TomTheDP

On 11/12/2022 at 11:01 PM, kye said:

When I (eventually) get around to grading these, any objections to me posting the graded images in the thread?

No go right ahead 

On 11/13/2022 at 7:23 AM, OleB said:

Have played a little with the files. Seems that the white balance and exposure was different. But once that has been set approx. the same the fp files look astonishing close the the ones from the Alexa.

For the Alexa I have just chosen the ARRI Rec709 LUT built into DaVinci.

For the fp settings were as follows:

RAW settings: 

Blackmagin Design + Blackmagic Design Film

CST: In: DaVinci Wide Gamut & Blackmagic Design Film => Out: Arri Wide Gamut 3 & Arri LogC3

LUT: Same built in ARRI Rec709 LUT

 

The white balance was set the same on both cameras but yes they are a lot different in how they interpret it. The exposure looks a lot the same in REC709 but when switched to a log profile it looks a lot darker on the FP. 

Feel free to share stills of your grades!!

Resolution and dynamic range are lacking compared to the Nikon Z6 as well as low light performance. 

They give a nice image if the lighting is decent. Canon colors rock. 

7 hours ago, OleB said:

Tom, I would like to test the files this weekend. What have been the ISO settings of both cameras, expect that the Alexa was on ISO 800, the fp?

Cheers!

I believe both were at 640 iso 

9 hours ago, OleB said:

Thank you very much for your efforts, Tom! I am downloading the files right now and will check them asap. 🙂

3 hours ago, kye said:

I'm also downloading, but it'll take some time (both to download and to check them out!).

Thanks, this should be really interesting.

Do you have some thoughts after shooting it?  I presume you've had a bit of a look at how they compare?

I am excited to hear your guys thoughts. 

The white balance and tint was quite different on the two cameras. The Alexa has a bit more texture I would say, the sigma a little cleaner. 

@kye @OleB

Here are some comparisons. 

One with pure tungsten, one with a 5600k and RED light, one daylight shadows, and one daylight sun. Hopefully these are helpful. Worst part of this is uploading to the drive takes like 12 hours lol. 

https://drive.google.com/drive/folders/1qxyxV-fFrBnyAh66joyRlgc2J6Xwbn7J?usp=share_link

I actually did expose both cameras exactly the same as they looked similar at the same ISO. Outside shots were at F11-F16 and further adjusted with the shutter. Lenses were Meike 35mm on Alexa and Meike 50mm on the Sigma. 

28 minutes ago, kye said:

Yeah, I thought that 12-bit RAW sensor readout (and the corresponding 10bit log) was pretty much the standard on almost all cameras that have a log profile.  I thought that some cameras had a 14-bit readout but it wasn't that common.

Of course, that is just another feather in the cap for 5D with Magic Lantern - 14-bit RAW straight to the card.  Still a standout spec and stand-out image, 14 years later.

Yeah I know some shooters who still prefer the 5D over the likes of the Panasonic S1 or Pocket 4k. 

I am super happy with the performance of the Sigma FP which is only 12 bits. Maybe color depth beyond that isn't that important. Who knows 

Honestly 12 bit RAW isn't a RED or ARRI competitor. To me that is the difference between a lot of lower end cameras compared to Venice, ARRI or RED.

RED and Venice are doing 16 bit and the ARRI is 14 bit readout into a 12 bit log Prores or RAW. These other cameras are doing a 12 bit readout into 12 bit linear RAW. It isn't the same. Maybe I am being too spec obsessive though. If the image looks good you can't argue with it. 

47 minutes ago, OleB said:

That could be one approach yes. Depending how close the ISO ratings of both companies is it might give slightly different exposures though. Let us see 🙂

I'd probably do one exposing to what looks normal per camera, one exposing for highlights, and one where both are the exact same aperture/ND. 

I can say having used them together on jobs the Sigma is amazing in terms of image quality. The workflow is just a pain to where I wouldn't use it on most projects as a main camera. But it punches way above its price tag.

