Sigma Fp L adds long awaited phase-detect AF

[amweber21]

Another thing I noticed. Watching the dpreview on it he compares a couple of the video options. The internal is better than the a7R IV, but the cinemaDNG in 10bit look to have the detail level of the R5's 4K HQ. In 12bit it line skips again and softens. This may be a strange balance of crop vs bit depth to figure out.

[kye]

It looks quite interesting actually, if you're willing to shoot 1080p.

I think in 1080p, you get:

• Internal RAW in 8/10/12 bit
• Downscaling from 8K
• Digital cropping from 1.0 to 5.0 which are all downscaled from the 8K

That could mean you could make a tiny portable/pocketable setup, for example if you chose a 20mm lens you could use it like a 20-100mm.  Could be interesting for tiny covert setups.

I've been compiling a chart of DR tests and the FP seems like the cheapest and smallest way to get a high-DR camera.

[MrSMW]

I think ^this^ is the key to this camera.

You need 2 things:

A: To think out of the box a bit and not try and compare it with anything else and

B: Have a bit of a willful, left field spirit about you.

I like to think I am, more so than most, but at the same time, do play a little safe.

Just a few too many quirks and workarounds with this one for me.

[Chris Whitten]

Still loving my FP.

I don't get the advice to go Sony. I have no desire to switch to Sony.

I have been getting great results from my FP as a stills camera. Maybe you can't use it for action, but to say it is useless other than for landscapes is not correct at all. I have shot a lot of video on my FP - 10 bit CDNG to a small SSD. The quality is fantastic, very sharp, great colours.

Maybe the FP-L has a few too many quirks and compromises, but many of these same popular reviewers made the same criticisms of the OG FP. I think the FP is the best hybrid, capable of stunning photographs and excellent video. Autofocus, vlogging, pro-resets are not what it is about.

[TomTheDP]

6 hours ago, Chris Whitten said:

Still loving my FP.

I don't get the advice to go Sony. I have no desire to switch to Sony.

I have been getting great results from my FP as a stills camera. Maybe you can't use it for action, but to say it is useless other than for landscapes is not correct at all. I have shot a lot of video on my FP - 10 bit CDNG to a small SSD. The quality is fantastic, very sharp, great colours.

Maybe the FP-L has a few too many quirks and compromises, but many of these same popular reviewers made the same criticisms of the OG FP. I think the FP is the best hybrid, capable of stunning photographs and excellent video. Autofocus, vlogging, pro-resets are not what it is about.

I think the OG FP is a cool camera if you don't mind huge CDNG files. 

 

9 hours ago, kye said:

It looks quite interesting actually, if you're willing to shoot 1080p.

I think in 1080p, you get:

• Internal RAW in 8/10/12 bit
• Downscaling from 8K
• Digital cropping from 1.0 to 5.0 which are all downscaled from the 8K

That could mean you could make a tiny portable/pocketable setup, for example if you chose a 20mm lens you could use it like a 20-100mm.  Could be interesting for tiny covert setups.

I've been compiling a chart of DR tests and the FP seems like the cheapest and smallest way to get a high-DR camera.

Not sure how accurate that list is. The P6K has more dynamic range than the URSA 4.6k by a good bit. Also the S1H and S5 dynamic range aren't that far apart, maybe 1 stop at most. Same sensor, same codec, same log. 

[Andrew Reid]

On 3/26/2021 at 3:39 PM, NeuroJitsu said:

I think the tendency of youtuber reviews is to focus on specs and camera comparisons, rather than use cases and who/what the camera is intended for.  I guess it's easier, and controversy always gets the 'click'... I like Gerald's channel but this video missed the mark for me.

A lot of them are not even photographers.

Canadian guy in the DPReview video, forget his name, goes on about the EVF making it hard to grip. So add a grip then!

It's got 4 or 5 grips - I can list them off top of my head!! I have 3 different grips myself for the Fp and attach them at will.

Yet he... Chris or something... Is complaining that it's hard to hold a square box with no grip when you have a big EVF attached to the other side.

Not a photographer.

[amweber21]

7 minutes ago, Andrew Reid said:

A lot of them are not even photographers.

Canadian guy in the DPReview video, forget his name, goes on about the EVF making it hard to grip. So add a grip then!

It's got 4 or 5 grips - I can list them off top of my head!! I have 3 different grips myself for the Fp and attach them at will.

Yet he... Chris or something... Is complaining that it's hard to hold a square box with no grip when you have a big EVF attached to the other side.

Not a photographer.

It bothered me his whole segment. I'm sure they intentionally left it off just to complain about the handling. 

