Slog must not be as good as they claim

Ok, I have a suggestion for you to test this- very unscientific, and you'll lose the 10bit compression, I assume. But it should work:

Timur, I think it was you who posted the car and the wall footage, if I'm not mistaken. Please post an uncompressed frame grab of the Slog (non graded) and Rec 709 (Ungraded) for download. Douglas, try to go into photoshop (or lightroom or Aperture) with both and play with them. You'll immediately see the difference.
 
Last edited:
The advantages of S-log are available in 422 as well, meaning it's not necessarily a $6000 recorder required. In 422, portable recorders including the Samurai, Pix 240, Hyperdeck Shuttle, Ki Pro, Ki Pro Mini - among others, can be used.

Not to mention some creative tethering options for laptops or Mac Pro towers that let you record 4:4:4 directly to the computer with an expansion card costing as little as $600 to $1000.

There are a lot of workflow options out there, depending on what you're doing.
 
Well I 'm coming to Douglas's defense here. I understand the advantages of S-Log and am sure it will be incredibly helpful in many contrasty situations but every example I've seen on the web has been under pretty flat lighting to begin with and non of the results have impressed me as much different than I could get with a cine gamma . In fact they often have looked milky even after grading so that I felt 709 might sometimes have looked better. So what I would love to see just for yucks would be some footage on a sunny day, back lit especially and compare graded s-log with cine 1 on front lit stuff or cine 4 back lit and even yes good old 709.

By the way Douglas - probably 6-8K for a 422 recorder which will be plenty good enough.
 
Douglas Villalba said:
That is only the upgrade. You still need a 444 recorder. $6,000
Click to expand...

1. you can, and most users have been, recording s-log to a 4:2:2 recorder, such as any of the models you can purchase for around $2k. You can also record s-log without a LUT straight to internal sxs 8-bit, if you don't have an external recorder and want to do some light grading. Someone did a very nice test with grading very low contrast logarithmic source footage in 8-bit space and the results are super more higher neater better looking than you'd expect.

2. this thread is redonk.
 
Lenilenapi said:
Well I 'm coming to Douglas's defense here. I understand the advantages of S-Log and am sure it will be incredibly helpful in many contrasty situations but every example I've seen on the web has been under pretty flat lighting to begin with and non of the results have impressed me as much different than I could get with a cine gamma . In fact they often have looked milky even after grading so that I felt 709 might sometimes have looked better. So what I would love to see just for yucks would be some footage on a sunny day, back lit especially and compare graded s-log with cine 1 on front lit stuff or cine 4 back lit and even yes good old 709.

By the way Douglas - probably 6-8K for a 422 recorder which will be plenty good enough.
Click to expand...

They look like flat lighting conditions because they are being recorded with 14 stops of DR. ANYTHING will look flat when you can see all 14 at the same time. The point being, yo ucan then choose in post where you want the whites to clip, blacks to go black and you can have it as contrasty or flat as you choose. Pretend its REDCODE RAW. Similar idea. You choose how you want the contrast, and color to look later.

I believe doug said the XDcam in the "First Slog Test" i shot seemed to have a disparity in exposure, that the XDcam was too bright. Well, thats be cause we did it intentionally, and sometimes over exposed the XDcam to see what we could recover form the log footage. The point is that its the SAME shot. Same exact moment in time. The XDcam was rolling to card with a LUT applied, and the sLog was shooting to a cinedeck. we were able to recover and rescue the shots with the LOG footage.

You have to understand what you are looking at first.

xdcam.jpg

slog.jpg


These are frame grabs to show you how much highlight rentention you gain. 2.5 stops.

The camera was recently rated by Abel to have 13.5 stops of Dynamic Range. That as much as 35mm Film, and about 95% as much as an Alexa, spitting out 444 Dual Link uncomrpessed, Razor sharp 1080p in Slog, with a price tag of $16,900.00

Understand you're getting formula one performance at hyundai prices. But it actually performs like a F1 car....... At hyundai prices......
 
Im chipping in here from my stills photography background so Im kind of inexperienced but also have been working 4k and raw images for a very long time

Both of those grabs look absolutely terrible milky and flat

I know the second one is unprocessed and ready go to work on

So I went to work on it in PS adding kick to the non roof area

Thats exactly what Id do with a still

Could anyone expand on the motion tracking post software that would be required to do this in video post

it seems to me without motion tracking selective post in place wide DR can do little but help one correct mis exposure

----

A second point - I say these images look terrible - is there another issue going on in terms of color space or something, because all of these images look flat on my 'computer' monitor

when I watch my 5d footage on a 'telly' it is far more saturated and punhcier

Are we therefore examining work corrected for broadcast outupts on the wrong monitor type /colour space and drawing the 'milky' conclusion unfairly?

Punch added to the image..

slog.jpg
You need to do post selectively treating the ceiling differently..
slog 2.jpg
 
I dont find this thread redonk or douglas as troll. I also see milky images and his thread opening asked for punch. I think he just asked for more punch and nobody is showing any because they dont like punchy ( which I understand:-) so could a kind soul take a heart and apply a crunchy punch to some slog screenshot to make it look like HVX cine gamma?
 
morgan_moore said:
Im chipping in here from my stills photography background so Im kind of inexperienced but also have been working 4k and raw images for a very long time

Both of those grabs look absolutely terrible milky and flat

I know the second one is unprocessed and ready go to work on

So I went to work on it in PS adding kick to the non roof area

Thats exactly what Id do with a still

Could anyone expand on the motion tracking post software that would be required to do this in video post

it seems to me without motion tracking selective post in place wide DR can do little but help one correct mis exposure

----

A second point - I say these images look terrible - is there another issue going on in terms of color space or something, because all of these images look flat on my 'computer' monitor

when I watch my 5d footage on a 'telly' it is far more saturated and punhcier

Are we therefore examining work corrected for broadcast outupts on the wrong monitor type /colour space and drawing the 'milky' conclusion unfairly?

