Support Board
Date/Time: Sun, 22 Dec 2024 05:50:01 +0000
Bollinger band study broken with Study OHLC input when using any of the BB OHLC, HLC, or HL aver
View Count: 1524
| [2015-05-18 06:57:14] |
| i960 - Posts: 360 |
|
Only open, high, low, or last settings work. Use of anything involving the remaining 4 common settings (of which the default is HLC average) causes the bands to explode.
|
 sc_ohlc_overlay_bb_0.png / V - Attached On 2015-05-18 06:56:40 UTC - Size: 94.71 KB - 571 views sc_ohlc_overlay_bb_1.png / V - Attached On 2015-05-18 06:56:50 UTC - Size: 78.9 KB - 574 views |
| [2015-05-18 07:23:17] |
| i960 - Posts: 360 |
|
You're seriously marking this as a user discussion. Are you that spiteful? This is a BUG. The study is simply not working correctly using a data source that I'm following YOUR directions on for setting up a high precision spread. How the heck am I even supposed to make this work with the normal HLC average setting?!! You can see it right there with the screen captures that its blowing up when I use the most basic default BB setting. It's not even remotely logical or normal tbat the BB study should do this nor is it even an issue with zero values or anything like that. You can see the OHLC values right on the chart. Date Time Of Last Edit: 2015-05-18 07:25:00
|
| [2015-05-20 04:48:53] |
| |
|
We do not understand exactly what you are doing. It is not a bug. There is something with the configuration that you are using. There may be a way we can solve it, but we do not understand the underlying problem. We will need to get a Chartbook to understand the configuration. Provide a Chartbook with a minimum number of studies showing the problem. It will be a while though before we can get to this. This is an advanced usage question that is not always covered within the scope of our support. 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 |
| [2015-05-20 05:19:14] |
| i960 - Posts: 360 |
|
All I'm basically doing is adding a bollinger band to a "high accuracy" spread as per the instructions offered on the Sierra Chart website. In my particular case the input to the OverlayOHLC study is a Spread - Butterfly study but it'll act exactly the same using Diff (single line). It's basically the OverlayOHLC study with bollinger bands. I'm also having another issue seemingly related to this where I have an alarm on a horizontal line within the region that has an OverlayOHLC study and it keeps getting triggered every single new bar with values of 0.00000 crossing the line (which is obviously not the case). It triggers the alarm like clockwork, every new bar. I feel these two issues are connected. This most definitely resembles a bug. I'll upload both chartbooks in a bit. Date Time Of Last Edit: 2015-05-20 05:20:14
|
| [2015-05-20 06:05:01] |
| i960 - Posts: 360 |
|
Okay here's the chartbook. #1: NQ, 30 sec, with ratio - single line study against chart #2. #2: TF, 30 sec, just price. #3: Two 1m OverlayOHLC studies, input is #1:ratio, one region with BB using last value and another region with BB using the default HLC average (which doesn't work). There's two regions just to show what I'm talking about. I don't normally use these in 30s/1m timeframes, but reduced the time window to demonstrate the alarms firing off all the time. I have a feeling that this general issue is related to the OverlayOHLC study using a spurious value of 0.0 somewhere internally (which is why it fires phantom alarms and possibly why the BB study freaks out). I also considered that the BB study with the default HLV average as input might have issues in general with any kind of line chart - but then I tested it with a simple price chart using a line for the graph draw type and it had no problems. It's specifically with the OverlayOHLC study that it has problems. I also uploaded a capture of the alert window. Note: the alerts with reference to 60m are from the chartbook I actually use but the ones below which are 5m and 1m are from this chartbook I made just to demonstrate the issue. Date Time Of Last Edit: 2015-05-20 06:10:48
|
open_ohlc_bb_issue_0.png / V - Attached On 2015-05-20 06:03:27 UTC - Size: 72.49 KB - 511 views open_ohlc_bb_issue_1.png / V - Attached On 2015-05-20 06:03:36 UTC - Size: 400.21 KB - 465 views  open_ohlc_bb_issue.cht - Attached On 2015-05-20 06:09:07 UTC - Size: 110.23 KB - 527 views |
| [2015-05-27 03:07:59] |
| |
|
We will be looking into this.
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 |
| [2015-06-01 00:59:42] |
| i960 - Posts: 360 |
|
Thanks.
|
| [2015-06-01 05:18:20] |
| |
|
OK we see why this happens. It is now fixed: http://www.sierrachart.com/image.php?l=1433135841138.png There was a problem setting the average values for the last array elements for the Overlay OHLC study. 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 |
| [2015-06-19 23:28:50] |
| i960 - Posts: 360 |
|
Was this supposed to be fixed? Using both version 1261 and 1264 I see the exact same issue using the same chartbook originally attached in this thread.
|
| [2015-06-20 15:35:05] |
| |
|
The solution to this was released. We did observe the problem, saw the cause and corrected it. Very puzzling why you would still see it. We will test again. 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 |
| [2015-06-20 15:44:49] |
| |
|
We see what has happened. The change was inadvertently reverted. This change will be released in the next release. 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: 2015-06-20 15:45:02
|
| [2015-06-20 21:25:36] |
| i960 - Posts: 360 |
|
Thanks, appreciated. BTW: I did notice one area of inconsistency that might be beneficial to quickly fix before the next release as its straightforward. The "use data from latest bar" option in the diff (bar) and diff (line) studies is missing from the ratio (bar) and ratio (line) studies. |
| [2015-06-27 04:26:47] |
| i960 - Posts: 360 |
|
So I've validated that the original overlay study issue has been fixed. However, there's another issue where overlay + ratio study and a day with missing data results in bollinger band problems. It's not a data file issue as those have been redownloaded. For whatever reason there's simply no data for that day (2013/10/10). The chart capture demonstrates the different behavior of an overlay of a diff vs an overlay of a ratio. The latter has severe problems whereas the diff doesn't. It might have something to do with 0-0==0 whereas 0/0==undefined. Also, it would be good to make the ratio study mirror the diff study in having the same option to use data from the latest bar for consistency sake.
Date Time Of Last Edit: 2015-06-27 04:27:29
|
| overlay_ratio_problem.png / V - Attached On 2015-06-27 04:26:42 UTC - Size: 157.08 KB - 340 views |
| [2015-07-05 00:35:06] |
| |
|
When we went to look for "use data from latest bar" we cannot find it on any study. Something like this is familiar but we do not see it anywhere. What is the exact name of one of the studies where you see this? 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 |
| [2015-07-12 02:31:44] |
| i960 - Posts: 360 |
|
Sorry, the exact setting is "Use latest source data for last bar." It was added per this thread: Problem with High Accuracy Spread Chart But was forgotten for the Ratio (single line) study. Additionally, the "Zero output when non exact data match" from Difference (bar) flag is also missing from the Ratio (bar) study. In summary: Ratio (single line) is missing "Use latest source data for last bar" Ratio (bar) is missing "Zero output when non exact data match" And presently Ratio (single line) does strange things when there is missing data and a bollinger band is used (as can be seen in the chart capture I uploaded in post 13). |
To post a message in this thread, you need to log in with your Sierra Chart account: