sF|: are motherboard with integrated radeon graphics worthy of buying these days or should i put my money on motherboard with geforce 8200
Xemanth: i'm searching motherboard with nice integrated graphics for my gf's mum
Xemanth: i have too many bad memories from ati graphics chips that i would put my money on geforce
Xemanth: but are these days ati graphics worth buyig or are they only paper weight
starkmjolk: AMD/ATI is the clear leader in integrated graphics
starkmjolk: for me it's a no-brainer
starkmjolk: both software and hardware wise
Xemanth: http://www.gigabyte.fi/Products/Motherboard/Products_Overview.aspx?ClassValue=Motherboard&ProductID=2758&ProductName=GA-MA78GM-S2H
Xemanth: is 780G good chipset :)
Xemanth: and how good is Radeon HD 3200 in linux usage
Xemanth: can i get Xv and direct rendering working and fast 2d
Xemanth: woot http://www.phoronix.com/scan.php?page=news_item&px=NjY0MA
Xemanth: http://www.phoronix.com/scan.php?page=article&item=amd_780g_linux&num=1 maybe i read this article first
shital: how can i install .rpm or .deb or tar.gz packages on Splashtop??
shital_: how can i install .rpm or .deb or tar.gz packages on Splashtop??
Kano: hi michaellarabel ,do you want xrandr output from etch? Because your testscript uses the fallback here
Kano: hi shital_
shital_: Hi Kano
Kano: did you customize your menu too
shital_: yess i did it
shital_: Kano , now i want to install Java on it and want run some programs
Kano: java is huge
Kano: 1.4?
shital_: yess
Kano: that was a bit smaller
Kano: 1.5/1.6 are twice as big
shital_: ohh ok
Kano: as browser plugin
Kano: ?
shital_: yaa, and also want to run java applications
Kano: will see what i can do
shital_: ok,
shital_: Kano, the basic Splashtop how much flash it requires to install?
Kano: well the original ssd has got only 243 mb size formatted, 255 raw
Kano: you can expect 14-15 mb for java
Kano: if you want 1.6 more than 30
shital_: ok
shital_: how i can check, what all the packages installed on Splashtop?
Kano: well every sqx file has packages
shital_: ok,
shital_: then u mean to say that, if i want to install any package then i should just do not more than pasting that sqx file into ASUS.SYS directory right?
Kano: if the package is already done yes
maligor: why would anyone want splashtop?
Kano: it is fun to play with ;)
maligor: you could just get puppylinux or some other small version
shital_: Kano, i tried ur script to create USB stick it has give this output http://pastebin.com/m257c0c52 is this correct??
Kano: shital_: well you did serveral things wrong
Kano: your stick was mounted before
shital_: yess it was
Kano: that was wrong
shital_: ok
Kano: reconnect stick
Kano: do NOT mount
Kano: rerun script
shital_: ok fine
shital_: Kano: http://pastebin.com/m6285e8d4
shital_: Can u see that?
Kano: what did you use as device?
shital_: My 2GB Pen drive
Kano: show me the line
shital_: kano : /dev/sdb1
Kano: wrong
Kano: totally wrong
Kano: use /dev/sdb
shital_: ok i expected
Kano: the script has to repartition the stick
Kano: it creates 2 partitions
Kano: a small 240 mb one + rest as data
shital_: ok
shital_: Kano: This is my script http://pastebin.com/m22610ae0 i am using
Kano: the script is fine - as it is my script, you just used the wrong device
shital_: ok fine
shital_: Kano: Now the O/p is http://pastebin.com/m22899050
Kano: shital_: show me: fdisk -l /dev/sdb
shital_: ok
Kano: maybe your stick has 2 kb not 512 byte sectors
Kano: then the stick will never boot
shital_: http://pastebin.com/m45985652
shital_: ohh
shital_: can u see that?
Kano: java will be really huge
Kano: 21 mb
Kano: hmm basically 512 byte sectors
Kano: maybe your sfdisk uses differnet options? i use the one from debian etch
shital_: ok
Kano: you use suse right?
shital_: yess
shital_: sfdik help page is http://pastebin.com/md20c193
shital_: this is sfdisk help page
Kano: you can remove the fdisk part and partiton the stick manually too
Kano: so java package ready
Kano: java 1.4 however
Kano: 21 mb...
Kano: so huge as 26 other plugins i made before
shital_: ok
Kano: btw. i use the global firefox install way, splashtop usually copies the plugins into the home, which is useless effort
shital_: kano, how i can conver any package into sqx format??
Kano: you dont convert it
Kano: alt least there is no direct way
Kano: debina packages like java use lots of symlinks into /etc/alternatives, thats what you have to create yourself
Kano: as it uses update-alternatives
shital_: ok ok
Kano: when you extract packages which dont use update-alternatives it is much easier of course
Kano: basically you extract em, then you fix the vactl file
shital_: yess i know, we can change it to different versions
Kano: do you have got coreutils .sqx?
shital_: no
Kano: you would need that for javaws
Kano: well there is no javaws in path in the sqx i sent you
shital_: Kano: this coreutils gives basically the utilities like shell, file, text utilities right??
shital_: some of these we can even get in Busybox
shital_: which is already there in Splashtop
Kano: well it is more easy to use a sqx for it
Kano: do you need javaws?
shital_: what for it is used?
Kano: ok, you dont
shital_: yess its for java web start right?
shital_: send me
Kano: ok, then i send you the file again with working javaws link
shital_: is this the new one with javaws?
Kano: yes, old ones hass javaws too,but not in path
shital_: ok
shital_: please send it to me once again
Kano: it is very unlikely that you need it
shital_: ok
Kano: cant you run the script with kanotix?
Kano: live mode would do
Kano: i dont want to figure out what your sdisk does differently
shital_: ok i will try now
Kano: http://debian.tu-bs.de/project/kanotix/preview/KANOTIX-2007-THORHAMMER-RC7.iso
Kano: http://debian.tu-bs.de/project/kanotix/preview/KANOTIX-2007-THORHAMMER-RC7.iso.md5
Kano: use that, you can install it and get my os ;)
Kano: use
Kano: sudo -i
Kano: fix-unionfs
shital_: ok
Kano: if you need to install things in live mode
shital_: is this the iso bootable of Splashtop??
Kano: that is kanotix, my own distro
Kano: can be booted from usb stick too ;)
shital_: ok ok
shital_: then i can directly right it CD or USB then i can use it right?
Kano: well use it as cd first
shital_: ok its downloading now :)
Kano: i have got a script for usb stick, it needs a partitioned stick
shital_: give it to me
Kano: next script will partition it automatically, but will need a working sfdisk, which you dont have...
shital_: ohh i have got installed know?
Kano: http://kanotix.com/files/fix/thorhammer-usb.bash
Kano: well you can try the script
Kano: use gparted to create 2 partitions on your steick
Kano: both fat32, one 800 mb and the rest as 2nd
Kano: then you can use the thorhammer usb script with first partition
Kano: it uses the iso image as input
Kano: that script even auto install freedos on stick too, for bios flashing...
shital_: ok
shital_: ur script wont partition the stick??
Kano: the thorhammer script does not
shital_: ohh ok ok
Kano: you just may need to remove one security check
Kano: [ "$(stat --format "" "$TRG")" != "floppy" ] && echo "target is not a removable device" >&2 && exit 5
Kano: in case of problems put a # in front of this like when it is not detected as removeable
shital_: ok
shital_: to do USB boot, the PC must support USB boot right?
Kano: shital_: not necessaryly, but of course it is more easy
shital_: ok
Kano: the other way is starting the kernel (and initrd if needed) from hd
shital_: yaa, that is from grub command line
shital_: ??
Kano: for splashtop you would start the ce_bz file as kernel
Kano: which you have to copy to hd of course
shital_: ok
Kano: hi michaellarabel
Kano: shital_: did you try java in vbox?
shital_: not yet
shital_: i am trying ur usb script with some other iso file
Kano: this only works with kanotix
shital_: hey but i installed Gentoo Live CD
shital_: using ur script
shital_: i got this output http://pastebin.com/m160421d4
Kano: it is not for gentoo
shital_: ok i need to change some thing then
Kano: you can try with the thorhammer ios
shital_: give me link to download it
Kano: [13:46:28] http://debian.tu-bs.de/project/kanotix/preview/KANOTIX-2007-THORHAMMER-RC7.iso
Kano: [13:46:30] http://debian.tu-bs.de/project/kanotix/preview/KANOTIX-2007-THORHAMMER-RC7.iso.md5
Kano: shital_: with java 1.4 input is somehow not possible, did you see that problem too, when used in the browser
maligor: why would anyone use java 1.4?
