Not to kick a man when he's down but I would have to agree here. It's got nothing to do with .Net whatsoever and everything to do with the new 2D graphic API (I forgot which one) that was chosen. Obviously we don't see these problems on NT7, do we?
And yes, absolutely need to agree with jcrza on this - I have been in computing since the early 1980s and drawing a crosshair fluidly wasn't an issue past 1990 or so. Don't you guys find it silly that we may have to upgrade our hardware in order to run a charting API without skipping frames?
Quite frankly, this should not even be an issue in this day and age. We are using super computers as phones these days and a skipped framerates wouldn't even be acceptable on mobile platforms. I completely understand the hesitation to make drastic changes and delay the final release. Marketing at NT would freak out at the mere mentioning and we've been using NT7 since what - 2010? I can't remember - it's been this long....
I'm not posting this to fan any flames of discontent. But I would like to encourage everyone to take a step back and look at this from an objective perspective. NT Team: How do you think the (growing) competition is going to use this? I can hear it now: "Sure you can try to use NT8 if you don't care about sluggish charts."
Meanwhile we have community driven tools like Quantopian and QuantConnect grow by leaps and bounds. They are still in the early beta phase but I expect both efforts to aggressively push into NinjaTrader's market niche going forward. This cannot be ignored. While we are worrying about skipped frames on NT7 those guys are implementing a community driven, open source, cross-platform algorithmic trading engine that runs in the cloud. Backtests that may take hours or days on NT can be performed in mere minutes or hours. It's a completely different ballgame - assuming they can pull it off and support futures and forex markets. How is NT going to compete on such a level let's say three to five years from now?
Personally I have made a huge investment in NT over the past eight years. All my systems, all my indicators and tools are running on NT7 and migration of only a subset of that will take me weeks and perhaps months. Lacking an automated migration tool for my existing codebase I don't find the idea of spending all that time just to face graphic issues like this particularly appetizing. And quite frankly I am tempted to consider if my time may be better spent porting it to Quantopian or QuantConnect. I rather would not but reports like these do not strengthen my confidence in the long term potential of NT8.
It took over 5 years (correct me if I'm wrong) from the announcement of NT7 to the early beta release of NT8. On NT7 I am still forced to support .Net 3.5 which is hopelessly outdated. How long until NT8 is in a similar situation? Will I have to wait until 2020 or later to see NT9? And if the drawing API is showing fundamental issues at this early stage - as opposed to showing a significant speed *increase* - then I am curious as to what my incentive is.
In closing - the devil quite often lurks in the detail. This is a very important detail that may haunt you guys for many years to come. If it can not be fixed then I propose that a different graphic API is being considered. I know that sucks but end users don't care about the technical details. Let's not forget that trading applications are mission critical and traders are not known for being patient or forgiving if they feel that their platform is slowing them down.
Food for thought...
Comment