recorded using the XDCA-FS7 puts any metadata into the ProRes files which Resolve can read and interpret appropriately.
 
For example Sony XAVC-I with a LUT baked in should be interpreted as Video, and S-Log XAVC-I should generally be interpreted as Data.

Just as an aside to that, the nature of Sony's LUT scaling is that on the output 0=0%IRE, 1.0=10%IRE, but actually the internal MLUT system of the F5,F55 and FS7 copes with output values up to 1.09, ie you can actually bake-in extended range curves (or feed them out of the SDI).

So in the absence of other specific information, assuming Video for ProRes is the safest bet. But that assumption is wrong if the ProRes is S-Log, and you want to apply a Sony LUT.

Or S-Log2 or C-Log, which I gather is one of the reasons why log curves have been moving towards staying within legal range (LogC, S-Log3, V-Log etc) even if it seems to be wateful of the highest 80 or 100 10-bit values. Not so much just a ProRes thing as protecting the top stops from accidental amputation whatever the capture / post workflow.
 
Last edited:
Hi,

If one goes FS7 HDMI --> 7Q+, whether 4K, 2K, or 1080, what is highest frame rate you can record?

Thanks!
Directly linking the two:
DCI 4K24p
UHD30p
2K60p
1080p60

Using an intermediate box (AJA HA5-4K, not the XDCA extension):
DCI 4K60p
UHD60p
2K60p
1080p60
 
Last edited:
Bumping this thread because I've got a shoot in a few weeks that could use 240 fps and perhaps raw w/ the Ody 7Q...
Because of the hassles from the get-go of shooting raw w/ the FS7, I've stayed away. Just hoping Mitch and others here might shed any light on the V3 firmware "fix" to the raw signal coming out of the FS7 extension unit ?? Am I better off sticking with Slog3 and 180?
Thanks for any input (perhaps this deserves a new thread??)
 
Back
Top