Day: November 29, 2008

NFL Week 13 Picks

OK, so we both went 2-1 on Turkey Day. I guess the Eagles still do have some life left in ‘em, and the Cards really aren’t the same team on the road. Also a good FF day for me, since I started Westbrook.

Anyway, on to the rest of this week…

Sunday 1:00
Niners @ Buffalo(-7)
First game on the board, and it’s a tough one. The Niners have shown some signs of life lately, but west coast teams traveling east to bad weather and a 1:00 start haven’t impressed of late. I’ll give the points.

Ravens @ Cincinnati(+7)
I don’t see the Bungles getting more than a couple field goals here. Baltimore

Colts @ Cleveland(+4.5)
Indy is clearly on the rise, the Browns are still confused. Give the points.

Panthers @ Green Bay(-3)
I know the game is in Green Bay and all, but this seems to be the wrong team favored here—I was expecting a spread of +2, not -3. I’ll go with Carolina and the points here.

Dolphins @ St. Louis(+9)
Stephen Jackson may be back, but Bulger might not be. Rams are in a tailspin either way. That’s a lot of points but I see no reason to think the Dolphins won’t cover.

Saints @ Tampa(-4)
Useless trivia: in this matchup, the road team is 8-3 against the spread in the last 11 and the underdog is 9-4 against the spread in the last 13. Seems pretty good to me, I’ll take the Saints.

Giants @ Washington(+3.5)
Even with Burress out (moron), the Giants are the class of the league. I have the G-Men covering.

Sunday ~4:10
Falcons @ San Diego(-5)
Too many points. Even if Atlanta loses, they’ll make it a squeaker.

Broncos @ New Jersey Jets(-9)
WTF is wrong with Denver? How do you lose to the Raiders at home? Ugh. Jets run the ball 35 times between Jones and Washington, gain 250 on the ground, and cover.

Steelers @ New England(-1)
OK, so the key to the Steelers vs. the spread is Willie Parker. They say he’s most likely going to play this week, so on that basis I’m taking the Stillers.

Chefs @ Oakland(-3)
KC allowed the Bills like a bajillion yards of offense last week, so maybe even the Raiders can score on ‘em. I’ll give the points.

SNF
Bears @ Minnesota(-3.5)
The winner of this game will have a clear lead for the division, thus, Vikings lose.

MNF
Jaguars @ Houston(-3.5)
This is three points for home field plus a half-point for how awful Jax looked last week. I think I’m going to go with the Texans here.

AppleTV 2.3 and Handbrake 0.9.3 Notes

OK, so I upgraded my AppleTV to firmware version 2.3 and… now most of my movies don’t have any audio. Yikes! What happened there?

Well, to save space, I had encoded all my movies with just the AC3 track (that is, the raw Dolby Digital) and no AAC track. Apple says that’s a no-no, and I guess with firmware 2.3 they finally actually mean it. Bummer. To solve this problem, there are two choices:

Option One
Manually add an AAC audio track. This actually isn’t that hard. Generate an AAC .mov file with about one tenth of a second of empty audio in it. Open it in QuickTime Pro, copy. Open the offending movie file, make sure the marker is at the beginning, paste in the moment of silence at the beginning, and save it. This may take a while if the file is large, because it has to re-write the entire file. This is a minor pain in the ass, but it’s doable.

Option Two
Re-rip the movie. This is obviously a more time-consuming option, but since Handbrake 0.9.3 just came out, it’s not totally unreasonable to consider re-ripping anyway. And the new version is supposedly much better. The big win is that HB now handles input sources other than the DVD, which is really nice if you have video in some other format already, but most of what I have is on DVD anyway.

So, what does HB offer for that? Well, new H.264 encoding which is supposedly much better. The default “AppleTV” setting used to work with an average target bitrate of 2500kbps, but now it’s set for a constant quality encode of 59% quality. This is an improvement on two fronts: first, it produces much smaller files, and second, it takes much less time to encode because this is a constant, single-pass process.

The problem is that the results aren’t quite as good. I spent a day playing around with this—actually, I just re-encoded the same movie over and over again at different settings while I was doing other things—to figure out what I liked. 59% constant quality (the default) produces noticeably more compression artifacts, particularly banding in gradients. I found a really hard frame in The Fifth Element to use as a test case: scene six where the general steps into the blacklight. Half the scene is white and half of it is a purple color, and the wall to the right of the general is a purple gradient which is hard to get right. I tried 70% constant quality and the still frames right around there still showed banding, which my 0.9.2 rip did not. So I tried ripping with 2500kbps and I still got a little banding in the stills. This was a little disconcerting, since that’s more or less the old setting. Then my brother reminded me to watch the moving video, and sure enough, the banding shows up at 59% but not at the other settings. Classic reminder that moving video and still frames are not equivalent.

However, 70% doesn’t save file space and the old 2500kbps doesn’t save time. So I backed down to 65% constant quality. This produces files that are 10-20% smaller than the old 0.9.2 rips, but of course much faster since this is constant single-pass. And, the banding is gone at this level, so the video quality is acceptable.

So, for most of my movies I’m going with option one because it’s still faster than re-ripping. However, I am planning on re-ripping all my longer movies with the 65% constant quality setting to save a little space, since my half-T drive is almost full.