F35: Los Angeles 1991. 10mins. F35, just released online

I haven't chimed in for months but I had to add my little 2 cents ...F35 is DEFINITELY SHARP...Maybe it was the lenses being used. All my images were always sharp
 
I haven't chimed in for months but I had to add my little 2 cents ...F35 is DEFINITELY SHARP...Maybe it was the lenses being used. All my images were always sharp. Btw, Macgregor you and your team MURDERED LA91 Bro...Cant stop watching it.
 
This isn't about sharpness, this is about smoothness.

The F35 was a great camera back in 2008, the best, but it doesn't have the smoothness around edges as an Alexa, Red, or F65 that are the generation that exist now.

Here's a test that shows this. I digitally enlarged the images 50% more than the 200% enlargement that existed. And this is pulling off someone's vimeo h.264 clip

amira_150 copy.jpgf35_350 copy.jpg

first is Amira, second is F35. The Amira is actually already magnified more than the f35, but shows less anti-aliazing.

Look at the glasses.

The F35 has elements of breaking, of not being smooth in how the glasses connect.
The Amira does not.

That to me marks a digital camera. When you can see digital artifacts that clearly.
 
Last edited:
MeesePiece said:
This isn't about sharpness, this is about smoothness.

The F35 was a great camera back in 2008, the best, but it doesn't have the smoothness around edges as an Alexa, Red, or F65 that are the generation that exist now.

Here's a test that shows this. I digitally enlarged the images 50% more than the 200% enlargement that existed. And this is pulling off someone's vimeo h.264 clip
.
Click to expand...

I don't have a horse in this race and read with interest. But seriously?

- You take an H264 clip (unknown compression settings and certainly not the same between).
- The source format is unknown, RAW? 422? 444? This will affect everything.
- You scale it, not using nearest neighbour, but some scaling algorithm that again isn't showing the source but adding softness and artefacts.
- You compress it and pop it only embedded so it's JPEG compressed.

I can't argue about the cameras, but i can point out your comparison is flawed and you can't make a point using this approach.

If you're going to compare, go back to the sources, take an uncompressed frame. scale using nearest neighbour then put in a uncompressed format and a link to it. Then that's a platform which you can judge sharpness. But don't forget to ensure that the DP has nailed PERFECT focus on both frames, especially if the ear is out of focus because the aperture is WFO. And the talent is stuck to a chair and is immobile. Tell them to not breath too.

cheers
Paul
 
Sure . From a CML test. First three images are at 400% zoom in from a DPX file in resolve. Next set of images after that is 300% zoom.
Save the images to see them larger, to really see the difference.

400_redone1.4.1.jpg400_d211.6.1.jpg400_f351.5.1.jpg_d21_300.jpg_f35_300.jpg_redm_300.jpg

Look at the hair. Her face. The Red One it is smooth. Smooth and round.
The F35 it's breaking up. The Arri D-21 is in my eyes, the middle. THe F35 is not as smooth or round of an image.

This is from a CML test - DPX files
http://www.cinematography.net/Hannover/Hannover contrast tests.html

I think smooth gets us closest to the inherent resolution of celluloid, my gold standard.
Without the smoothness, it feels digital.

Dynamic range, highlight handling, skintones - these are also give-aways of digital.
 
Last edited:
MeesePiece said:
But seriously Paul.
I don't have time nor do I have the source files, nor can I ask the person again to do the test.
If you want to do a test, fine, go for it.

Be my guest.
I don't have the money nor access nor time right now to a F35 and an Amira.
This is as close as I can get.
Maybe someone can pull a still from CML DPX tests - but they don't have the F35 probably in any comparison setting.
This was the best I could do.

No data, no attempt to have data or comparisons.
That's great.
Just let SJX pull up some screengrabs from some blu-ray movie site and say, "hey look how good the F35 is and how bad the F65 is" -
that's just great scientific research!
Click to expand...

We're all busy. My point is simply that you cannot use that method to make a definitive point because the methodology is fundementally flawed.
Say that you 'feel and that you believe' the F35 is soft, but you can't make a point with broken data.

You're clever and experienced enough to know that there's practically a limitless number of things that can affect the perceived sharpness of an image. Only in some artificial testing environment could you make a quantitive statement. In fact end to end from set to screen there's such an enormous number of processing that can effect each frame that these kinds of comparisons are rocky at best.

For example, you mentioned in your other thread using film convert set to Alexa for your F65 samples (which look wonderful). Well what's that doing to your huge F65 gamut? At best it's clipping it and at worst it is forcing a gamut remap that may actually be genuinely messing your colours. Depending on what FilmConvert is doing internally. When you start getting empirical with comparisons then you need to start detailing very exact workflows too. Colour science at this level is complicated, prone to errors and impossible to compare on uncalibrated displays.

Personally i think they're both awesome cameras and stunning work has been done with both. I've seen some stunning work from iPhones too :)

cheers
Paul
 
MeesePiece said:
You are right, Paul.
I modified my previous post and took from a DPX still from a CML test.
Click to expand...

Whilst that's the right direction you would need to do a lot more work on the comparison than that for it to prove a useful point (which i'm not sure is even worth proving, i think we're literally splitting hairs here in terms of sharpness)

But

- Again, you can't really scale something in resolve using a bilinear/sinc scale because it will introduce softness, you would need 'nearest neighbour' (or the images posted here have been scaled down, whatever happened they're not pixel accurate)
- I don't know how those DPXs were created, they're not camera originals
- Sharpness is more about contrast in this case and you have 3 images with very different contrasts

But what you're pointing out now is noise handling. And the exposures look different as well as white balance. The Red looks like a log image. If you were just talking sharpness, eyeballing the images i'd be tempted to say that focus on the contrast lines of the chair to the right of her head and F35 looks sharper but that's probably just contrast talking.

Again it's not comparing like for like.

But seriously to make your point it's a lot of work to get those images into the same place to draw a useful comparison. I really don't think anything is clear from the new examples, far from it.

cheers
Paul
 
Paulcurtis this is waste of time. He is in desperate mission to prove F35 is not as good as F65.
MeesePiece I can't believe you bring this BS to Mac's movie thread too. :Drogar-Angry(DBG): Also stop bringing me in your posts. I stop replying to your thread when I saw you talk nonsense. I at least post highest quality blu-ray framegrabs of both cameras. You post some 240p framegrab from youtube trailer to say F35 look bad. Not even real movie but trailer. From youtube! I didn't reply after this because I see it was waste of time. So please stop bringing me in your posts. You are a funny man. You say BS in public and then send me PM to say sorry in private. If you have to say sorry please say in public. But I'm ok if you only stop bringing me in your posts. Go use your F65.
Sorry for this mess Mac.
 
Yes Paul,
Those tests were contrast tests - and their methodology seems to be they pulled DPX files from the Arri D21 and F35 and converted the Red M to a log format.
Their test has the same focal length, but the red was at a different angle, they all were.

My point is that I created the jpegs from the DPX files, all same conditions. They all got the same treatment. I don't have time or money to do a formal test. I have seen enough, even from these flawed tests, to understand what I like and don't like about the imagery from the cameras tested.

SJX,
I apologize to you, profusely.

I brought it here, because you brought it here, SJX. Don't you remember? You wrote something about someone on another thread thinking the F35 is not sharp enough.

You also talk nonsense. You posted some jpgs from a blu-ray website. I posted some jpgs from movie trailers. And funnily enough, one of your examples of how great the F35 is, was from "friends with benefits." A film which had the frame grab with the horrible sky blow out - remember that?

And then I posted stills from a short I shot and directed. What did you think of them, SJX? I care deeply what you think.

I own a F35 and a F65 and in one month, an Alexa Mini. I know what each of their strengths and weaknesses are.

SJX, be prepared. You shook the bee hive. Just as I shook the F35 bee hive. Except I am prepared for the backlash. I own the cameras.
 
Last edited:
MeesePiece said:
Sure . From a CML test. First three images are at 400% zoom in from a DPX file in resolve. Next set of images after that is 300% zoom.
Save the images to see them larger, to really see the difference.
.
Click to expand...

Focus is off on the F35. Below is an 800% crop from the .dpx files, Red on top F35 bottom. You can see the curtain is in focus for the F35 and the blanket (on the focal plane of the subject) is soft.

red-crop.jpg
f35-crop.jpg
 
I know, it's a contrast test. I couldn't find a real-world resolution test. It's the best I could do.
With that knowledge, then with focus off, you would think the f35 would have smoother handling of her face and her hair would hold more.

Again and again, I'd love to set up a real-world resolution test between F35 and Alexa but I don't have the time - if anyone else wants to.
Until then, I still stand by my statement that from the flawed scientific tests that were conducted that I found, the F35 does not have the inherent resolving power than even the Red One M has, and certainly not as much as an Alexa, Amira, Kodak 5219 film stock scanned properly or current Red Dragon have.

The F35 has excellent dynamic range handling as you see in this test and beautiful motion and very nice skin tones, but lacks in resolving power that in my opinion makes sense why the cinematography community has not reached back , and embraced the F35, now that with the odyssey, it is a highly functional camera to use on any film shoot.

In fact, the DP who used the Genesis on Apacalypto, wanted to use it on that Angelina Jolie Disney film, Maleficent, did tests, but ultimately the Alexa won out. And he was a die-hard Panavision Genesis guy. I quote:

After conducting comparison tests between the Arri Alexa, Panavision Genesis and Sony F65, the filmmakers decided the Alexa best suited Maleficent’s extensive visual-effects requirements. Any misgivings Semler may have felt about abandoning the Genesis — his camera of choice since Click (2006) — quickly dissipated. “I believe that images from the Genesis still look the closest to film, but only by a tiny degree,” he says. “I loved using the Alexa; it’s a great camera. I was able to slightly shift the color temperature in-camera, usually by 100 to 200 degrees — a simple and subtle but incredibly effective tool that I used a lot.”

Opps! He liked the Alexa only slightly worse than the Genesis, but more than the F65!
Opps. The F65 is terrible!
 
MeesePiece said:
I know, it's a contrast test. I couldn't find a real-world resolution test. It's the best I could do.
With that knowledge, then with focus off, you would think the f35 would have smoother handling of her face and her hair would hold more.

Again and again, I'd love to set up a real-world resolution test between F35 and Alexa but I don't have the time - if anyone else wants to.
Until then, I still stand by my statement that from the flawed scientific tests that were conducted that I found, the F35 does not have the inherent resolving power than even the Red One M has, and certainly not as much as an Alexa, Amira, Kodak 5219 film stock scanned properly or current Red Dragon have.

The F35 has excellent dynamic range handling as you see in this test and beautiful motion and very nice skin tones, but lacks in resolving power that in my opinion makes sense why the cinematography community has not reached back , and embraced the F35, now that with the odyssey, it is a highly functional camera to use on any film shoot.

In fact, the DP who used the Genesis on Apacalypto, wanted to use it on that Angelina Jolie Disney film, Maleficent, did tests, but ultimately the Alexa won out. And he was a die-hard Panavision Genesis guy. I quote:

After conducting comparison tests between the Arri Alexa, Panavision Genesis and Sony F65, the filmmakers decided the Alexa best suited Maleficent’s extensive visual-effects requirements. Any misgivings Semler may have felt about abandoning the Genesis — his camera of choice since Click (2006) — quickly dissipated. “I believe that images from the Genesis still look the closest to film, but only by a tiny degree,” he says. “I loved using the Alexa; it’s a great camera. I was able to slightly shift the color temperature in-camera, usually by 100 to 200 degrees — a simple and subtle but incredibly effective tool that I used a lot.”

Opps! He liked the Alexa only slightly worse than the Genesis, but more than the F65!
Opps. The F65 is terrible!
Click to expand...

The resolving power is fine. All those dpx files show is that the Red was focused on her face and the F35 wasn't. You are saying things and offering images to back up what your saying, but those images are not doing that.
I just as easy could point at the curtain above and say look how poorly the Red resolves detail.

What the images do show is that the F35 is noisy, which has been discussed and known about since day one. It's part of the camera's charm. The Red being 'smoother' does not make it more filmlike as you claim. The Red is less film like.
 
So I'm new to the forum, but why does there seem to be drama in multiple threads about what camera is best? Since this is one of the few, if not the only, online community for the F35, shouldnt this be a place for sharing footage, tips, tricks, and providing support from our mutual experiences? If we really want a shoot out thread, lets do a shoot out thread, but lets not hijack someone sharing cool stuff to dive into a "which camera is best" circle jerk. It's a great way to drive away members and kill a community.
 
rphill100 said:
so i'm new to the forum, but why does there seem to be drama in multiple threads about what camera is best? Since this is one of the few, if not the only, online community for the f35, shouldnt this be a place for sharing footage, tips, tricks, and providing support from our mutual experiences? If we really want a shoot out thread, lets do a shoot out thread, but lets not hijack someone sharing cool stuff to dive into a "which camera is best" circle jerk. It's a great way to drive away members and kill a community.
Click to expand...

Amen.

​
 
It's a forum for the F35/F65. And I'm merely trolling SJX.
I apologize for the dust.

Here's why the F35 is a wonderful camera.

-Beautiful CCD sensor does not subsample any red, green, or blue channels.
-Can record 12-bit uncompressed from it
-beautiful global shutter
-wonderful skintones and dynamic range - holds up the highlights incredibly well.
-it is cheap as heck right now- a 250k camera is now only around 6k on ebay!
-has incredible natural resolution for a 1080p image
-It can compete, head to head, with the Alexa.

It's an amazing camera, thanks in part to being co-developed with Panavision.
Members on this forum, most specifically MacGregor have gotten some truly stunning imagery with this camera.

Many on this forum swear that it is as close as you can get to film with a digital camera. Including the DP who shot Apocylpto,
In fact, the DP who used the Genesis on Apacalypto, wanted to use it on that Angelina Jolie Disney film, Maleficent, did tests, but ultimately the Alexa won out. And he was a die-hard Panavision Genesis guy. I quote:

After conducting comparison tests between the Arri Alexa, Panavision Genesis and Sony F65, the filmmakers decided the Alexa best suited Maleficent’s extensive visual-effects requirements. Any misgivings Semler may have felt about abandoning the Genesis — his camera of choice since Click (2006) — quickly dissipated. “I believe that images from the Genesis still look the closest to film, but only by a tiny degree,” he says. “I loved using the Alexa; it’s a great camera. I was able to slightly shift the color temperature in-camera, usually by 100 to 200 degrees — a simple and subtle but incredibly effective tool that I used a lot.”
 
You must log in or register to reply here.
Back
Top