Sigma FP - EL zone

[@Another-Layer]

Thank you for your testing. What makes things confusing for many is the concept of the 18% grey moving up or down depending on iso used. It’s also counterintuitive. If you look at Sigma’s chart for dynamic range, at iso 100 there’s only 3 stops above 18% grey so it will clip at EL orange, +3 stops. It simply cannot capture the ranges above 3 stops so orange is the clipping point. However keeping it at iso 100, you can capture 9.4 stops below 18% grey and capture all the stops with natural gradients to black. Here’s the counterintuitive part, if you film puffy clouds at iso 100 and properly expose them, when you pixel peep, you’ll notice the gradients of grey are limited and don’t have a smooth gradient in the grey. Move the iso to 800 and properly expose (nd required) the same shot and pixel peep again you’ll see nice gradients in the grey. There’s 6.09 stops in the areas above 18% grey at iso 800. Moving the iso up for bright areas with gradients is super counterintuitive. Blackmagic does the same with their cameras. So study what your going to film, if much of what’s in your shot resides in the darker gradient (the green and blues in the EL chart) below 18% grey, use iso 100 or 200 and bring in lighting. If much is in the lighter gradients (in the yellow to reds in the EL chart) above 18% use iso 800 or above with nd filtering. I hope this explanation helps.

[@IonCosmina]

Hi, thank you for your explanation! It's very thorough and useful! We've had a few comments asking for outdoors footage and I'll add tests that illustrate what you're saying.

[@smallbrownfox594]

Spot on, and that’s how I use mine. Good explanation 👍🏼

[@jacqueslombart5242]

Hello and thank you for all these tests which allow us to identify the limits of our little FP that we love so much. Your conclusions make perfect sense!! The El Zone scale is evenly distributed (+6 and -6), so it only works correctly for ISOs with a similar distribution, like 800 and 6400. For me, it's perfect because I only use 800-1000 and 6400-8000, which therefore have a uniform distribution of dynamics. It's much nicer to compose the frame, because it puts the very crushed Rec 709 that the FP offers as a visualization "in the middle", it just allows you to see something in the viewfinder, especially in low light; because if you want to protect the highlights at 100 - 400 ISO, you have to pose very dark. To simplify the use of the FP, Sigma should have offered a base ISO of 800, because there are many users who abandon the FP because the Rec 709's vision of the screen is too dark when working at ISO100-400; and many users estimate exposure by eye and with the exposure indicator; and only at ISOs which give an equal distribution of dynamics there are not too many bad surprises!! Using false colours all the time, (unfortunately indispensable with the FP and its linear Raw DNG + Rec 709 crushed for visualization), is a complicating factor that many users try to circumvent; but often it leads to disappointments, because the Rec 709 display is only used for framing and not at all for estimating the correct exposure. And for that, Él Zone does not simplify the use any more, because it is also subject to specific exceptions and conditions of use, such as using only the sensitivities that provide a fair dynamic repartition. In short, we have not yet finished getting crazy to exploit the full potential of FP and its magnificent image; it always requires application and reflection to be rewarded!!! Thank you again for sharing your passion!! Just keep going, if your heart tells you!!

[@IonCosmina]

Hi! Thank you for your positive feedback and encouragement, it means a lot :) ! Very cool observations about the distribution of dynamic range in ISO ranges! I didn't think about the results I found in those terms and it helped me understand how the EL zone scale works better. I looked for a white paper or a patent for EL zone in order to see how it is processed from the sensor, but have yet to find it :). What happened with sigma's "standard" false colors is that now it shows the correct colors according to the RAW file, not the REC709 image from the monitor. I'm not sure if it was always accurate, or they corrected in this firmware update. I just tested it last week. But but there were a few comments in previous videos that were saying false color clips too soon. I found this reddit thread from 6 months ago where this guy did a few tests of his own and said the same thing - he looked at ISO 100 www.reddit.com/r/sigmafp/comments/wmr8ap/fp_false_colors_are_wrong_indicates_clipping_way/. I hope this makes sense, I'm almost done with a video with tests on Sigma's False colors, I'll post the link here when it's done. Another useful thing for exposing the FP/ FP L correctly is hooking it up to an Atomos Ninja V and setting the monitoring to "native". That shows a pretty good approximation of the RAW file.

[@Chrispy429]

Thank you for providing extremely informative and concise videos on the FP/FPL. There are way too many videos that are filled with unnecessary fancy intros and irrelevant data. Kudos from Seattle. Keep them coming. Much appreciated.

[@IonCosmina]

Thanks for watching and your feedback :) ! If you have FP L though, aparently it behaves differently, check @Anson & Co.'s comment below.

[@zezic]

We need EL Zone + 2 types of zebras over it which shows warning areas and blown highlights.

[@IonCosmina]

That's better way to do it than switching between false colors and EL zone ! I'd stick to the blown highlights zebra since hypothetically warning zones should be shown correctly in the EL zone. BUT, does the zebra show true clipping for all the ISOs???? :)) I never checked those

[@julioalvarez3788]

Wow. Good to know. Thanks for this.

[@nickgrant5800]

Why no log? - It’s the main reason most people aren’t using this camera. Seeing what you’re filming on screen is the most basic feature of any film camera.

[@kingpossie]

I’ll refer anyone who bashes this camera without proof to your channel. It’s a beast of a camera - but as I’m no pro I’ll stick to my EOS M with Magic Lantern (crop mood firmware March 2023) Lack of an EVF would maybe sway me to a secondhand S5. That addon EVF is pricey.

[@IonCosmina]

I wouldn’t spend on the evf. I would pay for any decent monitor/recorder instead.

[@sebastianquintanilladoming8807]

@@IonCosminalike an Blackmagic VA 12g ?

[@IonCosmina]

Yes! The VA 12g is very good for the fp.

[@nyambe]

Fantastic . This proves EL zone reveals the truth about the sensor. You are better of using 800 ISO outside. FP lower ISO are artificial. Never understood this concept until now! Thanks

[@IonCosmina]

you're very welcome! I actually found out about ISO 800 from other people that commented on sigma FP videos, so in this sense I'm lucky with a great, engaged community.

[@UriRa]

Filmed already with EL-Zone on the street and in the studio. Very comfortably! The distribution of brightness in the frame and the overall contrast of the frame are intuitively clear. Your test is exciting, but very similar to looking for a black cat in a very dark room.

[@IonCosmina]

Hi, thanks for your comment! I design these tests as "stress tests" so as to push the footage to the limits. I also agree with you about EL zone being a good tool in a "normal" environment - when your subject isn't exposed close to the clipping points and you don't care about clipping some highlights in the background. But I do think you make a good point about just "looking for a black cat in a very dark room". I will try to add footage of more normal settings as well.

[@UriRa]

@@IonCosmina, The translation is probably not very accurate, Russian is figurative. I meant that you are looking for something ghostly, something that does not exist, or something that does not affect the work process. To me, EL-ZONE, is more than an exposure meter and finding the point of going beyond the limits of the dynamic range, a primitive zebra would be enough for that. EL-ZONE is an artistic frame control tool, in my understanding, because it visually gives you the ability to shape the image in a light or dark key, control softness or contrast of the picture. By dividing the scale into warm and cool colors, relative to neutral gray, it is very easy for the operator to perceive everything and shape the image as he wishes. Your work looks like a formal attempt to find technical parameters in the relationship, which in practice you will not apply, most likely because the system is protected by a large margin. I don't know how to miss in mov, let alone raw..... Translated with www.DeepL.com/Translator (free version)

[@IonCosmina]

I do get your point. And I think you are right in what you say

[@UriRa]

@@IonCosmina, Thanks!)

