Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

For those who have used them: How does Red compare to the Genesis and CineAlta?

RED never claimed 4K, they say you get around 3.2 due to OLPF (which is necessary for any electronic imager). But other than Sony e.g. they are not electronically sharpening the image at all, that's up to the post guys. Maybe they didn't ?

Regards,

Uli
 
Yes, how was this Red footage processed? Rec709 Gamma? RedLog? Very important to get full range from the camera.
 
Nova, that you mention that RED has crushed blacks worries me a lot. We record below black, and never have I see crushed black unless intentionally done so in post processing. Unless a test is done by named people with a described methodology, no matter whether the result is good or bad, it's meaningless.

Graeme
 
re film out camera test in Budapest

re film out camera test in Budapest

Nova Invicta said:
Some DOPs in Hungary not associated to any camera company shot tests on Kodak film, Red, Genesis, D-20, F23 & F900R cameras in controlled conditions going from a room where the lights were progressively turned off to a panning shot to a girl sitting at a table lit from the left by a window & daylight.
The post was done by a very respected post house with one of the worlds best software experts for post.
Film looked the best, followed by Genesis, followed by the Arri D-20 etc. The Red camera which I was unable to find out which version of software it was supporting showed the least exposure latitude and had crushed blacks. The color chart at the beginning of the take was exposed well and the Red blacks were black unlike the F-23. I was disappointed that for a 4K camera it did not jump out as being superior to cameras that are 2K regardless of the price considering the bashing of other brands going on here.
You can be dismissive of these test but given its independence from manufacturers Red has a road to travel before the 4K claims can be justified.
So for those making rash claims without direct comparisons I would say do proper testing using people who dont have a particular axe to grind first.
Click to expand...

I was also just at the screening and I must say that using the methodology that was used for these so-called camera tests made it completely useless. Different lenses were used, the lighting changed dramatically between tests, the actresses missed their marks changing everything, even the focus was noticeably off in a few instances. In my opinion a camera test where such variables were not taken into account rendered (despite everyone who was involved best intentions and the truly excellent facilities) a waste of time and does a disservice more than anything as it leaves those not experienced enough with an inaccurate impression. Thank you to those who allowed such a screening to take place, especially those who donated their cameras and the post facility for the film out and 2k projection but I was not alone in wanting to see at least one well exposed fixed shot using the same lenses (at least in the cameras allowing full S35mm lenses) and then maybe we could have drawn some sort of conclusion about the different looks these cameras provide.
 
acehole said:
Thats the old grading trick! It would definitely be cool for AE to have redcine controls on the footage where you can adjust layers from raw..

Or does Scratch do this?
Click to expand...

Yes, SCRATCH does it now.

But on the DR point again.

Why, I see repeating posts about 8-9 stops DR again and again.
Gentlemen, just take your camera and take a shot of your room at 8AM or 4PM so you see part of window and doorway. Like in ISO test shot. Expose so histogram is wall to wall without clipping.
Then convert REDRAW in REDCINE to 16 bit TIFF. Load in AE. Start Color Finesse full interface in effects (under Synthetic Aperture). On the right lower corner you have color sampler and Match color RGB numeric readout. Switch the sampler to 16 bit resolution. Using color sampler pick the white spot that still has texture read the number.

31871 in my case
Pick the black spot that still has the texture and acceptable noise.
in my case 27
Divide 31871/27 = 1180

Well 10 stops is 1024. If you don’t get the 1000 in your test I am inviting you all to the French restaurant to explain you how to get this number around 1000. You pay.
Again and again I am able to get blacks below 30 with careful adjustment of the histogram near the blacks. Remember, if you miss left side of histogram just by 10 or 20 (4 bits in blacks is 16) then you lose 4 stops here.
 
Andrew,

You are technically correct. But, your results very likely have shadows deep in the noise floor of the RED One. And that is where the dynamic range of RED One becomes an interpretive dance.

How much noise in the blacks is a D.P. interested in aesthetically and technically? Zone II? Zone III? Under what lighting conditions is noise worse? Etc.

While the RED One technically delivers near 11 stops of dynamic range, a D.P. may only wish to use 9 stops in order to keep the resulting blacks cleaner. Thus he/she will lift shadows with fill light while protecting highlights until the scene fits within the desired range. Blacks may then be crushed in post to eliminate the noise altogether. (And vise-versa for film.)

This technique is nothing new and any D.P. worth their credentials is prepared to light a scene to suit the technical and aesthetic constraints of the recording medium, be it film, video, or a RAW camera like the RED One.

Best,

Michael
 
Andrew said:
Yes, SCRATCH does it now.

But on the DR point again.

Why, I see repeating posts about 8-9 stops DR again and again.
Gentlemen, just take your camera and take a shot of your room at 8AM or 4PM so you see part of window and doorway. Like in ISO test shot. Expose so histogram is wall to wall without clipping.
Then convert REDRAW in REDCINE to 16 bit TIFF. Load in AE. Start Color Finesse full interface in effects (under Synthetic Aperture). On the right lower corner you have color sampler and Match color RGB numeric readout. Switch the sampler to 16 bit resolution. Using color sampler pick the white spot that still has texture read the number.

31871 in my case
Pick the black spot that still has the texture and acceptable noise.
in my case 27
Divide 31871/27 = 1180

Well 10 stops is 1024. If you don’t get the 1000 in your test I am inviting you all to the French restaurant to explain you how to get this number around 1000. You pay.
Again and again I am able to get blacks below 30 with careful adjustment of the histogram near the blacks. Remember, if you miss left side of histogram just by 10 or 20 (4 bits in blacks is 16) then you lose 4 stops here.
Click to expand...

Andrew,

Does a studio at 12 o'clock with studio doors open.
A person standing with a backlight coming in from behind (created by reflector).
A "darker" side of the face, which had a fill (also bounce).
An exposed side of the face (here also bounced)

Framing was knees and up and it included parts of the studio doors (which were in shadow) and the outside with buildings and sky.

We had complete control of what all the different elements were in according to exposure.

It was then shot with increments of half stops on the iris scale loaded, through REDCINE log DPX 4K to Lustre.

Does that sound like an ok test for you? (this was only a part of the test of course).

I'm sorry Andrew but to me, 2 other DP's and the grader, the preference were around 8 stops of DR, anything else was not acceptable to us.

Fredrik Callinggard
 
Well, every manufacturer likes to mention photosites on the imager for those who are easily impressed by numbers, and Red could even talk about 4.5 K then…

But whatever system you have, true resolution (measued as MTF for example) is well below that, since any rasterized imager needs OLPF. Did you hear Arri mention resolution after filtering for the D20? Or Sony for the F23?

Numbers of photosites are pretty useless for evaluating the true quality of the overall system.

Have a look at the clips from Panavision. They explain the issues of the resolution through the optical/electronical system quite well, including the importance of having several MTF values to judge perceived overall sharpness. Even if they are bashing Bayer-sensors when they get to that part – and don't mention the possible issues with non co-sited stripe filtered imagers like their own, the other parts are quite interesting.

Regards,

Uli
 
Stephen Williams said:
Hi,

http://www.red.com/

4K RED ONE

Stephen
Click to expand...

Wait a minute 4K is the format 4096 pixels per horizontal line.
Anybody has problem with the fact that RED has 4096 pixels per line?
When we are talking about resolving power we are not talking about pixels we are talking about Line Pairs /mm RED and has bit more than 70 lp/mm on the sensor. (74X2X24mm = 3552 lines)

Now if we include the whole chain starting from the chart, lenses, sensor, compression, transcoding, to the final print after post, we are getting more then 50 lp/mm ( one line pair is one black line and one white line)
Sensor has bit more then 24mm horizontal size, so 24X50x2= 2400 easy recognizable lines across the horizontal line for the whole system. Show me a manufacture that delivers such tests?

So don’t mix the format nomenclature with the resolving power.
4K is format not 4K lines per width of picture.

Like 1920x1080 will never shows you 1920 lines along the horizontal line on your TV.
 
The only real issue that seems to be a downside is dynamic range. But there are SO many upsides....

Hopefully the sensor upgrade next year will significantly add to DR. In the meantime, you just have to shoot carefully and become a RAW/Post expert. :)
 
Andrew said:
Any links for lp/mm chart to final print test?
Click to expand...

I really don't know. But just because something isn't on the Internet doesn't mean it's not available. Call them.
 
Tom said:
The only real issue that seems to be a downside is dynamic range. But there are SO many upsides....

Hopefully the sensor upgrade next year will significantly add to DR. In the meantime, you just have to shoot carefully and become a RAW/Post expert. :)
Click to expand...


This I agree 100%
If you don't have experience with RAW just take a course or buy good book.
To get 10 stops DR from RED you have to know what you are doing.
I will train focus assistants to be RGB histogram/ISO/temperature watchers as well.

Though, after one month of playing with it, it is like driving a car, comes somehow automatic.
I guess we have ability to rewire our brains to different circumstances and enviroments.
 
mmost said:
I really don't know. But just because something isn't on the Internet doesn't mean it's not available. Call them.
Click to expand...

Call them? They will not return my call, I already have RED.

I read everything what they have on theirs website and did read many posts on many user groups and somehow nobody did end to end test for Delsa, you know “so many variables involved”
Well make the best case real life test and post the results , including the setup pictures, and good test description.

Maybe one day............

Beat this, end to end RED lp/mm test.

This is RED RAW compressed 12:1 or so and transcoded to TIFF -> transcoded to jpg, add 15 lp/mm just for that. The rest is OLPF and lenses




805_1208915809.jpg
 
laperal@info.com.ph said:
No comparison!

Tell you what... I would gladly trade your Red One for three (3) sets of my HVX-200.
Click to expand...

Sorry that Panasonic won't buy back your HVX-200 in 100% !!! You may sell HVX in 50% value that means 10 X HVX-200 to exhange RED Body on street deal.

Stew.
 
weirdcrew said:
Andrew,

Does a studio at 12 o'clock with studio doors open.
A person standing with a backlight coming in from behind (created by reflector).
A "darker" side of the face, which had a fill (also bounce).
An exposed side of the face (here also bounced)

Framing was knees and up and it included parts of the studio doors (which were in shadow) and the outside with buildings and sky.

We had complete control of what all the different elements were in according to exposure.

It was then shot with increments of half stops on the iris scale loaded, through REDCINE log DPX 4K to Lustre.

Does that sound like an ok test for you? (this was only a part of the test of course).

I'm sorry Andrew but to me, 2 other DP's and the grader, the preference were around 8 stops of DR, anything else was not acceptable to us.

Fredrik Callinggard
Click to expand...


Any single TIFF to look at?
Looks like I will have good meal in the French restaurant:-)
 
You must log in or register to reply here.
Back
Top