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 :)

Build 21 - 4.5k Rc28/36/42

Shawn Nelson

Shawn Nelson

Well-known member
Joined
Dec 28, 2006
Messages
4,595
Reaction score
1
Points
0
Age
43
Location
Portland, OR
Jerry Turner and I set out tonight to test Build 21 and see if we could observe any differences in RC28/36/42.

The sun had gone down so we couldn't do trees blowing in the breeze, but we did go outside and do a high contrast scene, light shining on the trees, shadows, etc. See what you think, I'm making the raw 4k tiffs available below

http://www.nelsonentertainment.com/tests/Street_Build21_4-5k

Here are the pics. Note that 'RC48' should say 'RC42', mainly because I'm an idiot.

EDIT: Adding thoughts
Overall I was amazed by how little difference there was. Right away it's made me feel okay about shooting RC28 again (back in my day, before you other late number whipper snappers showed up, thats all we had, and by god we appreciated it! :-). Amazingly, I think noise looks better in RC28 as it's softer, look at the underexposed van rear on frame left, do a comparison.

I also did test indoors, on a female face closeup, outdoor whip panning trees, etc. Again, I noticed fine detail SLIGHTLY less sharp, it almost felt like 1/8 or 1/16 classic soft. In fact, and I havent done this yet, i'm willing to be that if I'd had a 1/4 or 1/2 classic soft in, the detail change would have been impossible to detect.

Only test remaining is outdoors in full sun with trees blowing in the wind.

Conclusion: If I have the medium and plenty of it, I'll go ahead and record RC42 just because it's there, or if I'm doing an outdoors widescren shot I'l definitely try to get it, but for most shots RC28 is great.
 

Attachments

  • Street_RC28.jpg
    Street_RC28.jpg
    42 KB · Views: 0
  • Street_RC36.jpg
    Street_RC36.jpg
    42.1 KB · Views: 0
  • Street_RC42.jpg
    Street_RC42.jpg
    42.1 KB · Views: 0
Thanks for putting these up, Shawn.


At 100% I didn't notice anything obvious; but zoomed in to 200% there's definitely a difference between them.
 
Shawn... after the test and seeing the results, what will you shoot in the future?

Thanks for doing the test.

Jim
 
Craig Ryan said:
Thanks for putting these up, Shawn.


At 100% I didn't notice anything obvious; but zoomed in to 200% there's definitely a difference between them.
Click to expand...

At that point you'd be seeing the influence of the scaler of whatever your image viewer is
 
Jannard said:
Shawn... after the test and seeing the results, what will you shoot in the future?

Thanks for doing the test.

Jim
Click to expand...

I did a bunch of tests, I just didnt want to host all the pics :-). I did this same RC 28/36/42 test in 4.5k 2.33:1 indoors with closeup on a face, indoors shooting through glass, outdoors whip panning against trees, extreme motion blur, etc.

The result? I was amazed by how little of a difference there is from RC28 and RC 42 (ignoring 36 even). In fact, in the underexposed areas I PREFER the RC28 because the noise is softer, less detailed.

Conclusion: In broad daylight, wide angle scenic shots I'll go for RC42, any other stuff i have no problem using RC28. A human face also looked better in RC28, the detail was SLIGHTLY less detailed, felt almost like a 1/8 Classic Soft.
 
Yes!! Mysterium is deeper and deeper and deeper.

Much more info in medium grays. We´re gonna update this week.

Thanks for post.
 
Oh Jim, something else, I am COMPLETELY in love with 2.33:1. Its weird that its only a <9% difference but it feels huge to me. I knew I'd love it in a nice scenic wide, but what surprised me was how much I love it in a closeup interior dialogue shot! I put the 85mm RPP on it and framed up my lovely wife and was amazed by how much more perfect the 85mm feels than before, because I'm still just as close vertically, but now the frame feels more open, not as claustrophobic and even more cinematic.
 
I noticed that too, its noticeable going from 28-42 only minor from 36-42
 
Absolutely no lighting changes at all, any observed differences is strictly a RC issue, we didnt even touch the focus.
 
500+ views and 7 people commented? Do the other 400+ of you think its a bad test? say so.

jeff, it should work fine, I was using my first gen reddrive, but test on the ground to be sure!
 
hey shawn, thanks for doing the test, I wouldn't know where to start testing the different bitrates, so I'm more interested in your commentary on the matter than offering my own questions/comments. Prolly a lot of other folks feeling the same.

I will offer this though. I'm very interested in your comments about RC28 being softer. I was afraid that RC28 would possibly appear harsher do to increased noise. Is it possible that the "new" noise from the new color science (NCS) works more attractively with the lower bitrate?
 
Thanks Brandon, added this to the first post

Adding thoughts
Overall I was amazed by how little difference there was. Right away it's made me feel okay about shooting RC28 again (back in my day, before you other late number whipper snappers showed up, thats all we had, and by god we appreciated it! :-). Amazingly, I think noise looks better in RC28 as it's softer, look at the underexposed van rear on frame left, do a comparison.

I also did test indoors, on a female face closeup, outdoor whip panning trees, etc. Again, I noticed fine detail SLIGHTLY less sharp, it almost felt like 1/8 or 1/16 classic soft. In fact, and I havent done this yet, i'm willing to be that if I'd had a 1/4 or 1/2 classic soft in, the detail change would have been impossible to detect.

Only test remaining is outdoors in full sun with trees blowing in the wind.

Conclusion: If I have the medium and plenty of it, I'll go ahead and record RC42 just because it's there, or if I'm doing an outdoors widescren shot I'l definitely try to get it, but for most shots RC28 is great.
 
cool, if that true, I'd probably PREFER to shoot everything RC28 :)
 
Shawn Nelson said:
Oh Jim, something else, I am COMPLETELY in love with 2.33:1. Its weird that its only a <9% difference but it feels huge to me. I knew I'd love it in a nice scenic wide, but what surprised me was how much I love it in a closeup interior dialogue shot! I put the 85mm RPP on it and framed up my lovely wife and was amazed by how much more perfect the 85mm feels than before, because I'm still just as close vertically, but now the frame feels more open, not as claustrophobic and even more cinematic.
Click to expand...

Yep- very true in deed. That is the way I felt the first time I threw on an anamorphic lens! 4.5k @ 2.33 is great. I've loving the fact that I do not have to crop any pixels away when shooting spherical.
 
Shawn, given a FTP and an account on my server, would you mind to upload 16bit TIFFS of all the tests you did?

I have troubles finding the subtleties in the JPEGs.
 
Jonas said:
Shawn, given a FTP and an account on my server, would you mind to upload 16bit TIFFS of all the tests you did?

I have troubles finding the subtleties in the JPEGs.
Click to expand...

Sure! I did upload a set of 3 TIFFs that are available in my first post, i dont expect anyone to be able to tell from JPEGs, those were only for quick reference
 
Shawn Nelson said:
In fact, in the underexposed areas I PREFER the RC28 because the noise is softer, less detailed.
Click to expand...
That's because it's acting like a low pass filter.

I'd still recommend pushing data rates as high as you'll tolerate because that image concern of your can easily be fixed in post.
 
You must log in or register to reply here.
Back
Top