Bike Forums

Bike Forums (https://www.bikeforums.net/forum.php)
-   "The 33"-Road Bike Racing (https://www.bikeforums.net/33-road-bike-racing/)
-   -   My geek thread (https://www.bikeforums.net/33-road-bike-racing/688664-my-geek-thread.html)

Flatballer 01-29-14 11:08 AM

And if you like math, and assume bitcoin can somehow take over the world, you can figure out what each bitcoin would have to be worth. There's approximately $70 trillion US dollars worth of "money" in the world today. If you spread that over 21 million bitcoins each bitcoin is worth around $3.5 million. Makes it a pretty good bargain at $1000.

Fat Boy 01-29-14 11:38 AM

It seems terribly risky, but in that risk is a substantial upside. Will it take over all currency? I highly doubt it. It could become a reasonable sized player, though. If a big company like Amazon, Walmart or, hell, even Starbucks started to accept it for payment, it could quite quickly become 'mainstream'. The ability to ship it all over the world without transaction fees could be a real advantage.

It's kind of one of those 'toothpaste out of the tube' type of deals, I think. Even if Bitcoin itself doesn't ultimately succeed, I think some sort of virtual currency may very well be a common part of our future.

mattm 01-29-14 12:15 PM

Quote:

Originally Posted by Flatballer (Post 16450146)
and assume bitcoin can somehow take over the world

Does anyone really think that's a possibility?

ovoleg 01-29-14 12:25 PM

Quote:

Originally Posted by mattm (Post 16450356)
Does anyone really think that's a possibility?

world governments will step I'm sure. I think it'll exist in some form or another, just like silkroad and piratebay.

Flatballer 01-29-14 12:35 PM

http://www.wired.com/wiredenterprise...01/chrisdixon/. Here's one VC guy who thinks it'll be worth $100k. Whether that's just him running a long game pump and dump I dunno.

I don't think anyone thinks people in poor countries will be using bitcoins instead of US Dollars anytime soon, but certainly there are people who think it will be the major form of payment online, which is a good portion of overall money transfer.

I think it's a bubble, but I think as a means of transfer online it has some value, just not $1000+. It won't go to 0, but it won't be worth a ton either.

furiousferret 01-29-14 12:54 PM

SharePoint 2013 is a pain. That is all.

mattm 01-29-14 01:21 PM

Quote:

Originally Posted by furiousferret (Post 16450497)
SharePoint 2013 is a pain. That is all.

+1 I hate it!!

Sharepoint is bloated and stupid. The URLs to documents are way too long!! Whatever company that made it needs to redo it..

Fat Boy 01-29-14 03:23 PM

Quote:

Originally Posted by Flatballer (Post 16450427)
http://www.wired.com/wiredenterprise...01/chrisdixon/. Here's one VC guy who thinks it'll be worth $100k. Whether that's just him running a long game pump and dump I dunno.

I don't think anyone thinks people in poor countries will be using bitcoins instead of US Dollars anytime soon, but certainly there are people who think it will be the major form of payment online, which is a good portion of overall money transfer.

I think it's a bubble, but I think as a means of transfer online it has some value, just not $1000+. It won't go to 0, but it won't be worth a ton either.

I think the developing countries would love something like Bitcoin. When you have governments that are complete clowns with how they deal with their currency the people are constantly looking for other forms of money that are less volatile. Maybe Bitcoin or something similar would be a good option. The problem right now is how to spend them, but if it starts being more readily acceptable, then there are some interesting future scenarios.

Like any currency, people have to have faith in it. That's the biggest hurdle. Who knows if it'll get over that one.

Creakyknees 01-29-14 03:26 PM

Quote:

Originally Posted by mattm (Post 16450591)
+1 I hate it!!

Sharepoint is bloated and stupid. The URLs to documents are way too long!! Whatever company that made it needs to redo it..


https://www.box.com/

yer welcome

mattm 02-03-14 11:23 PM

http://www.lighterra.com/papers/modernmicroprocessors/

Quote:

Okay, so you're a CS graduate and you did a hardware/assembly course as part of your degree, but perhaps that was a few years ago now and you haven't really kept up with the details of processor designs since then.In particular, you might not be aware of some key topics that developed rapidly in recent times...
  • pipelining (superscalar, OoO, VLIW, branch prediction, predication)
  • multi-core & simultaneous multithreading (SMT, hyper-threading)
  • SIMD vector instructions (MMX/SSE/AVX, AltiVec)
  • caches and the memory hierarchy

As a software-only guy, this is interesting reading.. granted I won't read the whole damn thing!

Flatballer 02-05-14 07:18 AM

Any car guys, there's a new game I found through Jalopnik (awesome car website if you don't visit it). It's called Automation: The Car Company Tycoon or something like that. It's not complete yet, but they're releasing it in stages. The demo is pretty good and lets you play around a lot, for $25 you can buy the full version and it will get updated as they finish. Should be done late 2014, and they've been pretty good about sticking to their promises and there's lots of buzz and money coming in, so they should be fine.

