Sage

14 minutes ago, Parker said:

Thanks Sage! What I do to simplify the workflow in FCP is to add the full stack of pre and post LUTs that I need, make sure they're all in the proper order just once, and then I simply save the whole thing as a custom effects preset, so that I can just drop my single "S1 Alexa 24p" effect onto a clip and it's 95% of the way there. Definitely makes it much easier to use, and I've also been able to share this effects preset file with my assistant editors when they're handling my footage, so that I don't even have to try and explain the order or have to make them read your explainer pdf. 

As for the anamorphics, all of the shots I shared above were with the aivascope 1.5x (amber flare version) and contax zeiss taking lenses.

That's a good idea. The Pdf can be a lot to take in for the uninitiated 🙂 With anamorphic, I've wanted to experiment with that for a long time, but have stayed spherical for the time being. I think there are some good complete m4/3 anamorphics now that aren't too expensive (relative to most).

Really nice Parker. What anamorphics do you use? I'm now working on a software that will integrate all steps (WB, Exp, ABL etc) to vastly simplify the workflow (and make other things possible)

Emotive Color V5 'SPECTRUM'

The V5 'Spectrum' update is ready. The V5 emails are going out now (let me know if you haven't received it). It is the culmination of four years of engine development, and I hope you enjoy!

Emotive Color Component ('EC709')

• EC709 component (Full/Theatrical) rebuilt with new engine, allowing accurate, uniform far gamut projection of Arri display table
• Arri Rec.2020 now used as the color target (projected within EC709 envelope); results in deeper saturated red and blue that are otherwise clipped in Arri's Rec.709
• High Scale is now Full Scale (unlabeled); a direct replacement for LogC+Rec709, with notable advantages (aesthetics and gamut preservation)
• New Classic 'Soft' and 'Hype' variations (like V3-era 'Soft' and 'Main')
• Refined natural saturation model

Core LogC

• Fully realized uniform far gamut projection from known lower saturation samples (solves issue that caused saturated orange to twist towards red)
• New rolloff math; allows grayscale reconciliation of high saturation brightness values (only available for interpolated .cubes; not achievable in Matrix)
• Added final, fourth calculation pass (post-smoothing), which allows absolute accuracy to known sample points in final render (while retaining smoothness)
• Corrected an issue that allowed smoothing algorithm to clash with a 'safety lock' on grayscale hue, that could result in red-green banding near the grayscale (especially with 33x under Trilinear Interpolation); now fully smooth regardless of grayscale

Misc.

• New Ceiling Modifier [POST] files; allows variable highlight compression
• New 'Natural' profile [PRE] for GH5 (shares settings with Cine-D: -5 Sat, 200 Iso)
• Remade all PREs (picture profiles / exposures)
• Revised Pdf; Important: DaVinci Resolve - Enable 'Tetrahedral Interpolation' under Project Settings > Color Management > Lookup Tables   (allows 65x performance with 33x cubes; complete gradient smoothness - FCPX has this as default)
  • Remade Premiere PREs with new measurements (to reflect 2021 Premiere's improved codec color); no 8-bit VLog or Cine-D PREs needed
  • Significant revision to Premiere Pdf guide; Advanced layering section is no longer relevant, use Lumetri's luma curve (improved significantly)
  • Gradient Test is now 16-bit png (compatible with Premiere) rather than tif
• Entire Matrix PowerGrade now printed directly from the engine (complete reverse-engineering of .drx format):
  • Precise values entered directly into the curve interfaces, at maximal point density
  • Improved engine gamma accuracy enables significantly improved base matrices
  • New LogC Exposures feature (true exposure gammas extracted from sample data)
  • New 4200K PowerGrade variant

If you have any videos or stills you've made that you can share, I'd love to see them!

-Alex

Working on it at this very moment (and a raft of updates simultaneously). Its a generational shift rolled in. I rewrote all code since the last 4.2 update; it's been a very technical winter 🙂

It's a good question; I don't recall there being the same flagging issue. I am in the midst of going back through all cameras for a new update, and I'll test this. The key is, does the LogC variant place black level at 8 Ire, or notably lower (should sit at 8 )

On 3/13/2021 at 11:15 PM, TomTheDP said:

@Sage

Should the luts work just as well for Prores Raw on the S5 and A7S3?

From what I've seen, ProRes Raw is slightly different. I'm planning to get a newer Raw recorder to see if a Pre is warranted, but I think it is usable.

For thoroughly tested LongGOP that maintains complete functionality with pristine imaging, I would highly recommend LPowell's FlowMotion v2. He really did top notch work there.

On 2/17/2021 at 7:10 PM, RCV said:

The legend Sage. I'm guessing with the new update for gen 5 coming to the bmpcc4k, we are covered by your last update!

It does have Gen 5 support - I will be revisiting all G5 profiles after release to be sure that they are in line with the latest from BMD (along with a few other adjustments)

Interesting; what comes to mind is that the latest files allow projection below zero. This only happens in theoretical space (i.e. below legal LogC / VLog), and projects sans clipping (an offset immediately after can bring sub-zero data into view).  This area shouldn't be encountered in real footage, so it may be the higher Isos which change the lower data range (S1a is optimized for VLog 640 Iso, 100 Cine-D, 400 HLG). Also, luma grade maneuvers in advance of the Pre/conversion is another way to bring this range into view.  I used to have an artificial rolloff to 0 black for theoretical black projection, but this caused problems downstream, so I removed that cap with the latest files. For noise, increased color separation will amplify noise, despite completely smooth tetra interpolation, if the footage has noise macroblocks. Higher bitrate codecs and lower Iso is optimal for clean imaging. Can you send me .tifs of the raw footage to noodle around with?

And to you all - here's to a healthy '21!

On 12/15/2020 at 5:28 PM, TomTheDP said:

Shot this with the S1 using emotive color LUT. 

Lol @ the ending

@Attila Bakos Very cool, I was on a 3770k; I switched to Ryzen last year for this very purpose. Now I'd like to swap out to one of the newest Ryzens, but they are sold out everywhere. I'm doing something atm wherein I throw enormous amounts of data in, and the renders can easily go upwards of an hour at 100% cpu. My data files are usually around 1Mb per each; these are going upwards of 50Mb atm

@zerocool22 CST is a versatile tool that allows whitepaper transforms between any two spaces. These are the manufacturer specified color spaces. Emotive is a measured conversion between two cameras (or profiles etc), such that it factors in sensor response to a light source and internal camera processing (of WB, profile, etc). Emotive has simulations of sensor LogC solely using Resolve color tools (sans CST) called Matrix. Both the interpolated conversions (with Tetrahedral interpolation enabled) and the Matrix LogC simulations are non-destructive, while Arri's Rec709 (that can be added to Matrix LogC) is partly destructive (of extreme Led gamut). CST can be either destructive or non-destructive, depending on the transform specified and relevant gamut.

CST + Arri 709:

Emotive Color

On 12/8/2020 at 3:46 AM, Attila Bakos said:

Yeah, it's pretty far from being accurate. I wonder how they do their conversions on the fly, and what precision they use. When I build my camera matching LUTs it takes 1-2 minutes with all cores used to the maximum to build a size 65 LUT, they do it in a second or so.

It must be pre-baked data sets of some sort ('assemble on site'). What Cpu are you rockin'?

On 12/8/2020 at 4:02 AM, Simon Young said:

@Sage Seems like I will be buying your LUT:s after all 🙂 With regards to Fuji, is there a pre to convert eterna footage like the one you've made for cine d?

Yes, with default settings, except set to DR400 (for photos), FLog is preferable for widest gamut

I gave it a spin a few weeks ago; sometimes it works decently, and sometimes less so. Here's a comparison of GH5 and Arri, as well as their gradient charts:

5 hours ago, seku said:

@Sage: i'm a happy user of your conversion for the GH5 and A7s3. I remember you talking about the releasing the same color science you used in A7s3 for the GH5 too, is it out yet?

thanks 🙂

Not yet - I've been on an experimental binge for the last month (not pertaining to the Alex conversions), and I've had a major breakthrough there. But that update will be releasing for sure!