redeeman: that a question you don't want to know the answer to
redeeman: thats*
maligor: 1.6 or even 1.5 is much preferrable
pprkut: well, j2me is still 1.4
Kano: maligor: it si smaller
Kano: 21 mb for splashtop as sqx
Kano: is not really small,but much less than 1.6
Deanjo: curses as he just realized he forgot to save his lirc conf files before upgrading the os on his HTPC
Kano: then you get a backup file .dpkg...
redeeman: what i find most annoying is that you cannot do foreach in j2me
redeeman: i mean wtf, it's not like that would require great big changes
pprkut: yeah, that's terrible
Kano: redeeman: do you use it on a phone?
redeeman: yes
Kano: what kind of app?
redeeman: Kano: i have written an IM app for j2me
redeeman: which actually only do irc now
Kano: ah irc on phone, nice
redeeman: i made it because of the insane prices of mobile data transfer
redeeman: it uses a specialized protocol aimed at using least bandwidth possible
redeeman: which includes a special text compression algorithm i designed
Kano: any homepage for it
redeeman: nope as soon as it worked, i lost interrest in making it "release worthy"
redeeman: since it uses a non-standard protocol it naturally requires a relay server which connects to real irc server
Kano: hmm right
redeeman: the compression algorithm uses statistical analasys to adapt to compressing words best possible, according to which words are most used ;)
redeeman: it's quite awesome
redeeman: Kano: when i made it, my mobile operator wanted 1.5 eur /mb
Kano: and now
redeeman: half
redeeman: which is still horrible
Kano: no flatrate?
redeeman: yeah for 40 eur/month i get 10GB transfer
redeeman: but since i only use like... at most 1mb/month
redeeman: i don't very much feel like doing that
maligor: redeeman, only 10?
redeeman: yes
maligor: 40 eur?
redeeman: yes
maligor: expensive
redeeman: yes
maligor: 5Mbit costs 35eur a month here
maligor: mobile data package that is
redeeman: unlimited traffic?
maligor: never seen any mention, so I guess it is
redeeman: i wouldn't be so sure :P
maligor: they'd be misleading the consumer if they didn't state it clearly
maligor: and that'd mean they'd get into deep trouble with the government
redeeman: it's not like they are very afraid of doing that
maligor: well, here they are
redeeman: if you have it, you should try
maligor: I don't really use data
maligor: the cheapest one is 10eur/month for 384kbit/s on mobile
Kano: so now i have got java 1.6 as plugin, huge 35 mb size...
maligor: which is a fantastic price for mobiles
maligor: even 128kbit would be enough for a mobile tho
maligor: they don't really have quotas here on normal broadband either
redeeman: we don't have bandwidth limit on normal broadband
redeeman: or well.. traffic limits*
maligor: jeez, the 790GX has 128mb ddr3 integrated?
redeeman: maligor: will it have fast usb and sata?
redeeman: the mobos
maligor: I expect so
maligor: SB750
redeeman: Deanjo: ?
Deanjo: yes?
redeeman: SB750, know anything about that?
Deanjo: Ya benched one yesterday, still shitty NCQ and USB
redeeman: ah that was that one :|
redeeman: i guess no amd for me then
redeeman: what if one disables NCQ, does it work then?
Kano: ncq is only for sata..
redeeman: yes i know
redeeman: and i meant, if one disables ncq, does sata work fast
Deanjo: Nope
Deanjo: not on write operations anyways
Deanjo: :P
Deanjo: read operations are ok on the SB750
redeeman: well i need both to be ok
Ivanovic: Deanjo: what does crappy mean?
redeeman: as i recall from yesterday, 45MB/s compared to 100
Ivanovic: that is: how much is the hd able to get with ncq
Ivanovic: which test/benchmark?
Deanjo: Crappy = 45% of the speed of a 780a
Deanjo: take your pick of benchmarks, hdtach, hdparm, iozone...
