Home > Technical Topics > Order Execution > The importance of 'ping' - latency/connection speed in trading.

The importance of 'ping' - latency/connection speed in trading.

  1. Hey,

    I was wondering what everyone's thoughts are on the speed which you are connected to the exchange.

    Say for example you are in Australia and you connect to Eurex to trade the DAX - Your 'ping' is around 250-350 - while not slow - the traders in Europe have a slight edge in data-speed.

    How important is this for the scalping trader - is it a huge disadvantage?

    Your thoughts...

    Thanks.
     
  2. search and read the relevant posts by FuturesTrader71, i think he says the ping latency is critical for scalping - fyi my ping latency to IB's swiss server is 45ms and to their U.S. server is 79ms, i am in Europe.
     
  3. Depends largely on your style. If you enter at market chasing breakouts it will certainly have an effect. If you enter resting orders waiting for the price to come to you it may be totally irrelevant.

    Trent
     
  4. Don't forget that:

    1) your reaction time is 250ms to 750ms depending on the day etc
    2) ping times are normally both ways so the latency is often half the ping
    3) if you use a stoplimit for your breakout it is on IBs (or whoever's) server

    All things to think about.
     
  5. If what you are doing is:

    a) observe the market-data and quickly react to its changes by hitting bids and lifting offers

    b) doing the a) w/ the computer and the decision logic runs on the computer in Australia

    you will be disadvantaged to the users who reside in Europe. You are at a disadvantage both ways: you get the market-data later, the orders you issue will get to the markets later.

    The ping-time is not a bad proxy to use to asses the size of the disadvantage.
     
  6. FWIW, when my ping times exceed 50ms, I am at a complete disadvantage to other scalpers.

    Dont forget, you're competeing with folks who are on fiber.
     
  7. speed is the most important element to scalping. you can still make money if your scalping skills make up for the handicap.

    i have a 1-2 second delay on my time and sales tape (due to crappy trading systems) and my NX's miss like 75% of the time so i miss most of my outs and get squeezed if i just got in the position. so yes, having slower connections can severely hurt your trading profits.
     
  8. Hmm - so it looks like for the Australian scalper:

    He should mainly be looking at the SFE contracts, perhaps those from Singapore/HK/Japan.
     
  9. dont know about eurex, but i would imagine ALL exchanges do what LIFFE does......

    all quotes and prints will appear at EXACTLY the same time on all screens WHERE EVER you may be in the world.

    im not a techie, so excuse me if my explanation isnt quite clear here....

    as you probably know, the Internet consists of POPs or points of presence. these are really local telephone exchanges. the exchange knows how long it takes for a ping to get to each pop, so all data is sent to the pops in terms of how long it takes the data to get to that pop. so the london pop will be sent its data a fraction of a second AFTER the ozzie pop to counter for the delay/latency the data takes to get to the ozzie pop v london pop v ny pop etc.

    i know this cos a friend of mine worked for the isp who had the liffe contract and worked on the architecture.

    i know this only deals with the outbound quotes & prints, not incoming orders, but i thought it may be useful to know that the exchanges (esp liffe) do go to a lot of trouble to assure level playing fields.

    maybe they allow a fraction of a second for all orders to come in, then process each batch as if they all have the same relative time stamp (kind of reverse to the outbound stuff) if you follow.

    hth.
     

  10. Hmm I'll do some research on that. That sounds great if thats how it's done.

    Thanks!
     
  11. i can't speak for liffe but it's not done that way elsewhere.
     
  12. can anything be done, on the client side, to decrease ping time?
     
  13. Try this website - http://www2.dslreports.com/

    Then go buy Cisco routers, high end modems and ask your ISP about guaranteeing service times.
     
  14. Yes, but rather expensive.

    I've not had any reason to look into this kind of stuff for a few years but I'd say that the info that follows would still be valid.

    You need a leased line to your broker's server. As part of the lease you may specify a maximum latency. The laws of physics (speed of light/electron flow) will determine the absolute minimum latency. The routing used by your service provider to get from your equipment to the far end is the next limitation. The latency requirement (average and peak) will determine how your provider engineers the route.

    Leased lines are priced by capacity (64 kbps, 128 kbps, ....1.544 Mbps, etc.), straight line distance between the two ends, latency (if non-standard) and level of service (reliability - do they need to provide alternate routing to the customer premises - VERY expensive).

    In general, I'd say the cost would be prohibitive for most at-home traders trading for themselves.

    Jack
     
  15. Be cautious using the term "service time" with a telecom firm. This normally refers to outages. That is service is available 99.99% of the time, 99.999%, etc.. It has nothing to do with latency (network delay).
     
  16. Use a shorter cable.
     
  17. Probably not. One thing you could do to check the performance of the client side (computer router/modem combination) is to ping or traceroute your ISP ie a mail server or time server at the ISP.

    On my machine I get about .5 msec to the router/modem, about 10 msec from modem to first external router and about 22 msec total (five hops) from the ISP's servers which are located 650 miles away.

    Seems very acceptable to me. I very much doubt that replacing the cheap Netgear modem/router with expensive Cisco kit would make any practical difference.

    The reality is that the data rates of ADSL or cable are pretty small beer in the larger scheme of things and low end kit should easily deal with them.
     
  18. Here is the reply from Euronect.liffe SVP US Technology:


    >>
    Not all exchanges do what LIFFE does. A few things differentiate LIFFE CONNECT(R) from other systems.

    1) Multicast transmission of market data.

    When an change occurs in the market (price/size), a multicast message is sent out via the API to all directly connected end-users. This message is sent out once and subsequently "heard" by all participants at the same time however with slight variance depending on where in the world the end user resides. It is pure physics in that a message sent from London to Paris arrives sooner than a message sent from London to Chicago. That being said, no other transmission method is as "fair" as one sent via multicast.

    In contract, a unicast sent message requires separate sessions with each individual end-user, which then follows that any one user could receive a market data message sooner or later than another end-user.

    2) Bandwidth Management

    In the event market data sent from the host to the end user consumes the available bandwidth at the end-user level, LIFFE CONNECT(R) will "throttle" the data for all market participants, ensuring that no one end-user is receiving information that is not available to other end-users. Other systems typically employ a per end-user queue, whereby if bandwidth for an individual member reaches capacity, that one user will be queuing messages while other market participants (who may have larger amounts of bandwidth) would be receiving messages.

    3) Dynamic release of market data

    LIFFE CONNECT(R) disseminates all market data as it becomes available, rather than "pulsing" or aggregating market data every quarter second.

    4) Full depth of book

    LIFFE CONNECT(R) shows all resting bids and offers above and below the best bid and offer.

    As far and I am aware, CME applies (1) and (3) above, Eurex (1) (4) and LIFFE/CBOT all of the above.

    Now to answer the below question specifically [edit: I asked to comment Fredbloggs post], there is no speed of light leveling employed on the network. Aside from the fairness components described above, we do not delay data to Paris so that it is in line with data arriving in Chicago a few milliseconds later. To do so is difficult, and relatively unfair in attempting to objectively determine what metric should be applied to the Paris delay. If immediacy is of great concern to a firm, it would be best for them to set up shop next to the host. However, I have yet to see a firm that cannot trade profitably from the US because they experience a 40ms difference in immediacy.
    >>
     
  19. lol - well so much for my mates version!!

    just goes to show we shouldnt believe everything we hear!

    cheers bernard.
     
  20. I'm too old to be a scalper, so maybe what I have to say doesn't apply.

    But it seems to me that if a trader can claim, "I'd be profitable if only someone didn't keep beating me to the fill by a quarter second", he's playing the wrong game. After all, it takes an average .70 seconds (700 ms for you math impared) to execute a voluntary motor response.
     
  21. Tested this on a bank stock that rarely trades.

    Put a bid on INET.

    The bastard jumping program (prob a market maker) shows up before I can even see my price on the inside.

    I have a fast connection (IB) and all this is very sub second.

    Scalping the inside price is almost impossible. You have to trade levels.
     
