Support Board
Date/Time: Wed, 27 Nov 2024 07:27:04 +0000
Trend Lines move, when scaling in or out.
View Count: 4553
| [2014-02-15 21:30:54] |
| User14266 - Posts: 225 |
|
When scaling a chart up or down, my trend lines move all over the place. They change positions, and cross at different price and times depending on where the scale is set. Please watch this short video. In the video, I am slowly scaling up and down using the Interactive Scale Range. http://screencast.com/t/oGemDs1qSq
|
| [2014-02-15 22:16:23] |
| |
|
Nothing new here. Refer to this help topic: https://www.sierrachart.com/index.php?l=doc/helpdetails34.html#RayDrawings We cannot offer any further help. 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 |
| [2014-02-15 22:37:04] |
| User14266 - Posts: 225 |
|
So basically, I can't use trend lines accurately in SC? This is a major disappointment. I'm a little surprised honestly. All of the advance features in SC, but the most simple one does not work accurately. I wonder why other platforms don't have this issue? How is this possible? Moving trend lines. Almost speechless. Okay.
|
| [2014-02-16 09:44:59] |
| |
|
We updated the documentation to explain the solution. This is an image we added: https://www.sierrachart.com/images/AdjustRayDrawingForStability.png We do take exception to saying that they do not work accurately. They are precise and work as designed and Chart Drawings are designed as best as possible. So what exactly is the problem? Mapping into the integer pixel coordinate system of a monitor. Other programs must have this kind of issue if you have set up the very same kind of line and if you are doing exactly what you are doing in Sierra Chart. And if not, then you are welcome to use another program. We know we are handling this properly and we know the source of this kind of issue. Anyway, it should be resolvable with doing what we said in the image. Moving trend lines. Almost speechless. The general answer to this kind of comment from a user is to explain why they should not move. Although that might be hard for you to answer because you do not know all of the things that we know although we try to explain that as much as possible in the documentation. Also in that video, we can only get a general sense of the problem. It is too overwhelming because it is a video that shows rapid movement and there are a large number of lines in there. We also do not know the anchor points of those lines or the types of drawings they are. We would actually have to get a Chartbook containing a single line and the instructions to replicate the problem in order to give a precise explanation of it. But in any case, we know in general what the source of this kind of problem is. There is no need to provide a Chartbook. 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: 2014-02-16 10:04:11
|
| [2014-02-21 21:38:54] |
| User14266 - Posts: 225 |
|
Other programs must have this kind of issue if you have set up the very same kind of line and if you are doing exactly what you are doing in Sierra Chart. And if not, then you are welcome to use another program. We know we are handling this properly and we know the source of this kind of issue.
These anchor points can not be moved. They are anchored at time intervals, and moving the anchor point is completely unrealistic.Would you believe me if I told you that ToS doesn't have this issue? Well, it doesn't, and the anchor points are the same. If you require proof, I can spend some time putting something together, same charts, same anchor points, and show you so you can see side by side that SC has this issue and ToS doesn't. You can't possibly believe that all platforms behave this way? You seem to think this is a request to defy physics. It's not, it's just a request that many of your users have. It's just a request for trendlines to stay put, like other platforms. You are right, I don't understand the programming or technical side, but apparently ToS does(or maybe they "Mapped into the integer pixel coordinate system of a monitor). You are also right, I am welcome to use another program. I should take that as a hint. Thank you for your reply. |
| [2014-02-21 22:42:03] |
| |
|
There are various scenarios which can cause lines to move. Are you working with Lines or Rays? Thinking about this some more, one thing that we realize is that a function we use to clip a line to a window using integer arithmetic, might cause a line to shift more than a pixel and we will do some testing. If this is causing a problem we can solve that. We are going to be looking into this now. The thing about this is that many of the scenarios we hear about, are not related to this. But it might be the problem in your case. Although we are not completely clear on the details of exactly what you are doing. It is hard to tell from the video. It's just a request for trendlines to stay put, like other platforms. There are certain scenarios though, where this is just simply not possible and the users need to understand that. Your scenario is not the typical scenario we deal with.
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: 2014-02-21 22:42:40
|
| [2014-02-21 23:10:22] |
| User14266 - Posts: 225 |
|
I have tried it with both rays and line(extended to right). Here is another video. It looks fast because of choppy video. But I am slowly scaling in and out. http://screencast.com/t/346Y962A To duplicate this video, plot a ray or extended line on a 5min emini ES chart: Anchor point #1: 1820.75 @ 6:25am. Anchor point #2: 1822.50 @ 7:30am. Simply scale up or down slowly, and you will see this. The future intersection of this ray, is not definite. Thank you for taking the time to look into this issue. |
| [2014-02-21 23:14:16] |
| |
|
We do not want to confirm anything at this point, but tentatively what we are seeing is the code which is designed to keep Rays stable was not being used properly after some programming changes in recent years. So this is part of the problem. And we are also looking at another issue.
