Anders Holck
Well-known member
- Joined
- Dec 28, 2006
- Messages
- 1,325
- Reaction score
- 1
- Points
- 0
Well, I'm sitting here at the Schipol Airport, waiting to catch a plane at 1900 hrs.
Besides the horrors of "The Hotel From Hell" (See this Thread) it was a realy exiting exhibition.
The greatest moments were seeing the excitement from a lot of broadcast minded people when they heard the RED workflow presentation and finally got it. Not that people wasn't sceptic anymore.
I saw CTL Friday, which was a great experience.
I must admit that I was really surprised to see sensor dust on 60-80% of the footage which wasn't reported before. And the CMOS readout was really suffering from shearing because of the readout technique. Especially the "Through the snipers scope" rush pan to the tank's canon made it look like it was made of jelly. As I had already seen some footage from the production cameras I knew this was fixed, I was just surprised no one noticed at NAB. I still think RED should have explained this, because I had to "educate" a few people on the progress after the presentation.
I later spoke with Graeme (very briefly, as he was a wanted man) and he told that the dust problem was improved as the OLPF was moved further from the sensor and the Sensor readout was fixed early on, but after the Peter Jackson test.
Regarding the sensor dust problem, we do have to take a bit care of what blows into the lens port. But a clean should manage it pretty fast.
I was at the P+S technique stand and was surprised to see that the SI2k had obvious image shear on fast pans. Congrats to the RED engineering team for solving this.
Played around with the old Redcine Build in the tent. The debayer algorithm was pretty bad (The newer algorithm was in RedAlert and it looked stunning.) but the interface was really nice, especially the project stage where you could arrange your shoths, a bit like the lighttable view in Apple's aperture.
Too bad no working LCD or Viewfinders was available to us Europeans, I know a lot of people would like to see how those performs.
I liked the menusystem and the interface layout. The multicontroller on the back seems like a good solution, but of course without engraved button labels on the beta cameras for the other buttons it was a little unintuitive at first.
I spend some time at the Apple booth trying to explain a few bugs that annoyed me. Don't think they really listened anyways.
I like the way the media naming scheme on the RED.
As I understood the current scheme is like this:
Every media has a cardfolder:
Cardfolder: Cameraname_Cardnumber_YYMMDD
In that folder the media files are placed:
Clipname: Cameraname_Cardnumber_YYMMDD_Clipnumber
Cameraname can be set from the menu. Cardnumer is automatically incremented when you insert a formatted card.
Don't know why Panasonic didn't got this right from the start but having every card named uniquely with a folder including cameraname, date etc. is the only way to do it. This is very nice, and it's working. No need for RED Genie, thank god. Maybe a small app to autodetect the cards, copy with verify and format. but thats all
Regarding the lenses, I was told that all lenses except the 300mm will get T stops, so both zooms and the prime set should get them.
What was obvious was that some company needs to make an angled hdmi connector though. I asked Belkin and Kramer, but they didn't seem interested.
Besides the horrors of "The Hotel From Hell" (See this Thread) it was a realy exiting exhibition.
The greatest moments were seeing the excitement from a lot of broadcast minded people when they heard the RED workflow presentation and finally got it. Not that people wasn't sceptic anymore.
I saw CTL Friday, which was a great experience.
I must admit that I was really surprised to see sensor dust on 60-80% of the footage which wasn't reported before. And the CMOS readout was really suffering from shearing because of the readout technique. Especially the "Through the snipers scope" rush pan to the tank's canon made it look like it was made of jelly. As I had already seen some footage from the production cameras I knew this was fixed, I was just surprised no one noticed at NAB. I still think RED should have explained this, because I had to "educate" a few people on the progress after the presentation.
I later spoke with Graeme (very briefly, as he was a wanted man) and he told that the dust problem was improved as the OLPF was moved further from the sensor and the Sensor readout was fixed early on, but after the Peter Jackson test.
Regarding the sensor dust problem, we do have to take a bit care of what blows into the lens port. But a clean should manage it pretty fast.
I was at the P+S technique stand and was surprised to see that the SI2k had obvious image shear on fast pans. Congrats to the RED engineering team for solving this.
Played around with the old Redcine Build in the tent. The debayer algorithm was pretty bad (The newer algorithm was in RedAlert and it looked stunning.) but the interface was really nice, especially the project stage where you could arrange your shoths, a bit like the lighttable view in Apple's aperture.
Too bad no working LCD or Viewfinders was available to us Europeans, I know a lot of people would like to see how those performs.
I liked the menusystem and the interface layout. The multicontroller on the back seems like a good solution, but of course without engraved button labels on the beta cameras for the other buttons it was a little unintuitive at first.
I spend some time at the Apple booth trying to explain a few bugs that annoyed me. Don't think they really listened anyways.
I like the way the media naming scheme on the RED.
As I understood the current scheme is like this:
Every media has a cardfolder:
Cardfolder: Cameraname_Cardnumber_YYMMDD
In that folder the media files are placed:
Clipname: Cameraname_Cardnumber_YYMMDD_Clipnumber
Cameraname can be set from the menu. Cardnumer is automatically incremented when you insert a formatted card.
Don't know why Panasonic didn't got this right from the start but having every card named uniquely with a folder including cameraname, date etc. is the only way to do it. This is very nice, and it's working. No need for RED Genie, thank god. Maybe a small app to autodetect the cards, copy with verify and format. but thats all
Regarding the lenses, I was told that all lenses except the 300mm will get T stops, so both zooms and the prime set should get them.
What was obvious was that some company needs to make an angled hdmi connector though. I asked Belkin and Kramer, but they didn't seem interested.