  22. I agree with GNOME; Maybe it's the wrong game. In my situation I have never had ping times under 1600 ms. It takes 250ms to ping my ISP who is less than a mile away. Consequently, I had to tailor style and trading product to what the system will allow.
     
  23. There's something very wrong with your computer, router, and/or modem, or (less likely) ISP. Even with dialup, ping times to your local ISP should be less than 150 ms or so.

    My cable connection runs 10-12 ms for the first couple of hops.

    My DSL connection is about 15ms.
     
  24. The phone system is 1950's vintage and once the signal makes it to the server it still has to transit the satellite to get out of here onto the grid in the real world.
     
  25. Quote from gnome:
    ... After all, it takes an average .70 seconds (700 ms for you math impared) to execute a voluntary motor response.


    Thinking that this number seemed kind of high, I threw together a simple program that prints a timestamp at random intervals, and then prints another one and shows you the difference after you hit a button.

    My response time was consistently 300-400ms. Some might say I'm old and slow, too :)
     
  26. Me too I guess, the way these guys are equpped and/or located. But I scalp anyway or quite short duration a lot down 900+ feet of rotten phone line through woods, then off island coast of Maine to Osaka and Sydney Exchanges and back. Feels like a second to fill? 1 1/2 seconds? 44kbps. When one way sat is working, which it isn't again tonight, I can't see that it's any faster for the actual trade part nor any faster than the cable set up in my daytime office. Osaka nikkei is not a small tick/low volume undertaking either yet I have never had a sense of unreasonable slippage on market orders. One of the things I do think about some on limit orders is where I might be in queue, whether price may wander away from me and the merits or not of giving up my place in that queue. Trouble for me is rarely technical, it's that I'm wrong on the trade in the first instance or have otherwise put myself in an unrewarding price area.

    This summer I read a book written at the time of the Barings blowup that said nikkei orders had to be typed and transmitted from within the city of Osaka at that time. So big traders worldwide were using squawk boxes to typists. I have no idea if this account is truthful.

    Geo.
     
  27. Without going into details: If latency affects your trading then your focus is wrong in this game.

    You cannot beat the automated boxes that are sitting on the floor of the exchange and can trade 50 times / second.

    We did extensive testing on a strategy that created its signals of 2 minute ES bars. And you know what? We could introduce 3 ~ 4 minutes delay before transmitting the signal and over 3000 trades it did not affect the results.

    :cool:
     
  28. I have nothing to add, except maybe to change it to "excessive latency."

    nitro
     
  29. well in the old days I owned an office full of very trigger happy and profitable scalpers. They (and me) all rightly sensed problems long before the clearing firm, the exchange, or the t-1 providers ever admitted to anything.

    Then when we got wise and told them what was going on they would say b.s humans cant react that fast.

    maybe. But good humans and trained humans do fire their synapses at much higher rates than non trained ones. Ivan Lendl and navratilove spent a lot of time training. Hockey goalies do the same. But even if you are average:

    We can sense that if we pullled the trigger and the quote showed up whether we should have that quote. In fact that type of feedback tells you a lot about the market and the participants.

    Once or twice you figured someone else pulled the tigger at the exact same time. but when it happened for more than a day. Office mutiny.

    As the owner I can tell you good scalpers can go ape shit over 100ms or less changes. (I know I could, but I could have been kidding myself) but the angry screaming traders in my office screaming at the computers were right even when I sometimes doubted them. One especially loud guy who was prone to belief in conspiracy was still correct every time we investigated his challenge.



    That fiber thing should concern the heck out of you if scalp size.
     
  30. Cant beat fiber or high end equipment - if you call Cisco and tell them what you're doing, they steer you to $2000 routers and up to grab the few ms you can.

    Ask them why big traders tech support always buys the latest and greatest, regardless of cost - cause they have to stay competitive!

    This game has changed from cowboy traders to cowboy techs and if you dont "get tech", they'll smoke you every day.
     
  31. Sorry, I misstated... I should have said "involuntary motor response"... as in a reaction to something (like seeing a quote and clicking off on it.... or stomping on the brakes when a ball rolls out in front of your moving car). The fastest involuntary response I can recall off-hand was some guy who won a quick-draw hand gun competition.... 0.29 seconds.... measured from the time a light signal came on and the bullet passed through an electronic target. That guy was so fast, he could put a bullet into a fella that "had the drop on him" before the other guy could react and squeeze off a round.
     
  32. Nonsense. Human psychology has stayed the same over thousands of years and it will stay the same until the end. Trading is the oldest profession.

    It is interesting to know that since the advent of the personal computer there have not been any more succesful traders than there were before the introduction of the computer.

    :cool:
     
  33. Exactly, statistics have proven traders that started trading electronically after being successful by simply calling their brokers by phone, following a small period of success they would perform worse overall.

    If the link come back to my mind I'll post it.
     
  34. Why I am jumping in I have no idea.

    But the 700 ms "involuntary motor response" can be lowered significantly via intensive training. I am in NY, and the World Cyber Games (the video game Olympics) was here this week.

    Some of top world ranked players (not one of them over age 22, I think) have shown to have "Action per second" (a manual initiated action) of > 10.

    Now, I haven't been playing any video games for 5+ years. But just watching the games have shown to me that the decision making in these games are not simple by any means, there are significant strategies / tactics to be followed, similiar to trading.

    Just seeing a weird parallel.

    Rufus

     
  35. Let it be said once more. In practically all cases the impact of internet ping/latency on trading is complete nonsense. This crap comes up regularly on ET threads.
    Most of the posters are daydreamers who have never smelled a trading profit in their lives.
    :p
     
  36. I beg to disagree.

    The fastest, biggest meanest PC, router etc etc is key to trading. No expense is too great for technology. Those who skimp on their equipment or connection are at a clear disadvantage. If the extra one or thousand means alot to you, you're not that serious about trading.

    IMHO, the advent of tech to trading has created a layer of traders who remain silent, somewhat successful and are competing well with their big buck colleagues.

    As for talking about profits, IMHO, thats simply bad luck and poor manners.
     
  37. ping means absolutely nothing on the floor.
     
  38. LRD did an article on some of the london arcades and prop shops.

    many of them now recruit via adds in video games mags. your explanation is one reason, the other is that they want kids who can sit and focus at a screen all day.



    lets face it though - this whole thing comes down to 1 issue and 1 issue alone......

    whats your strategy?

    a scalper is going to need every speed edge he can.

    average intraday trader (off 3m+ chart) probably doesnt and can make do with cabl or adsl

    swing trader probably get away with 56k dial up (do they still exist?)

    eod traders can phone in orders.

    warren buffet sends his orders in by getting a lawyer to draw up a contract which could last for months!!! the orders may then be sent via fedex or ups on the overnight!!! hows that for speed?

    you get the picture.
     
  39. I only use the best hardware. It's cheap, so why be stingy about it?
    I gave my honest considered opinion about ping/latency nonsense. Has nothing to do with manners.
    Tom cole, you don't seem to get the point: trading profitably is all tied up with computers.
    Silicon hardware matters a bit, but far more important is the performance of the computer between your ears!
    :p
     
  40. Fred,

    I just happened to stumble accross the following not unrelated to the idea of hiring fast kids to take care of the the trading:
    It may help a few in getting the picture.
    :)
     
  41. Gotmessner

    If you are just trading Eurex and/or SPI my suggestion would be to switch to the HK server. That should reduce your ping latency and number of hops by half.

    Good luck
     
  42. nono -- Thanks for your disparaging remarks ..

    Mod -- could you please banish this guy for being disruptive?

    pekalu-- Your strategy is in place by large hedgies who appear to flood the market with stock after a Cramer or some other guru issues a pronouncement - it simply washes out the weaker hands.
     
  43. Normally have a 128K ADSL but use dial up when travelling. Used to get away with 19.2K with the IB datafeed for RT but after moving to another datavendor 33K is required. Still good enough for backup though.

    :D
     
  44. :D :D :D
    tom cole,
    if you find anything that attempts to put you straight disparging or disruptive, you'll be hard pressed to ever turn a profit in the market place.
    I did what I could to help you.
    nononsense
    :cool:
     
  45. Mod-- This guy is increasingly trying, without any luck, to illicit a negative comment from me. He is very disruptive, insulting and should be banned from ET.

    nono-- IMHO, those who talk about profitability, like yourself, and comment negatively to others, tend to be either paper traders or have lost their money and simply cant stand to see others do well.
     
  46. :confused:
     
  47. tomcole - I agree that nononsense can be a pain in the neck but in this thread he seemed to be offering a valid opinion, which, as he says, has been discussed a bunch.

    You seem to be overly sensitive. I'd go back and review both your and nononsense's comments and see who seems to be acting petulantly
     
  48. Hi Whimsy,

    Why not agree with nononsense that tom cole can be a pain in the neck? About valid opinion: I think that over the years I offered many rather valid opinions, some indeed extraordinary useful. I admit that I also love to laugh at the truly nutty stuff - many do at ET. Whimsy, I'm not going to pompously insist that you be banned for this! Nono is against that kind of nonsense. :D

    In the case of this thread, I pointed out that the relationship between trading and ping/latency has come up many times in the past. Few bothered even to look at this information. Further, I strongly believe that the ONLY CRITERION TO JUDGE THE MERIT OF LATENCY IS PROFIT. I brought up this point and simply suggested that abstract reasoning about this aspect can only be kept up by individuals totally unfamiliar with the profit making side of trading. Please, correct me if I am wrong. Where did our tom cole show that he generates more profit with less latency? He did not. Hence, I assumed he must be unfamiliar with making a profit. You may find nononsense's logic rather down to earth, but it's the only way how to learn to trade profitably.

    Be good,
    nononsense
     
  49. OK, let's let the personal stuff rest.... This thread has been informative. I'd like to see it keep going. I hope this isn't too subtle.
     
  50. Sorry to be so late to the party, but I thought I'd offer my opinion....

    I've worked in IT for a bunch of different prop firms, and in all sorts of areas of finance. I feel it totally depends on your trading style. The traders I most recently worked with were developing a black box system and we had the ping times down to less than 10ms for a round trip (although that's not including the exchange time). Once you add in the exchange time, (most exchanges like the CME allow up to a second to match the order) it really doesn't matter as much as you think it does. If you really, truly are scalping, or trying to hit some weird spreads or something, then yes, it'll matter a bit. If you're doing that and you're at home, however, you had better know that you are already at a massive disadvantage. I suspect, however, that most "scalpers" are a mix of scalping and trend trading but just call themselves scalpers. Plus, if you're trading futures, the que will negate any good ping times you can achieve, and if you're trading equities on NYSE ping times don't mean anything obviously, but they will help on the NASDAQ.

    If you're trading from home, and you really think it's a concern, get a T1 unless you're on some weird island. It should be about $500 a month in most major connected cities (although i'm pulling that number out of my wazooo). DO NOT upgrade your router or anything like that, it's a waste of money. You're likely to get a 1-2 ms improvement on that, when you could get 4 months of a T1 for the same price and see a much bigger improvement. Also, try different ISPs and try to find one that shares the same facility as their backbone provider, if you can. The ISP we use (we direct connect to everything but the smallest of markets for us) shares a facility with EVERY major provider. We simply cross-connect to their cages for connectivity.

    Hope that helps,

    -The New Guy
     
  51. Hi New Guy,

    Your reasoning is quite solid. For traders not having a direct setup with an exchange, you still have to figure in the time required by the broker's computer to check the authorization and margin/funding status of the account.

    As I pointed out earlier, such discussions have been going for a long time already. Most millisecond-ping-dreamers always forget to account for these administrative processing times making a few miliseconds utterly insignificant under typical conditions.

    Be good,
    nononsense
     
  52. That's an interesting take... to be honest, I never considered that. I assumed that those parameters were loaded into my trading platform, and were checked as soon as I clicked the "make a bad trade" button, but now that you mention it, sometimes there is a spot of latency there.

    I'm guessing that peice of software resides on the gateways to the exchanges, within my brokers walls. The latency on that is probably tiny comapred to the internet latency just getting there (I don't direct connect for my personal account). But it is definately concievable if there is a seperate internet call to a seperate set of servers that it could double your latency (if that's how your BD is set up...).

    Thanks for the thought,

    -The New Guy
     
  53. My view is simply that if there are latencies along the way, the faster your order gets handled the better.

    I think its a very reasonable question to ask brokers about their systems, call centers etc.
     
  54. I did a ping test to IB data server

    ping gw1.ibllc.com

    I am on broadband

    Are these results good? I was told time had to be less than 1000ms, but what are the optimal values?


    Pinging gw1.ibllc.com [208.245.107.3] with 32 bytes of data:

    Reply from 208.245.107.3: bytes=32 time=1292ms TTL=107
    Reply from 208.245.107.3: bytes=32 time=96ms TTL=107
    Reply from 208.245.107.3: bytes=32 time=96ms TTL=107
    Reply from 208.245.107.3: bytes=32 time=97ms TTL=107

    Ping statistics for 208.245.107.3:
    Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
    Approximate round trip times in milli-seconds:
    Minimum = 96ms, Maximum = 1292ms, Average = 395ms

    Thanks
     
  55. For broadband, that's pretty good. 100 or less is very respectable, but the first one:

    Reply from 208.245.107.3: bytes=32 time=1292ms TTL=107

    is a bit of cause for concern. Do a ping -t and get a bunch of results (cntrl c to exit) and then you can see if that is a reoccuring theme over time. Even if it is, however, it might not be indicative of the problem because cisco routers (among others) will prioritize traffic and drop pings if they get busy. However, if they are that busy then there might be a problem anyway....

    Anyway, here are my results to yahoo. I beleive my set up to be about as fast as you can get right now.


    Microsoft Windows XP [Version 5.1.2600]
    (C) Copyright 1985-2001 Microsoft Corp.

    C:\Documents and Settings\comp>ping -t www.yahoo.com

    Pinging www.yahoo.akadns.net [68.142.226.46] with 32 bytes of data:

    Reply from 68.142.226.46: bytes=32 time=24ms TTL=54
    Reply from 68.142.226.46: bytes=32 time=27ms TTL=54
    Reply from 68.142.226.46: bytes=32 time=24ms TTL=54
    Reply from 68.142.226.46: bytes=32 time=26ms TTL=55
    Reply from 68.142.226.46: bytes=32 time=56ms TTL=55
    Reply from 68.142.226.46: bytes=32 time=25ms TTL=54
    Reply from 68.142.226.46: bytes=32 time=25ms TTL=54
    Reply from 68.142.226.46: bytes=32 time=25ms TTL=55

    Ping statistics for x.x.x.x:
    Packets: Sent = 8, Received = 8, Lost = 0 (0% loss),
    Approximate round trip times in milli-seconds:
    Minimum = 24ms, Maximum = 56ms, Average = 29ms

    As you can see, there's a high one in there too, but it's about double the average, vs ten times like in your example. If the majority of your times are 100 or less, you're probably doing very well.

    -The New Guy
     
  56. Try going to dslreports.com and checking out their resources - could be helpful!
     
  57. Thanks, the new ping test shows similar times, except again for the first one, but the average is 124ms
     
  58. FWIW, my API application in California, talking to an Assent server in NJ, takes 90-120ms between the time I submit an order to INET via the local API function and the time I get an "order live" ack back from the API event. This seems remarkably low to me, given that my ping times to the server are 85-90ms when the connection is "quiet". I'm certain that I'm timestamping accurately. I'm remembering that ICMP packets (used by ping) enjoy a much lower priority than other traffic, so it would appear that the actual times for "real" traffic must be quite low.

    Just the physical roundtrip distance (about 5500 miles), accounts for 30ms at the speed of light.
     
  59. Hi thenewguy

    How do you manage to get much better timings, are you not just using a modem/router and broadband ?

    Thanks
     
  60. Sorry, I should have clarified... I'm at a trading office, where I run the IT set up. We run on dual DS3's that have a very short distance to a DC that is a super POP that most everything we connect to is in also located in. From my home the times would be much, much different.