Deanjo: I see techreport found the same issue on their review today http://techreport.com/articles.x/15256/9
redeeman: it's really shitty
redeeman: i would gladly buy amd, but i just cannot do it under those conditions
Deanjo: it's been the Achilles heel of all the AMD chipsets
Deanjo: That and high USB cpu usage
redeeman: considering how cheap it is to buy intel, which WORKS, i don't see how amd can seriously expect people to buy their stuff
Deanjo: Well it's only the AMD SB that have the issue. Broadcom and Nvidia chipsets are right up there (if not a bit faster) then the intels
redeeman: yeah but theres no, NO way im buying nvidia, not even if hell froze over, unfroze, then evaporated into clouds, then rained down, and froze again
Deanjo: I've ran into chipset bugs in pretty much every chipset manufacturer, worse being via of course.
redeeman: i'd almost rather have via than nvidia
Exopaladin: if the ati drivers actually have everything implemented so that it doesn't cause any wine issues I'll probably end up with ati for my next card
redeeman: yeah well :P i'd like to buy a 4850 now, even if theres only proprietary drivers now, but the drivers are worse than nvidias, so NO
redeeman: so i bought an X1950 on ebay instead, which i, ironically, haven't had time to actually use yet :(
Deanjo: heh, seems like the "Foxxcon" bug was a AMI bug as they found the same issues in Asus and MSI motherboards.
Deanjo: http://izanbardprince.wordpress.com/2008/07/28/foxconn-says-acpi-issues-are-amis-fault-is-having-them-repair-the-code/
vadi2: If anyone is an upstream dev: http://ur1.ca/1yc
jramskov: redeeman: I thought ATI's new proprietary drivers had been improved a lot?
redeeman: jramskov: they may well have, but lots of stuff won't work
jramskov: :(
jramskov: I expect to buy a 4870 at some point
redeeman: i expect ill get to skip the 48xx unfortunately
redeeman: a shame, since i was more than ready to buy it
redeeman: if it werent for the unfortunate little issue of it not actually being compatible with stuff
mha: http://mhakali.deviantart.com/art/Katrina-94124342
Deanjo: lol, what a dumbass, http://gizmodo.com/5034122/guy-buys-999-im-rich-app-discovers-hes-just-dumb
redeeman: yes, he bought an iphone?! wtf?! how stupid is he? :)
Deanjo: I couldn't couldn't care less about phones. Just give me one that makes calls dammit
Deanjo: I've got computers for all that other crap
Deanjo: This gets me more excited, http://dkc1.digikey.com/us/mkt/beagleboard.html
jramskov: redeeman: sorry (got distracted by phone and more) - I'll be building a new desktop this year with either a intel quad penryn or maybe a nehalem (depending on when that's released) and I want to support AMD/ATI for their change @ open source drivers, so an 4870 card it is.
redeeman: jramskov: well... you just have to know what you are getting yourself in to, expect that the closed drivers won't work for alot of things
Deanjo: the opensource drivers don't work on a lot of things too
jramskov: I'll probably be using the open source drivers
Deanjo: Let's put it this way, by the time the opensource drivers are good enough for things like gaming and such, there will be another couple of generations of cards produced
Deanjo: and if your not into gaming, your wasting money on a highend card
redeeman: jramskov: do keep in mind that you won't get 3d on R600+ yet
redeeman: and it will probably be a while
Deanjo: my guess another 12-18 months
redeeman: fortunately R500 appears to work well
jramskov: I know - will not really be needing it either
michaellarabel: Deanjo: Hi
Kano: hi michaellarabel ,do you want my xrandr output? your php script uses fallback here
michaellarabel: That is from the xrandr that ships with Debian Etch?
Kano: sure
Kano: xorg 7.1.1
Kano: kanotix
michaellarabel: I have a fix already, but one second in the middle of typing an email.
Kano: btw. i have java 1.6 (and 1.4) for splashtop, just a bit huge *g*
Kano: 35 mb
Kano: maybe asus should have used a 512 mb ssd ;)
Kano: for usb stick or hd this does not matter,as you can use a bigger first partition
michaellarabel: Kano, http://phorogit.com/index.php?p=phoronix-test-suite.git&a=commitdiff&h=7d6e126284eeff61994ed253768a8d4e4782affc&hb=070ac56fdbf320998f401b4f9b93531b5b463124 should fix it
Kano: well i used the test.php in the board, did not use pts lately
michaellarabel: okay, np I am confident this fixes it without causing regressions for current xrandrs
Deanjo: Dammit, once your on a MS spam list your on forever
michaellarabel: Deanjo: Do you still have that lcdproc module you mentioned before for printing the command output from the terminal to the display? I want to write a PTS module for LCDproc support but it's been quite a while since last using lcdproc
Deanjo: It wasn't a lcdproc module, it was a g15daemon module
michaellarabel: Oh, how does the G15 operate on Linux?
Deanjo: really well
Deanjo: Full functionality of the keys with the daemon and lcd display works fine
michaellarabel: But does it create say /dev/g15lcd that you can then write to or do you know how that daemon is working?
Deanjo: Ya that's basically how it's done
Deanjo: I wish I could point you to the page that describes it in detail but it seems the pages @sourceforge are getting revamps
Deanjo: revamped
redeeman: Deanjo: is it a kernel driver doing that?
redeeman: and is it inkernel?
redeeman: anyway, i have a das keyboard ;)
Deanjo: it's not in the kernel
Deanjo: for the extra key support it depends on uinput though
redeeman: i personally don't really care for those fancy input devices
Deanjo: I actually use the extra keys for a lot of app hotkeys
Deanjo: Really handy for xchat when programmed to boot trolls :P
Deanjo: http://www.phoronix.com/forums/showthread.php?t=11889 <----- the ATI chipsets get thumped pretty bad in write performance
Kano: Deanjo: whats hdparm -t of that drive?
Deanjo: Timing buffered disk reads: 344 MB in 3.00 seconds = 114.52 MB/sec
Kano: hmm impressive
Kano: is it a highspeed drive or is your timer going wrong?
Deanjo: Just a plain old 7200 RPM Sata II
Kano: forget it
Kano: must be a timer error
redeeman: my new seagate disks have 115MB/s continously for atleast the first 100gb
Kano: but not for old ones
Deanjo: Copying from one drive to the next easily sustains 100MB +
Deanjo: on large copies
Deanjo: (GB at a time)
Kano: when it is old then it must be a too slow timer
redeeman: dd: 1048576000 bytes (1.0 GB) copied, 9.12921 s, 115 MB/s
redeeman: and it's 115 atleast up to 100gb :)
Kano: redeeman: he said his drive is _old_
redeeman: yeah :)
redeeman: mine fortunately arent
redeeman: please do a: d if=/dev/sdc of=/dev/null bs=1M count=10000 iflag=direct
Kano: my 500 gb samsung drive does not give more than 80 mb/s
Kano: and it is not that old
redeeman: i can test a 750 and 1tb samsung F1
Kano: SAMSUNG HD501LJ
redeeman: 10485760000 bytes (10 GB) copied, 94,9086 seconds, 110 MB/s <-- samsung HD103UJ
redeeman: interrestingly enough
Kano: if you see more than 100 for old drives it is really impossible
redeeman: the first ~100mb is like 50MB/s
redeeman: then it goes like 120MB/s some while
redeeman: and then goes down more linearly
Kano: hdparm -t tests read on first mb
redeeman: yeah but i don't care about first mb particularly
redeeman: im more interrested in a ~10gb test
Kano: sure, but his values are bad
Kano: for a really new drive possible, impossible for old drives
Deanjo: My seagate 7200.11's give slightly better numbers
redeeman: both seagate ES.2 1tb and 500gb appears to be same speed
Kano: http://www.hardware.info/en-US/productdb/viewbenchmarks/
Kano: his drive has a real speed of 65 mb/s
Deanjo: 10485760000 bytes (10 GB) copied, 86.25 s, 122 MB/s <---- seagates
Kano: everything else is cases by bad clock
redeeman: i have no non-enterprise(hah) seagates
Kano: Deanjo: you are benching completely wrong values!
redeeman: wow you were right kano
redeeman: 10485760000 bytes (10 GB) copied, 120,824 seconds, 86,8 MB/s <-- HD753LJ
Kano: Deanjo: clock=pmtmr
Deanjo: Same drives dump back @ 45 mb/s on a SB750
Kano: use that kernel option!
redeeman: i think i use tsc as clocksource
Kano: Deanjo: your clock is wrong how often sould i say that
Kano: you dont need to bench anything with that system, all values are wrong
Kano: 1/3 too slow clock you must have
Kano: therefor all values 33% faster than normal
redeeman: that would appear to somewhat invalidate the SB750 claims relative to that system
Deanjo: So explain how realtime copy to copy times are still drastically faster
Kano: Deanjo: your timer!!!
Kano: you can not bench anything
redeeman: well it's easy, Deanjo
Deanjo: Using external clock
redeeman: that seagate bench
redeeman: did it really take 120 seconds?
Deanjo: not even on the system
Deanjo: yes
redeeman: humh
Kano: Deanjo: try differnt clock option until you get about 60-70 with hdparm -t and your maxtor drive
Kano: befor discussing is useless
redeeman: i suggest use external clock
redeeman: and do a 10gb dd
redeeman: no matter the clocksource the kernel uses, dd will still do the same amount of work
Deanjo: brb
Deanjo: Timing buffered disk reads: 348 MB in 3.01 seconds = 115.48 MB/sec
Deanjo: that's with clock=pmtmr
Kano: hmm
Kano: well try differnt options then
redeeman: lol
Deanjo: Hell I get roughly the same speeds in windows as well
Kano: maybe a hardware problem
Deanjo: Then the only hardware that doesn't have the problem is the ATI chipsets
Kano: you can be sure when you replace the board, the hd speed will not drop so extremely
Kano: a few mb more or less with hdparm at max
Deanjo: As same speeds are consistant on intel and nvidia chipsets
Kano: maybe you are the only one with a super fast maxtor drive ;)
Deanjo: I've got 2 of them and they perform within 2-3% of each other
Deanjo: Timing buffered disk reads: 166 MB in 3.01 seconds = 55.11 MB/sec <--- same drive on a SB600
Kano: Deanjo: thats more logical
redeeman: HashMap>>> superHashMap =
redeeman: new HashMap>>>();
redeeman: priceless :D
Deanjo: SB600 4 GB iso copy = about 70 seconds, on the Nvidia about 40 seconds
Kano: Deanjo: your hdparm values with your nvidia board are impossible
Deanjo: Clocked using my watch, same disto, same procs, same ram
Kano: maxtor does not exist anymore and the old drives have never been so fast
Deanjo: The seagates do the same Kano
Kano: well until you get a more normal value dont benchmark hds
Kano: you bench only crap
Deanjo: Kano as I said before, the only chipset that drops is the ATI
Kano: and it is correct then
Deanjo: So then how do you explain the copy discrepancy?
Deanjo: Using a clock on the wall
Kano: write performance can be better using better caching/ncq, not sure if 30s are possible
Kano: but first corect read performance
Kano: did you run sync in your copy script?
Kano: that forces all data written from cache
Deanjo: just a straight cp command
redeeman: that's useless then
Kano: well without sync you bench the cache from linux
Kano: clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
Kano: [Deprecated]
Kano: Forces specified clocksource (if available) to be used
Kano: when calculating gettimeofday(). If specified
Kano: clocksource is not available, it defaults to PIT.
Kano: Format: { pit | tsc | cyclone | pmtmr }
Kano: it seems that clocksource is now better
Kano: http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.26.y.git;a=blob_plain;f=Documentation/kernel-parameters.txt;hb=26b33d467b50e0a721cd69610a5ddc38c3279f31
redeeman: cyclone, wtf
redeeman: cat /sys/devices/system/clocksource/clocksource0/available_clocksource
redeeman: im using tsc
Kano: cyclone on IBM x440
Kano: redeeman: good idea
redeeman: tsc hpet acpi_pm jiffies <-- those available
redeeman: which is best?
redeeman: hpet?
Kano: well bios can be set to hpet 64bit even
redeeman: i have that
Kano: win 32 bit skips 32 bit of it ;)
redeeman: haha
redeeman: lol that's why the bios says hpet 64 is for vista 64bit
redeeman: lol morons
Kano: maybe it is faster when using 32-bit mode,no idea
redeeman: but is hpet better than the cpu's tsc?
Kano: i dont think you can bench the difference ;)
Kano: i only can tell you want win uses ;)
Kano: what
redeeman: by that available information, i can only conclude that something different is the correct aproach
Kano: i guess on a good system all options work
Kano: cat /sys/devices/system/clocksource/clocksource0/current_clocksource
Kano: acpi_pm
Kano: here...
Kano: Deanjo: maybe you can try changeing it on the fly without reboot
Kano: hmm does not seem to work here
Kano: at least you know all possible options now
redeeman: shit
redeeman: it appears a disk has gone baserk in one of my servers
redeeman: and reports 46 degrees hot