[@IonCosmina]

:)

[@Adeloye1000]

Thank you for the work. It's sad that it seems like you're having to clarify on sigma's statements but here we are! Thanks again

[@IonCosmina]

Thanks for watching! It is a bit odd that they put so much thought into applying this feature and then just left these odd behaviours unnoticed or unclarified. The upside is that the false colors is working fine now! I didn't get to test it before last week when I filmed the footage for this clip but there were a lot of comments out there saying it clipped too soon. So I'm assuming they fixed it with this firmware update.

[@adeloyearts205]

@@IonCosmina hi again, can you explain what you mean by the false colours properly working now? Thanks

[@IonCosmina]

Hi, I mean it shows clipping (red) where it's supposed to according to the dynamic range chart for all ISOs. I just skimmed through the footage though, I think I'm right for all ISOs. I'll put out a video on this with test footage asap.

[@IonCosmina]

Sorry, I think my last answer was incomplete. I didn't test the false colors before this firmware update, but there were a few comments in previous videos that were saying false color clips too soon. I found this reddit thread where this guy did a few tests of his own and says the same thing - he looked at ISO 100 www.reddit.com/r/sigmafp/comments/wmr8ap/fp_false_colors_are_wrong_indicates_clipping_way/

[@Ansonandco]

Interesting to see that 100-800 do not show true clipping. But 800 does. There has been a lot of convo on the Facebook groups to shoot in ISO 800 and this may be another reason to support this argument

[@IonCosmina]

Thanks for the comment! I've noticed a few reddit threads as well saying the same thing about shooting ISO 800. I'm not sure if there's something special about ISO 800 (like the sensor captures the information better), or it's just the first ISO that clipps at +6.0 EV. For this test ISO1600 and 6400 "behaved" as well as 800. Also, do you know if the convos about ISO 800 were centered around false colors not showing the true clipping point fore ISO 100-800? It does now- for the whole ISO range that I looked at (100-6400). Sadly I didn't check false colors properly before this firmware update so I don't know if it was as screwed up as EL zone is now before the update or not.

[@Ansonandco]

@@IonCosmina I don’t know if it was false color. I know there was a white paper from Takuma that said to shoot ISO 800 for better highlight recovery, but that was before false color. That said, I just ran through ISOs in EL zone on my fp L and no iso had a true clipping point. None of the ISO levels showed white or salmon colors, even when pointed at the sun directly 😞

[@innstikk]

But EL ZONE does not show clipping at all. It shows the dynamic range of the image and noting more as far as I understand it.

[@innstikk]

​@@Ansonandco I think that just relates to shooting style. If one is used to a balanced highlight and darks latitude in other cameras, then 800 ISO makes things easier. Otherwise, you have to expose to protect the highlights which is probably less common in the industry. Just a though.

[@IonCosmina]

@innstikk It's true that the EL Zone doesn't show clipping. But I think what @Anson & Co. is saying that he already knows the clipping points from the dynamic range chart and EL zone isn't showing the colors that correspond to those clipping points. Cause he then says "None of the ISO levels showed white or salmon colors, even when pointed at the sun directly". So even if we're not bringing clipping points into discussion at all, if you point your camera at the sun (with no ND filter), at any ISO, the exposure level in the sun should be at least +5EV (which would mean EL zone should show the color salmon) or +6EV (EL zone should show white). What he's saying is that the EL zone from FP L shows less than that. So probably the sun there is red or yellow.

[@IliaKurochkin]

Eeeeee) i was waiting this)

[@IonCosmina]

:)) drop me a comment if you do a live with it, I'll add the link in my description. I've filmed footage of the "standard" false colors as well, I think I'll release this one first before going back to finish the CDNG vs BRAW vs ProResRAW

[@melbango1]

Thank you so much for this!!!!!👏🙏🤟

[@IonCosmina]

Thanks for watching !

[@Whaever_1981]

Doesn't EL Zone interpret IRE values and convert those into stops? When you're at ISO 100 and 200, there's only about 3 stops above middle grey and after that the IRE values will hit clipping point. There's no way for for the EL zone to know this. Combine it with Zebra and all should be good

[@IonCosmina]

Yep, I found combining zebra and EL zone in those situations to be the best way too

[@rearranger]

Test footage for ISO 200 was shot in 8bit. Maybe that's the reason for inconsistency.

[@IonCosmina]

thanks for the suggestion!

[@MojoPapiFPV]

Makes sense that it can't accurately represent over or underexposures accurately outside of the native ISO range. ISO below 800 are still recorded at 800 and then poorly processed in-camera to be a lower ISO equivalent... basically its terrible and should be avoided.

[@mmihain]

Nu pot sa cred ca am gasit acest canal, nu prea am auzit de utilizatori Sigma FP in Romania insa sunt bucuros ca aparent nu doar ca exista alti utilizatori online si nu ma mai simt asa marginalizat cat si ca este un canal cu foarte mult focus pe camera asta. O sa urmaresc cu drag restul de clipuri pt ca nu am camera de mult timp si inca o invat.

[@IonCosmina]

Îmi pare foarte bine să întâlnesc pasionați de fp. Sperăm să o punem la treabă mai des ca să povestim aici. Mulțumim frumos!

[@SuchetB]

Love EL Zone.

[@IonCosmina]

It's very cool Sigma decided to invest in it 😃

[@SuchetB]

@@IonCosmina I will definitely learn how to use it - Thank you for the amazing useful content you are posting, I appreciate it very much - As a beginner of EL ZONE on SIGMA fp, should I be aiming for everything to be in grey? And any highlights to be okay in a maximum of Orange +2 and shadows to be no less than green -1???

[@shotbydatboi_jaywona]

I have created a lut to load into your monitor instead of buying another monitor

[@VIPAH]

Great to know the EL Zone has been integrated into the Sigma firmware. Have you tried using a light meter to judge your F stops and precisely see if that matches the EL Zone? Like if you are getting F2.8 on your meter for skin tone, would the EL zone match it for +2 ? I've seen the Small HD Smart series monitors use EL zone, but I have Canon R5C with False colour and waveform integrated.

[@IonCosmina]

Hi! I measured all the test footage with a light meter, yes ☺️! If you look on the left hand side where it says "measured EV", that was with the lightmeter. Thanks for the comment, I should have mentioned that at the beginning of the video. It's cool that they put False Colour on Canon cameras as well!the last Canon we owned was the C100, which only had waveform. Does it show the clipping points well?

[@VIPAH]

​@@IonCosminaYes the Canon shows the clipping points in red. Are you going to use a Small HD monitor with EL zone integration as well? I just don't know how the EL zone will determine cameras with more than 14 stops of dynamic range. Such as the Red Komodo X (16+ stops) and Arri Alexa 35 (17 stops). I think there still needs to be future firmware updates to support different cine cameras.

[@innstikk]

Thank you for the testing. Not sure I follow what you found. Did you find that +3 EV in the EL ZONE isn't +3EV? As I can see from your photos (although too fast) the EL ZONE looks to show the correct exposure. It doesn't change the clipping point and I don't see that it should? To me the example from Sigma shows to be correct for ISO 100. So could you explain why you say EL ZONE is not correct? Ah I think now I get what you mean, if the sensor clips it will give the clipped EV and nothing higher?

[@IonCosmina]

Hi! Thanks for your comments ! "if the sensor clips it will give the clipped EV and nothing higher?" Yep, that's exactly it, for all the ISOs except ISO 200 - this one's EL zone stops at mustard yellow (+3 EV) which is the clipping point for ISO100. If your image isn't clipped, the EL zone works just fine! I think it's the best exposure tool out there for exposing the face correctly and comfortably.

[@innstikk]

@@IonCosmina That is assuming that the graph is correct. I would trust EL ZONE more, but I have not tested it yet.