59 minutes ago, OleB said:

Tom to you have a Ninja V available as well? Because then you could upload the ARRI conversion LUT to the Ninja V to exactly monitor the camera. I would be very much interested in the result of ISO 800 from the fp against the Alexa. Theoretically they should be identical with the difference being the DR obviously. 

I could possibly borrow one. Would expose them at the same aperture? 

2 hours ago, OleB said:

I agree with everything you have said. What the fp does now using that ARRI LUT is extremely interesting, I have never seen a camera before where you get skin tones which are directly out of camera on the skin tone line of the vector scope unless you have set a totally wrong white balance. Usually that needs a lot more work. This point alone makes me very happy.

Working hard currently to get my hands on an Alexa for testing both cameras at ISO 800 side by side. Granted I will provide the results here once I have done so.

What I wanted to express again is my gratitude for all of your contributions on this topic. Without this forum I would never gotten to this point and I am extremely thankful for that.

I could do those tests for you next week and send you the files. 

The Sigma FP is really nice. Great color reproduction. It certainly doesn't have the dynamic range of an Alexa but that is ok. 

My biggest gripe with it is the micro HDMI which makes it scary for professional use. 

4 hours ago, newfoundmass said:

The thing I keep going back to, even after people talk about the benefits of being able to downscale it to get even better 4K, is that we really aren't even using 4K to its true potential and probably never will before it gets replaced by 8K. Even downscaling to get "better" 4K feels silly, because the difference is so negligible unless you're watching on a big enough screen. It feels weird to argue against higher resolutions, and I don't necessarily view it like that in my head, but this rush to 8K and beyond feels more like a way to charge us more for cameras than a genuine need or necessity. Give me a 4K camera that can film 120fps without any risk of overheating and with all the features we are genuinely asking for before trying to entice us with 8K that 98% of us don't need.

I mean it is clearly just a marketing push. They need a reason to market new cameras to people. This is especially true in the market that these lower end cameras are being sold to. Companies continue to gimp features and push new features that don't even function properly. 

$2500 for micro HDMI. 

RAW recording but we are gimped with Micro hdmi. I think I would rather go with the FX30 if I was to shoot RAW or use a monitor. 

Looks awesome man 

3 hours ago, A_Urquhart said:

While this is kinda old news and not strictly FX30 related but with the similarities between the FX3 and FX30......Greg Fraser is apparently shooting his next feature "entirely" on the FX3. Normally I would agree that what people are calling 'cinema cameras' really aren't cinema cameras but merely marketing BS and Greg does call it what it really is: a 'prosumer camera' but this is blurring the lines I suppose.

Go to 38:28 in

 

Interesting, you sometimes hear things like this but then see other cameras in the BTS. I remember seeing that this one film was shot with the RED Komodo and DZO optics and then seeing a RED Monstro with an ARRI Signature Zoom in some of the BTS. 

Sometimes it is true though. Steven Soderbergh shot "Let Them All Talk" entirely on the RED Komodo. It was a very indie film but it featured Meryl Streep, sooo. I think it was shot with all natural light.

I generally favor RED cameras over prosumer Sony stuff for cinema. I heard something about the FX30 being able to do 16 bit RAW capture from an Atomos recorder, that could definitely put it close to a RED Komodo. No global shutter but probably better high ISO performance. I would also think state of the art auto focus could be an interesting feature to have. 

I think high resolution is really more for applications like 3D or 360 cameras. 

Storage has gotten much cheaper and will continue to get cheaper. Cropping in post is useful, not terribly useful but also why not have the option. It is sort of like shooting in RAW, it has become a why not for a lot of people as it's so easy to do now vs 10 years ago. 

If 8k becomes cheap enough to shoot and process easily then most people will opt for it over 6k, 4k, or HD. 

Now as a delivery format I don't see much practical need for anything past 4k. As more and more people become tech savvy there will be less room for companies to push BS that makes no practical sense. OR maybe I am becoming too optimistic about humanity progressing lol. 

