At work, I’ve been getting a Gentoo/PPC system up and running on a fairly new (~half-year old) 2x800MHz/1.2GB Quicksilver PowerMac. It’s actually gone surprisingly well, the only real difference is fiddling with the mac-fdisk and yaboot stuff, but besides that, the install was pretty straightforward. I was sort of disappointed (not really surprised, though) by the not quite blazing compile times. I decided to run a little comparison against my Athlon 800MHz/512MB home system (not very scientific, my home system is running quite a few more daemons and is compiling a more featureful and larger kernel)

In anycase, here are some numbers for running a make clean bzImage modules modules_install on both systems:

Athlon PPC
real    13m35.243s

user    12m14.290s

sys     0m50.400s
real    14m34.575s

user    13m40.770s

sys     0m56.640s

Yep, and this is running with ‘-j3’, both CPUs were being used. This actually seems to fall in line with the bogomips report, I guess. Here’s the /proc/cpuinfo for the systems. My home server:

processor       : 0

vendor_id       : AuthenticAMD

cpu family      : 6

model           : 4

model name      : AMD Athlon(tm) processor

stepping        : 2

cpu MHz         : 800.059

cache size      : 256 KB

fdiv_bug        : no

hlt_bug         : no

f00f_bug        : no

coma_bug        : no

fpu             : yes

fpu_exception   : yes

cpuid level     : 1

wp              : yes

flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca 

                  cmov pat pse36 mmx fxsr syscall mmxext 3dnowext 3dnow

bogomips        : 1595.80

And the Mac:


processor       : 0

cpu             : 7450, altivec supported

clock           : 799MHz

revision        : 2.1 (pvr 8000 0201)

bogomips        : 797.90



processor       : 1

cpu             : 7450, altivec supported

clock           : 799MHz

revision        : 2.1 (pvr 8000 0201)

bogomips        : 797.90



total bogomips  : 1595.80

machine         : PowerMac3,5

motherboard     : PowerMac3,5 MacRISC2 MacRISC Power Macintosh

detected as     : 69 (PowerMac G4 Silver)

pmac flags      : 00000000

L2 cache        : 256K unified

memory          : 1280MB

pmac-generation : NewWorld

Now, at this point, if you’re still reading, you might be wondering, what’s the point? Honestly, I just did this little test mostly out of personal curiousity, after sitting around during the initial kernel compile (it took almost half-an hour without the j flag). Any developer or techie who uses both Mac and PCs can tell you that the Macs are much slower, hardware and software-wise (and blown out of the water when dollar/processing is factored in). Still, it was interesting to see my 2yr old hand-me-down-parts server beat out the much ‘beefier’ Mac (remember, the entire Gentoo system is custom-compiled on gcc3.2 w/ -O3 mcpu=7450 -maltivec -mabi=altivec -mpowerpc-gfxopt CFLAGS. It’s about as fast as I think it can get).

Also, in case you’re wondering, the Mac system was originally ordered w/ OS X Server (10.1) to be a workgorup file server, but after continually crapping out (in addition to being a bear to compile/install packages for), my co-worker tossed in the towel. I have nothing against Macs or OS X. It’s a nice OS, just crap as a server. Oh, and Jaguar is still slower than molasses on my 400MHz TiBook. The prettiness wears off when you get slowdowns scrolling.

Excellent summary of latest developments in CD DRM and related biz: Summary of the state of play. Related, reposting of WSJ article (and interesting attached thread) on SACD. (Interesting explanation of DSD encoding on SACD)

To read: How Do People Evaluate a Web Site’s Credibility?

Gee, can you guess what I spent today doing? Well, I also watched all of the Season 4 Reboot episodes. Extra fun: Official Site, A Minor Glitch, AMG: Overanalysis Theater, stuff from alt.tv.reboot: 1, 2, 3.

Went to the West Hollywood Halloween dealio last night. Pretty large crowd (several hundred thousand people, apparently) and some great costumes. I spent most of the time at one of the stages of a ‘band’ (Mutaytor) that my friend spins fire for. Also thought I’d mention, for those in the LA area, that RESFEST is coming to town this weekend. I bought tickets for the Cinema Electronica, Chris Cunningham Retrospective, and By Design shows. Also, all day Saturday is SCALE, the Southern California Linux Expo.

The LA Freewaves kicked off their wan^H^H^Hmemefest today. Xeni did a little writeup in today’s Wired News. I love this quote:

“Before, artists had to figure out how to access costly facilities for editing and output,” she said. “Now, $5,000 buys you just about everything you need to produce high-quality video artwork.”

For those unfamiliar with the subject area, ‘high quality video artwork‘ is by and large an oxymoron.

Redirect in response to POST transaction

Automatic redirection of a POST transaction to GET is entirely feasible, using status 302 or 303. Although the specification for 302 mandates clients to verify the redirection with the user, this mandate seems to be generally ignored.

Note, that if you’re doing an HTTPS -> HTTP redirect, the only way not to get an ‘insecure’ warning message seems to be by using a meta refresh. Going through a lynx shows that this is what Yahoo! does (interesting stuff in the trace). Too bad I didn’t think to look through lynx an hour ago.