Page 1 of 1

Updated: Instructions and example comparison of market data feeds (sticky post)

Posted: Mon Jul 09, 2018 11:53 pm
by Serg
Update: There are more images and video examples of such comparison HERE

HOW TO
 
  • Connect Bookmap to multiple data vendors and subscribe to the same instrument with each of them, e.g. ES futures.
  • Detach the charts by double-clicking on them (when finished, simply close the chart to attach it back to the main window) and align them as you prefer
  • Use the Synchronize charts button on the toolbar of one of the charts. This makes it the master chart and the rest will follow its settings including in Move mode, allowing to zoom-in further. The chart will stop following once it's moved or zoomed by mouse directly.
  • To make it even easier to compare, configure identical manual lower & upper cutoff on all the charts. To inherit settings of one chart by another, use right click on the chart -> Chart settings -> Inherit...

It's the same procedure for either Real-time or Replay.

GOOD TO KNOW

That time axis on Bookmap chart shows when the data was received by Bookmap. Here is why: Timestamps in Bookmap. Even if it's later by 100-200 milliseconds than exchange timestamps, trader couldn't know (and act accordingly) about the market events earlier.

Geographical distance to the exchange matters. Data vendors may throttle or aggregate the data, especially if client's internet connection is too slow to handle the data. They may also distribute the data from servers thousands kilometers away from the exchange where the data is originated.

A little reminder: the ideal latency would be your distance from the exchange, divided by speed of light c = 300,000 km/s. In other words, each 300 km of distance 'cost' 1 ms of latency if delivered with speed 1c. In practice, cables slow down the communication to 07-0.8c, the signal travels through numerous switches, and not via the shortest way. A rule of thumb for a very good connection speed would be 0.5c, i.e. each 150 km 'cost' 1 ms of latency.

Number of data subscriptions in Bookmap also matters because markets are typically correlated and intensive activity happens in bursts. For a cleaner experiment, subscribe only to data feeds you wish to compare, and reduce network usage by other software including other trading platforms.

EXAMPLE

The data was recorded from Ukraine with a 1 Gb/s connection, but via a WiFi.

Image

Bookmap was subscribed only to 2 data feeds of ES.

Re: Instructions and example comparison of market data feeds (sticky post)

Posted: Tue Oct 16, 2018 7:57 am
by ttwatrader
Hi Serg,
good summary. I would like to know what data feed belongs to which picture. Can you please add a label at each picture to differentiate and understand better what you want to show.
Thank you.

Re: Instructions and example comparison of market data feeds (sticky post)

Posted: Sun Nov 04, 2018 1:30 pm
by Serg
Hi, sorry for late response. These two are Rithmic (on top) vs CQG.

Re: Instructions and example comparison of market data feeds (sticky post)

Posted: Tue Dec 11, 2018 6:20 pm
by kahwai
Is there any US stocks feed that can support strategies? The dxFeed is blocking it.

Re: Instructions and example comparison of market data feeds (sticky post)

Posted: Mon Dec 17, 2018 1:00 pm
by Serg
kahwai wrote:
Tue Dec 11, 2018 6:20 pm
Is there any US stocks feed that can support strategies? The dxFeed is blocking it.
There are no other US stocks feeds available. But we are trying to evaluate the options to allow this with dxFeed. We are not allowed to allow exporting data, i.e. access to "non-display" data. One of the options to do it would be for us to review the code and sign it as permitted. But it's not ideal because the strategy may contain intellectual property that the trader doesn't want to expose.