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 |
| [2014-02-22 01:24:51] |
| Handle - Posts: 36 |
|
I will add that I experience this issue as well. In particular with Pitchforks when I scale in our out the Intersection of the lines with price moves.
|
| [2014-02-22 10:59:15] |
| |
|
We do not want users to start getting too many ideas from this thread, in regards to issues with chart drawings moving. In the case of Pitchforks, that kind of issue is documented here and what to do about it: https://www.sierrachart.com/index.php?l=doc/doc_Tools.html#Pitchfork (Refer to the section about Secondary Anchor Point) What this thread is about may or may not be related to your particular scenario. And also when you are drawing Chart drawing which ends in the fill space with variable timeframe bars, then you can definitely get movement for various reasons . All of this is already documented. In our testing, the particular function that extends a Ray drawing and keeps it stable, did not provide any benefit even though it was not being used. A program like think or swim is a completely different kind of program because it is built on a completely different graphics platform. Those guys have it very easy compared to us. All of the underlying details are handled already for them with the underlying graphics platform that they work with. We have to do all of this stuff manually and understand the math and the details behind it. And when we rely directly on the Windows GDI to clip a line to a window instead of using our own function, the very same line shifting occurs. So if you think we are not doing this well, the Windows GDI suffers the very same effect and there is a technical reason why this occurs that cannot be easily overcome. Anyway, we have the problem now resolved. However , this is only in regards to the particular issue raised here in the first post and no other. The issue had to do with the final clipping of the line to the chart region. This is now based upon the chart scale rather than the monitor pixel coordinates. Nevertheless, since the fact that a monitor is an integer coordinate system, there is a slight amount of shifting that does still occur when changing the scale that we observed. This does not occur with small-scale changes. It is typically only when expanding the price bars significantly using Interactive Scale Range. However, in our test scenario it is much more stable with the change made. And it is for this reason, that we insist, that other programs must have this effect even if you do not perceive it. It still is there, even if it is small. 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: 2014-02-22 23:34:59
|
| [2014-02-22 23:36:01] |
| |
|
The prior response has been updated to make a couple of points more clear. 1096 has been released. Give it a try. For the most part we would expect the problem to be resolved. Let us know. 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 |
| [2014-02-23 02:34:09] |
| User14266 - Posts: 225 |
|
It looks much, much better. However, here is my observation about the remaining shifting..... A ray that is touching the left or right edge of the chart is stable. If the ray is touching the top or bottom of the chart, there is still shifting. See video. Watch as the two red rays(marked with a white dot) transition from the side of the chart(where they are stable) to the bottom of the chart(when they start shifting). This is repeatable, on the bottom and top of the screen. http://screencast.com/t/nLzGJYSWXrey If you could apply your fix to the other axis, that would be great! What you have done is a large improvement. Thank you. |
| [2014-02-23 09:44:27] |
| |
|
Working on this. Must have something to do with the precision of the functions. We are also rewriting the formulas. In testing, we did see something like this after an extreme scale change, but we cannot reproduce it. Anyway we will figure it out. 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: 2014-02-23 09:45:07
|
| [2014-02-24 09:58:31] |
| |
|
We made some additional changes. May or may not help. Update to the latest revision by selecting Help >> Download Prerelease.
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 |
| [2014-02-24 17:31:25] |
| User14266 - Posts: 225 |
|
It is still behaving the same way as the last video I posted. Rays/line ending on the left or right edge of the chart are stable, rays/lines ending on the top or bottom are still shifting. I do not see a difference in the amount of shifting. Thanks.
|
| [2014-02-24 17:42:03] |
| |
|
We will think about this, but really we do not see how we can make this any better. It is the basic problem of mapping floating-point values into an integer coordinate system. There simply is an inherent limitation that exists that cannot be overcome, but we will consider if there is something we have overlooked. 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 |
| [2014-02-24 18:38:13] |
| ganz - Posts: 1048 |
|
SC Support Линия не сохраняет расположение относительно своих якорей даже после Snap to ... (High\Low в данном случае) До масштабирования http://www.sierrachart.com/image.php?l=1393266630524.png после http://www.sierrachart.com/image.php?l=139326666938.png Как известно через две заданные точки всегда можно провести прямую. Может быть возможно обеспечить, чтобы хотя бы те трендовые линии, которые привязаны якорями к уже существующим объектам сохраняли привязку к собственным якорям относительно тех объектов? спс. |
| [2014-02-25 01:58:32] |
| |
|
What you have drawn appears to be a Ray. To ensure that the line stays exactly at the high where you put the small square, so no matter how you change the scale it remains there at least in regards to that one particular point, you need to set that as the anchor point. Right-click on the Ray close to where you want to set the anchor point and select Adjust Drawing and then set the Anchor point exactly where you want.
