I use the following code in my renko strategy and testing on the latest API demo account is now the same as if I ran the strategy on a live account. i.e. 1 bar late on order entry. (known problem) ..what i want to know is the way i have it coded is right ? Is it possible to get latest demo to backtest same as current live api backtests ??

 public void onBar(Instrument instrument, OfferSide offerSide, PriceRange priceRange, IRenkoBar bar ) {

        try {
            if (!isActive(order)) {
                order = null;
            } else if (order.equals(IOrder.State.CREATED)) {
                return;
            }

            boolean sellSign = false;
            boolean buySign = false;
            
            
            
            

             Object[] output = indicators.calculateIndicator(feedDescriptor, new OfferSide[]{offerSide}, 
                    "JAMES_INDI",
                    new IIndicators.AppliedPrice[]{IIndicators.AppliedPrice.CLOSE}, 
                    new Object[]{20},
                   100, bar.getEndTime(), 0);
                   
            if (!(output[0] instanceof double[])) {
                return;
            }
            
           
            double firstvalue = ((double[]) output[0])[98];
            double secondvalue = ((double[]) output[0])[99];
            double thirdvalue = ((double[]) output[1])[98];
            double fourthvalue = ((double[]) output[1])[99];
            
            print(secondvalue, bar.getEndTime());
            print(fourthvalue, bar.getEndTime());

            if (firstvalue ==3 && secondvalue  >3 ) {
                print("buying");
                buySign = true;
            }
            if (thirdvalue == -3 && fourthvalue <-3 ) {
                print("selling");
                sellSign = true;
            
            }

thanks

This you can find out by back-testing.
There have been fixes in renko generation between JForex-API versions of LIVE and DEMO. Meaning that the renko back-testing results between the two may differ. The results should match once the current JForex DEMO version will get released in LIVE.

the way i have it coded backtests very slowly on Demo ... Is there a better way to code for the latest Api?

Renko and range bar feeds in general take longer time to generate (as opposed to timeframe-based candle sticks). Try different Historical Tester settings in order to speed up your backtests (e.g. by deselecting Visual mode and changing time period and tick generation methods).

thanks for the quick answer... I have tried all those .. my strategy tests at good speed in the live API version but is untestable in demo API

I ask if we have to change the way is coded as the strategy in this post viewtopic.php?f=112&t=47731

tests at very good speed in Demo API and Live API

See if the indicator calculation time is the bottleneck, e.g.

long start = System.currentTimeMillis();
Object[] output = indicators.calculateIndicator(//...
console.getOut().println("indicator calculated in: " + (System.currentTimeMillis() - start) + " milliseconds");

ok will try that thanks...

I have this in the strategy as well

Object[] stddev = indicators.calculateIndicator(feedDescriptor, new OfferSide[] {OfferSide.BID}, "STDDEV", new IIndicators.AppliedPrice[] {AppliedPrice.CLOSE},
new Object[]{timePeriod, nbDev}, shift);
double result = (Double) stddev[0];

and is producing this error in console

29.08.2012 15:26:23.720 WARNING connect.ce ] White Label Image is not received
29.08.2012 15:26:27.843 WARNING impl.cm ] There is not enough data to calculate value for STDDEV
29.08.2012 15:26:28.561 WARNING impl.cm ] There is not enough data to calculate value for STDDEV
29.08.2012 15:26:29.387 WARNING impl.cm ] There is not enough data to calculate value for STDDEV
29.08.2012 15:26:30.308 WARNING impl.cm ] There is not enough data to calculate value for STDDEV
29.08.2012 15:26:31.142 WARNING impl.cm ] There is not enough data to calculate value for STDDEV


just took it out is now backtesting is fine in demo api in terms of speed... how can i make it work with stddev?

Please provide an example strategy which replicates the case. Also please post a precise launch scenario.

stddev calculation time is around the 1000 millisecond if I use

new Object[]{timePeriod, nbDev}, shift);

if I use

new Object[]{timePeriod, nbDev}, 1);

calculation time drops to 0 - 1 millisecond

I have added it to this strategy you wrote..mine is based of this so would be very similar if you can get this to work... and can you make inversefisher call in progress bar... Thanks

The strategy you posted uses a custom indicator, the source code of which you did not provide. For more efficient assistance always please try to keep the example strategies as compact and simple as possible. Also when we ask you to provide a launch scenario, please describe if you launch the strategy with live data or historical data, what are the settings of historical tester, etc.

Here is the strategy which we tested both in Historical Tester and with live-data, and we could not replicate the problem that you reported:

package jforex.feed.test;

import com.dukascopy.api.*;
import com.dukascopy.api.IIndicators.AppliedPrice;
import com.dukascopy.api.feed.*;
import com.dukascopy.api.feed.util.*;

public class RenkoStdDev implements IStrategy {

    private IIndicators indicators;
    private IConsole console;

    @Override
    public void onStart(final IContext context) throws JFException {
        indicators = context.getIndicators();
        console = context.getConsole();
        IFeedDescriptor feedDescriptor = new RenkoFeedDescriptor(Instrument.EURUSD, PriceRange.TWO_PIPS, OfferSide.ASK);
        context.subscribeToFeed(feedDescriptor, new IFeedListener() {
            @Override
            public void onFeedData(IFeedDescriptor feedDescriptor, ITimedData feedData) {
                try {
                    Object[] stddev = indicators.calculateIndicator(feedDescriptor, new OfferSide[] { OfferSide.BID }, "STDDEV",
                            new IIndicators.AppliedPrice[] { AppliedPrice.CLOSE }, new Object[] { 5, 1.0 }, 0);
                    console.getOut().println(String.format("____ \n Completed renko: %s \nstddev=%.5f",feedData, stddev[0]));
                } catch (JFException e) {
                    e.printStackTrace();
                }
            }
        });
    }

    public void onTick(Instrument instrument, ITick tick) throws JFException {}
    public void onBar(Instrument instrument, Period period, IBar askBar, IBar bidBar) throws JFException {}
    public void onMessage(IMessage message) throws JFException {}
    public void onAccount(IAccount account) throws JFException {}
    public void onStop() throws JFException {}
}

this is the results i get in demo back testing

2012-09-05 01:16:33   indicator1 calculated in: 754 milliseconds
2012-09-05 01:16:33   stddev=0.00023
2012-09-05 01:16:33   Completed renko: StartTime: 2012-08-27 06:00:03.044 EndTime: 2012-08-27 06:04:07.173 O: 1.2504 C: 1.2506 H: 1.2506 L: 1.2504 V: 365.57 FEC: 188
2012-09-05 01:16:33   ____
2012-09-05 01:16:32   indicator1 calculated in: 748 milliseconds
2012-09-05 01:16:32   stddev=0.00015
2012-09-05 01:16:32   Completed renko: StartTime: 2012-08-27 05:56:13.224 EndTime: 2012-08-27 06:00:03.043 O: 1.2502 C: 1.2504 H: 1.2504 L: 1.2502 V: 683.51 FEC: 328
2012-09-05 01:16:32   ____
2012-09-05 01:16:31   indicator1 calculated in: 736 milliseconds
2012-09-05 01:16:31   stddev=0.00015
2012-09-05 01:16:31   Completed renko: StartTime: 2012-08-27 05:52:20.886 EndTime: 2012-08-27 05:56:13.223 O: 1.2502 C: 1.2502 H: 1.2502 L: 1.25 V: 610.6 FEC: 291
2012-09-05 01:16:31   ____
2012-09-05 01:16:30   indicator1 calculated in: 739 milliseconds
2012-09-05 01:16:30   stddev=0.00028
2012-09-05 01:16:30   Completed renko: StartTime: 2012-08-27 05:51:11.076 EndTime: 2012-08-27 05:52:20.885 O: 1.2504 C: 1.2502 H: 1.2504 L: 1.2502 V: 143.52 FEC: 70
2012-09-05 01:16:30   ____
2012-09-05 01:16:30   indicator1 calculated in: 741 milliseconds
2012-09-05 01:16:30   stddev=0.00023
2012-09-05 01:16:30   Completed renko: StartTime: 2012-08-27 05:41:20.643 EndTime: 2012-08-27 05:51:11.075 O: 1.2506 C: 1.2504 H: 1.2506 L: 1.2504 V: 1053.84 FEC: 483
2012-09-05 01:16:30   ____
2012-09-05 01:16:29   indicator1 calculated in: 737 milliseconds
2012-09-05 01:16:29   stddev=0.00016
2012-09-05 01:16:29   Completed renko: StartTime: 2012-08-27 05:36:41.873 EndTime: 2012-08-27 05:41:20.642 O: 1.2508 C: 1.2506 H: 1.2508 L: 1.2506 V: 364.04 FEC: 153
2012-09-05 01:16:29   ____
2012-09-05 01:16:28   indicator1 calculated in: 737 milliseconds
2012-09-05 01:16:28   stddev=0.00008
2012-09-05 01:16:28   Completed renko: StartTime: 2012-08-27 02:42:45.319 EndTime: 2012-08-27 05:36:41.872 O: 1.2508 C: 1.2508 H: 1.251 L: 1.2508 V: 7805.92 FEC: 3672
2012-09-05 01:16:28   ____
2012-09-05 01:16:27   indicator1 calculated in: 731 milliseconds
2012-09-05 01:16:27   stddev=0.00015
2012-09-05 01:16:27   Completed renko: StartTime: 2012-08-27 02:35:57.541 EndTime: 2012-08-27 02:42:45.318 O: 1.2506 C: 1.2508 H: 1.2508 L: 1.2506 V: 469.56 FEC: 208
2012-09-05 01:16:27   ____
2012-09-05 01:16:27   indicator1 calculated in: 622 milliseconds
2012-09-05 01:16:27   stddev=0.00010
2012-09-05 01:16:27   Completed renko: StartTime: 2012-08-27 02:23:27.255 EndTime: 2012-08-27 02:35:57.540 O: 1.2506 C: 1.2506 H: 1.2506 L: 1.2504 V: 741.7 FEC: 344
2012-09-05 01:16:27   ____
2012-09-05 01:16:26   indicator1 calculated in: 627 milliseconds
2012-09-05 01:16:26   stddev=0.00015
2012-09-05 01:16:26   Completed renko: StartTime: 2012-08-27 02:19:55.763 EndTime: 2012-08-27 02:23:27.254 O: 1.2506 C: 1.2506 H: 1.2508 L: 1.2506 V: 454.43 FEC: 228
2012-09-05 01:16:26   ____
2012-09-05 01:16:25   indicator1 calculated in: 722 milliseconds
2012-09-05 01:16:25   stddev=0.00015
2012-09-05 01:16:25   Completed renko: StartTime: 2012-08-27 02:10:08.662 EndTime: 2012-08-27 02:19:55.762 O: 1.2504 C: 1.2506 H: 1.2506 L: 1.2504 V: 908.34 FEC: 436
2012-09-05 01:16:25   ____
2012-09-05 01:16:25   indicator1 calculated in: 728 milliseconds
2012-09-05 01:16:25   stddev=0.00015
2012-09-05 01:16:25   Completed renko: StartTime: 2012-08-27 02:09:07.254 EndTime: 2012-08-27 02:10:08.661 O: 1.2504 C: 1.2504 H: 1.2504 L: 1.2502 V: 258.88 FEC: 123
2012-09-05 01:16:25   ____
2012-09-05 01:16:24   indicator1 calculated in: 725 milliseconds
2012-09-05 01:16:24   stddev=0.00027
2012-09-05 01:16:24   Completed renko: StartTime: 2012-08-27 02:07:16.060 EndTime: 2012-08-27 02:09:07.253 O: 1.2504 C: 1.2504 H: 1.2506 L: 1.2504 V: 306.78 FEC: 148
2012-09-05 01:16:24   ____
2012-09-05 01:16:23   indicator1 calculated in: 721 milliseconds
2012-09-05 01:16:23   stddev=0.00024
2012-09-05 01:16:23   Completed renko: StartTime: 2012-08-27 02:02:20.262 EndTime: 2012-08-27 02:07:16.059 O: 1.2504 C: 1.2504 H: 1.2504 L: 1.2502 V: 642.5 FEC: 319
2012-09-05 01:16:23   ____
2012-09-05 01:16:23   indicator1 calculated in: 621 milliseconds
2012-09-05 01:16:23   stddev=0.00028
2012-09-05 01:16:23   Completed renko: StartTime: 2012-08-27 02:02:20.260 EndTime: 2012-08-27 02:02:20.261 O: 1.2506 C: 1.2504 H: 1.2506 L: 1.2504 V: 25.13 FEC: 15
2012-09-05 01:16:23   ____
2012-09-05 01:16:22   indicator1 calculated in: 723 milliseconds
2012-09-05 01:16:22   stddev=0.00028
2012-09-05 01:16:22   Completed renko: StartTime: 2012-08-27 01:59:39.518 EndTime: 2012-08-27 02:02:20.259 O: 1.2508 C: 1.2506 H: 1.2508 L: 1.2506 V: 423.06 FEC: 210

What issue are you reporting? Indicator performance of shift=0 over feed has already been registered here:
viewtopic.php?f=16&t=47923