MultiCharts: Trading Software for Automated Trading and Backtesting

I know you know that there are problems with the n-day charts, and I know they're on your "to fix" list, but I thought it might be helpful to post a picture illustrating some of the issues.

Attached is a chart (screen grab and workspace) showing EURUSD in the 1-day scale, and each of the possible 3-day permutations on the same horizontal axis.

What I'd expect to see is a set of evenly-spaced candles in each 3-day chart. In fact - even ignoring the daylight-saving glitch that causes the daily bars to be double-spaced in the winter months, the bars are spaced unevenly. The nominally 3-day charts all consist of a jumble of 1, 2, and 3-day bars.

The problem appears to stem from the 3-day charts using a 7-day week (and counting Saturday and Sunday as two working days) while the daily chart excludes the weekends. In any case, it means that the n-day charts don't really make any sense, because the horizontal axis of any single chart isn't linear.

Hi MKlein,

Thank you for reporting the n-day problem. It is a known issue and will be fixed in future versions of MCFX (in about 1 - 1.5 months).

Regards.

Thanks Marina.

For reference, could you post a list of known bugs in the current version? At least that would save multiple reporting, and it would alert users to problems of which they might not otherwise be aware. Until you look closely, an n-day chart looks valid.

Hi MKlein,

At the moment we don't have a list of bugs for the latest version of MCFX. The bugs that we are aware of are quite minor and would be of no interest to the users.

Your reporting, on the other hand, is very important for us because it helps us find the bugs that we haven't been able to identify in tests.

Regards.

At the moment we don't have a list of bugs for the latest version of MCFX. The bugs that we are aware of are quite minor and would be of no interest to the users.

If you include this n-day problem in that description, I really can't agree. On cursory inspection the charts look correct, but on closer analysis they're false in every respect because the horizontal axis is inconsistent. That means that all indicators and candlestick patterns derived from them are also false. Essentially they're useless. MCFX users should really be made aware of that. The annoying thing is that this bug was introduced in a fairly recent version, prior to which they worked well.

Your reporting, on the other hand, is very important for us because it helps us find the bugs that we haven't been able to identify in tests.

Thanks. I'll do my best to keep the reports coming in.

At the moment we don't have a list of bugs for the latest version of MCFX. The bugs that we are aware of are quite minor and would be of no interest to the users.

If you include this n-day problem in that description, I really can't agree. On cursory inspection the charts look correct, but on closer analysis they're false in every respect because the horizontal axis is inconsistent. That means that all indicators and candlestick patterns derived from them are also false. Essentially they're useless. MCFX users should really be made aware of that. The annoying thing is that this bug was introduced in a fairly recent version, prior to which they worked well.

Your reporting, on the other hand, is very important for us because it helps us find the bugs that we haven't been able to identify in tests.

Thanks. I'll do my best to keep the reports coming in.

Hi MKlein,

If we identify enough major bugs to make up a list, we will publish it.

Regards.

Thank you for reporting the n-day problem. It is a known issue and will be fixed in future versions of MCFX (in about 1 - 1.5 months).

I wonder, do you have an update on progress with this bug?

If we identify enough major bugs to make up a list, we will publish it.

Wouldn't it be a good idea to publish such a list, even if it is a list of zero elements? It would be a benefit to your customers to see which issues are known, which are being fixed, and an estimated timeframe for each; and just think of the propaganda advantage of an empty list!

"There are no known bugs in the current version of MCFX!"

Imagine that!

I wonder, do you have an update on progress with this bug?

...or if not, do you know when you might?

Thanks.

I notice a possible similar problem with intraday charts:

I have a 2 second chart and a 5 min chart of EUR / USD.

Often, horizontal space between 5 mn candles is uneven: sometimes there is more space, sometime less. I'm not sure if it is the same cause or if it is when 2 sec data is missing.

I notice a possible similar problem with intraday charts:

I have a 2 second chart and a 5 min chart of EUR / USD.

Often, horizontal space between 5 mn candles is uneven: sometimes there is more space, sometime less. I'm not sure if it is the same cause or if it is when 2 sec data is missing.

I think it's related, but not quite the same effect. If there's no price movement in a given time period, MCFX doesn't record a bar. That means that many of your 2 second bars will be empty, and therefore not displayed. If you select "show empty periods" (Format Window / X - Time Scale / Show Empty Periods) the 2-second chart should be spaced linearly, and that should correct the spacing of the 5-minute bars - I think.

A related problem (but not quite the same) that affects my charts is the correspondence between 12-hour and daily charts. In principle, each day should be able to be divided into two 12-hour periods. However there are often stray ticks that occur between nominal market close on a Friday and nominal market open on Sunday evening. These weekend ticks and the inconsistent handling of the Sunday open mean that the regular correspondence between 12-hour and 1-day charts is often broken at the weekend by the insertion of spurious short 12-hour bars. The weekend of 13-14 December is an example, where EUR/USD shows two short 12-hour bars for the 14th, but no daily bar.

Thank you for reporting the n-day problem. It is a known issue and will be fixed in future versions of MCFX (in about 1 - 1.5 months).

I wonder, do you have an update on progress with this bug?

If we identify enough major bugs to make up a list, we will publish it.

Wouldn't it be a good idea to publish such a list, even if it is a list of zero elements? It would be a benefit to your customers to see which issues are known, which are being fixed, and an estimated timeframe for each; and just think of the propaganda advantage of an empty list!

"There are no known bugs in the current version of MCFX!"

Imagine that!

MKlein,
N-Day problem has been fixed. Please test it and let me know how it works.

N-Day problem has been fixed. Please test it and let me know how it works.

From a cursory inspection, that looks fine. Thank you.

I'll test more thoroughly later today.

N-Day problem has been fixed. Please test it and let me know how it works.

From a cursory inspection, that looks fine. Thank you.

I'll test more thoroughly later today.

You are welcome!