[Sean Oneil]

[Dave LaRonde] “I’ve done it in After Effects, but I bet you can do it in Motion, too. Create an old Pong-style animation: a circle caroms off all four edges of the screen in a 2-second cycle…. you can loop it. Add motion blur if you like. Now here’s the trick: render two versions, one in the proper field order and one with NO fields… render the entire frame. Now bring the files into FCP, loop them several times and compare the motion. You won’t believe the difference.”

You’re mistaken. NTSC does not change 60 times per second. Only 30. Individual fields are impercievable. You might see them as artifacts on your FCP canvas. But that is just because it is not being deinterlaced properly. A consumer progressive-scan television will deinterlace it properly.

You are correct about the flicker. A CRT monitor will “flicker” 60 times per second. But you only see 30 images per second. And again, this only applies to CRT displays.

An LCD monitor (or DLP, plasma, LCoS) usually has a refresh rate of 60hz. But it’s inpercievable because it doesn’t flicker. Unlike CRTs which “draw” lines one at a time, LCDs display a constant, continuous light output. So whether the refresh rate is 35hz or 200hz – no matter what you will only see 30 images per second (or 24 if it was shot on film). And flicker doesn’t equate with this kind of display.

As far as your experiment in regards to the human perception of motion and frame rates. An animated Pong is not a valid method of comparison. In fact animated anything isn’t a useful way of demonstrating it. You need real people as subjects. You need depth. You need photography of real life. You need the ability to compare and contrast (subconsciously) the difference between what you see on screen and what you would see in real life. That is how the “film look” and “video look” is derived. When John MaClaine scratches his nose, it looks a lot different than when Al Bundy scratches his nose, and when your buddy does it in real life. Comparing ways Aladdin scratches his nose won’t give you a comparison to actual reality which is fundamental to the film vs. video look. Simple animation doesn’t mimic real life.

Sean

[Sean Oneil]

[Herb Sevush] “but when converted to video with a 3/2 pulldown it is no longer 24”

You’re wrong. Yes it is still an actual 24fps even if it’s converted to a 30fps signal. The extra 6fps is made up by duplicate fields and/or frames. If an actor moves across the screen, your brain STILL only sees 24 unique images per second. THAT is the film look.

Sean

[Sean Oneil]

I’m sorry but I have to adamantly disagree with you guys.

24fps is what gives the film look. Not because of flicker or progressive-scan or anything like that. It’s the actual framerate. It doesn’t matter if it’s 1080psf 23.98 or NTSC w/ 3:2 pulldown. The actual framerate is always 24 frames per second.

It’s not rocket science. Our brains are trained to interpret 24fps as cinematic. The lower frame rate adds to the suspension of disbelief, but more importantly it reminds us of the movies. When we see 24fps, we are reminded of when we first watched Jaws, Indiana Jones, etc.

30 frames per second reminds our subconscience of live sports, soap operas, the news, cheesy sit-coms, game shows, etc.

That is why 24fps IS the film look. 30fps IS the video look. It’s because of the motion.

[Sean Oneil]

Two things.

1. On the consumer end, it makes NO difference. 60i and 30p source footage will look identical. They’re both 30 frames per second and they’re interchangeable with one another. If a consumer has a progressive-scan display, all 60i signals get converted to 30p anyways. And vice-versa (if there was such thing as a 30p DVD).

2. “Film Look” means 24 frames per second (23.98). That’s it. That’s what it means. Shooting 30p is no more “cinematic” than shooting 60i. That’s not necessarily a bad thing. It’s just not cinematic.

In digital post, the only advantage 30p has over 60i is that it’s easier to work with on computers. That’s about it. Sure, maybe certain cameras behave differently when set to “30p mode”. Maybe the CCD captures more light. But those are just quirks unique to a particular camera – not something you use to definitively compare two formats. Not when we’re talking about the overall big picture.

Sean

[Sean Oneil]

Thanks. I posted on the forum weeks ago but haven’t heard anything. I have Wasabi IP-SAN v2.3.2. The i386 version.

Essentially, I get really slow write performance when using a hardware RAID-5 (I have a 3ware 9550sx controller with write cache enabled). Speeds are fine if I create a RAID-0.

The other (more important) issue is that during large file transfers, the Finder will lock up. The diagnostics log shows “timedout” errors.

Both of these problems are exagerated when I try to use multipath connections. While my read speed goes up to 175 MB/s, write speed drops down to about 15 MB/s when using RAID-5. With RAID 0 I get 175MB/s both directions. But the timeout error occurs no matter how the disks are stripped, and it occurs much faster when using multipath connections.

Also, whenever I unmount a drive, the log shows an error “UNKNOWN SCSI OPCODE 0x1b”. Not sure if that means anything.

[Sean Oneil]

Thanks, but I didn’t purchase globalSAN. I have a different Wasabi-based iSCSI target which I purchased over a year ago (before I even heard of SNS). I’m trying to use your initiator with it, but am having some issues.

I’d gladly pay for support, but one of your sales guys told me it isn’t possible.

If I knew the node and portal settings which are compatible with your initiator, it would be a major help in troubleshooting my problems.

[Sean Oneil]

I use them. Haven’t seen any issues so far.

[Sean Oneil]

[Tom Wolsky] “We all get to beta test it. “

Exactly the problem IMO. They have more than enough resources to thouroughly test every possible workflow scenario.

[Sean Oneil]

Surprised no one’s mentioned this yet.

Has anyone tried reconnecting the render files?

Sean

[Sean Oneil]

[Tom Wolsky] ”
To be honest, we don’t much care whether you use FCS or not.”

You should care.

Out here in LA, these things stick in the minds of people. Including clients.

Even if they fix this tomorrow, a year from now clients will think Final Cut is a cheap solution that makes you re-render everything.

It’s just one more “bullet-point” that people will drop when discussing Avid vs. FCP. The stink will last forever.

Apple is making a KILLING now, especially with the iPhone. The amount of money and resources they have – this kind of thing should never happen. And it’s not like they have an infinite amount of hardware environments they have to test on. That’s why Macs are supposed to be so reliable. There are only a handful of hardware possibilities.

Sean