Support Board
Support Board
Search Board
Control Panel
View My Posts / Threads
Direct MessagesDate/Time: Thu, 09 May 2024 20:33:56 +0000
spreadsheet conversion problem
View Count: 2221
| [2016-12-05 19:15:30] |
| joer - Posts: 41 |
|
I want to try out a new computer to see if it does back testing faster than my current laptop. I am running verion 1307 because I have old spreadsheets. First I tried to install 1307 on the new pc but it said it couldn't install the files so I aborted. Then, with version 1493 installed on the new pc, I copied over a chartbook and spreadsheet from my old pc but when I opened the chartbook it said it couldn't convert the old spreadsheet to the new spreadsheet format. Other than starting from scratch and manually rewriting my spreadsheets, is there a way to get version 1307 installed on my new computer or resolve whatever it is that is keeping your conversion program from working?
|
| [2016-12-05 23:30:09] |
| |
|
What problem specifically did you have installing 1307? If you attach your Old Spreadsheet here, we can convert it for you. It can be privately attached. Here are instructions: https://www.sierrachart.com/index.php?page=PostingInformation.php#AttachFile Sierra Chart Support - Engineering Level Your definitive source for support. Other responses are from users. Try to keep your questions brief and to the point. Be aware of support policy: https://www.sierrachart.com/index.php?l=PostingInformation.php#GeneralInformation For the most reliable, advanced, and zero cost futures order routing, *change* to the Teton service: Sierra Chart Teton Futures Order Routing |
| [2016-12-06 00:10:18] |
| joer - Posts: 41 |
|
When I tried to install a previous version, I used Help, Install previous version, clicked on 1307... from list that came up, clicked agree to license agreement, program folder, and whatever other Next came up and then received a box with red X that said: Error opening file for writing: c:\sierrachart.exe, click abort to stop install, retry or ignore. I hit retry and got same message, ignore and then c:something studies.dll with same message, so I aborted. Here is copy of message log when release 1493 tried to convert my spreadsheet: Launching C:\SierraChart\SpreadsheetFileConverter.exe with parameters: "C:\SierraChart\Data\jrcontest.scwbf" "jrcontest.scss" 591504 -culture=0 | 2016-12-05 18:49:05 clZ16 [CV] 10 Min #5 | Reloading chart. | 2016-12-05 18:49:05 clZ16 [CV] 10 Min #5 | Waiting for historical downloads to complete with the IDs: 10, | 2016-12-05 18:49:05 clZ16 [CV] 10 Min #5 | Currently no data in file to display. Will start historical data download and real-time updates. File: C:\SierraChart\Data\clZ16.scid | 2016-12-05 18:49:05 Spreadsheet conversion process running. | 2016-12-05 18:49:05 Message from the spreadsheet file converter: Failed to load C:\SierraChart\Data\jrcontest.scwbf. Unhandled exception: Could not load file or assembly 'SpreadsheetGear, Version=6.0.3.190, Culture=neutral, PublicKeyToken=39c186f5904944ec' or one of its dependencies. The system cannot find the file specified. | 2016-12-05 18:49:05 * Unable to open C:\SierraChart\Data\jrcontest.scwbf because the spreadsheet file conversion failed. | 2016-12-05 18:49:05 * Launching C:\SierraChart\SpreadsheetFileConverter.exe with parameters: "C:\SierraChart\Data\jrcontest.scwbf" "jrcontest.scss" 657040 -culture=0 | 2016-12-05 18:49:07 Spreadsheet conversion process running. | 2016-12-05 18:49:07 Message from the spreadsheet file converter: Failed to load C:\SierraChart\Data\jrcontest.scwbf. Unhandled exception: Could not load file or assembly 'SpreadsheetGear, Version=6.0.3.190, Culture=neutral, PublicKeyToken=39c186f5904944ec' or one of its dependencies. The system cannot find the file specified. | 2016-12-05 18:49:07 * Unable to open C:\SierraChart\Data\jrcontest.scwbf because the spreadsheet file conversion failed. | 2016-12-05 18:49:07 * Launching C:\SierraChart\SpreadsheetFileConverter.exe with parameters: "C:\SierraChart\Data\jrcontest.scwbf" "jrcontest.scss" 722576 -culture=0 | 2016-12-05 18:49:10 Spreadsheet conversion process running. | 2016-12-05 18:49:10 Message from the spreadsheet file converter: Failed to load C:\SierraChart\Data\jrcontest.scwbf. Unhandled exception: Could not load file or assembly 'SpreadsheetGear, Version=6.0.3.190, Culture=neutral, PublicKeyToken=39c186f5904944ec' or one of its dependencies. The system cannot find the file specified. | 2016-12-05 18:49:10 * Unable to open C:\SierraChart\Data\jrcontest.scwbf because the spreadsheet file conversion failed. | 2016-12-05 18:49:10 * Launching C:\SierraChart\SpreadsheetFileConverter.exe with parameters: "C:\SierraChart\Data\jrcontest.scwbf" "jrcontest.scss" 198306 -culture=0 | 2016-12-05 18:49:12 Spreadsheet conversion process running. | 2016-12-05 18:49:12 Message from the spreadsheet file converter: Failed to load C:\SierraChart\Data\jrcontest.scwbf. Unhandled exception: Could not load file or assembly 'SpreadsheetGear, Version=6.0.3.190, Culture=neutral, PublicKeyToken=39c186f5904944ec' or one of its dependencies. The system cannot find the file specified. | 2016-12-05 18:49:12 * Unable to open C:\SierraChart\Data\jrcontest.scwbf because the spreadsheet file conversion failed. | 2016-12-05 18:49:12 * Socket (27) | Shutdown and closed. | 2016-12-05 18:49:12 |
| [2016-12-06 00:24:00] |
| |
|
In regards to: Error opening file for writing: c:\sierrachart.exe, click abort to stop install, retry or ignore. I hit retry and got same message, ignore and then c:something studies.dll with same message, so I aborted. This means Sierra Chart is running. You need to exit from all running copies of Sierra Chart. If you still have trouble restart your computer. We do not need to see the Message Log in regards to the spreadsheet conversion process. That is is not going to help us help you. We have offered to convert your spreadsheet files for you. You can attach them privately here following the instructions: https://www.sierrachart.com/index.php?page=PostingInformation.php#AttachFile Sierra Chart Support - Engineering Level Your definitive source for support. Other responses are from users. Try to keep your questions brief and to the point. Be aware of support policy: https://www.sierrachart.com/index.php?l=PostingInformation.php#GeneralInformation For the most reliable, advanced, and zero cost futures order routing, *change* to the Teton service: Sierra Chart Teton Futures Order Routing Date Time Of Last Edit: 2016-12-06 00:24:33
|
| [2016-12-06 13:32:24] |
| joer - Posts: 41 |
|
I did have Sierra Chart running on my old computer when I was trying to install on the new computer. So I exited that and then the install of 1307 with old spreadsheets was successful on the new pc. I followed instructions in the documentation about copying over certain files, then copied over one of the chartbooks and spreadsheet I am working on. Then I ran the same backtest on both computers and got significantly different results. For just one month the total profits differed by more than 50% between the 2 computers. Looking thru the trade activity logs, I see this was caused by just a few ticks here or there. For example, one trade had a difference of 3 ticks between the entry point for the exact same trade which caused one to be stopped out and the other to be profitable. While most trades are exactly the same, these small differences snowball into making complete trades that followed to not be triggered or extra trades to be taken. Over the month this turned into a large variance. My question is, does the processor speed or hard disk speed affect the results of your back testing software or is there something I missed when setting up my new computer?