[Andrew Reid]

Yeah it is as if they thought, "well it comes like this so must be reviewed like this". What does he think the quarter inch thread is for? A beer holder? It takes about 10 seconds to attach the official Sigma grip. The box format is about the modular flexibility and they barely touch on it.

Would have been easy for Sigma just to do yet-another-clone of a Sony A7. Like Nikon are.

Props for having the guts and imagination to step outside the box.

Just shows up these YouTubers for how uncreative they are as artists that they can't even be bothered to attach a wooden hand grip.

[KnightsFan]

I have to agree with Gerald a bit actually. It's not a bad camera, but it's expensive considering the feature set. I almost exclusively use an EVF for photography, so with the cost of the (really cool and well designed) EVF, it's the price of an A7r4, and more than a typical S1r on ebay. I'd have to see tests in photo mode I suppose, but the slow readout and lack of mechanical shutter make it less than ideal for photos also. That's maybe not a deal breaker... but with the a7r4 at the exact same price I don't know why I would get an fp L.

Then on the video side, I wouldn't consider anything without 10 bit, log, or 4k60 anymore. Panasonic set the full frame expectation that high. Anything less, and it's purely a photography device. So it comes out to not really doing what I want for photos or videos, and not being less expensive either.

The other concern that Gerald mentioned was a lack of enough buttons and dials, which is clearly visible. You can add a grip, but you can't add dedicated wheels for ISO, shutter, and aperture.

On the positive side, though, the fp L sits in a unique space for being very small with very high MP. So if those are your concerns, then fantastic! I hope they find the audience for that. I guess I'm more the audience for the original fp since I don't need a 60 MP sensor. I also truly believe that Sigma is very innovative and will continue to make better products, so maybe by the time I can get back out in the real world, they'll have an fp 2 that will tick every box for me.

[amweber21]

If you wanted both photo/video, the a7rIV will do better on the photo side but worse on video. 8 bit 4K30p and equal rolling shutter. If the fp L didn't offer raw video it wouldn't be very attractive to me, but it does, and in a method I don't need any recorder to get. I feel this is their closest 50/50 camera. The original fp being mostly video, and their next camera (when they figure out the foveon) will be ONLY photo. 

On the customizing part, I read it has added the ability to set custom buttons: "Most of the fp L’s buttons and the rear dial can be customized for quick access to a variety of settings" Here's a picture of the menu. I'm not sure if we can finally add iso to the rear dial, but I know you can the crop zoom function. Maybe someone knows?

[KnightsFan]

1 hour ago, amweber21 said:

If you wanted both photo/video, the a7rIV will do better on the photo side but worse on video. 8 bit 4K30p and equal rolling shutter. If the fp L didn't offer raw video it wouldn't be very attractive to me, but it does, and in a method I don't need any recorder to get.

Fair point, but I would never shoot uncompressed raw video on anything so as far as I am concerned the fp is 8 bit only. Ridiculous data rates well beyond the point of diminishing returns. I used to shoot Raw on a 5D3, but back then it was HD so only 80 MB/s, and there were no photo/video hybrids with anything else decent. The recording options on an S1/S5/S1R or even an XT3/4 on the APS-C end are much preferred.

I'd welcome a photo only camera if it did it really well with a reasonable price, like a foveon with good ergonomics. I'd probably get one of their old foveon cameras just for fun if they used L mount actually.

[kye]

12 hours ago, TomTheDP said:

Not sure how accurate that list is. The P6K has more dynamic range than the URSA 4.6k by a good bit. Also the S1H and S5 dynamic range aren't that far apart, maybe 1 stop at most. Same sensor, same codec, same log. 

I went back and re-examined the two curves and it look like you're right, and that there are complications in the way that they tested the outputs from the UMP, to do with NR or something I think. Personally, I only built this table for my own purposes and the UMP isn't an option that I'd consider due to its size, so it's not a big deal for me..

Looking at the waveforms, it seems like the P6K has maybe 1-2 stops more DR than the UMP 4.6K G2.  

Regardless, I found the tests at C5D to be useful, and definitely better than the manufacturers (often heavily inflated) claims!

[amweber21]

I am curious the improvement in dynamic range of the fp L over the original. According to a review it was measured at 12.0, up from the 11.3 of the fp. This was in 12bit stills image, but I would think it should translate to the same in 12bit video. It is better than what the a7R IV is measured at and far cleaner at 12,800.

[noone]

On 3/27/2021 at 10:06 AM, amweber21 said:

 

Here's the official dynamic range for given iso values:

If that is correct, that is fantastic at higher ISOs.      8 plus stops at ISO 102400?   Seriously?    Even the original A7s only has around 7.5 stops of DR at that ISO.     Others A7Riv about 6.5 stops, A7iii just over 7 stops, Nikon Z6 and Panasonic S1 and Canon R6 all about 6.5 stops and those are the better cameras at that ISO.

I am happy to have that as the highest setting on my A7s but would prefer to go lower than about 80,000 when possible.

Am I missing something with that chart?

[Caleb Genheimer]

I’m kinda with Gerald on this one. If it was a grand or so cheaper, I think this would be a very different discussion, but as it sits, it’s in a pretty tough bracket, where (IMO) being a bit out of left field in size and shape don’t set it apart enough to ignore technical deficiencies versus the other options. 
 

I hope they’re still building this platform in 5 years, because at that juncture, I think we’ll be looking at a very different playing field, where even the “alternative” brand offerings will have more solid feature sets. I’m absolutely in love with their size/shape/styling. If either of these had an open gate video mode, I’d already own one just on principle. 
 

Blackmagic REALLY needs to take notice with their “pocket” cameras. That 6K Pro literally made me instantly say “oh.... oh no.” Because it made it abundantly clear that BM thinks their pocket body is still an appropriate design. 
 

I want to like the FP/L really bad, but I just can’t quite bring myself to jump on board when it’s knocking on the door of buying something like the S1H.

[TomTheDP]

11 hours ago, kye said:

I went back and re-examined the two curves and it look like you're right, and that there are complications in the way that they tested the outputs from the UMP, to do with NR or something I think. Personally, I only built this table for my own purposes and the UMP isn't an option that I'd consider due to its size, so it's not a big deal for me..

Looking at the waveforms, it seems like the P6K has maybe 1-2 stops more DR than the UMP 4.6K G2.  

Regardless, I found the tests at C5D to be useful, and definitely better than the manufacturers (often heavily inflated) claims!

Yeah I find their dynamic range readings helpful sometimes. I think where it fails is reading the shadows properly. The readings become inaccurate when noise reduction comes into play and of course how much noise is acceptable is also a subjective thing. This is where it gets really confusing as the URSA shadows really fall apart especially due to the fixed noise pattern and the URSA is also weaker in the highlights. So I have no clue why the reading for the P6K is so low compared to the URSA.

 The P6K does about 2 stops more in the highlights than the S1H or A7S3. The question is, are the Pocket shadows that much worse than the S1 or A7S3 to warrant such a low rating? I'd say the dynamic range on all three cameras are very similar. IMHO the only camera that really punches above the rest in terms of dynamic range is the C70. The S1, A7s3 and P6K are all great but also all in the same league or at least relatively close with the P6K and S1 in the lead by a little. 

I think the original FP is a compelling camera. CDNG is probably the best form of video recording(highest quality*) 12 bit CDNG Full frame is pretty compelling. The lack of full HDMI is a bummer though. I'd love to test the camera with an under and over test to see how it actually compares to the S1 or A7S3. I'd imagine it should be similar as they all use Sony sensors. But maybe Sigma did something weird to gimp it. 

[deezid]

On 3/26/2021 at 5:51 PM, Thomas Hill said:

Toward the end of the review, Huff tests the firmware 1.0 AF in video and it looks pretty good, better than Panny and far better than the original fp.

 

Holy crap, that's probably the most oversharpened video I have ever witnessed my entire life. 😳

[deezid]

1 hour ago, TomTheDP said:

IMHO the only camera that really punches above the rest in terms of dynamic range is the C70.

The S5 (S1H/S5 also) outperform it in usable latitude. Sure IMATEST loves heavily denoised footage like from the A7sIII or C70, but the IQ is quite bad on these cameras in comparison. The Pocket 6K (Pro), S5/S1H/S1 and Komodo are the best cameras in that regard under $10,000. 

Here's a nice exposure comparison chart from Slashcam:

[TomTheDP]

19 minutes ago, deezid said:

The S5 (S1H/S5 also) outperform it in usable latitude. Sure IMATEST loves heavily denoised footage like from the A7sIII or C70, but the IQ is quite bad on these cameras in comparison. The Pocket 6K (Pro), S5/S1H/S1 and Komodo are the best cameras in that regard under $10,000. 

Here's a nice exposure comparison chart from Slashcam:

I didn't know the C70 used a lot of noise reduction. The FX6 seems more organic.

[deezid]

9 minutes ago, TomTheDP said:

I didn't know the C70 used a lot of noise reduction. The FX6 seems more organic.

The FX6 is probably the better choice if you want it match it with a proper cine camera. This camera has a really nice looking image. 
The C70 looks quite mushy although not as bad as the Pocket 4K (only Braw) or A7sIII (ProRes Raw fixes it).