Looks like a failure at the encryption stage (this rarely happens, a lot of heavy lifting on the back end). Download it once more and all the files should be present. Let me know how it goes!

16 hours ago, hypp said:

@Sage Have you thought about making Emotive Color for the new iPhones given that they record in 10-bit now? I understand that the footage is limited by other factors but I think some sort of Emotive Color even if it would be limited, would be really welcomed by iPhone users that use it for content creation. 

My sister may be getting one, so I'll give it a look. The ideal would be an app to record directly to LogC sans sharpening/processing etc. I've only dabbled in Xcode though. I've also got some neat experiments I am going to delve into this winter. 

1 hour ago, austinchimp said:

Yeah I think the skintone is much better with the S1a lut. It's from a hike near Benidorm in Spain.

It reminded me exactly of the Canary Islands

@austinchimp Flipping between them, it is quite a pleasing improvement! Where is that location?

@Emanuel Thank you! I'll take a close look. Up next, I intend to do some side experiments that I've been waiting forever to try out  : )

@jack jin Feel free to send an email with the dpxs

@Emanuel I was concerned with the news of the R5 overheating; I haven't followed the latest, but if viable (via firmware update perhaps) it could be done

@Victor With the new cameras, the side by sides with the Arri are added to the back of the Match Gallery; the other camera examples (especially using v4.2/3) are a very good indication of how it will work on the new (universal space)

@austinchimp Indeed, it was quite impressive. I believe he mostly used GHa Tungsten Soft (V3 era). Each new camera is a separate thing (that's where the work comes in), though updates are at no cost. What's your email?

Regarding Soft: its going to be returning in an update soon - in new form.

@Emanuel I had heard that it was to use the GH5S sensor. If so, there is a reasonable chance that the color will be the same. If not, then it could be measured. I didn't know about the Osmo, but it is something that could be explored. There were some cool experiments I wanted to mess around with, so a deep dive may be warranted there next.

@HockeyFan12 Very cool, I haven't read that one yet. I did notice S1 moire on saturated fabrics, and in that moment felt the value of the S1H. Here are the S1 fabrics compared to the Alexa:

S1

Alexa

@jack jin Indeed, if it was Day used for tungsten or artificial light, that would explain it. In his case, he has GHa-S,  so I think he is referring to the S1 variant

@austinchimp The way GHa deviates on the S1 feels oddly earthy and 'filmy'. S1a restores that Arri vibrant technicolor quality (1:1 to the other measured LogCs). With each new camera, I add side by sides to the Match gallery. I had the S1 on loan, but I'm considering the S1H, because its quite a good camera (Intra, 180 deg, Olpf are vital in for me)

This is a really cool video sent to me by Tom C, of GHa Density on the S5 that I was impressed with:

@mirekti Can you send me an email with dpx(s) of the raw? It should match GHa-S on the GH5S (assuming same optics/settings and neutral WB/reference light source)

@HockeyFan12 It makes sense now why they retired that one.. I suppose they calculated a matrix with a single triad and put it out there (film is quite non linear to boot)

On 10/18/2020 at 11:08 PM, HockeyFan12 said:

Arri also had a similar mode on the Alexa that they discontinued, it was used when the Alexa was to be intercut with film.

I found the Arri film matrix numbers, and used the Matrix drx printer code to inject them into DaVinci. I'm not sure what to make of it.. it is a lot more saturated for sure. Here are the numbers and the lut/powergrade:

1.271103    -0.284279   0.013176

-0.127165   1.436429   -0.309264

-0.129927  -0.510286  1.640214

Arri Film Matrix Download

On 10/19/2020 at 7:46 AM, Hicham said:

Glad to hear that your Dad is recovering !

And cool to hear that the S series conversion is finished. When do you plan to release this new S Series Lut Pack ?

It lives! The Sony a7s III conversion is now complete as well.