|
| [2016-12-06 17:39:11] |
| |
|
Most likely the difference has to do with the underlying data in the chart. How are you running the back test specifically? Sierra Chart Support - Engineering Level Your definitive source for support. Other responses are from users. Try to keep your questions brief and to the point. Be aware of support policy: https://www.sierrachart.com/index.php?l=PostingInformation.php#GeneralInformation For the most reliable, advanced, and zero cost futures order routing, *change* to the Teton service: Sierra Chart Teton Futures Order Routing |
| [2016-12-06 19:45:25] |
| joer - Posts: 41 |
|
I am running AutoTrade System Replay Back Test with a continuous futures contract - Volume Based Rollover, 5 minute chart. My spreadsheet uses attached orders and user exits. I am using the SC Historical Data service for data with intraday data storage time unit of 1 tick.
|
| [2016-12-06 22:51:15] |
| |
|
There may be some difference with the underlying data. Make sure there is a tick by tick data configuration in each instance of Sierra Chart: Tick by Tick Data Configuration And re-download the data in both charts you are back testing from. Also refer to: https://www.sierrachart.com/index.php?page=doc/Backtesting.php#BackTestConsistency Sierra Chart Support - Engineering Level Your definitive source for support. Other responses are from users. Try to keep your questions brief and to the point. Be aware of support policy: https://www.sierrachart.com/index.php?l=PostingInformation.php#GeneralInformation For the most reliable, advanced, and zero cost futures order routing, *change* to the Teton service: Sierra Chart Teton Futures Order Routing |
| [2017-01-30 23:01:38] |
| joer - Posts: 41 |
|
It seems that I get the same error messages with any spreadsheet I try to convert to the new format. I appreciate your offer to convert my spreadsheets and am attaching 3 of them here.
|
Private File Private File Private File Attachment Deleted. |
| [2017-01-31 00:55:13] |
| |
|
The converted spreadsheets are attached.
Sierra Chart Support |
Private File Private File Private File |
| [2017-02-01 23:22:53] |
| joer - Posts: 41 |
|
Thanks for converting the spreadsheets. I started working with one of the new ones and the first thing I did was run a backtest to compare with results from the old version. I have 2 copies of Sierracharts on my computer, one with an old release that supports the old spreadsheets (1307)and one with a recent release (1510)for the new spreadsheets. I followed your documentation for installing multiple copies and copied all the files that were listed there. So I set up a backtest with 1 month worth of data and ran the Auto Trade System Replay Backtest with nothing else running on my system. I ran each test separately. The good news is the test results were the same for the old and the new spreadsheet. The bad news is the old test took less than 2 minutes but the new test took more than 6 minutes. The time difference was so large that I made sure nothing was running in the background and reran the new spreadsheet test 2 more times but the results were the same. I'm sure this cannot be normal. Actually I was hoping the backtest time would be faster with the new spreadsheets but at least the same. Anything I can try to fix this problem? Date Time Of Last Edit: 2017-02-02 04:16:24
|
| [2017-02-02 04:16:43] |
| |
|
We need to know what specific Spreadsheet you are using and we need the Chartbook you are using and we will have to look at this and see if there is anything we can do.
Sierra Chart Support - Engineering Level Your definitive source for support. Other responses are from users. Try to keep your questions brief and to the point. Be aware of support policy: https://www.sierrachart.com/index.php?l=PostingInformation.php#GeneralInformation For the most reliable, advanced, and zero cost futures order routing, *change* to the Teton service: Sierra Chart Teton Futures Order Routing |
| [2017-02-02 04:46:59] |
| joer - Posts: 41 |
|
The test I ran is with sheet4. The chartbook and spreadsheet are attached.
|
Private File Private File |
| [2017-02-02 07:52:28] |
| |
|
We will look this over as soon as we can but allow us about 3 to 4 weeks.
Sierra Chart Support - Engineering Level Your definitive source for support. Other responses are from users. Try to keep your questions brief and to the point. Be aware of support policy: https://www.sierrachart.com/index.php?l=PostingInformation.php#GeneralInformation For the most reliable, advanced, and zero cost futures order routing, *change* to the Teton service: Sierra Chart Teton Futures Order Routing |
| [2017-02-03 03:40:10] |
| joer - Posts: 41 |
|
I appreciate you looking into this and will await your findings. In the meantime can you tell me if it is normal for backtesting of the new spreadsheets to take significantly longer than the old ones?
|
| [2017-02-03 04:12:40] |
| |
|
It could be. We would not have expected that with performance improvements we did about more than a year ago. This is why we are going to look at the Spreadsheet you are using and see what can be done to improve upon the performance. Sierra Chart Support - Engineering Level Your definitive source for support. Other responses are from users. Try to keep your questions brief and to the point. Be aware of support policy: https://www.sierrachart.com/index.php?l=PostingInformation.php#GeneralInformation For the most reliable, advanced, and zero cost futures order routing, *change* to the Teton service: Sierra Chart Teton Futures Order Routing |
| [2017-02-04 11:03:30] |
| |
|
We started to have a look at your spreadsheet. And we will be able to make performance improvements. We are not sure how soon that will be. Some improvements can be made this month and other improvements will take longer.
Sierra Chart Support - Engineering Level Your definitive source for support. Other responses are from users. Try to keep your questions brief and to the point. Be aware of support policy: https://www.sierrachart.com/index.php?l=PostingInformation.php#GeneralInformation For the most reliable, advanced, and zero cost futures order routing, *change* to the Teton service: Sierra Chart Teton Futures Order Routing |
| [2017-03-19 20:03:09] |
| joer - Posts: 41 |
|
Just wondering if you've made any progress with the painfully slow back testing with new spreadsheets. I also notice significant differences between back test results and replay results. Attached is a Trade Activity Log from a spreadsheet system for trading. I have data loaded from 2/16/2017 to 3/17/2017. First I ran an Autotrade System Replay Back Test for all the data, then I ran Replay from as close to the beginning of the chart as I could get. The trades at the end of the activity log from 2/16/2017 to 2/24/2017 (created by the Replay) are much different from the results for those same days at the beginning of the log (from Autotrade System Replay Back Test). I changed nothing in the spreadsheet or the chart. Results like this combined with the back testing taking 3 to 6 times as long as with the old spreadsheets make it impossible for me to use the new spreadsheets.
|
Private File |
| [2017-03-21 03:38:52] |
| |
|
Some progress but not completely. That could be many more months. We are doing our best. But we have a lot to do. You can evaluate whatever improvement, if any in your case, simply by using the latest version. In regards to the different back test results, answering that question is not within the scope of our support for all of the reasons explained here: Automated Trading Management: Troubleshooting Automated Trading System Behavior What you need to do is to read this section and use one of the example trading systems that we provide to see if you still notice inconsistencies. https://www.sierrachart.com/index.php?page=doc/Backtesting.php#BackTestConsistency And you also need to perform a manual replay using the same settings on the Chart Replay window as the automatic replay. This is explained at step 7 here: Auto Trade System Back Testing: Replay Back Testing - Automatic Sierra Chart Support - Engineering Level Your definitive source for support. Other responses are from users. Try to keep your questions brief and to the point. Be aware of support policy: https://www.sierrachart.com/index.php?l=PostingInformation.php#GeneralInformation For the most reliable, advanced, and zero cost futures order routing, *change* to the Teton service: Sierra Chart Teton Futures Order Routing Date Time Of Last Edit: 2017-03-21 03:48:24
|
| [2017-03-22 15:20:22] |
| joer - Posts: 41 |
|
So far I have not noticed any improvements in the backtesting run time. Regarding my question about different backtest results, I did not encounter different results when running the same backtest. The example trade activity log I sent you contained results from an automatic backtest and then for a few days at the end of the log the results were from a manual replay. The settings were the same except for the speed. As I stated: "First I ran an Autotrade System Replay Back Test (this is what I believe you are referring to when you say automatic replay) for all the data, then I ran Replay (this is a manual replay) from as close to the beginning of the chart as I could get. The trades at the end of the activity log from 2/16/2017 to 2/24/2017 (created by the Replay) are much different from the results for those same days at the beginning of the log (from Autotrade System Replay Back Test). I changed nothing in the spreadsheet or the chart." Please note on the activity log I attached previously where the automatic replay had 2 trades on 2/17, 2 trades on 2/22, 1 trade on 2/23 and no trades on 2/24 while the manual replay (results at the end of the activity log) had 1 trade on 2/17, 2 trades on 2/22, no trades on 2/23 and 2 trades on 2/24. I don't see how this can be a problem with my spreadsheet since I changed nothing yet get different results in the automatic replay vs. the manual replay. I have seen this with tests from different spreadsheets, not just the one I sent you. |
| [2017-03-22 20:13:17] |
| |
|
So far I have not noticed any improvements in the backtesting run time.
What version are you testing with? The current version is 1531. In response to: I did not encounter different results when running the same backtest. The example trade activity log I sent you contained results from an automatic backtest and then for a few days at the end of the log the results were from a manual replay. The settings were the same except for the speed. As I stated: "First I ran an Autotrade System Replay Back Test (this is what I believe you are referring to when you say automatic replay) for all the data, then I ran Replay (this is a manual replay) from as close to the beginning of the chart as I could get. The trades at the end of the activity log from 2/16/2017 to 2/24/2017 (created by the Replay) are much different from the results for those same days at the beginning of the log (from Autotrade System Replay Back Test). I changed nothing in the spreadsheet or the chart." Please note on the activity log I attached previously where the automatic replay had 2 trades on 2/17, 2 trades on 2/22, 1 trade on 2/23 and no trades on 2/24 while the manual replay (results at the end of the activity log) had 1 trade on 2/17, 2 trades on 2/22, no trades on 2/23 and 2 trades on 2/24. I don't see how this can be a problem with my spreadsheet since I changed nothing yet get different results in the automatic replay vs. the manual replay. I have seen this with tests from different spreadsheets, not just the one I sent you. Our answer remains the same: In regards to the different back test results, answering that question is not within the scope of our support for all of the reasons explained here: Automated Trading Management: Troubleshooting Automated Trading System Behavior What you need to do is to read this section and use one of the example trading systems that we provide to see if you still notice inconsistencies. https://www.sierrachart.com/index.php?page=doc/Backtesting.php#BackTestConsistency And you also need to perform a manual replay using the same settings on the Chart Replay window as the automatic replay. This is explained at step 7 here: Auto Trade System Back Testing: Replay Back Testing - Automatic The settings were the same except for the speed. You need to use the same speed and that is documented. And also specifically what settings were the same?Please understand, there is no way we can possibly analyze this. That is a very costly and difficult effort that is not within the scope of our support unless we know there is a problem we need to look into. The information provided is not a basis for that. You need to do the comparison using a Sierra Chart provided trading system and see if you notice any differences between an automatic and manual replay back test. Not your own. Sierra Chart Support - Engineering Level Your definitive source for support. Other responses are from users. Try to keep your questions brief and to the point. Be aware of support policy: https://www.sierrachart.com/index.php?l=PostingInformation.php#GeneralInformation For the most reliable, advanced, and zero cost futures order routing, *change* to the Teton service: Sierra Chart Teton Futures Order Routing Date Time Of Last Edit: 2017-03-22 20:17:04
|
| [2017-03-23 00:13:20] |
| |
|
Improving performance is very important to us and that task is still an active task for us but we just have not been able to spend recent time on it. We will also test a basic automated trading system using Spreadsheets today and see if we notice the inconsistency you are seeing. This new Spreadsheet Input also may be helpful to eliminate the inconsistency: Spreadsheet Systems, Alerts and Automated Trading: Minimum Chart Bars Required for Entry Exit Cell Evaluation Sierra Chart Support - Engineering Level Your definitive source for support. Other responses are from users. Try to keep your questions brief and to the point. Be aware of support policy: https://www.sierrachart.com/index.php?l=PostingInformation.php#GeneralInformation For the most reliable, advanced, and zero cost futures order routing, *change* to the Teton service: Sierra Chart Teton Futures Order Routing |
To post a message in this thread, you need to log in with your Sierra Chart account: