what is the maximum size for a VI

[pnielsen@gplains.com]

I am currently using 2009. My VI is 1.8 MB. I am having problems. Time to make a chang is 15 seconds. Sometimes saving a file won't work. I have tried increasing the swap file.I also have 2 GB of memory. I will be uploading to 2010. I keep adding to the VI as the designer wants nmore features. This is a nightmare for code changes. Changes are now weekly. There are no test specs?? I also have a serious bug in my PC. It keeps flashing tring to load windows ?? I will add 1 gig of memory. Maby I need a better PC. It is 2,4 Ghz with 2G of memory.

I am having my PC rebuild and and reload 8.2 to support 8.2 and 2009 and 201o. My PC is dead as far as I am concerned. IS can;t fix my PC till next week. I am to frustrated with the code chages on thefly.

Thanks

Philip 

[Ben]

The max size of the file is limited by the file system you are using.

1.8 Meg is not THAT big.

Let IT fix your PC and then try again.

Ben

[Dennis_Knutson]

Instead of adding more memory, perhaps you need to rethink the design of the VI. You should be using a lot of subVIs and you should not be saving front panel arrays, graphs, etc with large amounts of data. You could also think about dynamically calling VIs.

A correct architecture will probably mean much more than memory/faster cpu/etc. Those are generally just band-aids on a fundamental problem.

Edit - oops, I read 1.8MB as 1.8GB. Ben's right. If 1.5MB is correct, then that is not that large and the problem might be with the pc.

[Hooovahh]

---

Dennis Knutson wrote:

Edit - oops, I read 1.8MB as 1.8GB. Ben's right. If 1.5MB is correct, then that is not that large and the problem might be with the pc.

---

Sweet Jesus!  1.8GB would be a HUGE VI.

It doesn't sound like you have this problem, but some times if you have lots of images on the front panel they can be saved in raw image format similar to BMP which causes the VI to balloon to a pretty large size.  If that were the case I would expect you had a VI that was in the order of 20MB or more.

[Ben]

---

@Hooovahh wrote:

---

Dennis Knutson wrote:

Edit - oops, I read 1.8MB as 1.8GB. Ben's right. If 1.5MB is correct, then that is not that large and the problem might be with the pc.

---

Sweet Jesus!  1.8GB would be a HUGE VI.

It doesn't sound like you have this problem, but some times if you have lots of images on the front panel they can be saved in raw image format similar to BMP which causes the VI to balloon to a pretty large size.  If that were the case I would expect you had a VI that was in the order of 20MB or more.

---

If you ever have an afternoon to kill, the code I posted in this thread once run (minimize FP when running otherwise you will need a day or a really mean machine) I could not save it without clearing the display.

I guess 150 slices of CAT-scan rendered in 3d is just a little too much.

Ben

BTW:

We also disected the brother of your avatar from LAVA in that thread. I have also done the 3d on that image as well becuase I really wanted to turn the skull so I could see the guy's face.

[Hooovahh]

---

@Ben wrote:

---

 

If you ever have an afternoon to kill, the code I posted in this thread once run (minimize FP when running otherwise you will need a day or a really mean machine) I could not save it without clearing the display.

I guess 150 slices of CAT-scan rendered in 3d is just a little too much.

Ben

BTW:

We also disected the brother of your avatar from LAVA in that thread. I have also done the 3d on that image as well becuase I really wanted to turn the skull so I could see the guy's face.

---

Brother of my Avatar?  Herb Powell? (I know who you mean)

It doesn't surprise me that someone can use 1.8GB or RAM on a LabVIEW project, but what was mentioned.  What was mentioned was the size of the VI (I assumed hard drive space).  So when I saw that the mistake was made and thought it was 1.8GB instead of 1.8MB, I tried to imagine what a 1.8GB LabVIEW VI would look like.  Which is why I mentioned images on the frontpanel.

[Philip2001]

The file is 1.7X mb. The PC has been very erratic for a long time. The Windows deskview pop up menue pops up with a message installing?? It sits there until I do a control delete. Then my PC runs OK. However througout the day the message appears.When Lab View is open it always appears. This may may be also be causing problems. I have a time line so I opted to work with the erratic PC. The PC was dead this morning. So IT will now have to fix the problem and reload the OS. Thanks for the responses. I was begining to be afraid of a program size. The company that I work for likes daily changes. I find it difficult to have structure software when the changes impact the whole VI.

Thanks

Philip

[johnsold]

Philip,

Perhaps you should address both the "daily changes" and the program size issues by converting your program to use lots of subVIs.  Then you can work on the subVI which needs to be modified for the "daily change" without affecting the rest of the program.  It also makes most of the VIs (subVIs) much smaller.

Also make sure you have backups.  When you modify VI or subVI do a Save As... with a different name.  So Random.vi becomes Random.2.vi or Random.2a.vi, or whatever naming scheme you decide to use.  Keep copies of the originals so if the person making the decisions decides he likes the old way better, it is easy to go back without reprogramming a VI to do what it did yesterday or last week.

A cleaned up computer and a cleaned up program!  You will be in good shape!

Lynn

[Philip2001]

Some portions of my VI can be pulled out and set up as seperate VI'S. Every night I put the latest program on the company network. Backing up software is a programmers first rule. I have als copied some of my VIS to my flash drive. If I had more time I would pull my VI apart and create seperate VIS.

Thanks

Philip

[BBullard]

My project with drivers is 450Mbyte.  The Project Build worked fine in LabVIEW 2009 SP1.  After converting to LabVIEW 2011 SP1 the Project Build takes 10x as long as it did for LabVIEW 2009.

The Project appears to be free of errors, and mass compiles without errors.

Any suggestions about how to improve.