Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
in this thread I am giving you a short movie and 3 screenshots from its for showing, what is scary in this market data feed from Continuum for me. The instrument is Crude Oil, CL, and this scary movements I had seen at least at 28. of April 2016. This video now are from today, the 12. of March 2016.
Please: I know, what spread expansions are. And that, what you can watching here are not typical spread expansions. In this short video you can see, that in one case the bid-line is jumping (second 21), in another situation the ask-line is jumping (second 43) and in a third case both are jumping together (second 26).
This is not a price moving. Its a very fast jump and back for a reason.
I want to know, REALLY TO KNOW - EXACTLY, what the holy grail that is. Its Continuum? Its an algorithm, that f***s the traders special in crude oil? Or its an cable problem, that is only in special times doing crazy things? Can I trust Continuum fully? Do I get my money back from Ninjatraderbrokerage in case of filling by this strange irregular movements (losts or stopping out more than 60 ticks) AGAINST me?
Best regards,
Renkotrader
Can you help answer these questions from other members on NexusFi?
@Renkotrader, it looks like a kind of artifact, not a real bid/ask.
You should try to:
- download the market replay for this specific day and check if it was "real". As it will come from another data source it will confirm it did not happen on the NYMEX itself
- create a simple indicator which will log these events (on a 1 tick chart, bid (or ask) for bar 0 x% higher/lower than bid (or ask) for bar 1)
@Pure charts: no, I have no test by a pure chart without indicators. And really only in CL and and only sometimes I had seen its. Do you think, that its possible, that an indicator is creating this problem?
@market data: now not at trading computer, back in some days. I hope, that I will find this. In NT7 there are the option for this, isnt it?
@Creating indicator: sorry, I dont know, how and what.
In Ireland I had a similar (possibly the same? but I was using ES mostly) problem with random price moves. It was NT 7 and in 2014.
It was (in my case) my particular e-fibre broadband technology was corrupting packets coming from Chicago. NT is not good with error correction. Not a problem specific to my home but the ISP nationwide, confirmed by their support and my tests in 3 locations. On a different ISP no problem and then I moved to Colombia (same laptops, no issues).
Other times when the market was busy an ATM order would be partially submitted (and accepted live), e.g. missing stop or target.
For certain you need to test using a network cable and not on Wifi (I'm sure you have).
You might try a continuous ping logged to a text file e.g. ping -t YourCQGserver > ping.txt
NT brokerage support would certainly help seeing your video but they would in no way actually admit that NT was worse than other leading apps on a poor connection (and it is). It was bordering comical how they avoided saying this, I suspect it was support policy.
They did (surprisingly helpfully) offer to "pilot customer" connect me to CQG's gateway in London but I discovered the issue was my ISP(then moved to Colombia). I am not sure if NT brokerage still route everything to Chicago or they have opened gateways in Europe but support could give you your current servers or netstat -n may show them.
Hmm, that's very very strange.
There are error corrections in multiple places: on the broadband size (fiber or DSL or whatever) and then in the Ethernet packets themselves (Frame check sequence). I assume there are also CRC in the data feed provider API itself so I don't see how the broadband link could be the root cause.
I f the payload is corrupted (the useful data in the network frame) then it's CRC will failed and the packet will be dropped at layer 2, higher layers (operating system) won't see it.
So I assume your ISP was corrupting packets and was sending valid CRC what a shame, that's criminal!
@Renkotrader first I would say, the problem I describe is only a possibility to check, but NT support are helpful and you have video evidence.
@sam028 Indeed strange. I doubt it was a false CRC issue though there were many bad packets, it was at the application level inside NT when dealing with a high error rate connection. Perhaps a threading or buffer size issue? Everything else perfectly fine, Web sites, torrents, data file downloads (6GB ISOs etc, perfect, not damaged files). Sierrachart running in parallel (on CQG data) perfect when these glitches/artifacts appeared in NT.
Eircom's e-fiber ran the last section over old copper phone lines however the speeds were remarkably fast. Their support confirmed that certain gamers also encountered a problem and it was 'nature of the technology'. They are rolling out fibre to house nationwide now to eliminate the copper leg and going to 1Gb standard.
I was an IT consultant and I took a long time to accept it (and it cost me $$). PathPing showed various problems even leaving the house (in 3 homes around the country).
A guy in Australia confirmed the same issue with his ISP & NT and a friend in the US had it recently though this was probably down to faulty equipment (certainly we had ATM orders with no Stop submitted twice and other issues). NT has a hand in it though as something is poorly written.
I know to anyone technical it seems... strange... easy to think could have been this or that. NT support very obviously knew (by how they skirted it on the phone and in email) but were not going to admit it.
This was very frustrating however they did offer to connect me (or try to arrange) via a European CQG Gateway so I have to give them credit for that. It may have reduced the problem however my move to Colombia fix is much better.
A trading PC should be used for nothing but trading...Not running in tandem with all that junk mentioned above. In fact, posting to websites on your trading PC should be frowned upon. Get another PC to have fun on. (To download torrents and movies, etc). They are cheap. A full PC package with a monitor these days costs what, 500 bux? Treat your trading like a business, your trading PC as a piece of the business, and you will get farther along I reckon'.
Good advice however.... IT consultant.. ? just kidding. A few minutes to kill waiting for someone...
In truth you can get away with a lot of crap running no problem but certainly keep it simple if your not an old pro and clearly nothing running when diagnosing a problem. Also no WiFi as this can lead to congestion and losses, use an RJ45 cable.
The best thing anyone can do is enough ram (8Gb+) and a decent SSD. This really smooths out the bumps with wait states and locking etc. It makes a world of a difference in NT 7. Excluding your NT folders from AV active scanning is also a good idea and was Ninjatrader's advice also.