2021-03-08 20:25
egg changed the topic of #kspacademia to:
https://git.io/JqLs2 | Dogs are cats. Spiders are cat interferometers. | Document well, for tomorrow you may get mauled by a ネコバス. | <UmbralRaptor> egg|nomz|egg: generally if your eyes are dewing over, that's not the weather. | <ferram4> I shall beat my problems to death with an engineer. | We can haz pdf | Logs:
https://esper.irclog.whitequark.org/kspacademia
01:00
egg|anbo|egg__ has joined #kspacademia
01:02
egg|anbo|egg has quit [Ping timeout: 189 seconds]
01:35
e_14159 has quit [Ping timeout: 189 seconds]
01:36
e_14159 has joined #kspacademia
02:24
<
whitequark >
why... does that tweet not load
02:27
<
raptop >
oh, you deleted/redrefated
02:30
<
whitequark >
it's a new tweet
02:30
<
whitequark >
not an old one
02:32
<
egg|anbo|egg_ >
!wpn whitequark
02:32
* galois
gives whitequark a XSLX with a vstab attachment
02:32
* egg|anbo|egg_
pets whitequark with a cat
02:36
<
raptop >
It's probably fine to use an excel file as a vertical stabilizer
03:13
egg|anbo|egg_ has quit [Remote host closed the connection]
09:31
egg|anbo|egg has joined #kspacademia
09:32
egg|anbo|egg has quit [Remote host closed the connection]
10:42
egg|anbo|egg has joined #kspacademia
12:07
egg|anbo|egg has quit [Remote host closed the connection]
12:09
egg|cell|egg has quit [Ping timeout: 204 seconds]
13:32
egg|anbo|egg has joined #kspacademia
14:07
<
galois >
title: [astro-ph/0307029] Nuclear-Powered Millisecond Pulsars and the Maximum Spin Frequency of Neutron Stars
14:08
<
mofh >
turns out there's an upper-bound (when the centrifugal pseudo-force & the self-gravity equal, which seems to be around 0.3 milliseconds/pulse).
14:23
egg|anbo|egg has quit [Remote host closed the connection]
14:48
egg|anbo|egg has joined #kspacademia
14:58
egg|anbo|egg has quit [Remote host closed the connection]
15:03
egg|anbo|egg has joined #kspacademia
15:04
egg|anbo|egg has quit [Remote host closed the connection]
15:04
egg|anbo|egg has joined #kspacademia
15:11
<
egg|anbo|egg >
!acr -add:EMIRS Emirates Mars InfraRed Spectrometer
15:11
<
galois >
Definition added!
15:35
egg|anbo|egg has quit [Remote host closed the connection]
16:52
<
egg|anbo|egg__ >
!wpn
16:52
* galois
gives egg|anbo|egg__ an owl
16:53
<
raptop >
The owl takes your loved ones hostage, and forces you to learn a language
16:54
<
raptop >
(Also, EMIRS is very much in line with astronomers naming things)
17:39
<
raptop >
!choose radioshack|radioshark
17:39
<
galois >
raptop: Your options: radioshack, radioshark. My choice: radioshark
17:43
egg|anbo|egg has joined #kspacademia
17:47
<
galois >
title: [2105.08722] ENCORE: Estimating Galaxy $N$-point Correlation Functions in $\mathcal{O}(N_{\rm g}^2)$ Time
17:52
egg|anbo|egg has quit [Remote host closed the connection]
18:07
<
galois >
title: [2105.09338] Stars with Photometrically Young Gaia Luminosities Around the Solar System (SPYGLASS) I: Mapping Young Stellar Structures and their Star Formation Histories
18:10
egg|anbo|egg has joined #kspacademia
18:42
egg|anbo|egg has quit [Remote host closed the connection]
18:59
<
galois >
title: [2012.15226] Isostasy with Love: II Airy compensation arising from viscoelastic relaxation
19:43
egg|anbo|egg has joined #kspacademia
19:48
* egg|anbo|egg
purrs at whitequark
19:53
<
egg|anbo|egg >
gaaaaaaaaaaah my benchmarks say that andps is slower than serially-dependent mulsd-addsd-subsd in the critical path, WTF aaaaaaaaaaaaaa
19:54
<
egg|anbo|egg >
(meanwhile IACA says it should be 10 cycles faster, and so do the agner fog tables)
19:54
<
egg|anbo|egg >
whyyyyyyyyyy
19:57
egg|anbo|egg has quit [Remote host closed the connection]
20:01
egg|anbo|egg has joined #kspacademia
20:04
egg|anbo|egg has quit [Remote host closed the connection]
21:13
<
egg|anbo|egg__ >
iximeow: do you happen to have any idea how latency why wat
22:39
<
iximeow >
oh that seems.. odd
22:41
<
iximeow >
i would also expect it to be ~10 cycles faster. it's also slightly easier to schedule (uops.info says it can go on port 5, while mulsd/addsd/subsd are ports 0 or 1)
22:42
<
iximeow >
egg|anbo|egg__: reg-reg operands? or reg-mem?
22:43
<
iximeow >
(or, reg-reg with an operand recently loaded from memory?)
22:52
egg|anbo|egg has joined #kspacademia
22:56
egg|anbo|egg has quit [Ping timeout: 189 seconds]
23:05
Iskierka has quit [Ping timeout: 204 seconds]
23:18
egg|anbo|egg has joined #kspacademia
23:21
<
egg >
iximeow: there is one constant in both cases, either memory operand or loaded beforehand then reg-reg
23:21
<
egg >
when I poke the compiler into preloading it for the and it makes things... slower
23:23
<
egg >
(memory = L1 almost certainly, benchmarking the function in a loop, feeding its result into its input for total latency)
23:24
<
egg >
(Sandy Bridge fwiw)
23:57
egg|anbo|egg has quit [Remote host closed the connection]