The basic gist is that you build a car company from the ground up, over the years, as technology progresses you get access to new stuff, you buy different factories to help you, etc.

The difference from most tycoon games is that it's extremely involved. The first thing you do is make an engine. And you don't just pick components and stick them together. You pick the block type (I-4, flatplane V8, crossplane V8, or I6 are the current choices, but they'll be expanded later, I4 is the only one available in the demo), block material (iron, aluminum, AlSi, Magnesium), crank material, connecting rod material, piston material. Every choice has a cost, it affects the economy, rpm limit, weight, build time (man hours), money, etc. It's extremely well modeled, such as pistons made of a different material than the block can lead to a lower MTBF or lower fuel economy (too much blowby). Then you pick the heads, valve configuration, etc. Turbo(s), if you pick one or two, you can choose the compressor ratio and everything. Fueling type (carb or injection, double barrel or four, common injection, direct injection, etc), intake type, etc. Then you dyno the engine and see if it blows up. Then you tweak for hours getting it right. Then you build a car to put it in (similarly involved, but not as finished yet as the engine builder). Eventually there will be a campaign where you try to make the whole company by building your car lineup and stuff and updating it through the years.

There's also challenge things, right now only engine challenges, where it gives you a list of restrictions, and a list of requirements, and a little story to go along with why you're building that engine, and then you try to make an engine that meets the requirements. It's pretty awesome.

Enthalpic 02-06-14 01:07 AM

Played with Scratch again tonight and changed the physics of my little orbit program. Learned about drawing and cloning. It's kind of fun if you get a neat orbit.

http://scratch.mit.edu/projects/17567062/

waterrockets 02-06-14 09:01 AM

Quote:

Originally Posted by Flatballer (Post 16469278)
Any car guys, there's a new game I found through Jalopnik (awesome car website if you don't visit it). It's called Automation: The Car Company Tycoon or something like that. It's not complete yet, but they're releasing it in stages. The demo is pretty good and lets you play around a lot, for $25 you can buy the full version and it will get updated as they finish. Should be done late 2014, and they've been pretty good about sticking to their promises and there's lots of buzz and money coming in, so they should be fine.

The basic gist is that you build a car company from the ground up, over the years, as technology progresses you get access to new stuff, you buy different factories to help you, etc.

The difference from most tycoon games is that it's extremely involved. The first thing you do is make an engine. And you don't just pick components and stick them together. You pick the block type (I-4, flatplane V8, crossplane V8, or I6 are the current choices, but they'll be expanded later, I4 is the only one available in the demo), block material (iron, aluminum, AlSi, Magnesium), crank material, connecting rod material, piston material. Every choice has a cost, it affects the economy, rpm limit, weight, build time (man hours), money, etc. It's extremely well modeled, such as pistons made of a different material than the block can lead to a lower MTBF or lower fuel economy (too much blowby). Then you pick the heads, valve configuration, etc. Turbo(s), if you pick one or two, you can choose the compressor ratio and everything. Fueling type (carb or injection, double barrel or four, common injection, direct injection, etc), intake type, etc. Then you dyno the engine and see if it blows up. Then you tweak for hours getting it right. Then you build a car to put it in (similarly involved, but not as finished yet as the engine builder). Eventually there will be a campaign where you try to make the whole company by building your car lineup and stuff and updating it through the years.

There's also challenge things, right now only engine challenges, where it gives you a list of restrictions, and a list of requirements, and a little story to go along with why you're building that engine, and then you try to make an engine that meets the requirements. It's pretty awesome.

Ack! last thing I need is another addiction cool though

Quote:

Originally Posted by Enthalpic (Post 16471935)
Played with Scratch again tonight and changed the physics of my little orbit program. Learned about drawing and cloning. It's kind of fun if you get a neat orbit.

http://scratch.mit.edu/projects/17567062/

Nice. Seems like most of my clones draw some cool stuff. I didn't know about the SHIFT- advanced mode stuff. Coooool.

Enthalpic 02-07-14 01:10 AM

Crap this "game" is too much fun. By "game" I mean figuring out how to make simple games... this one was easy.

http://scratch.mit.edu/projects/17671257/

waterrockets 02-07-14 08:33 AM

Quote:

Originally Posted by Enthalpic (Post 16474917)
Crap this "game" is too much fun. By "game" I mean figuring out how to make simple games... this one was easy.


http://scratch.mit.edu/projects/17622493/

:) All those 80's style games seem to be so easy to make with Scratch. This was actually my first learning project on it (using my daughter's account). 40 minutes to make this -- same game I made in a week in college.
http://scratch.mit.edu/projects/15947387/

mattm 02-07-14 12:41 PM

Speaking of 80's games, that "Flappy Bird" iOS game is supposedly pulling in $50k - per day!!

Per day... I mean wow.

Haven't played it, but it looks pretty simple.

waterrockets 02-07-14 01:19 PM

Agreed.

It's a bit of a crap shoot for the developers. There are a ton of little games out there that are no lower tech, but they lack the balance and addictive play-ability. Still there are probably 100s of these well-designed games that never catch on either. I guess it's just like every other part of the entertainment industry with respect to tons of productions and relatively few that snag the attention of millions. Certainly the challenge is not a technical one though.

Bad Land is a pretty simple little game that I started playing a couple months ago when I'm waiting in line or whatever. Great feel, and I could probably write about 80% in Scratch w/out too much trouble (scrollers are a bit of a pain in Scratch though).

Enthalpic 02-07-14 07:22 PM

Quote:

Originally Posted by mattm (Post 16476109)
Speaking of 80's games, that "Flappy Bird" iOS game is supposedly pulling in $50k - per day!!

Per day... I mean wow.

Haven't played it, but it looks pretty simple.

I tried the android version. I would say it's far from "balanced." It is stupidly hard, I guess people like frustration.

From a programming point of view it is very simplistic.

http://scratch.mit.edu/projects/16975457/

Enthalpic 02-17-14 08:51 PM

Still playing around on Scratch, it's been one month now and I've learned a lot.

Here is my attempt at schooling / flock behaviour. Their behaviour is not predetermined; it's a bunch of randomizations and rules that say they must stay close -but not too close- to each other and that they should try to align their direction.

http://scratch.mit.edu/projects/18077844/

From what I've read very similar simple rules control fantastic things like this. I would love to see this in person.
http://www.youtube.com/watch?v=iRNqhi2ka9k

waterrockets 02-17-14 10:23 PM

Very cool stuff. These steering behaviors are a lot of fun to play with, as you can build them from the ground up and get some really natural looking behaviors. If you can start by modeling pretty close to how a fish can move in the water, then have each fish steer through that interface while obeying the school, flock, or swarm behavior, it all comes together really nicely.

Good work.

Enthalpic 02-17-14 10:43 PM

The steering keys still work (arrows, hold r to swim fast), try herding a bunch of clones into a ball then charging and scaring (spacebar). The scatter effect I wanted works OK.

waterrockets 02-18-14 07:29 AM

Yeah, that is a cool effect. Easy to ball them up in the corners too :)

mattm 03-02-14 11:07 PM

Write code to generate random numbers in a range.

int myrand(int min, int max);

But, you can't use rand() or any library methods for generating random numbers..

I wrote some basic code to do it but there's gotta be a better way.

slynkie 03-03-14 07:25 AM

Quote:

Originally Posted by mattm (Post 16542961)
Write code to generate random numbers in a range.

int myrand(int min, int max);

But, you can't use rand() or any library methods for generating random numbers..

I wrote some basic code to do it but there's gotta be a better way.

You need something to seed the generator. I've never written a lib function for this but iirc, some common textbook examples are a) mouse cursor input and b) microphone input. I doubt lower level language lib implementations use these, though..

Off the top of my head - barring access to some input device you could use for seeding, I'd look for a high resolution time function, and use a digit from each time sample for your random #. If your timer isn't high enough resolution to get (different) digits quickly, try some small sleeps.

mattm 03-03-14 02:31 PM

Quote:

Originally Posted by slynkie (Post 16543368)
You need something to seed the generator. I've never written a lib function for this but iirc, some common textbook examples are a) mouse cursor input and b) microphone input. I doubt lower level language lib implementations use these, though..

Off the top of my head - barring access to some input device you could use for seeding, I'd look for a high resolution time function, and use a digit from each time sample for your random #. If your timer isn't high enough resolution to get (different) digits quickly, try some small sleeps.

Yeah at first I tried using milliseconds as my seed but it wasn't fast enough; sleeping wouldn't be an option. (this is just for interview practice btw, but they wouldn't like the sleep option)

I also toyed with using uninitialized integers/strings as my seeds, but it's not reliable enough.

Quote:

Originally Posted by waterrockets (Post 16543618)
Is this where you're trying to derive your own generator for the challenge, or are you actually trying to solve this problem?

http://www.codeproject.com/Articles/...ber-Generation

Yeah something like the "Multiply with Carry" method or the "XORshift" method seems to be the best approach.

Xorshift:

Quote:

uint32_t xor128(void) {
static uint32_t x = 123456789;
static uint32_t y = 362436069;
static uint32_t z = 521288629;
static uint32_t w = 88675123;
uint32_t t;

t = x ^ (x << 11);
x = y; y = z; z = w;
return w = w ^ (w >> 19) ^ (t ^ (t >> 8));
**
That code pretty much blew my mind. I understand the syntax, just not how one would arrive at it!!

I wrote a basic rand method using CPU clock cycles and the Middle-square method, but I doubt it would pass the "Diehard" test.

In one of my recent interviews I was asked to generate GUIDs without using a built-in rand() method, which brought this problem to mind.

If I ever get asked it again I'll have a better answer! (which would not be memorizing one of the fancy methods, but at least being familar with them and simpler options like the Middle-square method)


All times are GMT -6. The time now is 08:35 AM.