Notices
Vishnu Performance - California [Visit Site]

Xede datalogging now possible on Mac and Unix

 
Thread Tools
 
Search this Thread
 
Old Jun 1, 2005 | 09:56 AM
  #31  
crusin_lancer's Avatar
Evolved Member
iTrader: (10)
 
Joined: Aug 2002
Posts: 751
Likes: 0
From: On the Roads
Originally Posted by donour
Nobody's interested. Look, look! shiny pictures:




d

OOOooooo...... I likey.... bundle? I've been so busy I haven't been able to touch pyXEDE after we talked over pm. Looks good though.
Old Jun 1, 2005 | 10:08 AM
  #32  
jj_008's Avatar
Evolved Member
iTrader: (8)
 
Joined: Nov 2003
Posts: 2,064
Likes: 0
From: Salem, OR
Originally Posted by donour
Is anybody interested in a windows release? I was fooling around with a windows XP box this morning and verified that my code does indeed run on windows without any modification.

d
I'm interested. Ok, I downloaded the latest version of your site. Now, how do I use it?

Last edited by jj_008; Jun 1, 2005 at 10:27 AM.
Old Jun 1, 2005 | 12:24 PM
  #33  
donour's Avatar
Thread Starter
Evolved Member
iTrader: (6)
 
Joined: May 2004
Posts: 2,502
Likes: 1
From: Tennessee, USA
I added windows installation instuctions.

http://www.donour.com/cars/pyxede


Everything should work on windows, but it's not thoroughly tested so it might crash occasionaly :-p. If people want to try it out, by all means give me some feedback on what causes it to go kablooey. As per earlier mesgs, I don't have access to a windows machine to test it.

d
Old Jun 1, 2005 | 12:56 PM
  #34  
jj_008's Avatar
Evolved Member
iTrader: (8)
 
Joined: Nov 2003
Posts: 2,064
Likes: 0
From: Salem, OR
Originally Posted by donour
I added windows installation instuctions.

http://www.donour.com/cars/pyxede


Everything should work on windows, but it's not thoroughly tested so it might crash occasionaly :-p. If people want to try it out, by all means give me some feedback on what causes it to go kablooey. As per earlier mesgs, I don't have access to a windows machine to test it.

d
I will try this when I get home tonight on my Windows 2000 machine w/ my serial to USB adapter.
Old Jun 1, 2005 | 05:33 PM
  #35  
Rushin's Avatar
Evolving Member
iTrader: (4)
 
Joined: Sep 2004
Posts: 191
Likes: 0
From: Redondo Beach, CA
Originally Posted by donour
Everything should work on windows, but it's not thoroughly tested so it might crash occasionaly :-p. If people want to try it out, by all means give me some feedback on what causes it to go kablooey. As per earlier mesgs, I don't have access to a windows machine to test it.
d
got a chance to run version 0.4 on windows last night. Its amazing how high the sampling rates are compare to xmap datalogging!! I had to make a minor change... had to do with a division by 0 during torque calculation. The time value was the same in two consecutive samples. I guess the computer is too fast ha?? heheh
Old Jun 1, 2005 | 07:06 PM
  #36  
jj_008's Avatar
Evolved Member
iTrader: (8)
 
Joined: Nov 2003
Posts: 2,064
Likes: 0
From: Salem, OR
The program kept locking up on me. I couldn't get it to datalog either. I kept getting an error. Also, the road dyno would collect data, but at the end it would say bad run or WTF.
Old Jun 2, 2005 | 07:20 AM
  #37  
donour's Avatar
Thread Starter
Evolved Member
iTrader: (6)
 
Joined: May 2004
Posts: 2,502
Likes: 1
From: Tennessee, USA
Originally Posted by Rushin
got a chance to run version 0.4 on windows last night. Its amazing how high the sampling rates are compare to xmap datalogging!!
You ain't seen nothing yet. I'm working with chiptorque to increase the sample rate dramatically.


I had to make a minor change... had to do with a division by 0 during torque calculation. The time value was the same in two consecutive samples. I guess the computer is too fast ha?? heheh
This shouldn't happen with the dyno code. Timestamps come from system time which is at the submicrosecond level on most systems. Memory must be getting fouled up or something. I'll look into it.

d
Old Jun 2, 2005 | 07:27 AM
  #38  
donour's Avatar
Thread Starter
Evolved Member
iTrader: (6)
 
Joined: May 2004
Posts: 2,502
Likes: 1
From: Tennessee, USA
Originally Posted by jj_008
The program kept locking up on me.
Ouch. As I said, the software hasn't been tested on windows. Anybody want to help with win32 testing. I have virtually no access to that platform.

I couldn't get it to datalog either.
Did you get a mesg? Did it lock up? Or did nothing happen?

I kept getting an error. Also, the road dyno would collect data, but at the end it would say bad run or WTF.
Blame this one on lack of documentation. The road dyno code is currently dependant on a bunch of cockmame heuristics that need to be improved. For example if RPM ever drops during a run, it removes the whole run. It doesn't tell you this, it just does it.

Expect things to improve.


d
Old Jun 2, 2005 | 12:28 PM
  #39  
jj_008's Avatar
Evolved Member
iTrader: (8)
 
Joined: Nov 2003
Posts: 2,064
Likes: 0
From: Salem, OR
Originally Posted by donour
Ouch. As I said, the software hasn't been tested on windows. Anybody want to help with win32 testing. I have virtually no access to that platform.


Did you get a mesg? Did it lock up? Or did nothing happen?


Blame this one on lack of documentation. The road dyno code is currently dependant on a bunch of cockmame heuristics that need to be improved. For example if RPM ever drops during a run, it removes the whole run. It doesn't tell you this, it just does it.

Expect things to improve.


d
The program lock up immediately when I first started it. Then one time I got it to start showing RPM then when I set it up to do a road dyno it would start datalogging RPM then it would stop at the RPM limit and show bad run. On the datalogging I would get a destination error.
Old Jun 2, 2005 | 12:40 PM
  #40  
donour's Avatar
Thread Starter
Evolved Member
iTrader: (6)
 
Joined: May 2004
Posts: 2,502
Likes: 1
From: Tennessee, USA
Originally Posted by jj_008
The program lock up immediately when I first started it.
Disheartening, I'm sure. Did you say you were using windows 2000? Perhaps there is something funny going on with wxPython and that version of windows.

d
Old Jun 2, 2005 | 12:52 PM
  #41  
jj_008's Avatar
Evolved Member
iTrader: (8)
 
Joined: Nov 2003
Posts: 2,064
Likes: 0
From: Salem, OR
I am using Windows 2000 w/ a slow *** USB adapter.
Old Jun 2, 2005 | 12:54 PM
  #42  
donour's Avatar
Thread Starter
Evolved Member
iTrader: (6)
 
Joined: May 2004
Posts: 2,502
Likes: 1
From: Tennessee, USA
Originally Posted by jj_008
w/ a slow *** USB adapter.
But fast enough to work with with xmap, no?

d
Old Jun 2, 2005 | 12:55 PM
  #43  
jj_008's Avatar
Evolved Member
iTrader: (8)
 
Joined: Nov 2003
Posts: 2,064
Likes: 0
From: Salem, OR
Yes, but it takes like 3-4 min to upload or download a map.
Old Jun 2, 2005 | 01:38 PM
  #44  
crusin_lancer's Avatar
Evolved Member
iTrader: (10)
 
Joined: Aug 2002
Posts: 751
Likes: 0
From: On the Roads
I just ran this program during my lunch break. The pyXEDE is a really nice program, but I had some problems with the road dyno and logging. I would run the road dyno and everything worked till the very end. I got a "ZeroDivisionError: float division". I took a screen shot if that helps any. When I was datalogging I got a permission denied error which is weird because I was logged in as Administrator on my laptop. I'm gonna work on the permission denied as it is somewhat strange. I am very impressed with the sampling rates. Nice job Donour!!

EDIT: By the way. I'm running a Windows 2000 machine and I had no problems what so ever with the program locking up or freezing. I will be testing it later with a XP machine. Also I tried plotting the example dyno and it tells me "the system can not find the path specified" Do I need to change the path somewhere in the source?

Last edited by crusin_lancer; Jun 2, 2005 at 01:43 PM.
Old Jun 2, 2005 | 02:59 PM
  #45  
Rushin's Avatar
Evolving Member
iTrader: (4)
 
Joined: Sep 2004
Posts: 191
Likes: 0
From: Redondo Beach, CA
Originally Posted by donour

This shouldn't happen with the dyno code. Timestamps come from system time which is at the submicrosecond level on most systems. Memory must be getting fouled up or something. I'll look into it.

d
the timestamps on my win xp machine only had 100th of a sec precision. i changed it so it throws out the sample if the previous one had the same time. That got rid of the division by zero error.

I also had to tweak it to fix the "bad run" error.. hope this helps. Its a great program and I wanna see it working in windows! (im sure im not the only one)



All times are GMT -7. The time now is 03:38 PM.