Tweet
I have just re-downloaded & re-installed 7.5, and the following continue to happen with BackTest:
1. Backtest refuses the Time Template I give it, and just makes up one of its own which is similar to nothing I have on file -- I need to test 20 days and it gives me 6 when I have nothing close to a 6-day Template, etc. When I finally get it to take the correct template, I don't know how I did it -- it seems to be making up its own mind randomly;
2. Frequently starts the Daily Trading Summary, item 1, page 1, with lots of (random?) stuff in the "...Rolling Period..." columns, thereby making totals and comparisons totally useless. I suspect it's analyzing days PRIOR to the top-line date, and just not telling anybody about it;
3. Changes output from run-to-run with identical data and identical studies;
4. Started "HIDING" the Formula Output Window, & will only show it after a LONG period of the cursor sitting where the data is SUPPOSED to be (the top ribbon appears, but no body), then disappears again when the cursor leaves the data section, then takes an eternity (up to 60 seconds) to REappear when the cursor is moved back over the data area (same data, it didn't need to rethink anything -- why'd it take a minute?).
The above seems to have started spasmodically a couple of days ago, and reached the point where BackTest is totally unstable and untrustworthy. It matters not whether I use new, under-development code or old, tested, unchanged studies. Did I do something that broke it? Did somebody slip in a new, buggy aplha version somehow? What's going on?
Any comments or advice from "out there"?
1. Backtest refuses the Time Template I give it, and just makes up one of its own which is similar to nothing I have on file -- I need to test 20 days and it gives me 6 when I have nothing close to a 6-day Template, etc. When I finally get it to take the correct template, I don't know how I did it -- it seems to be making up its own mind randomly;
2. Frequently starts the Daily Trading Summary, item 1, page 1, with lots of (random?) stuff in the "...Rolling Period..." columns, thereby making totals and comparisons totally useless. I suspect it's analyzing days PRIOR to the top-line date, and just not telling anybody about it;
3. Changes output from run-to-run with identical data and identical studies;
4. Started "HIDING" the Formula Output Window, & will only show it after a LONG period of the cursor sitting where the data is SUPPOSED to be (the top ribbon appears, but no body), then disappears again when the cursor leaves the data section, then takes an eternity (up to 60 seconds) to REappear when the cursor is moved back over the data area (same data, it didn't need to rethink anything -- why'd it take a minute?).
The above seems to have started spasmodically a couple of days ago, and reached the point where BackTest is totally unstable and untrustworthy. It matters not whether I use new, under-development code or old, tested, unchanged studies. Did I do something that broke it? Did somebody slip in a new, buggy aplha version somehow? What's going on?
Any comments or advice from "out there"?
Comment