mrchicity

This is a known issue that only affects early revisions of Ryzen CPUs under specific workloads, so it's unlikely to appear in new ones like the mobile Ryzen. If you're affected, AMD will RMA your chip for free.

And AMD isn't the only manufacturer to have such issues. Intel had a bug in Skylake where it would crash under heavy load with Hyperthreading enabled. Modern CPUs are complicated beasts.

Check whether your chip is affected by the segfault bug. It affects Ryzen CPUs, but not Threadrippers or Epycs. It's a free RMA: https://www.reddit.com/r/Amd/comments/6ubmd1/ryzen_compilati...

If your system is totally new, try checking your RAM with memtest too.

My Threadripper build had similar issues in Linux and Windows. Ended up being my PSU. I tried to save $30 buying a refurb one from Corsair. My system would lock up occasionally at totally stock settings and even shutdown on its own. Works perfectly now after I swapped it out.

FWIW, I'm really happy with the Threadripper/X399. Most of the early Ryzen bugs are worked out. It's a great value for the performance. Compiling large C or C++ code bases is insanely fast.

Still, I'd be wary of buying AMD's first gen mobile offering. When you buy a laptop you're buying a whole system and battery life, no glitches after sleep/wake, heat, noise, etc. trump raw performance. Intel has way more experience there. This Radeon iGPU also sits in a weird market segment. People who don't game/render/3D model on the go do just fine with the weak Intel iGPU. People who really care about that stuff can get a big laptop with faster dGPU or just do 3D on their desktop. So I guess their target is people who casually game but want a thin-ish laptop?

NB: I don't doubt AMD's ability to make a decent mobile platform. Just Intel has shipped hundreds of millions more units, so they've seen every "system crashes after sleep when re-pairing with $OBSCURE_BLUETOOTH_DEVICE" issue on earth. On a desktop you can always swap hardware, but on a laptop you're stuck with what you get.

Panther was a one-man shop that did nothing but market manipulation. Yes, they used automation but for every Panther, there are 1000 manual traders being fined by CME for spoofing every year. Automation doesn't imply state of the art HFT. The Flash Crash spoofer was using a Trading Technologies system that any small trader can rent with a few scripts loaded in.

Look at real players with hundreds of employees. None of them make their money from market manipulation. It's all from low-latency arbitrage. If you're fast, you can make way, way more money doing legitimate trading than market manipulation.

Layering, Spoofing and Momentum Ignition are not "HFT Strategies." They're illegal market manipulation techniques, most often used by manual traders. You don't need to be ultra fast to bully prices around or enter large non bonafide orders. I've never heard of a legitimate proprietary trading firm (i.e. one that pays a salary and hires highly qualified people, not a boiler room operation) intentionally using these techniques. Most are in the arbitrage, market making or statarb business.

It would be much more interesting to see the author dive into which events lead to fast price changes. How quickly are price shocks on other markets are reflected on GDAX? How are changes in outright contracts like BTC/USD and ETH/USD reflected in ETH/BTC? Do USD denominated pairs drive price discovery more or less than EUR denominated ones, and does this change when US based traders are asleep? Lots of stuff to look at here.

This is evidence that we should invest more in early childhood enrichment and healthy food for the poor, so everyone can reach their potential. Genetics determine the capability of your engine, but you still need to feed it with gas.

It doesn’t justify lowering the bar for hiring, creating quota systems, or engaging in positive discrimination.

Of course. There's rarely a single cause for anything.

But in today's world, real systematic discrimination in hiring and employment isn't likely to be the main driver at places like Google. When you're selecting for individuals several standard deviations above the mean in ability, even very slight shifts in the population mean dramatically affect the number you'll find at such a threshold. This is a touchy third-rail subject, one we'd probably do well not to discuss much, but when the alternative is constantly being told that you're racist or prejudiced, people will naturally want to fight back.

If the city police or local factory had extremely lopsided racial representation, I'd be much more inclined to believe it's due to prejudice. Any average person can do these jobs. Even if there are differences between groups, all groups would have a huge number of people around the average range.

And I'm not saying this out of some attempt to claim supremacy, justify prejudice, or advocate for eugenics. My family is from Portugal. If I had to wager, I wouldn't bet on Portuguese being at the top of the heap intellect-wise, on average. Look at who's winning the Nobel Prizes and Fields Medals--it's not people like me most of the time. Still, we should treat every person as an individual, with respect and dignity.

