View Flash XDR Time Lapse Sequences Here
 

Hi Friends:
Please follow the link here to stream a flash media video of my first timelapse attempt with the Flash XDR. I have several other tests to upload, so I'll be posting those in this thread as well. This is one frame every 7 seconds shot on my Canon XL H1 to a Sandisk Extreme IV 16 GB CF Card. Video is made up of a whole bunch of tiny 3 second video files put together on an Avid Media Composer timeline (Ver 3.1.3.2) and exported via a QT Reference to Sorensen Squeeze (Version 4.5.700). FLV file was encoded using the Sorensen Spark codec.

Test

Thanks Mark,

Besides the Star filter what else did you use to keep exposure right while the sun entered the frame.

No Star Filter
 

Hi Paul:
I used no Star Filter on the Canon Xl H1. The star effect is what the iris diaphram leafs produce as they close down. I set the exposure on automatic and used the built in maximum Neutral Density filter setting built inside of the lens at 1/32 ND. The gain setting for the video signal was set manually @ 0 db gain. White ballance was set at Sunlight. I deliberately left in the *speed jump* to show that this must be addressed somehow, or you *must* take your sequence in ONE (1) recording session and record longer than what you need so the speed jump can be cut off the end of your Time Lapse Sequence.

Thanks Mark,

So the speed jump is from the Nano? An you are saying the jump can be cut out if you record longer then needed? Just checking since I have still not had a chance to do a time lapse with my Nano.

I have no nano
 

Hi Paul:
I don't have an nano. I own its big brother the Flash XDR. I'm saying the speed jump can be cut out if you record your timelapse in One (1) session *ONLY,* and record longer than what is required. Dam Keaton says that the speed jump is needed to rapidly close the file, or else it wouldtake 9 hours plus to close a timelapse MPEG data chunk.

Makes sense thanks Mark.

Dear Paul,

The timelapse works the same on the Flash XDR and the nanoFlash.

If you record a long time sequence, there will be one "Speed Jump", frames recorded at the very end of the last clip in the sequence, which will be up to 1 GOP, or up to 15 frames at regular speed, up to 1/2 second of video at regular speed.

If you start and stop your timelapse, then you will get one of these each time you stop.

We recommend recording your timelapse in one continuous recording, then recording at least one second longer than what you need, then slicing off the last second of video at the end of the last clip.

We realize that this may require AC Power, or a UPS, or a large battery rig for very long, uninterrupted timelapse sequences.

Thanks Dan,

For my entire time lapse in the past I have always started and stopped with plenty of extra room in the clip so I can cut the best out. I think this is normal practice so you have options in any clip shot.

So here's a timelapse for you. Well sort of. I've been considering going out into the outdoors and getting some timelapse shots just for the fun of it. Since my camera will kill a Dionic 90 in about 50 minutes, I needed some cheap portable power. Dionic 90s are more than $400 each.

So I started considering these portable 12v things that can jumpstart your car. They also have 12v power ports, in addition to being able to jumpstart a car. I looked at a couple of them and they were cheap enough to give it a try. There was a black and decker at home depot for $99. And then I came across this one at Costco for $65. Cheap enough.

So before I dragged it out to get some timelapse shots, I needed to know how long will the battery powering my camera.

What better way than to test the timelapse capability, than to do a timelapse.

I used the iphone as a slate, and added a voltmeter to get an accurate reading of the power level over time. The little $65 battery powered my camera for over 3 hours. Darn good enough to do some decent sequences. In three hours, the voltage dropped from 12.70v to 11.8v, at which point my camera shut off.

I haven't done enough testing to say whether or not I recommend doing something like this. My first thought is, this probably isn't a real good way to power a camera. It's darn cheap though and you can't ignore that.

PSX Timelapse Testing

Dear Aaron,

Thanks for posting.

I assume that you were using the DC output to power your camera.


Some of these have built in AC Inverters, but they almost always are "Modified Sine Wave".

They usually advise against powering sensitive electronics (and their power supplies) from a "Modified Sine Wave" inverter.

However, the DC output of these units are fine.

You're right Dan. Cheap inverters produce no where near a sine wave. Those inverters can ruin electronics. I cringe hooking up my MacBook pro power supply to an inverter, and that only costs $70 to replace when it blows up. No way I'm connecting my camera to one.

More Behind The Scenes Time Lapse On Location in Downtown Montreal
 

Hi Shooters:
I posted some more behind the scenes clips Flash XDR shooting in downtown Montreal @ Philips Square. Please Stand By

All I see are behind the scenes, where's the actual timelapse sequences!

Where's the Beef ?
 

Hi Aaron:
I'm editing the footage now. I should have it up late tonight. Yah, sorry about that.

Well you DID say it was "behind the scenes."

:o)

Here are the Actual TL Sequences
 

......Here are the actual time lapse sequences. Unfortunately they are not useable due to the timelapse interval being to great.
Please Stand By

I was bored so I made a timelapse this afternoon


Since my boredom was also combined with laziness, I only ventured about a thousand yards from my house to get this footage of some cars driving by. One of these days, I'll shoot something interesting.

There is an hour's worth of timelapse to start the clip. About 40 seconds on playback. Then there are some random clips at the end of the sequence. I just grabbed them as I started packing up the rig to head back home. Some of them are with fast shutter.

I started color correcting but I only got as far as "good enough". As always on vimeo, you can download the full HD original. 36MB.

The Viper is So Increadibly Sharp !
 

Hi Aaron:
Even with the smoothing effrect of the average Vimeo Post, your footage out of the XDR and your Viper is increadibly sharp. What was your time lapse settings ? (I use the 220 I-Frame mode exclusively). Your slow mo stuff looked great, only, I noticed the vehicles close to the camera looked ever so slightly stroboscopic in their motion. In other words, the temporal fluidity at distances looked normal, but less so at closer ranges.

Hi Mark. I think the XDR fixes all timelapse at 220/I-frame.

The final segment was shot 720/60p then conformed to 24p for 2.5x slomo. The clip is on the fcp timeline at 800% for a few moments then finishes at 50%. The motion blur as the cars ge close is pretty big. Shutter was fast as possible. The strobing you see is optical flow creating new frames using heavily blurred frames as the source.

Hi again Mark. I just watched the video on vimeo and on my system. The strobing effect is definitely exaggerated on the vimeo video compared to the original here.

Vimeo Colors Effect Result
 

Hi Aaron: OK. Good to know. Aaron, what do you think of the new Sony SRW 9000 Camcorder ? It does all the way up to uncompressed HDCAM *SR* @ 4:4:4 I wonder what this looks like from that camera ?

Hi Mark. I just went to Sony's site and read all the specs for this camera. It looks like a VERY sweet rig. I'm not sure I understand the hype though, unless you're doing live, broadcast or other quick turnaround work. For cinema applications, the Sony F35 still seems like a better option.

The S-log Gamma feature seems interesting but it still seems like there is no way to get an unprocessed signal out of the camera. The RGB modes are processed signal. It's also a little disappointing that the 4:4:4 RGB and other advanced modes require extra boards.

You can bet though that the picture out of an SRW 9000 will look very very good. Uncompressed HDCAM SR 4:4:4 would look very good.

My camera also shoots all the frame rates as that camera, except 1080p60. I like my physical shutter like a real film camera too (until the motor stops working someday). My camera can also do 4:4:4 RGB (which is a processed signal) like the SRW-9000 but my camera can also do 4:4:4 FilmStream, which is an unprocessed signal (raw sensor data). In Filmstream mode, there is no gamma curve applied to the sensor data, no white balance, no color matrix, no knee, no detail, no gain, I mean really, no processing. When shooting FilmStream mode there is no way to set any of those settings because they don't apply. Only raw sensor data is recorded 4:4:4, 10bit log. Focus, lens aperture, shutter speed, composition of picture. That's it. Think about it.

Oh yeah, Viper shoots cinemascope with standard lenses, using nearly the FULL sensor (unlike Red, which crops the sensor).

I'd love to have either one of the SRW-9000 or an F35 though. I could hire myself out at thousands of dollars per week.

Unfortunately the Viper market is shrinking. The biggest Viper rental company in LA dumped all the Vipers and the ones I know still rent the Viper aren't getting much money for it.

No Work for a 27 Megapixel Camera :-(
 

Hi Aaron:
It's sad to know you can't get much work with what is a flawless 27 megapixel camera ! However, with your Nano mounted slant back on your custom mount to the Viper, it's just like it's own camcorder of sorts. (??) I checked out the specs of the SRW 9000 this past weekend and it looks like the next cmera we need to buy. Just think, I could have the SRW 9000 as the *A* camera, and my XL H1 with the XDR attached recording in full 10 bit uncompressed as the *B* camera. Perfect ! ;-) No ? Yeah ! I could even by a duplicate lens like the one on the SRW 9000 and with an adaptor put that on the H1. :-) P.S. I wonder if the viewfinder on the SRW 9000 is a little higher grade than the XL H1 ?

I think if you had an SRW-9000, you would make it the A, B and C camera unless you had to shoot live. I don't think your Canon would get much use unless you had to get small.

I don't have an XL H1 but I heard the viewfinder is not very good quality. My camera's viewfinder is ok and I can view any channel in the viewfinder and it's b&w too. I still find it tough to get critical focus on ANY viewfinder though.

I set the backfocus on my camera last night and I had to hook up a 24inch full raster LCD in order to get it just right. When there is no LCD available though, I have to set it by viewfinder though!

Hi Aaron:
I was being visicious about my XL H1 versus the SRW 9000 ;-) As good as my XL H1 is, and it is fantastic when used with the uncompressed HD-SDI out, I certainly cannot compare this solution to an SRW 9000. Notice how Sony chose to use CCD's versus CMOS sensors in this camera ? I am beginning to wonder if CCD's are better overall than CMOS sensors ? I'll have to do some more research on this.

I think that RED and the DSLR guys have proven that CMOS, engineered correctly can be VERY good technology. Heck, I believe the Phantom even uses CCD if I recall correctly.

However, I'm sticking with CCD. The cameras I lust after most are all CCD.

You joke abut the SRW-9000 just like I would have joked about the Viper years ago. That's the beauty of this business though, in a few years, an SRW-9000 can be had for next to nothing. I love seeing super expensive (when new) cameras pop up on ebay every now and then.

Now that they don't have moving parts,...

Another Time Lapse Shot for Ep 2 of Please Stand By
 

Hi Everyone:
Here's a time lapse shot that's a keeper for inclusion into the opening sequence of Episode 2 of Please Stand By. This scene was taken from atop of Mount Royal from the lookout section. It was so windy atop of the Mountain that the gusts grabbed at the Flash XDR which acted like a sail attached to the rear of my XL H1. The XDR is rather large, so this gave the forces of nature something to grab at and the camera began to move up and down on its tripod mount slightly.

I corrected the up and down movement in Avid to remove the extreme telephoto bounce I got with each wind gust. In the future, I will bolt the camera directly to the mounting plate on the tripod head and not place the Anton Bauer Elipse battery between the camera and tripod mount. Mounting the Elipse battery to the bottom of the XL H1 works perfectly well for hand held shooting, but is not suitable for tripod mount work, because there is too much play in the plastic quick mount system for the camera and the battery. Dan's engineered XDR mount for the XL H1 allows for mounting an Anton Bauer battery right on the plate which holds the XDR itself, so this may be a better solution ultimately.

This scene was shot on Automatic exposure with the 1/32 ND filter engaged using the stock 20X optical zoom Canon lens at -3 dB gain with the dailight filter on.

Test Double clicking on the video screen will give you full screen video viewing.

EDIT: The time lapse interval was 1 frame every second.

Hi Mark. I tried to look at it but it didn't work. I just get a blue screen that says test.

EDIT: Looks like it comes up in Windows.

Viewing My Tests
 

Hi Aaron:
I just checked out my test with my iMAC and it works fine, but you need to download the Flip for Mac pluggin for Safari and Quicktime Player. You need the very latest version of Flip for Mac as well (Version 2.3.0.14). I'm using Windows Media Video prpfessional 9 codec with the special HD1080p codec. I use this because it is absolutely faithful to the Grey Scale Values of the original footage corrected in Avid Media Composer. You can download the free player components for Quicktime Player on your Mac Telestream Flip4Mac WMV - Overview

Mark. Extremely nice time lapse work! Beautiful.

Can you provide all the details as to how you setup the XDR menus.

