DataFeed Recommendations

Discussion in 'Data Sets and Feeds' started by Viper101, Jul 22, 2009.

  1. I would once again disagree.

    It's not just the kind of volatility that occurs around FOMC announcements . . . it's any time that volume and price activity surge. And given the high percentage of program trading activity in the market place today, the majority of volume occurs via these programs that more often than not, can create surges in price.

    A "filtered" snap-shot feed is processed and presented in a totally different manner of technology vs a tick-feed. The "lags" are significant.

    But that's just my opinion based on my experiences in the market place as a "scalper".
     
    #41     Jul 24, 2009
  2. What did you use as your "reference" feed?

    Please tell us how you "tested" the ZEN and DTN feeds, and please be specific.

    Thanks.
     
    #42     Jul 24, 2009
  3. Vertex

    Vertex

    Last fall I happened to be running a demo of Zenfire to compare against my live IB account during a Fed announcement. Actually, I was specifically comparing ES during a volatile announcement because I had read several statements like the one quoted above.

    I remember the details because there was a striking difference in the feeds. The Fed bar opened and spiked down very fast followed by a bounce back up. At this point the Zenfire chart showed a bar low of 2.25 points lower than the IB feed, and posted those lower trades to the T&S that IB missed completely. As I remember, price then dropped back down and closed much lower basically covering the missing trades on the IB feed chart. But still at one point in the bar IB was missing 9 ticks.

    Sure, last fall was extraordinarily volatile on a normal bar much less a Fed bar, but I saw first hand how a snapshot sometimes misses things. I also saw how live feeds don't necessarily lag, at least not all of them.

    I personally think the whole "never lags because its a snapshot" is just an attempt at selling a weakness as a positive.
     
    #43     Jul 24, 2009
  4. thstart

    thstart

    Here is information directly from the source:

    "...the reality is that market participants receive data from a variety of sources: sometimes directly from exchanges, sometimes from vendors that perform consolidation. Lime has observed over 10 times the latency between SIP data and direct exchange feeds in one case..."

    "Differences in the quality of market data infrastructure at the various providers as well as at each participant cause the timing of receipt of the data to vary widely across recipients, even if they are working with the same SIP or exchange feed. Some consumers of market data will receive that data within seconds of its publication by the Market Center; others such as clients of Lime Brokerage can receive the same data in a matter of milliseconds."

    I posted this on my blog:

    http://blog.thstart.com/2009/07/technologyhigh-frequency-trading-new.html

    http://blog.thstart.com/2009/07/technologyhigh-frequency-trading-new_24.html

    http://blog.thstart.com/2009/07/technologyhigh-frequency-trading-new_9673.html
     
    #44     Jul 25, 2009
  5. moarla

    moarla

    thats the fastet move you looked at...
    and i guarantie you you will never be able to trade the lowest low in such moves :))))

    but i can guarantie you also , that in9 of 10 ZEN lags in such moves

    (in such moves a snapshot feed can miss the lowest low, when the price was there for 0,2 sec, but the ZEN feed will be 1 sec behind the reality)
     
    #45     Jul 25, 2009
  6. That may be bullshit.

    Really.

    I had the same problem - zen fire lagging like hell on ninja.

    Happens I had T&S open, not on timed redraw but real time.

    Happens Ninja could not keep up (redwarding the T&S), and kept lagging behing. Ninja - not zen fire. CPU core fully loaded, the the UI runs on one core only. Well... not zen-fire to blame.

    One has to be carefull with whom to blame. Tick updates can be stressfull for the computer. Not the feed, but the software has to cope ;) And if not, it should degrade graceful - and inform the user. ;)
     
    #46     Jul 25, 2009
  7. moarla

    moarla

    i dont care who is not able to show me the real actual price:
    the software
    the provider
    the feed
    the hardware from the feed provider
    etc....

    i want to see the actual price :)
    IB with the system they use, gives me that oportunity
     
    #47     Jul 25, 2009
  8. Ah, the bliss of being an ignorant idiot.

    Sorry, but if it if the software USE ANOTHER SOFTWARE AND DONT PLAME THE FEED.

    You probably are one of the people still thinking Sadam had weapons of mass destruction.

    Blame the right side. Dont be an idiot. Take responsibilities for your action.

    Especially as for Ninja it is a nice little property that - per default -is actually set to not let that happen.
     
    #48     Jul 25, 2009
  9. Vertex

    Vertex

    My experience since switching would seem to invalidate your guarantee.
     
    #49     Jul 25, 2009
  10. I call complete BS on your posts.

    Moreover, you never answered my question on Page 7 as to what feed you used as a "reference" when you allegedly "tested" Zen and DTN and came to the conclusion that IB was far better and didn't lag.

    Another shill for IB on ET.
    Shocker.
     
    #50     Jul 25, 2009