    Thanks,

    -The New Guy
     
  61. In futures (ES) i found it was best to "SEE" the technical setup being formed, then put in your limit orders for entry and have the stop loss order ready to go as "AS SOON" as you get filled.

    Then i can not blame a missed trade on a ping being slow.:cool:
     
  62. For automated trading you can do this:
    buy a U1 rack machine, set it up and ship it to co-location service, See Paul Vixie's website
    http://www.vix.com/personalcolo/

    if you go with a good provider, located in NY, London, or wherever, you may be abler to cut down latency, and increase reliability. I find that even with a decent connection at home, you get dropouts where the network seems to go away, then come back a few seconds later.

    So, if your box does not need your intervention for every trade, only supervision, i would definitely go with this setup. The coloc costs from $50 a month upward.

    K
     
  63. This is pretty much true, if the traders who compete against each other use EXACTLY the same exact strategy.

    But, if you developed a successful system using dsl, dial up, smoke signal, etc....there is no cause to believe that being faster will make you more successful. In fact, it could screw it all up. There are elements of latency (relative latency) that are part of each system. Timing is so important in entering and exiting trades, and latency is an element of timing.

    If anyone here is a musician, I think that their knowledge of different timing and rhythm would help them understand better what I am trying to say.
     
  64. Quote from pairsarb:
    ...there is no cause to believe that being faster will make you more successful. In fact, it could screw it all up. There are elements of latency (relative latency) that are part of each system. Timing is so important in entering and exiting trades, and latency is an element of timing.


    I can confirm this. As I improved my system speed, there were situations in which I performed worse, and had to re-introduce some delay to take advantage of certain market characteristics.

    Imagine what a reduction in bandwidth/CPU/brain requirements we would all experience if the bots stopped trying to compete in single-millisecond timeframes and paid more attention to quality and strategy instead.
     
  65. Just to put a different spin on things, why not consider a faster platform, say X Trader through Velocity as opposed to J Trader which has been demonstrated to be 1/2 second slower. I don't know if this was mentioned, I didn't read the entire thread. I use X Trader and I am very happy with it and Velocity, it is lightening quick.

    Also FTP, dedicated fiber to the home, is available in a few markets, with more to come, Qwest has it here in Denver, in a few places. Go cable over dsl always, it is faster, unless some weird anomoly. Good luck.

    Additionally, if using cable modem, make sure you have a newer cable modem, I get faster speed now using a Motorola SB cable modem as compared to my outdated toshiba modem that was 4 years old. Apparently it can accept a faster speed, but don't quote me on that one.

    Make sure you have the fastest computer you can afford with alot of RAM. Lastly, if you are big time, consider getting dedicated T-1 or 2 bonded T-1's, 1 T-1 may be sometimes slower than cable, but it is only your bandwidth and you get better service response should there be a outage.
     
  66. Probably correct. There are quit ea few times where a very brief lag would result in a better fill. ie, a flush that attracts bids. If you are too fast selling the flush, you get the worst price.

    With a plain old cable modem and IB , I'm already subsecond. Who is trading a system here that depends on 1/10 second latentcy?
     
  67. latency is a funny thing; you need some slow down to attract bids and offers, but I would think that over all, the fastest ping times but also the best in breed speeds of some of the automated/high volume brokers would be the best combination.

    I think traders get caught up in single ping times, but really, trading electronically is the combination of where their server is (is it co-located?), the internal processing time of the broker (I think this is overlooked, dramatically, by the speed obsessed) and the ability of that broker to not only get those orders out to the ECN"s, but also what is done with those orders. Are they just sent, but not accepted? Have they been replicated anywhere so that the system can cancel or cancel replace?

    So, the latency needs to be low in sending, receiving, processing, really everywhere. It all adds up, and you really can't measure your loss to someone who may be quicker an on a more reliable system. Low latency and high reliability in tandem is probably the most important concern you can have. A simple ping test doesn't measure the efficacy of a high quality brokerage.
     
  68. what about the matching engine algorithm?

    a FIFO (first in first out) fills on time stamp priority - so latency would be an issue - esp if scalping.

    a pro-rata fills on size priority - so ping time is less of an issue. you can jump the queue so to speak - if you are big enough