DSP Robotics Support

Woohoo!

Finally some hope at the end of the road! Nice Mr Hollins, glad to hear the initial test results loaded with no problems!

great to hear this good news!

The BIG TEST is coming !

I've performed the 'transplant' from NuBeats' Marshal module into my project. I expanded the required INPUT/OUTPUTs to handle the specifics of the project. It was rather straightforward for the most part.

I immediately began testing, and everything worked great.

XML loaded and parsed, Save wrote the Config file to the 'User' folder [as a RawBinary] ... it then Auto-Loaded and stuffed all the pull-down selectors.

I went back and forth between 2 different patch XML files, and confirmed the system works correct.

Of course ... I have seen this before I've sent the VST out to BT'r ... we are now in the 'Waiting Room'.

... this is nerve racking ...

wow ... is this ever frustrating.

BT'r reports back that VST loaded.
Imported his XML file ... it parsed and display the results.

Hit the SAVE button to write the config file ...

it never Auto-Loaded. The file was written to the correct spot [he sent me the file], it just never re-loaded ... thus menu selectors remained empty.

This is the same RUBY module from the last test that was a VST success. I did expand the number of arrays from the test 4 to the needed 17. The file size is still tiny ... 1-2k. It's not that much data.

On my XP machine, with multiple testing, works and works ... like all the other times.

Do I need to increase the DELAY before Auto-load ... it's at 1.5 secs now.

I went into the schematic and added a manual LOAD button to trigger the RUBY module.

Still not able to load the file.

So ... since all of this routine is within RUBY, I not sure how to test if the file can even be found [as with the FIND primitive].

So at this point, I left guessing what is wrong !

This test was now done on both his systems, the Main Win 8.1, and a Win7 laptop ... it does not load the file it saved. It doesn't seem to matter that it works fine on my XP

Since the previous 'testing' app worked, and my project VST doesn't ... the only thing I can think to try is to pull the 'broken' Ruby out of my project and put it in the test app that worked The difference is the number of ARRAYs. 17 in my project ... 4 in the test. How this could make any difference ... not even a clue.

GUYs ... really, I'm questioning ... is there a BUG that I've just happened to hit on ??? How could such a fundamental process as loading a file we just saved, fail on 2 different machines ... yet work in 4 or 5 DAWs that I've tested on mine !

At first I thought, maybe, his main machine has a disk drive problem. Maybe a bad sector that we just happen to be hitting on, or a faulty disk controller ... but with 2 different machines with different OS's ... what are the chances.

CRAZY IDEA ... in another thread, the topic of SSE & SSE2 came up. TROG said [paraphrased] that he's not checked those boxes in years [hope I quoted that close]. I'm grabbin' at straws here ... If this routine would fail ONE TIME on my machine, I'd at least have a chance to maybe identify it

wow ...

Maybe this should have been my first question ....

? Will applications and VST plugins written with FS work on Windows 64 bit with 7 and/or 8 OS ?

if not ... where might I find the requirements listed.

Sincerely.
Thank-you.

Yeah, they work mate.
I have some host issues,
but it's nothing to do with the Win compatability really.
I have Win 7 & 8 both 64-bit.
I have run some 20 plugs,
some for extended periods of time.
hopefully u can solve your issues

nix wrote:Yeah, they work mate.
I have some host issues,
but it's nothing to do with the Win compatability really.
I have Win 7 & 8 both 64-bit.
I have run some 20 plugs,
some for extended periods of time.
hopefully u can solve your issues

Nix ... I am both happy and sadden by your most welcomed post. I have followed and respected your postings back into SM-Land, and I very much appreciate you responding to my thread. Thank-you.

I've just been in lengthy emails to my Beta-tester. I am left shaking my head in trying to understand the cause. I have tried to go step by step to inch our way through testing. We actually had things working. And then just as quickly nothing. If it was a single computer, thats one thing ... but to be in a situation that both systems, different OS's would both fail ... that points to #1 my code, 2. lousy computers, or #3 ... which I'm now taking off the consideration list.

The 'my code' part was graciously resolved thanks to Mr NuBeat. I may be a slack putter together of Prims and Code ... but His design worked perfectly on my system .... hey now wait a minute ... my previous 3 designs worked equally as well on my system Yeah, but NuBeat added cool features and took it to a whole new level !!!

It also worked on my Bud's system, both EXE and VST. UNTIL .... I expanded the container list of Arrays from the original 4 tested ... out to the 17 Arrays I needed in my project. I did this modification, hooked everything up and watched it work exactly as we wanted.

When my B-tester got it ... it all worked UP to the point of Re-Loading the file. This is where we are stuck ... or I am.

We are now in the middle of RUBY code ... which is great,and I can almost follow along But now I cannot tell WHY the file is not loading. Even adding a manual Load trigger does not seem to execute. And NO data [that was indeed Saved] is getting loaded. Either it can't find the file it just wrote, or some kind of data corruption is happening so early that the load is ignored.

I don't yet know how to code RUBY in the area of 'Error Management & Reporting'.

Maybe I should post the FSM file of the test module for other's to test ... I'm just a bit embarrassed that the look of the test is very ... hmmm ... amateurish, but we just wanted to confirm data moving correctly.

I don't know ... with much less sleep than I normally DON'T get ... this has been an unbelievable issue that I never considered so elusive.

Hey Nix !! really ... I do mean Thanks. In the long ... this is very good news to hear.

glad I can input at least a little.
I thought I'd add that FS itself runs solid as,
the only way to crash it seemingly is to do something wrong in Ruby perhaps.
Of course ASM meddling can bring down the entire system I understand.
Thanks for the feedback hint in the other thread.
hehe I've turned up a pun in that line.
It does seem to affect high frequencies particularly,
guitar feedback. Sounds like a good way to blow my speakers prolly.
Sorry for getting away from the point-
maybe you could fire the load trigger after some duration has passed?
Like 10 seconds?
Although a fuzzy remembrance u have already tried this.
Sometimes the duration just has to be long enough.
Also what about a manual trigger instead of auto-loading?
--just to test
Will read the thread again and see if I can understand this better.

in my last example i postet all methodes are written in a rescue block which outputs an errormessage if an error is happening - does it show something when trying to load the file?

if not, what could happen on windows like trog said, you need to compare different scenarios

i prepared the load methode with some if statements about the file before trying to load the file
and the loading process it self is inside a error rescue block so an error message should be shown if happens while loading
maybe it shows something interesting