Page 2 of 5

FPGA library

Posted: February 10th, 2012, 6:53 pm
by farmer
QuoteOriginally posted by: outrunnetworking announcements: 100 Gbps to the End User; Ultra-low-latency MicrowaveThere is no way you can go 100Gbps without a repeater. And here it is:QuoteWhat that means is that even though NeXXCom?s microwave technology needs 35 to 40 hops between the critical New York/New Jersey and Chicago markets, it?s still faster than fiber, Benti said. The company claims to cut more than 3 milliseconds from the fastest fiber networks on that route.From my understanding of propagation, if I use three redundant frequencies, and don't aim for such a high data rate, I can do it in a single hop. You don't need a high data rate, you need a fast data rate. The only problem is the geeks at the convention told me I should be looking at something from this company:SCS packet controllerI worry the latency inside the demodulator at both ends is too big. So my first guess is you need a custom barebones modulator controlling three transceivers at once. And then put all the error correction inside an i7.I am working on learning to use FPGA's as transceiver controllers. My main goal for the convention is to get the right antennae. But I think without initial testing, my understanding of the problem is too weak. There is so much crazy crap for sale there, it is not a good way for a novice to shop. I got some cool stuff already and I will only go back there again if I get bored before Sunday...

FPGA library

Posted: February 10th, 2012, 9:02 pm
by Traden4Alpha
QuoteOriginally posted by: outrunWhat I've read you can't get beyond 30 mile in a single hop (don't know why). You also need line of sight.The ultra-high frequencies used by these systems only propagate in clear air. For a spherical Earth, you need 50m-high towers to get a clear LOS between towers. Even trees and rain can attenuate the signal. A larger dish antenna, higher emitted power, or better receiver can improve the distance, but I'd bet there's a diminishing return vs. just building another tower.

FPGA library

Posted: February 10th, 2012, 9:30 pm
by Traden4Alpha
QuoteOriginally posted by: outrunwhat about laser instead of microwave?Much much worse. I'd expect near zero transmission beyond a few kilometers during any kind of fog, snow, rain, dust, smog, etc.Longer wavelengths are better WRT atmospheric attenuation but worse WRT bandwidth (which means worse latency for a given message length).

FPGA library

Posted: February 11th, 2012, 11:18 am
by farmer
QuoteOriginally posted by: Traden4AlphaI'd expect near zero transmission beyond a few kilometers during any kind of fog, snow, rain, dust, smog, etc.Even if it only gets through once, that's money.

FPGA library

Posted: February 11th, 2012, 1:44 pm
by Traden4Alpha
QuoteOriginally posted by: farmerQuoteOriginally posted by: Traden4AlphaI'd expect near zero transmission beyond a few kilometers during any kind of fog, snow, rain, dust, smog, etc.Even if it only gets through once, that's money.Indeed! Yet why mess with lasers if they have less range than microwaves?

FPGA library

Posted: February 11th, 2012, 2:09 pm
by Traden4Alpha
QuoteOriginally posted by: outrunQuoteOriginally posted by: Traden4AlphaQuoteOriginally posted by: farmerQuoteOriginally posted by: Traden4AlphaI'd expect near zero transmission beyond a few kilometers during any kind of fog, snow, rain, dust, smog, etc.Even if it only gets through once, that's money.Indeed! Yet why mess with lasers if they have less range than microwaves?Ok, what about masers then? LOL! How about just plain WiFi? I see that the current world record for WiFi is now 382 km.

FPGA library

Posted: February 11th, 2012, 9:32 pm
by farmer
I am not doing arbitrage between New York and Chicago. But if I were, I would have the radio system running in the first month. The fact that there are people doing arbitrage between New York and Chicago who don't have a proprietary radio system really makes me wonder about people. What the fuck kind of dull idiots must these people be? Are they broke? Is there no money in it? What the fuck is wrong with them?

FPGA library

Posted: February 13th, 2012, 11:42 am
by farmer
You could get people who own billboards at high points along route 80 to let you put in a repeater. You could try to do it in one hop, and then just use whatever signal gets there first.If you get a signal that the the price in New York was one tick above the price in Chicago 0.00000008 seconds ago, it doesn't really matter if you missed the 100 previous ticks.You could tolerate a pretty high error rate...0re322565yia32re32_256xyz4re32_2565yz432re32_2565yziamreal_56.5_xyz <- $$$$$$$$$$$$$$$$$$$$$$$$$$132re32_2565xy

FPGA library

Posted: February 13th, 2012, 1:44 pm
by Traden4Alpha
QuoteOriginally posted by: farmerYou could get people who own billboards at high points along route 80 to let you put in a repeater. You could try to do it in one hop, and then just use whatever signal gets there first.If you get a signal that the the price in New York was one tick above the price in Chicago 0.00000008 seconds ago, it doesn't really matter if you missed the 100 previous ticks.You could tolerate a pretty high error rate...0re322565yia32re32_256xyz4re32_2565yz432re32_2565yziamreal_56.5_xyz <- $$$$$$$$$$$$$$$$$$$$$$$$$$132re32_2565xyActually, each successful communication yields very limited expected profits defined by the instantaneous liquidity and the price differential over the system's latency advantage.

FPGA library

Posted: February 13th, 2012, 2:33 pm
by farmer
QuoteOriginally posted by: Traden4Alphaeach successful communication yields very limited expected profits defined by the instantaneous liquidity and the price differential over the system's latency advantage.In forex and spu's, the instantaneous liquidity is more than a few dollars, no?