Pretty cool, sounds like downsampled 8k to 6k isn't that feasible yet. I still think 8k is overkill for both video and photo. 6k sensors are awesome. 

8 hours ago, The Dancing Babamef said:

the what off the where? I didn't know the A7S3 shoots also R3D 🤔

 

Compared to a RED shooting R3D

Interesting. Probably why Prores RAW off the A7S3 still looks weak compared to Redraw. 

3 minutes ago, FHDcrew said:

I wholeheartedly agree, and I wish I had done that with my original post.  My original point was that natively, when looking at the RAW files, downsampled 1080p doesn't look much softer than native 4k.  The difference is small enough not to matter unless viewed side-by-side.  The difference is small, and as a result one can use a tiny Atomos Ninja Star with a Nikon Z6 to gain 10 bit log without needing a large bulky external recorder/monitor, thus making the Z6/Star combination very compelling given the price.

I agree, it is a great setup. 

16 minutes ago, The Dancing Babamef said:

I was hyperbolizing it a bit but what I said is true to an extent. 4K to 1080p is what you do when you are playing the file locally but uploading to Youtube it's better to leave it as 4K and add a slight NR to it. 

 

Let's say you are recording in camera, downsampling to HD from 4k in camera, like the original poster did. Even though your file is natively HD I still think rendering it in 4k, which makes YouTube upload it in 4k will give you a better result. The footage itself won't look any sharper, but the extra 40mbps YouTube gives 4k makes a difference. 

I am not saying you disagree, just clarifying my thoughts/experiences with YouTube. 

my 2 cents 

14 hours ago, kye said:

When did I state that YT doesn't degrade the image capture?

Were you saying that downsampled 4k to HD uploaded to YouTube in HD will look better than a 1080p sensor shooting 1080p video uploaded to YouTube in HD?

I'd test that out but I don't have any working SD cards for my BM micro. 

1 hour ago, kye said:

My comments are about resolution, in the context of downsampling, not bitrate.

Yes, but if your comments about downsampling were true, then it wouldn't matter if God himself uploaded a video to YT, it would still be........

But, it's not, and therefore, it isn't.

Shallow DOF makes images look higher res, which in my opinion makes that example a poor one if you are saying YouTube 1080p compression doesn't degrade your initial capture. 

I haven't found the 600 to be too heavy but the 1200 is a tank. If you don't need the light you don't need it but try setting up 4x 300w lights right next to each other to try to imitate a single light source. 

300x lights seem to work for interview stuff even if its strongly backlit. Trying to light larger areas is a different story or trying to push light in from outside on a sunny day. 

6 hours ago, kye said:

There's a lot of opinions about how good YT is or isn't, but I'm not so sure.

Here's a video from ARRI that is "only" uploaded in 1080p, but just looks fantastic...

I own both the OG BMPCC and BMMCC cameras which are native 1080p sensors, and even shooting in RAW or Prores HQ and processing them in post, I still seriously struggle to get an image as detailed as the above, even though the above has been seriously compressed by YT.  The BM 1080p cameras have a slightly softer pixel-to-pixel transition, simply because they're not 1080p 4:4:4, whereas downsampling cameras are all going to be 1080p 4:4:4, and YT has enough quality at 1080p to show these differences.

With my GH5, the difference in resolution between the 4K mode that's downsampled from the 5K sensor and the 4K 1:1 mode is definitely noticeable, even though the 4K 1:1 is a very small crop and even if you adjust ISO, SS and aperture to create the cleanest and sharpest images possible.

Shallow depth of field with nice lighting and super sharp lenses is maybe not the best representation of the struggle of HD compression on YouTube. Throw more movement, shadows, longer DOF and it really starts to fall apart IMO, at least compared to 4k. 10 vs 50mbps is going to make a difference. 

I have also noticed some content such as big studio film trailers seem to have a different quality HD than the standard uploader. 

5 hours ago, FHDcrew said:

Did you mean downsampled 4k to HD is hard to distinguish from 4k?

Yes this is what I meant to say.