Bug? y-position graph of an object depends on its position
11 posts • Page 1 of 1
Bug? y-position graph of an object depends on its position
Hello
It is strange that the graph of the y-position of three exactly equal squares is different.
Is this a bug?
It is strange that the graph of the y-position of three exactly equal squares is different.
Is this a bug?
- chedas
- Posts: 7
- Joined: Sat Jun 15, 2013 11:16 am
Re: Bug? y-position graph of an object depends on its position
Please attach a copy of your scene so that we can take a look at it.
-
Xray - Posts: 500
- Joined: Sun Jun 17, 2012 6:12 am
- Location: USA
Re: Bug? y-position graph of an object depends on its position
Xray wrote:Please attach a copy of your scene so that we can take a look at it.
See, for example, the disc at 0,-3. Even the two discs at the same y-coordinate have different time signals (better to see in excel).
Thank you
- Attachments
-
- Test_2.phz
- (22.08 KiB) Downloaded 23 times
- chedas
- Posts: 7
- Joined: Sat Jun 15, 2013 11:16 am
Re: Bug? y-position graph of an object depends on its position
Well, there are a couple of things that I noticed about your scene. First of all, you have the sim frequency set to 5000! I'm not sure if that's even valid or technically possible for most computers. The other thing is, your motor assemblies are mounted with springs, and that's what's causing the high frequency oscillation in the two upper graphs. When I glue the black square (box) to the background, which effectively eliminates the springs, then the oscillations go away, and the plot becomes a flat line. Other than those two things, the plots all looks the same to me.
-
Xray - Posts: 500
- Joined: Sun Jun 17, 2012 6:12 am
- Location: USA
Re: Bug? y-position graph of an object depends on its position
I think the amplitudes are all the same, but the graph is more zoomed in for the objects that are near y=0.
-
Ivan - Posts: 202
- Joined: Tue Oct 06, 2009 2:01 pm
- Location: Croatia
Re: Bug? y-position graph of an object depends on its position
it's irrelevant.Xray wrote:Well, there are a couple of things that I noticed about your scene. First of all, you have the sim frequency set to 5000! I'm not sure if that's even valid or technically possible for most computers.
I think that's what is intendedXray wrote: The other thing is, your motor assemblies are mounted with springs, and that's what's causing the high frequency oscillation in the two upper graphs.
his point is that while all three boxes are identical, only the upper two boxes show oscillations. The reason for which has been discussed previously, and I will say it again, in a very annoyed tone,Xray wrote: When I glue the black square (box) to the background, which effectively eliminates the springs, then the oscillations go away, and the plot becomes a flat line. Other than those two things, the plots all looks the same to me.
ALGODOO
DOES NOT
SUPPORT
SUB-CENTIMETER
PHYSICS.
GET OVER IT!
if ANY change is below 0.01m, Algodoo EXPRESSLY denies the ability to acknowledge it, indicate it, or act upon it, and you should feel honored, not entitled, if it does.
- jon_joy_1999
- Posts: 233
- Joined: Fri Dec 09, 2011 12:51 am
Re: Bug? y-position graph of an object depends on its position
jon_joy_1999 wrote: his point is that while all three boxes are identical, only the upper two boxes show oscillations. The reason for which has been discussed previously, and I will say it again, in a very annoyed tone,
ALGODOO
DOES NOT
SUPPORT
SUB-CENTIMETER
PHYSICS.
GET OVER IT!
if ANY change is below 0.01m, Algodoo EXPRESSLY denies the ability to acknowledge it, indicate it, or act upon it, and you should feel honored, not entitled, if it does.
CALM DOWN, MAN!..... In order for someone to "Get over it", they first need to understand what it is that they need to get over. I'm not so sure that chedas understood that his issue is caused by Algodoo not being able to support sub-centimeter physics. I did not think of that as being the problem either. And it doesn't explain why there are different plots between the three geometries.
-
Xray - Posts: 500
- Joined: Sun Jun 17, 2012 6:12 am
- Location: USA
Re: Bug? y-position graph of an object depends on its position
You don't seem to have understood my post.
-
Ivan - Posts: 202
- Joined: Tue Oct 06, 2009 2:01 pm
- Location: Croatia
Re: Bug? y-position graph of an object depends on its position
Ivan - You are correct. I deleted the existing plot for the mechanism located at 0,-3 and opened up a new one with the same configuration as the other plots, and now it looks exactly like the other plots when I run the sim. So, it was simply a matter of misinterpretation by chedas. There really is no difference in the behaviors between the three mechanisms.
Thanks again for your help!
