2024-09-14 03:04:26 unjust: It actually slipped one day; it came today instead of yesterday. I still consider that good. And so far it's great. I've got the "main strokes" of my setup in place. Got Dropbox running and synched, mosh installed, this irc interface working, etc. A few little things to do, but mostly all set. 2024-09-14 03:05:00 Oh, and moved my .ssh folder over from the old system lock stock and barrel, and all the keys "just worked." I did have to delete known_hosts, though, which makes sense. 2024-09-14 03:05:22 nice KipIngram, how well is the hardware supported by the default install so faR? 2024-09-14 03:05:38 I haven't noticed anything not working yet. 2024-09-14 03:06:15 There is some labeling on the touchpad, though, and I can't decide if it's really part of the touchpad or a sticker meant to be removed. 2024-09-14 03:06:34 None of the pictures show that, so that says "sticker." I don't want to pick at it too hard and damage something, though. 2024-09-14 03:06:41 I sent them a message and asked. 2024-09-14 04:00:43 Is it a RISC-V based machine? 2024-09-14 04:03:25 begin ... while ... if ... then ... until ... else ... then ... am i going to forth hell 2024-09-14 04:06:05 if you're being greedy with flow control, maybe 2024-09-14 04:07:41 https://en.wikipedia.org/wiki/Inferno_(Dante)#Fourth_Circle_(Greed) 2024-09-14 04:09:39 nicely played 2024-09-14 04:10:54 No, it's Intel. 2024-09-14 06:22:54 One thing I haven't figured out so far is how to check the fan speed from the command line. 2024-09-14 06:23:05 The lm_sensors package on this machine doesn't display that. 2024-09-14 06:23:34 I maintain a small little window in the bottom left of my console workspace that shows me a bunch of stuff like that. 2024-09-14 06:40:19 if the fan controller has an i2c interface, it may be something exposed somewhere in /sys, like /sys/bus/i2c/devices/, if it the kernel already knows about it 2024-09-14 11:13:13 The Forth circle of hell? 2024-09-14 15:28:20 wordlists vs vocabularies: fight 2024-09-14 16:50:54 I use neither :) 2024-09-14 17:05:00 how does retro do it? 2024-09-14 17:06:07 oh that's right, you have sigil prefixes that are kind of like namespaces. so you probably just have everything reachable from a global dictionary 2024-09-14 17:47:09 dpans: @ 2024-09-14 17:47:10 <[bot]dpans > 6.1.0650 @ "fetch" CORE 2024-09-14 17:47:10 <[bot]dpans > ( a-addr -- x ) 2024-09-14 17:47:10 <[bot]dpans > x is the value stored at a-addr. 2024-09-14 17:47:10 <[bot]dpans > See: 3.3.3.1 Address alignment. 2024-09-14 18:04:51 zelgomer: yes, a single global dictionary and prefixes for namespaces 2024-09-14 19:02:47 X-Scale: I'm not sure there are RISC-V-based laptops yet? 2024-09-14 19:04:03 https://www.omgubuntu.co.uk/2024/06/framework-laptop-risc-v-motherboard-announced 2024-09-14 19:05:26 https://store.deepcomputing.io/products/dc-roma-riscv-laptop-ii-with-octa-core-cpu 2024-09-14 19:07:13 wow! what CPU is the DC-Roma thing using? 2024-09-14 19:08:08 presumably not the same StarFive JH7110, since it's 2GHz 2024-09-14 19:08:24 and 8 cores instead of 4 2024-09-14 19:11:59 https://www.cnx-software.com/2018/11/02/sifive-7-series-risc-v-cores-e76-s76-u74/#u7-core-ip-series-u74-and-u74-mc-cores says the U74 core in the JH7110 is about 2.5 Dhrystone MIPS per MHz, while http://www.roylongbottom.org.uk/Raspberry%20Pi%20Benchmarks.htm#anchor5a says the Cortex-A53 in the Raspberry Pi 3 is about 2.95 Dhrystone MIPS per MHz 2024-09-14 19:12:24 https://www.omgubuntu.co.uk/2024/06/the-worlds-first-risc-v-laptop-running-ubuntu "It’s built around a SpacemiT K1 SoC, which has 8 64-bit RISC-V cores running up to 2.0 GHz." 2024-09-14 19:14:02 but *which* RISC-V cores? They might be much slower (or faster) per cycle 2024-09-14 19:16:16 it's nice to see the SiFive cores getting some love 2024-09-14 19:16:37 T-Head had been getting all the attention 2024-09-14 19:19:16 the page says "Fans of Framework laptops have come to expect performant chips from the likes of Intel and AMD, but the StarFive JH7110 is far from hat — it’s a fairly slow, unpowered chip for driving a full-fledged desktop computing or running any particularly intensive workloads" but to me it sounds pretty decent 2024-09-14 19:19:42 "CPU – 8-core X60 RISC-V processor with single-core performance equivalent to about 1.3x the performance of an Arm Cortex-A55" https://www.cnx-software.com/2024/05/10/banana-pi-bpi-f3-sbc-spacemit-k1-octa-core-risc-v-ai-soc/ 2024-09-14 19:20:14 yeah, I just found https://docs.banana-pi.org/en/BPI-F3/SpacemiT_K1, which says "Octa-core RISC-V AI CPU,provided 50KDMIPS CPU computing power and 2.0 TOPS AI computing power.The computing power for single core in K1 is 30% ahead of ARM A55" 2024-09-14 19:21:05 50'000 Dhrystone MIPS 2024-09-14 19:21:20 3.66 DMIPS/MHz https://www.spacemit.com/en/spacemit-x60-core/ 2024-09-14 19:21:31 impressive 2024-09-14 19:22:14 and it has RVV 1.0 256-bit vector operations 2024-09-14 19:22:27 which of course are not in the Dhrystone benchmark :) 2024-09-14 19:31:23 fifty thousand VAXes in your backpack 2024-09-14 19:31:54 the RISC-V instruction set is pretty nice, although the actual physical instruction encoding is kind of bletcherous 2024-09-14 19:33:44 there aren't that many instruction word formats, but the bits are shuffled around in a seemingly deliberately sadistic way 2024-09-14 19:34:22 ARM is a bit like that too 2024-09-14 19:44:19 no, ARM is the other extreme 2024-09-14 19:45:09 Thumb not only has a zillion formats, it's state-dependent 2024-09-14 19:45:33 but all those formats have the fields in a sensible order 2024-09-14 19:45:54 and the pre-Thumb ARM instruction format, you don't even really need a disassembler, you can just read the hex 2024-09-14 20:25:00 the riscv instruction format is to make the routing easier in hardware