Grampa_of_Oliva_n_Eden
-
Posts
2,767 -
Joined
-
Last visited
-
Days Won
17
Content Type
Profiles
Forums
Downloads
Gallery
Posts posted by Grampa_of_Oliva_n_Eden
-
-
QUOTE(Catweazle @ Mar 17 2007, 11:50 AM)
When I once got pain after EXTREME wiring I bought http://www.swopper.de/' target="_blank">this chair and an intuos3 pen tablet from Wacom. These tools and some exercises for back and shoulders was exactly what I needed!Are sure that is a chaid and not an amusement ride?
Ben
-
QUOTE(Mike Ashe @ Mar 16 2007, 04:43 PM)
...What if you preallocate your strings and then copy in and out of those? Won't that prevent the memory manager from having to reallocate?I belive so.
The bottom line is watch you memory usage.
I had a customer that was crashing after a period of time. I revised the code to work in-place and it stopped crashing.It ran fine for months. The customer added a text field to the control cluster and when the recipe was large the RT target wolud eventually crash. Took the text stuff out. It worked fine again.
Another customer wanted to buffer events in the RT if the Windows app went down. This was implemented as a round robbin buffer and worked fine. Unfortunately I included the error cluster in the buffer entries. Provided the were no errors the app ran and ran. If a bunch of errors were encountered, the app would crash after running out of memory. I modified the code (after attempting many other things) such that the Source fields of the error sub-Cluster (that got stored in my round robbin buffer) was replaced with an empty string and the crashes went away.
What you optimally want to do ( going out on a limb here concidering the audience) is use all brown clusters. If you have to go pink, make sure the variable length buffer storage requirements are not changing. If they have to change, they change to smaller requirements.
Just sharing my RT battle scars.
Ben
-
QUOTE(jakestern @ Mar 16 2007, 01:34 PM)
...if one person adds a new element to the enum, then all subVIs need to be resaved. This involves checking out a potentially large number of VIs, often to do nothing other than resave....This is not neccesarily the case.
If the sub-VI's case structure that uses the enum has a "default" case, then they do not require saving. Yes LV will wan to save them but you do not have to.
Now if the sub-VI is such that a default case is intolerable, then the sub-VI will require editing to add a case for the new enum values. In this case they will have to be checkd-out to do the changes.
Ben
-
QUOTE(Beri @ Mar 16 2007, 10:03 AM)
Hi LAVA-Users,The goal is to implement an analyzer which is similar to the Ettercap which can be found here
Thank you very much for every hint on this in advance.
For a jump start see this thread on the NI forum
http://forums.ni.com/ni/board/message?boar...d=58185#M171976
Ben
-
regarding the kneck and shoulder thing.
I get that if I have to reach out for the trackball. By moving the trackball such that it is under my hand when I am sitting comforatbly, with my arms resting on the arms of my chair, I can eliminate that pain.
Think fetal position! We all spent 9 months in tha postion with problems. Organize your work space such that are as comfortable as possible.
Trying to help,
Ben
-
QUOTE(LV Punk @ Mar 16 2007, 06:28 AM)
4 bytes instread of 21+ bytes, and no conversion overhead.... :thumbup:Just transfering the time stamp as a number works fine. I have at least one application that has bee running at multiple sites for the last five years using that method. It also works across time time zones.
Added note: Avoid strings whenever possible on RT app that have to run forever. The can do very nasty things to your memory requirements.
Ben
-
-
First
I urge you to concider seeing a doctor. CT can disable you. I was waking up at night with shooting pains in my forearms.
Second
Mice are evil creatures. They chew your wrists up. Never touch one again!
Continuing:
I managed to code in LV for about 5 years without any trouble. Then I did a 6 month 12 hour a day marathon and my problems started.
I now use a hacked up Mouse Pad
http://forums.ni.com/ni/board/message?boar...3&jump=true
that I flipped over and re-inforced with a steel plate to hold my Kensington Expert Mouse. The track ball is held to the mouse pad using a pair of small bungy cords. This lets me put my trackball were my hand falls rather putting my hand were the track ball is. It also permits me to adjust the angle so the palm of my hand is at about a 35 degree angle relative to the plane of the floor.
The mouse pad straps onto the arm of a chair so my hand is in the same "rest" position it would be in if I was just sittng there with a beer in my hand.
NOTE:
The slight twist of the wrist required with flat mouses or horizontally mounted track balls causes the two bones in my lower arm to cross while track balling.
I also have a pull out adjustable tray that lets me postition the keyboard, again, where my hands are rather than the other way around.
I also wear a pair of wrist braces (the fancy ones with three straps a splint and padding). The first day I put them one my writsts started feeling better.
I alos discovered that once I had done the damage, that driving my Cavalier was casusing problems. Oddly enough, this is not the case with my pcik-up. So i have a pair in the car I use while commuting.
As it stands today, I am pain free and have been so for alomost a year now. It would have been longer but I found myself debugging an application on-site without any of my aides. At that time I was scared I would be forced into an early retirement.
Closing:
I am not a doctor and none of the above should be concidered as such. If the pain is affecting your work, take action, otherwise it will get worse.
Ben
-
I heard the developer of the beer launching robot interviewed and he was questioned about the shaking. He said "It's all in the catch." A soft touch on the catch and the beer is not shaken. I saw this demonstrated.
Intital version of the robot are expected to sell for about the price of 3 developers suites.
I hope he succeeds.
Ben
-
-
Note:
This is a cross post from the NI forum here.
http://forums.ni.com/ni/board/message?boar...0&jump=true
I asked Bill to post this Q here because I thought that LAVA could do a better job of answering this Q that we can do on the NI-Forum.
Thank you,
Ben
-
QUOTE(didierj @ Mar 7 2007, 08:05 AM)
In Switzerland the only (main) contractors are Alcatel (since January 1st named Thales) and Siemens. Actually the swiss federal Rail accept only one axle counter system for new projects and it is the one from Alcatel. So guess in which company I work
.But not in a 40year old suburb train at 80km/h (which seems to be more than this train was designed for).
Remember the Lounge is always open.
May the "Big Wire-wroker" smile on your adventures.
Ben
-
This may not apply but I should point it out anyway.
There seems to be a bug in how LV handles re-using buffers and the "bundle" function with an un-wired input (in LV 8.2).
LV will copy the data associated with the un-used bundle input when to should leave it alone. If the un-used bundle element is an array, this can really impact performance.
See this thread on the NI forum
http://forums.ni.com/ni/board/message?boar...=232654#M232654
to see more details.
Ben
-
-
Titou,
If you move the DB to another machine physical machine does it still keep sucking down memory?
Ben
-
QUOTE(crelf @ Mar 3 2007, 09:21 AM)
Whilst I agree that 6.1 was super (much better after the 6 debacle) I was always partial to 5.1.1...LV 5.1.1 = BV 2.1.1 (I believe)
That was before control references was it not? We had to do all of our GUI tricks in the top level VI and it often became difficult to seperate the app functional code from the GUI functions.
But stability-wise the LV core was very solid in BV 2.1.1.
Ben
-
I posted an example that may help you on the NI Forum in this thread.
http://forums.ni.com/ni/board/message?boar...=205891#M205891
It uses a picture for the labels and lets the graph do the rest of the work.
http://forums.lavag.org/index.php?act=attach&type=post&id=5096
Ben
-
HI Slugger,
Some of us can't get enough LV so I noticed you cross posted this Q on the NI forum in this thread
http://forums.ni.com/ni/board/message?boar...5&jump=true
Cross-posting is OK but please indicate as much so we are not duplicating our efforts (please).
Like I suggested on the NI forum, please post example code.
Thank you,
Ben
-
QUOTE(Tomi Maila @ Mar 1 2007, 11:14 AM)
I have reported many dozens bug reports since the release of 8.20 so I'm not entirely sure this is a stable product.Tomi
"Beauty is in the eye of the beholder." (W. Shakes.... ?)
I have not started using LVOOP (yet) so for me LV 8.2.1 Beta is a thing of beauty.
Ben
-
Yes go for it!
LV 8.0 is still useful (if you need to back-save from LV 8.2 to LV 7.1)
BTW: One of my administrators recieved an e-mail that LV 8.2.1 is due to ship soon. I concider that a public acknowledgement that it exists. So let me add that I (and a number of other Bug hunters) had a lot of problems trying to find bugs in LV 8.2.1 durring Beta testing. In fact I could not find ANY show stoping bugs. LV 8.2.1 will displace LV 6.2 as the most stable LV release (in my book anyway). Oh yes, and wait until you see the realease notes. Your jaw will drop!
Ben
-
My coffe hasn't started to work so this may be in left field...
How about using the PID to regulate the error between your set point and the actual and use gain scheduling such that when the error is large, the corse PID values dominate and when the error is small the fine PID dominates.
Like I said, pre-coffee idea.
Ben
-
QUOTE(orko @ Feb 23 2007, 12:00 PM)
Hello,The problem that I am fighting against now in my current project can be viewed in this simple VI:
Download File:post-3266-1172249715.zip (LV8.0.1)
Basically, the problem is a real time-burner in large projects which use typedef constants a lot (for instance in queued message handlers where you are converting variant data to their typdef'd form in multiple cases). Every time you change the typedef, no matter if you have autosizing turned off or not, the constant resizes and makes a mess of the block diagram.
I believe this may be related to the following post, but his problem seemed more related to labels:
http://forums.lavag.org/index.php?showtopi...t=0&p=24425
-- orko
This is a known issue.
I can only advise that you use a sub-VI to return the constant and use the sub-VI everywhere the typ-def-d constant would have appeared.
Ben
-
QUOTE(george seifert @ Feb 22 2007, 10:19 AM)
When I put an error probe on a cluster that's connected to a shift register it stays gray and won't show any values. I've tried it with the probes in several places along the wire (inside my case structure and on either side of the structure). Probes on clusters that aren't connected to a shift register work fine. Has anyone else seen this?George
Hi George,
I have probed clusters in LV 8.2 within the last week using LV 8.2 and it worked fine.
I have seen issue using the SDE in execution highlighting mode (which is just a fancy probe) if the VI was opened from the project but worked if not opened from the project.
Could you post code that demonstrates this, please ?
Ben
-
QUOTE(dsaunders @ Feb 21 2007, 12:09 PM)
I am glad that my Dad has excellent taste in music, allowing me to also grow up with Dire Straits. Now I want to go home and bring some Dire Straits in to work. :thumbup:Y'all are trying to make me* feel old aren't you?
My first LP was Led Zeplin 1.
Well the good news I can see retirement from were I stand.
Ben
* Yes, I am paranoid.
The 5th dimension
in LAVA Lounge
Posted
QUOTE(alfa @ Mar 19 2007, 12:03 AM)
I swear "it was coming right at me!"
If you had 5 foot tall rats with horns jumping in front of your car and knocking over your trash cans, you would shoot them too.
I do not care what Walt Disney thought, deer are NOT cute!
Ben