SoftwareGraphs (Excel or not)

  

Press Ctrl+Enter to quickly submit your post
Quick Reply  
 
 
  
 From:  Chris (CHRISSS)   12 Feb 2015 18:11 
 To:  ALL 1 of 19 
41396.1 
I've got two sets of data which I want to use on the same graph but not sure how to do it like I want. I can get one set working in Excel by using the scatter graph but I want the two sets on there.

The times are in milliseconds, both sets starting at different times and both received at slightly different times, so nothing matches up. How can I synchronise them on the graph?
Me
0/0
 Reply   Quote More 

 From:  koswix   12 Feb 2015 19:32 
 To:  Chris (CHRISSS)      2 of 19 
41396.2 In reply to 41396.1 
Is the time of interest?

 ▪                    
             ┌────┐    ┌────┐                      
          │    │    │    │ ▪                    
          │    └────┘    │                      
          │   ──┐  ┌──   │ ▪                    
   ┌──────┤    ▪    ▪    │                      
  ┌┘      │              │ ▪                    
┌─┤       └──┐  │  │  ┌──┘                      
│ │          │ ││  ││ │   ┌─┐                   
│ │          └─┼┤  └┴─┴───┘ │                   
│ │           ─┘│           │                   
│ │   ┌──────┐  └┬──────────┘                   
  │   │      │   │                              
  │   │      │   │                              
  └───┘      └───┘
If Feds call you and say something bad on me, it may prove what I said are truth, they are afraid of it.
0/0
 Reply   Quote More 

 From:  Chris (CHRISSS)   12 Feb 2015 19:37 
 To:  koswix      3 of 19 
41396.3 In reply to 41396.2 
I'd like them to both be in sync, yes.
Me
0/0
 Reply   Quote More 

 From:  koswix   12 Feb 2015 19:55 
 To:  Chris (CHRISSS)      4 of 19 
41396.4 In reply to 41396.3 
so you want event 1 in set 1 to graph at the same imaginary time point as event 1 set 2? Then just drop the time table and replace with an numeric index. Select that index as the x-axis data for each data set.

 ▪                    
             ┌────┐    ┌────┐                      
          │    │    │    │ ▪                    
          │    └────┘    │                      
          │   ──┐  ┌──   │ ▪                    
   ┌──────┤    ▪    ▪    │                      
  ┌┘      │              │ ▪                    
┌─┤       └──┐  │  │  ┌──┘                      
│ │          │ ││  ││ │   ┌─┐                   
│ │          └─┼┤  └┴─┴───┘ │                   
│ │           ─┘│           │                   
│ │   ┌──────┐  └┬──────────┘                   
  │   │      │   │                              
  │   │      │   │                              
  └───┘      └───┘
If Feds call you and say something bad on me, it may prove what I said are truth, they are afraid of it.
0/0
 Reply   Quote More 

 From:  Chris (CHRISSS)   12 Feb 2015 20:03 
 To:  koswix      5 of 19 
41396.5 In reply to 41396.4 
Not sure. Maybe yes or no.

So I have say
568,19
1568,19
2568,19.5
3568,20

and

1234,5
2234,5
3234,5

So I want both the temps to be using the same time scale and for the points to be in the correct places. I'm not sure if I'm making sense. I'm sure it's easy enough to do.

Also I can't guarantee the same number of samples from each set, if the signal drops, so want the times to still be correct if one set is missing data.
Me
0/0
 Reply   Quote More 

 From:  99% of gargoyles look like (MR_BASTARD)   12 Feb 2015 20:20 
 To:  Chris (CHRISSS)      6 of 19 
41396.6 In reply to 41396.5 
Since x>0, you could try a log scale.

truffy.gifbastard by name
bastard by nature
0/0
 Reply   Quote More 

 From:  Chris (CHRISSS)   12 Feb 2015 20:26 
 To:  99% of gargoyles look like (MR_BASTARD)      7 of 19 
41396.7 In reply to 41396.6 
What would a log scale do? I don't remember them too well from school apart from making a curving increasing line.

I've attached a drawing of what I want. I'm sure its simple.

X and Y are both the same scale on each set, I just want each point on the graph to correspond to the correct x/y value.
Me

Attachments:
0/0
 Reply   Quote More 

 From:  99% of gargoyles look like (MR_BASTARD)   12 Feb 2015 20:26 
 To:  Chris (CHRISSS)      8 of 19 
41396.8 In reply to 41396.1 
Oh, hang on, I think I just figured out your problem. It's not the scale so much as getting both sets of data onto the graph. Correct?

If that's the case, create a table with four columns. Time1, Temp1, Time2, Temp2. Create the scatter plot with Temp1 vs Time1. Then select the graph, right click, and 'select data set' (or something like that, I'm doing this from memory). In the dialogue box create a new data set and set the x to be the Time2 column/boxes, and y to be the matching temp.

truffy.gifbastard by name
bastard by nature
0/0
 Reply   Quote More 

 From:  99% of gargoyles look like (MR_BASTARD)   12 Feb 2015 20:26 
 To:  Chris (CHRISSS)      9 of 19 
41396.9 In reply to 41396.7 
I thought the problem was the scale. See ^

truffy.gifbastard by name
bastard by nature
0/0
 Reply   Quote More 

 From:  Chris (CHRISSS)   12 Feb 2015 20:27 
 To:  99% of gargoyles look like (MR_BASTARD)      10 of 19 
41396.10 In reply to 41396.8 
That's what I want, yeah. I'll give that a go, thanks.
Me
0/0
 Reply   Quote More 

 From:  Chris (CHRISSS)   12 Feb 2015 20:29 
 To:  99% of gargoyles look like (MR_BASTARD)      11 of 19 
41396.11 In reply to 41396.9 
That's perfect. Thank you. I knew there'd be a simple solution but couldn't work it out. Thanks :)
Me
0/0
 Reply   Quote More 

 From:  Peter (BOUGHTONP)   12 Feb 2015 22:58 
 To:  Chris (CHRISSS)      12 of 19 
41396.12 In reply to 41396.11 
Another quick workaround for stuff like this would be doing two graphs and super-imposing in your graphics editor.

Obviously not ideal for ongoing changing stuff, but worth remembering you don't always to stick to a spreadsheet. Trend lines are a good example of something that can be a PITA to get right at times, but are easily drawn with vectors.

 
0/0
 Reply   Quote More 

 From:  99% of gargoyles look like (MR_BASTARD)   13 Feb 2015 07:27 
 To:  Peter (BOUGHTONP)      13 of 19 
41396.13 In reply to 41396.12 
quote: Peter (BOUGHTONP)

Trend lines are a good example of something that can be a PITA to get right at times, but are easily drawn with vectors.

Select data, right click, add trend line, select trend line that best fits the data. Gives you equation and R2 to boot. What's so difficult about that?

truffy.gifbastard by name
bastard by nature
0/0
 Reply   Quote More 

 From:  koswix   13 Feb 2015 08:18 
 To:  Chris (CHRISSS)      14 of 19 
41396.14 In reply to 41396.7 
Ah, right. Also thought it was the scale. 

 ▪                    
             ┌────┐    ┌────┐                      
          │    │    │    │ ▪                    
          │    └────┘    │                      
          │   ──┐  ┌──   │ ▪                    
   ┌──────┤    ▪    ▪    │                      
  ┌┘      │              │ ▪                    
┌─┤       └──┐  │  │  ┌──┘                      
│ │          │ ││  ││ │   ┌─┐                   
│ │          └─┼┤  └┴─┴───┘ │                   
│ │           ─┘│           │                   
│ │   ┌──────┐  └┬──────────┘                   
  │   │      │   │                              
  │   │      │   │                              
  └───┘      └───┘
If Feds call you and say something bad on me, it may prove what I said are truth, they are afraid of it.
0/0
 Reply   Quote More 

 From:  Chris (CHRISSS)   13 Feb 2015 08:40 
 To:  koswix      15 of 19 
41396.15 In reply to 41396.14 
I don't think I explained it very well.
Me
0/0
 Reply   Quote More 

 From:  koswix   13 Feb 2015 09:11 
 To:  Chris (CHRISSS)      16 of 19 
41396.16 In reply to 41396.15 
No, I don't think you did, Christopher :C

I also assumed that everyone knows you right click a plot area and choose select data to tell excel what to plot. 

 ▪                    
             ┌────┐    ┌────┐                      
          │    │    │    │ ▪                    
          │    └────┘    │                      
          │   ──┐  ┌──   │ ▪                    
   ┌──────┤    ▪    ▪    │                      
  ┌┘      │              │ ▪                    
┌─┤       └──┐  │  │  ┌──┘                      
│ │          │ ││  ││ │   ┌─┐                   
│ │          └─┼┤  └┴─┴───┘ │                   
│ │           ─┘│           │                   
│ │   ┌──────┐  └┬──────────┘                   
  │   │      │   │                              
  │   │      │   │                              
  └───┘      └───┘
If Feds call you and say something bad on me, it may prove what I said are truth, they are afraid of it.
0/0
 Reply   Quote More 

 From:  Chris (CHRISSS)   13 Feb 2015 09:37 
 To:  koswix      17 of 19 
41396.17 In reply to 41396.16 
:( I've either been naughty or you're my grandmother.

Select data, not add more data. They should make it more obvious.
Me
0/0
 Reply   Quote More 

 From:  Peter (BOUGHTONP)   14 Feb 2015 15:01 
 To:  99% of gargoyles look like (MR_BASTARD)      18 of 19 
41396.18 In reply to 41396.13 
> select trend line that best fits the data

You either have significantly more options than I, or have only ever dealt with boring data.

 
0/0
 Reply   Quote More 

 From:  99% of gargoyles look like (MR_BASTARD)   14 Feb 2015 18:19 
 To:  Peter (BOUGHTONP)      19 of 19 
41396.19 In reply to 41396.18 
My Excel gives me exponential, linear, logarithmic, 5 orders of polynomial, power, and several orders of moving average (I've never figured the point of that). I'm sorry that your Excel is broken Paeter. I'm also sorry that your data are not as boring as mine. You live an exciting life.

What are your views on pizza Napoli?

truffy.gifbastard by name
bastard by nature
0/0
 Reply   Quote More 

Reply to All    
 

1–19

Rate my interest:

Adjust text size : Smaller 10 Larger

Beehive Forum 1.5.2 |  FAQ |  Docs |  Support |  Donate! ©2002 - 2024 Project Beehive Forum

Forum Stats