XDR Time Lapse Settings
 

Hi John:
Thank you for the kind words :-)

The XDR Time Lapse menu settings are as follows..........

Video>Codec set to I-Frame>Scroll down to T-Lapse and put check in box>Set interval time,

You should see a Flashing "Time-Lapse" indication. I would also go into the System>Trigger set to Record button.

Hi Mark. I do not think you need to set the I-Frame or bitrate when doing time lapse. If I recall correctly, in order to do timelapse, you only need to put the check mark in the time lapse box and also set the time interval.

The I-Frame and 220Mb/s bitrate should be chosen automatically while time lapse is active.

When you uncheck the time lapse, the codec and bitrate settings should got back to what they were before you enabled time lapse. I can check this within the next couple of days to be sure.

Hi Aaron:
Yes, I believe this is correct. I didn't list these settings to John because they are taken care of automatically once you inable the check in the time lapse box.

Dear Friends,

At this time, I am not certain as to whether you need to set the bit rate or not.

(I think you do.)

I know that you do not need to set "I-Frame Only" as we force this internally.

I thought that we allowed you full control over the Bit-Rate so you could match this up with your card speed.

For example, a SanDisk Extreme III card can not playback 220 Mbps, but, in Time-Lapse, one can record to the same card in 220 I-Frame Only, since we do not write more than 1 frame a second.

But, playback would be at the 220 Mbps full speed and the card could not keep up.


In think we want you to set the Bit-Rate so you could choose a bit-rate that was approrpriate for playing back from the nanoFlash, if desired.

More Time Lapse Clips & Behind the Scenes Action On Location in Montreal
 

Hi friends:
Please see the new clips I shot last weekend in Lower Mount Royal Park Test

Dear Mark,

Thanks for posting.

I see that you are shooting one frame every two seconds.

I suggest you try 1 frame a second for one of your tests.

Time Lapse Interval Settings
 

Hi Dan:
As you can see from the three newest shots I have posted, the 1 frame every 2 seconds works well for the statue, but is still too fast for the Tam, Tam people. I will keep shooting there. (Dangerous location) - I think they already pick pocketed me and stole my old cell phone ! This group is made up of very itinerate musicians and homeless folks. One guy tried to unplug my XDR while rolling. The ganja and pot smoke was so dense one was loose and giggling endlessly within half an hour. I have to go back there to get the really wide andle long shots, but I will be much farther away from these folks, so it should be OK.

...If you folks want to use any of my shots from the timelapse I'm now shooting, then let me know in an email. All I ask is that you credit the source footage. In this case it is from the Internet Tv Series Please Stand By Epiode 2. I will be posting more shots this week.

More Time Lapse Clips to Check Out
 

Hi friends:
Here's some more wind swept time lapse. I steadied the second shot in Avid MC but the first shot isn't steadied yet and will be when it's cut into Ep 2 if we decide to use that clip. I need to change my Anton-Bauer Battery setup, since the Elipse battery is mounted between the tripod steel mounting plate and the camera. This is no good for tripod work, since there is too much play in the plastic mount for the camera on top of the Elipse battery. I don't know why they chose to use plastic in the quickmount slug which screws into the camera base. Plastic isn't as demensionably stable as metal. This setup works perfectly for hand held shooting however. Click onTest

Strange Time Lapse
 

Hi friends:
Here's another final timelapse I shot in lower Mount Royal Park this past Wednesday (Although I incorrectly ascribed it's shooting date to the first week of November in my on camera commentary). The specs in the clips are actually birds flying by and some tops of heads. I will digitally paint these out in After Effects later. However, there's something really strange about what I actually recorded frame by frame here. You cannot see it in this windows media video realtime playback, but if you go back and fourth in slow motion, then you can clearly see what looks like 2 flying white rods (Possibly a meteor passing in or out of the atmosphere over Montreal at the time ?).

I know these are not aircraft, because they are not in the correct approach or departure side of the mountain. All aircraft approach Montreal to land in the West Island either from the Southwest over Vermont State, or North West of the city and make a Right turn to approach overr East Montreal passing by the Right (North side of Mount Royal), so what is this ?? I am making a slow mo version of this clip now so you can see for yourself. I will post it later. Watch the time lapse clips and commentary by clicking on Test