That's great, Virtusa. Thank you. Did you do any such tests during more volatile/high volume periods? Would that change things? I did some test trading yesterday and followed the entire session with my left eye. Didn't experience any issues.
It is only data, so that does not take much bandwidth. Volatility can change a bit but not much as NT8 refreshes on 250ms. So the data coming in will be the same size I think as you get only an update every 250ms. It will be sooner a problem for your CPU then for your connection. I once lowered the refresh rate and in volatile markets, my CPU could not keep up with the calculations. I once traded from a hotel in Sofia. So very simple connection. Had no problems at all. Graphics would be much more demanding.So if you open Youtube you will need Mbs of bandwitdh. You will never have any problems with the 30Mbs connection. Plenty of bandwidth. If you install Networx V5.5 you can monitor it yourself. Runs on the background and needs almost no resources. https://www.majorgeeks.com/files/details/networx.html
So, I'm having some issues. Basically, my charts seems to be lagging a bit. Yesterday, an order triggered (I could hear the sound), but at the same time I could see with my very eyes that the price had not printed on my charts. According to the task manager - my system have ample resources free across the board. I also did another test on latency while trading and experiencing this and it looks similar to before as far as I can see. There's a spike similar to earlier, but I wonder if that could be triggered by the test itself: I'm not fully convinced it's the internet connection that's my issue as I did experience something similar briefly in the past a few times.
Maybe check in the tab YTELSE the chart of the CPU. How high is he going? You should try to have that chart running and wait till the problem might occur again. I had in past a similar problem. The cause was that I set the refresh rate in NT8 on 100ms with recalculate every tick, with the result that in fast markets and high volume the CPU could not follow with the recalculations. You can also test with freeware: https://www.ocbase.com/ Does not look like an internet problem.
This will sound really stupid to be reminded of but... The order type matters. It COULD explain the "sound" alert with no price print. Limit orders require a print (execution) at or better to trigger. Stop and MIT orders are triggered by QUOTE. Stop-limit orders are first triggered by QUOTE. NB: Exchange rules are not necessarily uniform across venues.
I will take a look tomorrow. Thanks. How do you adjust refresh rate in NT8? I had to Google it and from what I found it seemed to actually be a complaint that the refresh rate on NT8 is fixed at 250 ms and not adjustable without a custom indicator. Interestingly, I found a thread with many other users experiencing issues with lagging charts in NT8. Could you show me the change you did? I use a few indicators, but nothing that should be too consuming. And I have them all set at "Calculate on price change". Not on every tick. Thank you.
https://ninjatraderecosystem.com/user-app-share-download/tickrefresh/ I don't see the need for faster than 250ms, as that means already 4 times per second. I tried it but the calulations were lagging in fast markets. At 100ms, the CPU has to calculate already 2.5 times faster. And at that speed you can never see all the refreshed values at it means that valies change 10 times per second. You need a very powerful CPU for fast refresh rates. And surely don't open too much workspaces at the same time. Even at 250ms refresh rate you never know what price you will get unless you use a limit order. I always go for market price to be sure to be filled. If you put an order in the market, the refresh rate is irrelevant as the exchange will fill you at the price of the order. Your refresh rate has no impact on the fill.
This is the result of a test. Price was compared for different refresh rates going from realtime to 250ms. The best quote was 52.76 and the worst one was 52.77 For most traders it makes no sense to reduce the refresh rate. If the 0.01 makes a real difference in performance you should improve your trading system, not reduce the refresh rate. The fact that in this case the slowest version gives the best price, proves what I told in previous post: you cannot predict which refresh rate will give the best price as prices move up an down and very fast speed. Only when you trade at marketprice there can theoretically be an advantage as all other orders are in the market and will be executed independently from the refresh rate you use, as the execution is done at the prices at the exchange. So faster execution does not automatically mean better execution. At 1 ms, the CPU had to work 250 times faster, you will never be able to monitor all the prices that will pop up on your screen, and in the case of the test, execution will be worse than at 250ms.