|
PureBytes Links
Trading Reference Links
|
I have now analyzed several days of data from two ESignal
subscribers including days during which they conducted an
experiment for me. I asked them to access the Internet
aggressively for a half-hour during ESig data collection,
accessing large Web pages, sending and receiving e-mails to
themselves with large attachments, and downloading very
large files from fast FTP sites. I asked that they then
pretty much keep hands off for the next half-hour, then
another half-hour of aggressive use, and finally another
hands-off half hour.
One subscriber is receiving data using Metastock and the other
is using TS2000 and the results, as described below, apply
equally to both.
The results indicate that the aggressive concurrent usage
had no effect on data reception, to my pleasant surprise.
Collected S&P ticks during the experimental half-hour
periods were either 100% or with a single tick missed, and
for the one day when I checked this aspect, almost all of
the few missing ticks (about 4 or 5 for the entire day, if I
recall correctly) were missing also from my BMI/cable feed
showing that the omissions had occurred at a very early
point in the data pipeline and not during Internet
transmission or in the user's computer.
Carroll Slemaker
|