Punch added to the image..

View attachment 38479
You need to do post selectively treating the ceiling differently..
View attachment 38480
Click to expand...

Exatcly!

If you could be so kind to work with an 1080p still Slog and post it here with the "punch" for Douglas to see, it would be better. The example you did was excelent, but the Jpeg compression is terrible and its hard to see a before and after.... Anyone with SLOG would be so kind to post a 1080p still with, for example, a landscape with blue skies and white clouds, or an image like the one Sam graded, also full res, for Sam to do a quick photoshop grade and compare side by side Slog vs rec 709?

Sam, you would need Apple Color, Da vinci Reslove or Magic Bullet Colorista II (or is it III?). There are more expensive options like Smoke and others but these are the more affordable and work well.
 
Timur,
I fully understand the concept of S-Log and why it would look milky before being graded, and have no doubt that it can provide an enormous amount of additional dynamic range . No one is questioning that.

However I also agree with Douglas that I haven't seen good examples of what S log can do posted anywhere, especially compared with cine gammas and even 709 gammas for various kinds of lighting , especially contrasty outdoor lighting. Without trying to bring up specific posts, the examples I've seen of exterior shooting have often been done under conditions that were not contrasty hence his saying "only someone in the cold weather in Europe and North America may like/" The slog made it flatter for sure, but they weren't contrasty situations to begin with so my gut reaction was it added very little. And fact most of those examples were already graded and graded pretty flat as well.

The only test I've really learned anything from was Alister Chapman's pans inside a trade show where he compared graded s-log in 8bit 4:2:0 to 709. The tests you did in Las vegas were heroic and valuable, but the 709 stuff was admittedly overexposed so it was not an accurate comparison, and again it was not shot under the conditions that IMHO most need more dynamic range - sunny exteriors.

I think all Douglas and I are saying (hope I'm not putting words in his mouth) is we'd like to see some tests with S-log that really show us the goods. Something that will make us say yikes, I just have drop another $7000 of my daughters college fund into this camera.

I know tests are a pain in the butt though and its easy for me to ask someone else to do it.
 
Macgregor just posted some nice stills that one can play with in photoshop! That's IMO enough to do a quick grade and see the differences...

Edit- Nevermind, did a quick grade in Aperture, simple S Curve on the right one, the left one messed with color temperature and tint with added blacks and contrast. Another example to follow.
 

Attachments

  • Image3 (1).jpg
    Image3 (1).jpg
    73.2 KB · Views: 0
  • Image3.jpg
    Image3.jpg
    69.4 KB · Views: 0
  • Image3.jpg
    Image3.jpg
    15.4 KB · Views: 0
Last edited:
ahhh :-) very convincing indeed! thanks a LOT Sergio! even though I am not a Sony fan, this slog upgrade does seem to make sense.
 
I shot this in the hot brooklyn sun - harsh harsh shadows on the models face. Harsh.
and i pulled something out of it.

http://vimeo.com/27422285

you can download the 720p quicktime thru the link.

this was a grade I did in only FCP using the 3 way color corrector so you probably still think it's milky looking and all that and that's fine. but i just shot today for the history channel in s log and will post the stills.
 
Douglas, I wasn't trying to be disrespectful but I do think there is an overall lack of comprehension of what S-Log does or doesn't do based on the questions asked (not just by you).

S-Log does not make the picture brighter.
S-Log does not make the picture clearer.
S-Log does not make the picture contrasty.
S-Log does not make the picture have better colors.

S-Log does give more ability to save detail information while adjusting color and contrast in post. How much color or contrast and what that accomplishes is up to the individual. Personally I don't like the super contrasty crushed blacks of the 5D which seems to be what was asked for above.

Asking if S-Log makes the picture better reminds me a lot of when people would ask me if my Nanoflash made the picture better. For ordinary talking heads type of images the Nano does not make the images better to the naked eye. They may be better technically better, but not noticeably better.

However, add enough motion to in camera footage and 8bit 4.2.0 images will fall apart. Then the higher bit rates and color data of the Nano will be very noticeable. It seemed to make a tremendous difference about 10% of the time.

If S-log increases the cameras capability from 12 stops to 13.5 stops and you are shooting with the entire scene lit within 12 stops you won't see much if any difference in the final product. I would find it hard to justify S-log in a studio environment. (That may just be me and my expectations.)

On the other hand if you are following someone with a pan across an open door or window S-log can mean the difference between a blown out shot, or not, because you can save it in post.

Is S-log required? No. People have been working around cameras with far less dynamic range than these for years. I would expect that overall (similar to the Nano) that it will make a tremendous difference about 10% of the time. However, if the shooting environment is all high contrast shots it will make all the difference in the world.

What it may mean is that you don't lose an hour or two as your crew repeatedly adjusts flags, screens, masking windows, etc each day. On a twenty day shoot you may save two whole days of shooting time. Do you want a reputation to get things done on time and under budget? That's what it will mean to many people.
 
Last edited:
Duke love your reply
the one thing though shooting at s log is you expose different - there is no pressure to hit your highlights as your want them. so you are more in a controlled frame of mind. you are not color correcting in camera per say. it puts the control of your exposure more in post and frees your mind up to think more about overall how the lighting will feel.
 
timbook2 said:
ahhh :-) very convincing indeed! thanks a LOT Sergio! even though I am not a Sony fan, this slog upgrade does seem to make sense.
Click to expand...

Not a Sony fan either, but with the Epic-S looking like a never ending delay story, this cam is my main contender for a purchase in the next 2 months. I need a camera by the end of October!
 
You must log in or register to reply here.
Back
Top