EDIT: I did more experimenting with this and found that the magnitude of the scale changes depending on how far away from Y(0) you place the mechanism. So, at Y(0), the plot is zoomed in for a relatively smaller scale on the Y axis, and therefore, the oscillations appear to be higher than on the other plots that are situated away from Y(0). While the sim was running, I moved one of the mechanisms while watching what happens to its plot, and the result was rather surprising and unexpected. As I moved the mechanism higher or lower than Y(0), the amplitude of the Y axis changed significantly so that when the mechanism is only a couple of meters above or below Y(0) then you can no longer see those oscillations. It appears that the Y scale changes with distance from Y(0).
Would anyone else like to add to this?
Thanks again for your help!
EDIT: I did more experimenting with this and found that the magnitude of the scale changes depending on how far away from Y(0) you place the mechanism. So, at Y(0), the plot is zoomed in for a relatively smaller scale on the Y axis, and therefore, the oscillations appear to be higher than on the other plots that are situated away from Y(0). While the sim was running, I moved one of the mechanisms while watching what happens to its plot, and the result was rather surprising and unexpected. As I moved the mechanism higher or lower than Y(0), the amplitude of the Y axis changed significantly so that when the mechanism is only a couple of meters above or below Y(0) then you can no longer see those oscillations. It appears that the Y scale changes with distance from Y(0).
Would anyone else like to add to this?
-
Xray - Posts: 500
- Joined: Sun Jun 17, 2012 6:12 am
- Location: USA
Re: Bug? y-position graph of an object depends on its position
Thanks for your posts and help.
I really think there is a problem.
I changed the sim frequency to 2000 Hz and even to 500 Hz to satisfy some of your arguments. The bigger this frequency better. Naturally the limit is our computer. Even so the problem remain.
What I want to do is to use Algodoo to simulate machinery problems and obtain the vibrations signatures. In fact I already built very good models of some typical problems (for example ball bearing damages). So, The box is a model of a bearing, the disc is a model of a shaft and the springs are fundamental to model the stiffness and damping of the bearing. Please don't erase them or glue the box to the background.
Please try youselfs. Use the file I sent previously. Don't look for the graphs from Algodoo. You can't do much with them. Save them to excel and create in excel the graphs. You will see that the machine at -3,0 doesn't move but that is impossible because it is unbalanced the same amount as the other two. If you subtract both graphs of machines at the same y position you will see that there is a big difference.
I think Xray already notice that there is a problem!!!
Thank you again
I really think there is a problem.
I changed the sim frequency to 2000 Hz and even to 500 Hz to satisfy some of your arguments. The bigger this frequency better. Naturally the limit is our computer. Even so the problem remain.
What I want to do is to use Algodoo to simulate machinery problems and obtain the vibrations signatures. In fact I already built very good models of some typical problems (for example ball bearing damages). So, The box is a model of a bearing, the disc is a model of a shaft and the springs are fundamental to model the stiffness and damping of the bearing. Please don't erase them or glue the box to the background.
Please try youselfs. Use the file I sent previously. Don't look for the graphs from Algodoo. You can't do much with them. Save them to excel and create in excel the graphs. You will see that the machine at -3,0 doesn't move but that is impossible because it is unbalanced the same amount as the other two. If you subtract both graphs of machines at the same y position you will see that there is a big difference.
I think Xray already notice that there is a problem!!!
Thank you again
- chedas
- Posts: 7
- Joined: Sat Jun 15, 2013 11:16 am
Re: Bug? y-position graph of an object depends on its position
not alwayschedas wrote:Thanks for your posts and help.
I really think there is a problem.
I changed the sim frequency to 2000 Hz and even to 500 Hz to satisfy some of your arguments. The bigger this frequency better.
I've done thischedas wrote: Naturally the limit is our computer. Even so the problem remain.
What I want to do is to use Algodoo to simulate machinery problems and obtain the vibrations signatures. In fact I already built very good models of some typical problems (for example ball bearing damages). So, The box is a model of a bearing, the disc is a model of a shaft and the springs are fundamental to model the stiffness and damping of the bearing. Please don't erase them or glue the box to the background.
Please try youselfs. Use the file I sent previously. Don't look for the graphs from Algodoo.
I put it into a google docs spreadsheetchedas wrote: You can't do much with them. Save them to excel and create in excel the graphs.
chedas wrote: You will see that the machine at -3,0 doesn't move but that is impossible because it is unbalanced the same amount as the other two. If you subtract both graphs of machines at the same y position you will see that there is a big difference.
I think Xray already notice that there is a problem!!!
Thank you again
as I have said before, both in this thread, and in others, Algodoo DOES NOT support changes less than 0.001 meters.
the changes that are occurring in your scene are at most 0.000000633702 meters. this is not a bug, this is expected behavior due to the 32-bit single-precision floating point numbers that Algodoo uses
- jon_joy_1999
- Posts: 233
- Joined: Fri Dec 09, 2011 12:51 am
11 posts • Page 1 of 1
Who is online
Users browsing this forum: No registered users and 5 guests