Is that really true though? Where I've worked, almost everyone was from an upper class or upper middle class background, grew up in a handful of superzip suburbs like Atherton/Scarsdale or high COL cities like NY, stable family life, attended a top tier university, had similar cultural interests or hobbies, read the same news sources, even had similar vices. They looked like a Benetton ad, but all acted exactly the same.

Is hiring Bs who are exactly like your As, except in superficial characteristics like appearance, really bringing in diverse points of view?

How is that racist?

Racist is "I think all Mexicans are dumb, so I won't hire any or even give them an interview. I hate Mexicans."

Racism is not "Groups of humans evolved dramatically different physical characteristics, vulnerability to health conditions, ability to consume foods like lactose, and countless other characteristics. There's some evidence to suggest they may have different distributions of skills required to do this job well. Even if there are no innate differences, there are certainly cultures that prioritize learning the skills we require more than others.

I wouldn't be surprised to see different groups represented at different rates vs. the general population at this job. However, I'm smart enough to realize there are extremely skilled and extremely unskilled members of every group, and I'd gladly interview and hire from any."

Intent. How do you differentiate hacking from legitimate access? You're just sending byte streams to the server and it decided to give you access to something, after all.

The reason it's illegal or banned on normal exchanges is because it drives out legitimate customers and you end up with a den of thieves. Everybody looks at the order book to gauge liquidity and make decisions, not just bots. Users want prices to be relatively stable and tied to fundamental values. An exchange where manipulators spoof or wipe the order book in a high stakes game of chicken is unstable. Imagine an eBay with no protections against counterfeit goods, or a StubHub full of fraudulent tickets.

And spoofing is very easy for regulators to detect. The behaviors of market makers, who also cancel many orders, and spoofers, could not be more different. Market makers place orders for small symmetric size on both sides of the book. If they quote asymmetric sizes, they quote smaller bids when they're long, or smaller offers when short. They cancel bids after they buy, offers after they sell. Spoofers do the exact opposite. They place asymmetric sizes where one side makes up a large amount of liquidity. They place big size on the same side when long or short to press their trades favorably. They cancel buys after they sell, and sells after they buy.

Most brokers charge some all-in commission per share to clients. If you take away the rebates, brokers will simply raise commissions to compensate. Whether a fee is passed through or not is irrelevant. Brokers don't pass through direct costs of exchange licenses, servers, development time on their algos, etc. but clients still end up paying for them.

Odds are clients like the current system because they know exactly how much executing a 100k share block will cost in commissions. If they didn't, there are plenty of competitive equities brokers who will offer pass-through pricing.

And saying the shortest queue is best makes no sense. Most exchanges with short queues have laughably low market share. The lines are short for a reason, because they get serviced more slowly. Otherwise it'd be free money for fast HFTs to fill those lines up and get trades before someone on another exchange.

I've tried work-sample testing with middling results:

-The more hoops a candidate has to jump through, the more likely they are to bail out of your recruiting funnel. This is especially bad for college/postgrad recruiting when you aren't the #1 employer in your field. Everyone wants to work for the Googles and Facebooks of the world. It's hard getting someone to spend a couple hours for your startup job.

-People cheat. We usually issue a short coding project, grade for correctness, then do a code review over Skype or face-to-face. Many candidates turn in the exact same responses. I've even seen people cheat and have a friend do the Skype session with a totally different guy flying out. Do you proctor your test in a secure center? Use an online service to lock down their machine and record? Both are pretty invasive. Switching up the questions constantly is tough and makes your signal noisier.

-Industriousness and raw intellect trump skills/knowledge most of the time. Sure there's a baseline level of skill required to train someone quickly enough, like I wouldn't hire someone who didn't know basic data structures, but work-sample tests are often biased to those with a very specific background. I don't want employees who are great at doing what I need today. I want ones who will be great at figuring out what to do years down the line.

Because what good can come from discussing that topic? The only reasons to broach it are to a.) troll and/or b.) agitate for some horrific policy proposals.

