Username:
B
I
U
S
"
url
img
#
code
sup
sub
font
size
color
smiley
embarassed
thumbsup
happy
Huh?
Angry
Roll Eyes
Undecided
Lips Sealed
Kiss
Cry
Grin
Wink
Tongue
Shocked
Cheesy
Smiley
Sad
1 page
--
--
List results:
Search options:
Use \ before commas in usernames
Edit history:
Apjjm: 2012-09-14 03:37:04 pm
What?
System info:

OS: Windows 7 64-Bit
CPU: I5 760 @ 2.8ghz
RAM: 8 GB

Raw footage info:

Captured by playclaw
Resolution: 1024x768
Fps: 60


I appended the files together using Vegas (as well as applied gamma correction and ordering segments at this stage) and exported with lagarith lossless compression @ 30fps - these files were then encoded using anri 3.3.
I have created two tests - firstly I have a a standard quality test (LQ MQ HQ  IQ) with the gamma correction I used when uploading to youtube. However, gamma adjustment is an important factor in amnesia runs - the game normally would  have you play a low enough gamma so the dark areas are actually dark, however, as the run skips the lantern this makes it impossible to see some of the tricks, so a higher gamma should really be used (I ran with  gamma settings ranging from 1.5 to 2 - the capture software doesn't record the gamma adjustments made in game and captures with gamma 1). Therefore I have also created a longer video which is just the standard quality test video looped with some different gamma settings (LQ MQ HQ, with gamma at 1, 1.7 and 2.4 [which I find best for ytube uploads]).


Download links

Standard quality test (50s):


Same video with multiple gamma settings (150s):


In the clips i have selected two clips which I consider to be some of the brightest & darkest clips of the run. The trick after the first segment must be visible really (otherwise all the out of bounds tricks will be difficult to follow, and a good portion of the run is spent out of bounds). However, the in bounds clip at the end is one of the brightest areas should also obviously not look crap.
Thread title:  
thethrillness.blogspot.com
This is an interesting problem and solution. Do you have any unaltered captures we could look at?

Also to save you time with Lagarith exporting, look into frameserving from Vegas to anri: http://www.debugmode.com/frameserver/
i kind of like 1.7. any higher than that and the fog starts to look really bad (e.g. at 2:00).

normally i don't like "correcting" gamma at all since good monitors let you see everything anyway and it can't be undone losslessly and viewers can just do it themselves if they want. but not being able to see what you're doing a lot of the time is also not optimal and not everyone knows they can do it themselves.
Edit history:
Apjjm: 2012-09-15 02:47:03 pm
Apjjm: 2012-09-15 02:44:43 pm
What?
Quote from TheThrillness:
This is an interesting problem and solution. Do you have any unaltered captures we could look at?

Also to save you time with Lagarith exporting, look into frameserving from Vegas to anri: http://www.debugmode.com/frameserver/

I have all the unadjusted captures still, these are the files referenced by vegas. I don't have particularly fast internet at the moment so I'd rather avoid uploading large files. However, the first test (gamma=1) in the multiple gamma settings video is just as if I ran the capture footage through anri though - so that's unadjusted aside from compression (Though here is an unadjusted 7 sec file [no sound] for one of the darkest areas: https://dl.dropbox.com/u/64805489/SDA/storage2.avi). The frame-serving looks like it might save me some disk space so I will have a play about with that Smiley

Quote from nate:
i kind of like 1.7. any higher than that and the fog starts to look really bad (e.g. at 2:00).

normally i don't like "correcting" gamma at all since good monitors let you see everything anyway and it can't be undone losslessly and viewers can just do it themselves if they want. but not being able to see what you're doing a lot of the time is also not optimal and not everyone knows they can do it themselves.
I agree that 1.7 looks better than higher settings when considering the range of lighting conditions the game presents (and this is in the middle of the range of gamma settings I ran the game with - for what its worth). I could always provide two IQ (and perhaps HQ and lower) versions - one with gamma adjusted to 1.7 and one without any gamma changes - based off the assumption that people who really want the quality benefit of adjusting gamma themselves will want to use higher quality file(s) anyway (The full run actually includes the game's gamma adjustment and intro screens so viewers can use this as an aid if they want).

I have uploaded another sample with gamma adjustment at 1.7:

Same clip with no adjustment put through anri: