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)
I have been relatively vocal about. As someone above points out there are some really nice new features in the NT 7 platform, so I think it is going to be nice upgrade for most. I do however think that this particularly problem is a something that really holds it back. Instead of viewing it is as a pain and putting of the future they could actually use to help differentiate themselves. Their handling of tick and bid/ask data is still the archilles heal:
* They are finally getting daily bars for future contracts from ZF, and even then it only goes back 6-9 months
* Replay data, which is the only substitute for lack of historical bid/ask is finally going to be available for download and be able to play forward continuously - however that process of getting the replay data is one day at a time (a little painful).
* Bid/ask data is available, just no sequenced and timestamped properly (useless in other words). It wil be interesting to see the samples produced from this. When I get time I am going see if that data is usable in anyway compared to OnMarketData().
* Still no continuous contract for ZF conections - makes no sense to me
I'm thinking you can use Gomi's tools to record bid/ask and have it available the correct way, the useful way. A work around to be sure, but better than what NT came up with.
Just like piersh's Genetic Optimizer. It blows the socks off the one NT came up with.
Dierk with NT just contacted me again and assures me the beta is open to anyone requesting it.
He says he has spoken to the sales team about it, but suggests if you still have a problem getting approved to just PM him here on nexusfi.com (formerly BMT). His profile is here: https://nexusfi.com/trading_member/dierk-droth.html
I'd advise anyone who isn't already in the Beta program to wait before joining.
NT7 really isn't beta quality yet, and they're currently performing major surgery on it to fix some architectural deficiencies, which will surely introduce more problems. It's quite unstable. Right now they're putting bugs in faster than they're taking them out. Several major areas can't even be tested yet because they don't work (multi-series indicators, indicators-on-indicators, and multi-timeframe strategies).
Ray (the CEO) said in their forum that he thinks they're halfway through Beta (they are 4 months into it). He's dreaming (or spinning), no one has any idea when NT7 will ship, but there is no way they ship NT7 in 4 months unless they want a tech support and PR disaster on their hands.
They are clearly "QA-challenged", to put it mildly. They appear to have little if any internal QA; we, the beta testers, are their volunteer QA staff. I'm surprised they released B9, it broke a bunch of things and didn't fix anything that I'm aware of.... on second thought, I'm not surprised.
You'll know when NT7 is approaching release status when:
1) All functionality works, and no new functionality is being put it. (it's feature complete)
2) bugs are being fixed faster than they are being found (steadily declining outstanding bug count)
3) crashes become steadily less frequent (severity-1 bugs declining toward 0)
Only then will anyone be able to make a realistic prediction about when NT7 will ship. Then, a wider beta, and a lot of documentation, will get them to release status... hopefully this year.
I totally agree. Even with its complete lack of QA I still prefer NT 7.
Several of the betas have been pretty stable for what I use, but you need to be careful when upgrading to a new beta, as they have had a habit of breaking things that already work.
Also, if you watch the forums, there are a couple of key features like correct ordering of historical bid/ask/last that they should fix, but they are very resistant to because they know it would blow them out of the water due to their lack of any real development process. The problem is they are chasing a pipe dream of getting this out fairly soon, then cave on a feature and start the whole instability issue all over again. The other major issue they seem to have is not understanding their customers needs, so they think they have a neat new feature in, only to find out that it is not useful if they don't do it right.
NT7 is a perfect example of a love hate relationship, hopefully ending in love.
The problem is that nobody in that company is a serious trader. For example, it never occurred to them that there might be a problem if they locked the chart refresh rate to 1/2 second, instead of making it variable as it was in NT 6.5. It took a near revolt by most of the users on their forums to get them to change it and allow refreshes as low as 100 ms. Even now they have their time and sales refresh locked to 1/4 second and have made no motion to change that, despite the fact that a number of traders have complained about it on the forums. Time and sales refreshing at 250 ms is useless for tape readers, but they don't seem to get that, or seem to care. Initially, they completely changed how templates worked in NT 7 and claimed that you couldn't have chart templates the same way they existed in 6.5, and only after a method was pointed out on their forums by some users did they then concede that it was possible and added it in as a feature. In fact, half of the battles on the forum seem to be about features that existed in 6.5 and were simply stripped out in version 7 with no regard for how it might affect their users. A final example is the bid/ask historical data issue. It is completely useless to have bid/ask data unless it is synchronized with the actual trades, yet that never seemed to occur to them. I don't need to get into the fact that they haven't yet figured out how to make the UI update in more than one thread, which makes the program only partially capable of taking advantage of multi-core CPUs. The list goes on and on.
Sadly, there aren't many other great choices out there, and other platforms have their problems too... I have at least managed to find work arounds for most of NT's issues.
Do you have an example of a Windows UI updating in more than one thread? The message pump is in one thread IIRC, and WndProc uses that as well as all the windows.
I believe that you are supposed to run each window with its own thread. It is correct that you cannot update a UI component from a thread that did not create it, but you could create multiple windows and spawn a thread for each one. I think that this might be an example: