| --- Log opened Sun Dec 24 00:00:07 2017 |
| 00:01 | <@himi> | I'd rather keep old kernels around until I manually decided to clean them up than have them cleaned up automatically |
| 00:03 | <@gnolam> | McMartin: it does not. |
| 00:04 | <@himi> | I do kind of wish Ubuntu would release fewer kernel upgrades - it's rather hard to manage them in a server installation, particularly if it's a server installation you can't reboot freely |
| 00:17 | <&McMartin> | gnolam: Given that I literally just did this thing, with that command, with that level of deletion, I must ask you to clarify that statement. |
| 00:17 | <@gnolam> | "It does not work on any installation I have tried it on" |
| 00:18 | <&McMartin> | Okay. Stock Xenial over here, AFAIK. Several iterations of kernel, image, headers, etc, picked out, etc. |
| 00:18 | <&McMartin> | And the entries do seem to have vanished from the boot menu. |
| 00:19 | <&McMartin> | (I was trying to determine if you were asserting that deleting the packages doesn't actually free up any space) |
| 00:23 | <@himi> | . . . are you guys talking at cross purposes? |
| 00:23 | <@himi> | McMartin appears to be saying that apt-get autoremove takes out old kernels, gnolam seems to be saying that fedora doesn't automatically clean out old kernels |
| 00:24 | <@gnolam> | No, I'm saying that apt-get autoremove doesn't remove old kernels on my Ubuntu VMs, and never has. |
| 00:26 | <@himi> | That's weird - I literally just ran that on my laptop now and it deleted half a dozen old kernels |
| 00:27 | <@himi> | It won't delete the running kernel, or I believe newer kernels if you've got an old one running and newer packages installed but not used, but it definitely deletes old unused kernels |
| 00:33 | <&[R]> | <McMartin> Both of those statements are, IME, false <-- when'd they start shipping updates that don't require reboots? |
| 00:33 | <&[R]> | <TheWatcher> Yeah, win10 pro I'm on at the moment, 10d 2:37m uptime <-- 10d < 2w |
| 00:36 | <&[R]> | Also apparently it's actually monthly, seemed more frequent than that |
| 00:36 | <&McMartin> | [R]: Like, Vista? At the very latest? |
| 00:36 | <&McMartin> | And yeah, the big monthly updates will usually patch the kernel and require a reboot, but updates to basically any other system won't. |
| 00:36 | <&[R]> | Every update I've had on Windows has required a reboot |
| 00:36 | <&McMartin> | You may have disabled all the ones that don't. |
| 00:36 | <&McMartin> | The most common ones that don't are Windows Defender updates. |
| 00:37 | <&McMartin> | If you let it manage your sound/graphics drivers, those don't either post-Vista because they can hot-swap |
| 00:37 | <&[R]> | I just keep the defaults until the reboot requirements annoy me enough to disable it altogether |
| 00:38 | <&McMartin> | It is entirely possible that you are not paying attention to updates that don't require reboots, then. |
| 00:38 | <&McMartin> | W10 is louder about those |
| 00:38 | <&McMartin> | But yes, "~monthly" is the range I would expect |
| 00:38 | <&[R]> | It's "every second tuesday" |
| 00:39 | <&McMartin> | Slightly faster if you're aggressively altering aspects of the system, slightly lower if there's a slow month for patches |
| 00:39 | <&[R]> | Occasionally there's out-of-band updates, but those are pretty rare |
| 00:39 | <&McMartin> | Yep. "Patch Tuesday" |
| 00:39 | <&McMartin> | Yeah |
| 00:39 | <&McMartin> | Out-of-band updates also often do not require reboots |
| 00:40 | <&McMartin> | So if you aren't on Win10 or disabled notifications from Windows Update, you'll only know they even happened if you consult the update logs. |
| 00:40 | <&McMartin> | Because they'll show up as the Windows equivalent of a service-restart. |
| 00:41 | <&[R]> | But my experience with Windows is it wants 2-3 updates a month, occasionally a few days from each other. Though my means of determining this are: get on jobsite, open laptop (which I keep hibernated), and Windows does a long boot to install the updates leaving me waiting an hour on site to do the work I'm supposed to be doing. |
| 00:41 | <&[R]> | And that's with W10 |
| 00:42 | <&[R]> | W7 seemed to be quite a bit less douchey in that regard |
| 00:43 | <&McMartin> | That sounds like something's shitting in your driver space that normally doesn't. |
| 00:43 | <&[R]> | There was even a point where it was 2.5 hours, which is beyond excessive. |
| 00:43 | <&[R]> | Maybe |
| 00:43 | <&[R]> | This is with two seperate laptops |
| 00:43 | <&McMartin> | (Possibly relevant; My Win10 machine is a desktop and built from parts.) |
| 00:44 | <&McMartin> | (And as such has zero OEM crapware) |
| 00:44 | <&[R]> | One upgraded from W7, the other OEM'd with W10 |
| 00:44 | <&McMartin> | Yep. My desktop experience with W7 and W10 on that system have been roughly similar. |
| 00:44 | <&McMartin> | And it did both upgrade and OEM since its boot drive died the death after 8 years. |
| 00:45 | <&McMartin> | My core laptop is a 2012-era System76 laptop running Xenial. |
| 00:46 | <&McMartin> | (And the Fedora system is a NUC that has only recently been bequeathed a head.) |
| 00:46 | <&McMartin> | (Which started out 'beheaded' but that would mean the opposite, wouldn't it.) |
| 00:47 | <&McMartin> | And yeah, uh, on that thing |
| 00:47 | <&McMartin> | Not even installing the FCU edition took an hour |
| 00:47 | <&McMartin> | HDDs are slow, but I don't think they're *that* slow |
| 00:48 | <&[R]> | I really wish I were exagerating |
| 00:48 | <&[R]> | But 1130 to 1230 is an hour. And 0800 to 1030 is 2.5. |
| 00:48 | <&McMartin> | Yeah, I'm thinking more "is that drive healthy" |
| 00:49 | <&McMartin> | High Sierra took something like 45 minutes on an SSD, but that included a complete nondestructive change of the drive's filesystem |
| 00:49 | <&[R]> | Only issues it had were with Windows Updates, everything else was speedy quick (which is unsual for Windows) |
| 00:50 | <&[R]> | The one that is W10 OEM doesn't have overly long updates, but I get enough "we've just updated, all your shit is closed" instances to be annoyed |
| 00:51 | <&McMartin> | Windows Update had a hilarious performance bug where stuff took exponential time to work out what to apply, but I was under the impression that was another bug they'd fixed in Vista. |
| 00:51 | <&[R]> | Though, that system has had a few instances where the HDD got hammered making everything slow to a crawl. |
| 00:51 | <&McMartin> | (Which usually means 'fixed in 7' because fuck Vista) |
| 00:51 | <&[R]> | I don't think they've made hardware that can run Vista yet |
| 00:52 | <&McMartin> | I prefer to think of Vista as the paid beta for 7. |
| 01:08 | | Kindamoody is now known as Kindamoody[zZz] |
| 02:02 | | gnolam [quassel@Nightstar-hsn6u0.cust.bahnhof.se] has quit [[NS] Quit: http://quassel-irc.org - Chat comfortably. Anywhere.] |
| 04:41 | | Vornotron [Vorn@Nightstar-1l3nul.res.rr.com] has joined #code |
| 04:41 | | mode/#code [+qo Vornotron Vornotron] by ChanServ |
| 04:54 | | Degi [Degi@Nightstar-hn99vf.dyn.telefonica.de] has quit [[NS] Quit: Leaving] |
| 05:03 | <&[R]> | In bash, is there a simpler way to assign the value of the tab character to a variable than IFS=$(echo -e "\t")? |
| 05:05 | <&[R]> | Apparently it's $'\t' |
| 05:09 | | Vornlicious [Vorn@Nightstar-fkpnkj.sub-174-211-0.myvzw.com] has joined #code |
| 05:13 | | Vorntastic [Vorn@Nightstar-1l3nul.res.rr.com] has quit [Ping timeout: 121 seconds] |
| 07:05 | | celticminstrel is now known as celmin|sleep |
| 07:06 | < Vornlicious> | A math puzzle: consider the puzzle game "kakurasu", in which you have an n*n grid where some of the cells are "lit", and you must figure out which ones those are, given the sums of (1-based) column indexes for lit cells in each row, and vice versa |
| 07:07 | < Vornlicious> | Is there a pair of grids with the same set of clues? |
| 07:17 | < Vornlicious> | ... Yes there is: for 3x3, the clues are all 3s, and you can have either (1,1),(1,2),(2,1),(2,2), and (3,3) or its complement |
| 07:18 | < Vornlicious> | Similar things work for every size where triangle(n) is even. |
| 07:21 | < Vornlicious> | (and for those that don't you can just make the biggest row/column be anything) |
| 07:38 | | Vornotron [Vorn@Nightstar-1l3nul.res.rr.com] has quit [Ping timeout: 121 seconds] |
| 07:51 | | * McMartin sets up an application shell to test out the timing for the old City Connection shenanigans from earlier this month. |
| 08:12 | < Vornlicious> | \o/ |
| 08:26 | <&McMartin> | City Connection: *confirmed* |
| 08:26 | <&McMartin> | I'm doing a couple of trivial timing crunches just to make sure I'm not pointlessly wasting time, but that's at the level of "not being clever, just careful" |
| 08:29 | < Vornlicious> | I love it when I'm right |
| 08:30 | <&McMartin> | https://www.dropbox.com/s/2yfzfurcd5ecs95/vramblit.nes?dl=1 |
| 08:30 | <&McMartin> | This link won't last forever; I'm going to make it interactive so you can push it until it explodes |
| 08:34 | <&McMartin> | Which appears to happen at six lines. |
| 08:35 | < Vornlicious> | Rock on (I'll grab it when I get home) |
| 08:37 | <&McMartin> | It'll last that long~ |
| 08:38 | <&McMartin> | But will probably go away when the final version is published (which with luck will be part of the Bumbershoot 2017 compilation) |
| 08:41 | <&McMartin> | pfffff |
| 08:41 | <&McMartin> | I'm going to need to build that 2017 compilation on my RPi. |
| 08:41 | <&McMartin> | Since really creating the zipfile right means adding the Acorn filesystem metadata for the relevant RISC OS projects. |
| 08:45 | <&McMartin> | 2017 programs: ZX81 and ZX Spectrum Lights-Out ports; Smoking Clover for DOS; Coast to Coast, Statue of Liberty type-in port and the directory editor for C64; Colorbar demo and Target Acquired port for RISC OS; City Connection blit test for NES. |
| 08:46 | <&McMartin> | A new record for platforms targeted. -_- |
| 08:47 | <&McMartin> | (Me, last February: "2016 had completed projects for five platforms: C64, NES, Apple II, DOS, and the Z-Machine. That's a new record and one I'm unlikely to beat." |
| 08:48 | <&McMartin> | ) |
| 08:50 | < Vornlicious> | Things you should not write: "unlikely to beat" |
| 09:15 | | Kindamoody[zZz] is now known as Kindamoody |
| 09:22 | <&McMartin> | Hey, if it means things that are awesome... |
| 09:26 | <&McMartin> | Also for the second year running my post about making the CGA card jump and do tricks is my most popular single post by nearly a factor of two |
| 09:28 | <&McMartin> | For reasons I'm not at all clear on, "ZX81: On to Machine Code" is the most popular post I've had this year. |
| 09:28 | <&McMartin> | (That was posted this year) |
| 09:39 | <&McMartin> | Also, I forgot to charge time for Sprite DMA |
| 09:39 | <&McMartin> | Once you do that, 4 lines is the absolute maximum you can blit per frame. |
| 09:40 | < Vornlicious> | Man |
| 09:41 | <&McMartin> | Resolution for Bumbershoot 2018: Use some fonts that aren't Halogen |
| 10:15 | | gnolam [quassel@Nightstar-hsn6u0.cust.bahnhof.se] has joined #code |
| 10:15 | | mode/#code [+o gnolam] by ChanServ |
| 10:23 | < Vornlicious> | Perhaps use Rare Earth Metal instead |
| 10:24 | < Vornlicious> | Though I will say, considering your computer naming scheme, that is quite appropriate |
| 10:34 | <&McMartin> | That's why I named it that~ |
| 10:36 | <&McMartin> | (Halogen is my modification of the old art-deco-era font "Stop") |
| 10:38 | < Vornlicious> | Oh I thought you'd gotten it from somewhere |
| 10:58 | | Kindamoody is now known as Kindamoody|afk |
| 11:27 | | mac is now known as macdjord|slep |
| 13:05 | | Degi [Degi@Nightstar-hn99vf.dyn.telefonica.de] has joined #code |
| 14:14 | | celmin|sleep is now known as celticminstrel |
| 14:47 | | Vornlicious [Vorn@Nightstar-fkpnkj.sub-174-211-0.myvzw.com] has quit [[NS] Quit: Bye] |
| 14:47 | | Vorntastic [Vorn@Nightstar-1l3nul.res.rr.com] has joined #code |
| 14:50 | | Vornicus [Vorn@Nightstar-1l3nul.res.rr.com] has joined #code |
| 14:50 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 15:09 | | gnolam_ [quassel@Nightstar-hsn6u0.cust.bahnhof.se] has joined #code |
| 15:11 | | gnolam [quassel@Nightstar-hsn6u0.cust.bahnhof.se] has quit [Ping timeout: 121 seconds] |
| 16:57 | | gnolam_ [quassel@Nightstar-hsn6u0.cust.bahnhof.se] has quit [[NS] Quit: Gone] |
| 20:37 | | Vornicus [Vorn@Nightstar-1l3nul.res.rr.com] has quit [Ping timeout: 121 seconds] |
| 21:36 | | Kindamoody|afk is now known as Kindamoody |
| 21:44 | | McMartin [mcmartin@Nightstar-rpcdbf.sntcca.sbcglobal.net] has quit [Operation timed out] |
| 21:49 | | McMartin [mcmartin@Nightstar-rpcdbf.sntcca.sbcglobal.net] has joined #code |
| 21:49 | | mode/#code [+ao McMartin McMartin] by ChanServ |
| 22:40 | | celticminstrel is now known as celmin|xmas |
| 23:56 | | Kindamoody is now known as Kindamoody[zZz] |
| --- Log closed Mon Dec 25 00:00:09 2017 |