He's not an academic researcher. Don't you think it's odd for some old white guy to get worked up over minority test scores? Why write about this rather than basically anything else? Either he's a troll or profoundly unaware about how bringing this line of argument up may make people feel threatened. There are a lot of things that simply don't need to be discussed. It's like graphically describing your partner's episiotomy over dinner. Some things just inspire a visceral reaction.

In a two-party system, few people agree with every policy in someone's platform. I voted for Clinton but disagreed with many of her views and actions. Oftentimes people are voting based on a single very important issue or for the lesser of two evils.

I'm curious why you think supporting a ban like this is a "cancer"? It's heavy-handed and misguided, but religion is ideology. People see heinous acts worldwide perpetrated in the name of Islam. Some Muslims have extremely regressive beliefs on women's rights, gay rights, personal freedoms like drinking alcohol or even making loans that make even the staunchest Bible thumper look progressive. They see countries with larger Muslim immigrant populations like France in constant disarray, with little integration and once peaceful neighborhoods lit on fire. Even people with no initial ties to terrorism read up on this ideology online and commit lone wolf attacks.

So why do we want these people coming in? Maybe we have a principles-based belief in religious freedom and don't want to appear discriminatory. There are some more secular or Westernized Muslims who can live peacefully and contribute to our society. Helping refugees will help save innocent lives and undermine ISIS. Those are all great reasons, but it's disingenuous to pretend they come at zero cost. Some people do this analysis and say it's worth it, and others disagree. A sovereign state has a right to its borders and admitting immigrants who advance its interests.

I don't support the ban, namely for the reasons I gave above, but it's completely understandable why someone would. It doesn't make them a bigot or evil, just scared and perhaps lacking in empathy. The hardcore left does itself no favors in this regard, constantly parroting the "religion of peace" line and pretending there are no reasonable reasons to oppose immigration that they need to argue against.

Rolex is a fine brand. It's a "safe" choice in the mechanical watch world. Not super high-end. Not total BS schmaltz wrapped around an off the shelf ETA movement.

What it reminds me of most as far as products go is a vintage Land Rover, Toyota Land Cruiser, pre-Chrysler Mercedes-Benz, or air-cooled Porsche. You know the brand will support and service the watch forever and you can expect to find qualified technicians all over the world.

If you're buying a watch not just to show off today, but to pass down through your family, that's worth something. Rolexes generally hold their value well, are durable, keep the same style for decades, and have a relatively conservative look.

Exactly. Unless you can show that the price continues to move against your losing positions by more than their cost to liquidate, you shouldn't do it. Your screw-up was putting the bad trade on in the first place.

I'm especially surprised it's used in a model like this. Most quant funds use some type of portfolio risk management to allocate capital toward bets that with the highest expected return while controlling overall risk. That losing stock may be offsetting other risks in the portfolio.

As I mentioned in my other comment, a stop-loss is a very crude approach to risk management. It will help control the middle of your left tail, but the far left tail extreme events cannot be protected against by such a rule.

Markets are only zero-sum on a transactional level, in isolation. Yes for every buyer there's a seller and for every winner over some time scale, there's a loser.

But in the broader context of the entire real world, a well-functioning market is valuable in itself. It enables people to transfer risk inexpensively, instantly, automatically, and at fair prices. This competitive pricing mechanism also provides valuable signals to the real economy so people can invest in areas that need it.

Imagine a market with 4 participants:

-A farmer who grows corn and wants to lock-in a price for next year's harvest so he can invest in a new tractor and refurbishing his grain elevator.

-A baker who wants to lock-in his price for his corn purchases next year so he can invest in new ovens for his plant.

-A speculator who monitors price trends and takes risk intermediating between buyers and sellers.

-An exchange that provides a meeting place for these parties.

On Monday, the farmer places his orders with the exchange to sell his 2018 corn harvest. The baker is on vacation until the next day, so the farmer can't transact. However, the speculator knows bakers tend to buy corn around this time of year, so he buys the contracts from the farmer at a slight discount. The exchange also takes a fee from the farmer, and from the speculator.

The farmer goes off to the tractor dealer, and gets to work refurbishing his grain elevator.

On Tuesday, the baker comes to the exchange. No farmers are around, just the speculators who bought from them yesterday. Today is a lucky day for the speculators who can now sell to the bakers at a slight premium. Sometimes they misread the market and take losses instead. Again, the exchange takes a fee from the baker, and from the speculators.

The baker is comfortable funding purchases of his new ovens, knowing that he isn't at risk of corn prices rising next year, which would cut into his profit margins and make his plant unprofitable.

So far, the exchange has made money for providing a meeting place, and the speculator has made money for matching up buyers and sellers who arrive at different times. The farmer and baker each lost a bit of money by trading against the speculator--his profits are their losses, and vice-versa when he's wrong.

2018 rolls around. The cost of corn has risen due to import tariffs. The baker made money on the contracts and the farmer lost an equal amount. The exchange and speculator each made some money. Transactionally, this is actually a negative-sum game, because the exchange makes money no matter what, but it let everyone involved focus on their particular business and insulated them from risks they didn't want to bear.

Can't you say that about any profession, especially other forms of middlemen? Does it really matter if your grocer's motivation is having a nice house and good schools for his kids vs. feeding people and providing a market for farmers?

It gives a false sense of security in cases like this. The biggest losses you want to liquidate will always happen in cases where it's impossible or extremely costly to do so: company has merger announcement overnight while you're short, accounting scandal, bankruptcy, earnings surprise, fat finger order takes all liquidity out of the market. You will end up losing many multiples of your designated loss limit in the bad cases. The only way to prevent massive losses is to size your bets sensibly or insure against tail risk using options ahead of time, of course that comes at a cost.

By bugs and adverse selection I'm assuming you're talking about something like automated market making. For traders like that, a loss limit makes sense because of technology risk. If your order router has a bug like Knight's did, or your data feed gets stuck, you can lose a lot of money very quickly.

By general service do you mean a system where response time isn't especially important? Using TCP to fan-out messages in a PubSub pattern is going to waste network and compute resources in a serious way. It can easily exhaust capacity or result in convoluted, unreliable systems with layers of proxies for managing load.

FWIW when you talk about different requirements, I'm not sure you're correct. For traders, sure, oftentimes they can just dump old data if they fall behind because it's mostly irrelevant. However, stock exchanges have performance and reliability/availability as top requirements. They must record all transactions and can't crash or lock-up. Most use multicast internally and externally. They would never be able to scale and maintain a low response time with TCP.

Think about happens if indexes don't pre-announce their changes. Is their market impact larger or smaller than if people know about it and "front-run"?

Suddenly someone wants to trade 10x daily turnover in a small-cap stock right on the closing bell. It could be a dumb whale index fund, or it could be an insider trader who knows about a merger, or a sharp hedge fund who thinks the stock is undervalued.

If you're a liquidity provider or speculator, do you want to stand in front of this freight train at all? If you offset the huge liquidity imbalance, how much will you charge to cover your losses if it was an informed trader? Is it more or less than if you know it's an index fund?

Pre-announcing generally lets utilitarian traders execute their trades more cheaply: http://rfs.oxfordjournals.org/content/4/3/443.short

Remember, even if there is some edge in this type of anticipatory trading, competition will reduce its margins significantly. Imagine I announce tomorrow in the Wall Street Journal that I'll buy 100000 off-lease Honda Accords next week from whomever will sell them to me cheapest. Everyone rushes out to buy them on Craigslist or eBay, but only those who make the least on the deal end up selling to me.

If you invest in ETFs, you are certainly being helped by automated arbitrage traders. These products rarely trade more than a cent from their true value because computers can calculate the hedging basket almost instantly. Yes, they have a profit motive, but competition keeps their profits small: https://meanderful.blogspot.com/2013/01/hfts-dirty-little-se...

It always costs money to transact. The alternative to trading against an ETF arbitrageur's quote isn't a costless transaction at the fair price, it's manually paying the spread and transaction costs on hundreds of stocks and risking slippage.

FWIW: YC probably earns more than even the biggest automated trading businesses.

That "mere facilitation" is extremely valuable. Do you think stock exchanges themselves are valuable and worth paying for? They bring buyers and sellers together in a centralized place, merely facilitating a pre-existing desire to trade, taking a small cut in the process.

Imagine a world without them. When you want to buy a share of Microsoft, you could call all your friends seeing if they want to sell any or know someone who does. That could take forever or you might never trade. Search costs are real.

Much like the stock exchange helps buyers and sellers find one another in the same security, HFT helps buyers and sellers find one another across exchanges, securities or risk factors. They basically create a meta-market for you, at very little cost. Typical HFT profit margins are fractions of a cent per share traded, similar in magnitude to what the exchange earns.

To give a real example, imagine you want to sell Valeant Pharmaceuticals after their latest scandal. You only have access to the US markets. There's an abundance of sellers in New York pushing the price down, but over in Canada, the price is a little higher. Someone in Toronto just read the latest report, thinks the price will rise, and put a bid in, but he only has access to the Canadian markets. The two of you wish you could find one another, but you can't.

An HFT algo with real-time data from Canada and the foreign exchange rate places a bid on the NYSE, tightening the spread. You sell to them. They turn around and hedge their stock exposure by selling to the guy in Toronto, and trade an FX futures contract in Chicago. This is "mere facilitation", but it helped two traders transact at a better price than they would have otherwise, and forces convergence between two related markets to make them more efficient. Automation lets him run this same trade over thousands of securities for very low margins, making a few dollars in each every day. How is that not a great thing?

Yes it's an awesome job, and not just because of the money. I've worked for small shops that weren't very profitable and didn't really make a lot those years, but still loved the work.

I could see derivatives pricing being boring, not as much room for creativity there. The strategy/trading quant work is like a super open-ended science problem with a constant stream of new challenges as you try to outperform yourself or the competition.

FWIW, I don't think these are great questions for strategy quants. Some of the best people I've worked with a.) used very, very basic ML techniques b.) were just naturally smart/clever enough to engineer great features. and c.) had incredible attention to detail, didn't "hand wave" away the dirty work like checking/cleaning data or executing the model in practice.

I think the individual retail trader is almost always outgunned informationally when it comes to intraday trades. Most short-term price action is driven by order flow and cross-asset correlations, which machines are very good at trading. They are often net trading cost earners due to rebates and capturing bid-offer spreads. That means their win rate doesn't need to be as high, so they can pull the trigger on a trade before you can, just by having lower fees and superior execution. In addition to that, they're faster, more scalable, have more access to liquidity, and are more disciplined than humans could ever be. For the ones who trade off statistical correlations, they have the best data and armies of PhDs working on signals.

There are some event-driven fast trades still done by humans, like after news or responding to economic releases, but hedge funds have highly-educated people modeling the effects of an interest rate change or earnings release as their full-time job. This area is also becoming dominated by bots doing sentiment analysis.

I think the retail trader could have an edge in a few ways. One would be an illegal edge like inside information or market manipulation. Another would be finding illiquid stocks that proprietary traders and hedge funds won't bother with and trading using similar techniques. You could also look for extreme situations that model-based traders can't understand well due to lack of data, like a merger target breaking away. Those trades would be very risky though.

FWIW I'm a professional trader and never day trade my own account or pick individual stocks. My company allows it, but I don't believe I have any edge in doing so. I just buy and hold a portfolio of ETFs.

Reg NMS says you cannot code a system or intentional device to delay access to a quotation. Nasdaq had tried doing this on one of their smaller markets in software many years ago, but the SEC did not approve their proposal. Geographic or network latency is not regulated by Reg NMS. For example, there is a very small stock exchange that operates out of Chicago, which is many milliseconds away even by microwave vs. New Jersey where most of the equities market trades.

There are a lot of SEC comment letters claiming IEX's delay coil is an "intentional device"[1][2] but the SEC approved their exchange application because they consider the delay to be "de minimis." One area where the IEX delay differs from simple geographic latency is that IEX delays incoming orders, outgoing fills, and outgoing market data, but they do not delay market data from other exchanges. That difference makes it more difficult to "pick off" a stale order on IEX, but on the flip-side, it also means IEX orders get to free ride the price discovery happening on other markets without taking risk.

This delay also raises issues because under Reg NMS, brokers have to route orders to exchanges with the best price when trading through an entire price. Since IEX's data is lagged, it may be showing a quote that is no longer accessible. Brokers would face additional slippage/execution uncertainty if they need to route at this phantom quote and wait 700 microseconds to find out it wasn't there. Canada has exchanges with delays similar to IEX, but IIROC ruled that they are not "protected" quotes so brokers have discretion not to route there if they feel it would harm their execution quality.

1: https://www.sec.gov/comments/10-222/10222-300.pdf

2: https://www.sec.gov/comments/10-222/10222-16.pdf

I don't think people on the floor were necessarily all crooked or doing something illegal, but that market structure doesn't invite fierce competition. You'd see the same people every day and go to drinks with them. There was a real camaraderie between people on the floor, even if they were ostensibly competitors. Think of any community. Most people would probably work a little harder to have a nicer car than their neighbor, but they wouldn't want to see the neighbor's family starve to death in the process. By undercutting your rivals in the pits, you were literally taking away their mortgage payment, and you had to do it to their face. A lot of the locals in the pits were independent, so it was personal, not just business.

You also could see who was on the other side of your trade. Savvy floor traders would use that information to their advantage. If a sharp broker traded with you, you could hedge more aggressively or speculate in the same direction. Most electronic markets are anonymous these days so there's less information leakage.

In the pre-HFT era there were manual day traders doing spreading, arbitrage and short-term speculation based on the order book. A famous example would be the SOES bandits of the 90s. They would pick-off slow NASDAQ market makers and offer the shares back out on nascent ECNs like Island to make a few ticks.

Some of them worked for firms as professionals, but many traded on retail platforms similar to what you'd see from Interactive Brokers or TT these days. Still you are right that the average investor never traded in this fashion. At worst, the people HFT put out of work were trading as a profitable hobby similar to playing poker online.

HFT is mostly doing the same thing that banks, prop traders, specialists, or hedge funds were doing manually in the past. They risk their own capital to transfer liquidity across time (market-making), place (cross-market arbitrage) and product (trading on statistical correlations). That activity is valuable because it provides liquidity, the ability to buy or sell at a relatively fair price, where it is needed most and enforces structure on the market to keep it efficient.

The speed race between participants is probably not valuable, but it's the least bad form of market structure that exists. Somebody will do these trades, and everybody wants to, so you need some way to pick winners and losers. Having whoever sets the best price fastest win is less prone to gaming than other structures and rewards competition that drives down inefficiency (you have to do every marginally profitable trade because someone else will snatch it up before it turns into a really great one).

If you could put all the top HFT players in a smoky room where they agreed not to buy new faster microwave lines or high-speed switches, they'd certainly agree to. They're already faster than humans and most banks. They win on that front no matter what. All of this speed competition just changes their rank ordering amongst their peers and costs a lot of money.

This isn't true. HFTs are very active and profitable in markets where there are no rebates (some US stock exchanges and most dark pools, futures markets, most overseas exchanges). Even in markets where you have no fragmentation at all, you'll find HFT. The rebate just gets baked into how wide of a market you provide or how eager you are to pay the spread.

The main reason Reg NMS and Reg ATS drove a proliferation of HFT was breaking the NYSE specialist's monopoly on liquidity provision. Even before that companies like Tradebot where very active in Nasdaq stocks on various ECNs. Automated trading is simply cheaper and it drove manual market-makers out of business.

I understand it but still hate doing the dance. Negotiation doesn't end once you've started a job either. If you want good projects, promotions, higher compensation, you usually have to negotiate for those too. Sometimes managers recognize talent and take initiative, but it's easier to compensate the loudest voices or people who are credible flight risks. This can lead to a very toxic culture. It takes very mature and long-sighted management to pay for value before the employee complains.

I agree with the author's one point: The best possible thing you can have is an alternative. I remember reading through the leaked Sony emails where executives were discussing compensation packages for people making millions of dollars. They didn't care much about how much revenue a VP brought in, if they were a nice person, how many hours they worked, but they sure cared a lot about losing them to a competitor. Performance reviews, metrics, titles, etc. were just a pretext to pay whatever the boss decided was necessary to keep them.

FWIW, I think careers where you know your objective value and have a portable skill-set or book of business are the best in terms of employees capturing more of their surplus value. Google's revenue per employee is probably similar to a good trading firm, but no one Google engineer can recreate Search, so they take home much less of their value than a trader would. You see the same thing in sales roles, law partners, etc. where employees and management both know what they bring to the table, and know they could walk out the door to another company if compensation is unfair.