HNHacker News
TopNewBestAskShowJobs

noxvilleza

317 karma · joined November 11, 2016

E: bensteenhuisen [@] {google's mail site} . com

Technical Director @ Team Liquid. Live in Berlin, sometimes around Utrecht! I'm interested in statistics in competitive video gaming, mostly Dota 2; as well as mathematics, film, and trivia. I run datdota.com and windrun.io

submissionscomments
noxvilleza··on No, QuestDB is not Faster than ClickHouse
Oh dear. I did a brief search for 'adage on benchmarking' and only saw Rugg/Feldman benchmarks.
noxvilleza··on No, QuestDB is not Faster than ClickHouse
Is there an existing named adage for something like "if one creates a benchmark in order to rank general performance of some products, some of those products will ultimately sacrifice general performance in order to optimize for that benchmark"?
noxvilleza··on Scotland starts renewed case for independence
Obviously the status quo within the UK has changed a lot since the 2014 Scottish independence vote - the key aspect being Brexit. This change was sort of crucial to many of the key talking points in the vote: EU CAP subsidies for farmers, sole EU membership (now from a different prior state), border controls, etc.

The prior vote was relatively close (55/45) so it's reasonable to expect that 200k people might've flipped to the 'Yes' camp.

noxvilleza··on Getting a vanity phone number with four consecutive digits
At first I thought you might've slightly leaked your phone number here, but there's surprisingly many 10 digit prime numbers with a triple-7 in the middle (and obviously that inner 777 can't be next to a 7 else you'd call it a quadruple-7 [etc]), 1905394.

If you filter on only valid area codes, the number is a bit smaller (620350), and if you exclude Toll Free, Reserved, and Government area codes it's down to 567277.

noxvilleza··on Why isn’t there a decent file format for tabular data?
Unsure if it was added after your comment, but there is a brief comment on parquet (in that it's binary - so it can be annoying to view/edit).
noxvilleza··on Two workers are quadratically better than one (2020)
There's also some subconscious effect - you rarely remember when queues are fair to you, but you always remember when you get 'queue-screwed'.
noxvilleza··on End the Streaming Struggle with Plex
I basically never see Plex adverts (and I only use local streaming really). Where do you normally see them?
noxvilleza··on Winning Wordle
Even the worst starting word is guaranteed <= 1 average guess more than the optimal strategy though!
noxvilleza··on Some folks asked how our Super Bowl ad came to be, here is the quick back story
I thought that included the cost of the ~20 second broadcast cost, but now thinking about it that seems really low.
noxvilleza··on Run Your Home on a Raspberry Pi
One of the scary things (for me) when it comes to home automation and DIY smart home tech is that there's so many options that I feel totally overwhelmed. For the last few months I've wanted to do a small setup with sensors around my apartment that measure temp, humidity, air quality - and a few more for my upcoming chili pepper growing setup in the kitchen. There's so many options each with their own advantages & disadvantages - even on the air filter side there's pages of discussions about the various filters possible and how some don't collect particles below a specific size. I certainly miss the geeky/tech circle of friends I used to live near which would always have an friend (or a friend of friend) who was an expert on basically any matter!
noxvilleza··on The rising human cost of sports betting
> That's the AVERAGE win rate.

I thought that they were suggesting it's ~ the average winrate given optimal play.

noxvilleza··on The rising human cost of sports betting
One of the best ways of knowing is by analyzing a large number of bets made punters from multiple sports books - it's easier to spot irregular betting patterns in this regard.
noxvilleza··on The rising human cost of sports betting
The maximum win rates part is normally clear by the game (for example, for Roulette or Blackjack you can see odds) or otherwise regulated by local/government laws (e.g. for games of pure chance like slot machines).

> and otherwise fully detail their "we throw out anyone that wins" policy up-front

They can just ask the gambler to leave the premises - I cannot imagine some 'libertarian free market' solution which would force casinos to allow anyone to enter the premises and gamble (provided it's not clear discrimination). Being able to count cards isn't some sort of protected class, right?

noxvilleza··on Foobar2000
Ah, wasn't even aware of a fork (but makes sense!).
noxvilleza··on Foobar2000
What about Clementine [https://www.clementine-player.org/]? I used to love Amarok, so been using Clementine on Linux for the last few months.
noxvilleza··on Solving Wordle in 3.64 guesses on average, 99.4% of the time
Narrowing the search space greedily doesn't solve the problem optimally, because the size of the search space is just a heuristic. Some search spaces of size X might be more difficult to reduce further than another search space of size Y (for Y < X). This is more apparent in hard mode where you have a very limited guess-pool. The greedy heuristic approach would work fine if all possible 5-letter combinations were valid answers.
noxvilleza··on Solving Wordle in 3.64 guesses on average, 99.4% of the time
Adversarial strategies are another type of problem entirely (as in, if you're playing Word Mastermind). Wordle is a 1-player game, so I find hgibbs's comment totally reasonable. I would think that 'optimal' is more accurately his second definition - minimizing avg guesses but always winning (as in, losing is +inf guesses).
noxvilleza··on Wordle Is a Love Story
That's what I was clarifying with 'Yes, it's in the list of guessable words - however it's not a valid answer word'.

They have two lists of words: one is a list of possible answers, and another is a list of extra (valid) words which can be guessed (in addition to the words in the answer list). Sometimes it might be better to use a non-answer word as a guess: the best case gets worse (since you cannot win immediately) but the average case and worst case both get better.

noxvilleza··on Wordle Is a Love Story
Yes, it's in the list of guessable words - however it's not a valid answer word. You need to use the WORDLE word list in order to evaluate how good or bad your initial guess is, using another word list will provide distorted results.
noxvilleza··on Wordle Is a Love Story
> You are trying to use the word that once you get the match result back, it discards the most number of words.

This isn't strictly true either. Two N-sized subsets of words from a common initial set might have completely different difficulty in reducing further, because in the worst case there might not be a valid guess which nicely spreads the remaining words out among the 243 possible outcomes for that guess.

Set-size is a good heuristic, but it's just that - a heuristic.

noxvilleza··on Wordle Is a Love Story
I did some analysis on this last week [https://noxville.medium.com/raising-the-wordle-first-guess-b...], ROATE left the lowest average valid answers (60.424) but 195 in the worst case. RAISE was the second best average (61.0) but a much better worst case of 168 words. Both are just heuristics, there might be a better word than either if the game were solved.
noxvilleza··on Music Library and MusicBrainz Picard
Yeah I found out about the solution via that exact part of the docs, I'm just saying it seemed like a bug when I first came across it - and some users might not realize the default behaviour (for example, if you're importing a large preexisting collection you might not miss a few albums being deleted).
noxvilleza··on Music Library and MusicBrainz Picard
Love beets, used it now for a few years with foobar2000/cmus as my media player. A few tasks I've found somewhat irky (most recently handling the Blue / Green / Red / White / Teal / Black albums by Wheezer which are all officially just named 'Wheezer'), but otherwise very easy and accurate tagging interface.

I also just rsync my music library to an external SSD so that it's backed up, and up-to-date when I'm travelling.

noxvilleza··on Want to be an actuary? Odds are, you’ll fail the test
When I left high school I loved maths and stats, so I started a degree in Actuarial Science. Was totally bored by it by the end of first year - could not be motivated at all by it. There was no semblance of problem solving, it was primarily just applying a few types of formulae to a situation. Sometimes normalizing some input, sometimes adjusting something in a tiny way - but essentially the same stuff over and over again. Looked at past-papers for years 2 and 3, and it didn't any more interesting. I realized towards the end of first year that the idea of doing an entire degree and then 4-10 years doing board exams just seemed a sort of narrow path to take - few alternatives to swap to once you go too far. Ended up changing to Computer Science.
noxvilleza··on Ordering movie credits with graph theory
I wonder, now that there's a metric for what are the most "normal" credit structures, what the most abnormal structures are.

Also interested why they went with Trueskill over Glicko 2, since it's just 1v1 'encounters' anyway.

noxvilleza··on Magnus Carlsen retains title after winning lopsided match
Agree with all your first paragraph points. Although he's only an IM, I really like Eric Rosen - he's a pretty reasonable personality and is able to crack jokes and keep the conversation moving nicely. Unfortunately he was part of the broadcast with Danny Rensch. Giri just seemed downright rude to Judit. He ignored what she was saying in many instances, argued with her when he was listening, and failed to credit her when he predicted or calculated something incorrectly.

I'm also not a huge Maurice Ashley fan - but think he did a good job handling the press conferences. What was interesting was that for Karjakin/Carlsen, Jan Gustafsson was doing commentary, but also was part of the Carlsen analytics/prep/second team. Svidler's analysis is always enjoyable, but sometimes he seems bored with games and his language almost becomes 'this is over let's move on'.

noxvilleza··on Magnus Carlsen retains title after winning lopsided match
It's just "Elo", not "ELO" - it's not an acronym, but rather named after the guy who created it.

Dropping Elo after winning an event isn't really that 'wonky' - it's just a Bayesian update for underperforming your statistical expectations. The same thing happened when Carlsen won the Candidates in 2013, he went +5=7-2, but he was 2872 rating against a field of 2774, so 8.5/14 would mean he loses ~6 rating.

noxvilleza··on ProxyChains
Yeah same with me in university. They eventually started rate-limiting each SSH connection to ~50kB/sec, so I ran a bunch of separate SSH connections, made them as upstreams on a tinyproxy - and then used proxychains to force apps to use that tinyproxy tunnel (a lot of apps back then didn't allow you to configure a proxy).
noxvilleza··on Winamp source code leak
In foobar2000 it's kinda easy, Preferences -> Playback -> DSP Manager -> Downmix channels to mono [+]. (picture: https://i.imgur.com/FDhxh3j.png)
noxvilleza··on Acquisition of chess knowledge in AlphaZero
The way in which I think about it is that "learning" (in humans) often occurs is in the form of heuristics - certain rules or shortcuts which are not guaranteed to be optimal, but do a good job at simplifying (the search-state of) the problem. Instead of trying many simple comparisons between legal moves, we instead use a heuristic to score them.

Non-NN chess engines are really all about doing a good job at balancing these heuristics (like it says above). There's fundamentally an upper bound in how good the overall approach can be, simply because the underlying heuristics are defined and selected by humans: there could be many more heuristics that haven't been considered. This version of "learning" is really just removing human forced constraints which were introduced to help learning, by starting from a position without bias.

← PreviousPage 4 of 6Next →