nate,


i will be using some slow motion spots but the majority of the time will not be. thanks for the input. if you don't mind in your opinion what should i be shooting surf footage at? i am a little confused on the proper shutter speeds with my FX1. i shot the other day shooting with 1/60th shutter and i had motion blur all over the place. so i am now torn between my GL2 or my FX1 for shooting fast action spots like surfing. is there any site or information that you know of that you can direct me to that could shed some light for me?


joel

Are you sure you were really shooting at 1/60th? If you set the shutter speed by hitting the shutter button, and then it gets pushed again by accident and you happen to be in a low light situation, the camera will then bump it to 1/30th. Then it's smear city.

I find that I CONSTANTLY have to be watching the shutter speed on the camera, because that damn button will get pushed and then all bets are off. The way Sony handles those settings is way way wack.

I suppose you really could be shooting 1/60th and you think it's smeary, but other than the tiny bit of MPEG smear, I think 1/60th looks like every other camera on the planet.

yeah i am sure. i got some footage i'll post later. i never shoot below 1/60th.because of the strobing @ 1/30th.

oh yeah one more thing, if you don't mind. how did you shoot @ 120th? i can only get 1/125th.


joel

Hi Nate !

There are still interlace jaggies on the edges of outlines. It doesn't look like the downconvert and slowdown does a correct job. Good enough to look OK on that resolution, but not right.

I'm going to try to use your same idea and actually seperated the fields into full frame to finally put them together into a slowmo sequence. The methode I'll try to use will gets the clean fields, without jaggies.

I'll create a shake script to do that.

I'll post the result once it's done.

Frank

frank,

i'll bet you that his DV master is way better. remember you are watching a quicktime web version.


joel

I'm almost sure it's not. You wouldn't get that kind of interlace artifacts on the edge. Even with compression.

Frank

Here is the test I did as explained in a few messages earlier on this thread. No interlace artifacts and clean frames.

http://home.comcast.net/~chalbers/slow.wmvx

Download the link and take out the x at the end to play

Frank

frank,

i totaly see what you mean now. so what was your secret? and what was that hacky sac footage shot as?


joel

Way to go Nate!!!!! That is one awesome production there!! Congratulations!

Hello everyone !

So it seems that Nate DID use full clean frames. The footage was just 30 frames footage with 2:3 pulldown of his original 24 frames . That explains the interlace !

Now I feel stupid ! :-)

Frank

The hacky sac footage was HDV 60i footage taken with an FX1 that I found online.

I responded to the same bit on another board, but in the interest of defending my methods, I'll respond here too :-)

Frank, what you're seeing is the 2:3 pulldown applied to the slow motion footage AFTER the initial slowdown. My intent from the beginning was not to slow down the 30i to 30P, but to 24P. That way I get 2.5x slow motion as opposed to just 2x. This is also in line with how the Varicam post processes footage, and also if you shot with a film camera @60fps.

So anyway, I took the captured 29.97 full res Apple Intermediate Codec into After Effects. I brought them into a 720x480, 23.98 comp. I manually sized the clip down to fit. I set the Interpret footage dialog in AE to understand the AIC clips had upper field dominant, and also to conform to 12fps. This slowed down the clips to almost exactly 40%.

I then rendered out the clip, adding the 2:3 pulldown in the output module. I needed a 29.97 DV clip to intercut with the CF30 footage, which was downconverted in AE also, but to a 29.97 comp.

At that point I had all 29.97 DV clips to cut in FCP. The interlacing you see in the web movie is the 2:3 pulldown added in After Effects, and ONLY in the slow motion shots.

Trust me when I say that the only problems with the footage in the end is that there are bad compression blocks that jump around in a few of the shots, assumedly from errors on the HDV tape (but not the 15 frame dropout). Also, there is a little line twitter in places...the downconversion in After Effects is pretty sharp.

Thanks for the explanation !

I did it using a compositing package.

I doubled the amount of frames by changing the speed to 0.5.

frame 1 becomes frame 1 and 2, frame 2 becomes frame 3 and 4 ... etc.

Then I take out the upperfield and downsize it to DV. I do that for frame 1,3,5,7 etc.

I then take out the lower field and resize it to DV. That one I save out as 2,4,6,8, etc.

Off course, all that is not done manually. I use a compositing script in Shake.

That's about it

That looks very nice Nate. I get my camera on Tuesday, the more I see the more excited I get!

by the way, who is the band?