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 |
| [2014-02-25 02:09:37] |
| ganz - Posts: 1048 |
|
SC Support Никаких квадратиков отдельно не рисовал - это маркер якоря от этого же луча (Ray) - линия уплывает от собственного якоря. |
| [2014-02-25 02:16:18] |
| |
|
We understand. This has to do with the final mapping once again into an integer coordinate system from the floating-point value. There will be some amount of inaccuracy that exist. As monitor resolutions get higher, it will lessen the problem. As long as you are on version 1097, we do not see what more we can do. You need to think of it this way, if we try to address the particular issue you have pointed out, then we go back to the original problem and then you will point out a problem at the far end of the line where the ending point does not go exactly as you expect. You may not want to hear this, but we cannot help with this any further. We have done the best we can. If some other program happens to do this better in some way, then feel free to use it. And we also cannot really believe that other programs can do a perfect job with this. They simply cannot. 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: 2014-02-25 02:20:05
|
| [2014-02-25 03:35:08] |
| ganz - Posts: 1048 |
|
SC Support And we also cannot really believe that other programs can do a perfect job with this.
Если вам интересно, то прикладываю скрины из софта от мистера W. Linn. Именно этой проблемы там нет. Трендовые линии при масштабировании не плывут. Там есть некоторые другие недостатки и это просто пример того, что стабильные линии - это возможно. Речь только о линиях. До масштабирования http://s10.postimg.org/9b8qtgppl/irt1.png после http://s27.postimg.org/468fwrkxf/irt2.png You may not want to hear this, but we cannot help with this any further.
If some other program happens to do this better in some way, then feel free to use it.
Линия между двумя фиксированными координатам "гуляет" - вот такого действительно вряд ли где увидишь, имхо. Обратил ваше внимание на очевидную нелогичность раздельного от двух явно указанных якорей плавания линии. Ничего более. А вот чем мне пользоваться - вопросов не задавал - советов у вас не спрашивал. К чему такие болезненные реакции на рабочие вопросы? Это же ваша работа - и ваш профессиональный результат на выходе. По мне так: не можете - ну и ладно, так и запишем. удачи в делах. |
| [2014-02-25 03:49:43] |
| |
|
And we can produce a video right now if you want, showing numerous scenarios where Sierra Chart will not have this problem either. I have done something very similar to what you did with this other program, and did not see a problem. The image you posted from this other program does not prove anything because it is not an exact duplication. There are so many variables involved here. You should recognize that. The bar spacing, the resolution of the chart, the actual endpoints, the relationship between the two of them, the scaling. And many more variables. And more precisely it is not really about mapping a floating-point coordinate system used by the chart into an integer coordinate system used by the monitor. The main issue is that the coordinate system of the chart does not match the monitor. If you see a line drift away from its anchor point, it only occurs at certain points as the scale is being changed. What you find is that it may move a couple of pixels off and then go back. All of this is completely natural. 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 |
| [2014-02-25 03:55:17] |
| |
|
Do know that is not possible to draw straight on a monitor? Have you ever closely looked at a line on a monitor and realize that it is not a perfectly straight line. Edit: We are not talking about a vertical or horizontal line. A line that is at an angle, is not perfectly smooth on its edges and cannot be drawn between any two infinite points on the monitor. Refer to the attachment. 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: 2014-02-25 06:01:29
|
 AngledLineOnMonitor.png / V - Attached On 2014-02-25 06:01:26 UTC - Size: 1.06 KB - 502 views |
| [2014-02-25 05:14:48] |
| ganz - Posts: 1048 |
|
SC Support Мне понятно о чем речь. В будущем выберу время более детально изучить эту тему, чтобы понять почему в СЧ сдвиги линий более заметны, чем в приведенном выше примере стороннего софта. При тестах учту ваши замечания по исходным условиям. |
| [2014-02-25 06:14:47] |
| ganz - Posts: 1048 |
|
SC Support A line that is at an angle, is not perfectly smooth on its edges and cannot be drawn between any two infinite points on the monitor.
это понятно. однако ... с другой стороны на чарте имеется графический объект, н-р: метка якоря, ее координата известна - она никуда не уезжает и остается на указанной позиции относительно хая свечи. это может говорить о том, что для отрисовки линии в районе якоре не используется область/параметры самого якоря, те конечная точка линии в районе якоря не соответствует местоположению метки якоря при изменении угла наклона линии в процессе масштабирования. в этом и состояла суть изначального предложения: закрепить отрисовку линии жестко на существующих двух якорях - полученный в результате масштабирования угол относительно неизменной оси и заякоренного участка проецировать в перспективу. |
To post a message in this thread, you need to log in with your Sierra Chart account: