Scratch

I'm not sure how this script would not work, there might be something else effecting it.

Anyway, you don't have to tell the script to show the variables so much. However, that probably isn't why this script is working.

I tried recreating a more efficient script, here's what it looks like:

Use that and it should work.

Hmm... Maybe a mod should help. 

Hi Alec_2 and hmnwilson,

I could not reproduce the effect you are describing. Both counts first increased then decreased again. Could you perhaps post a project that contains the actual script that's not working for you, so we might examine it? Thanks!

Thanks for all the suggestions guys.  So far I've only had time to run single-stepping.  Somewhat unsurprisingly this shows the expected flow of control down to the end of the first loop - then it goes back though the loop indefinitly, not stopping at the limit value.

Another thought - still clutching at straws - could the 64-bit environment be setting the counts in, e.g. a 32 bit variable, but the loop test failing as the varible is being looked as if it were 64 bits, such that, for some of us, we never see the match condition being met?  I see that I can dump the call stack, but are there any other tools that might be relevant here?

This might be my own systems problem but when i tested it with BASIC scratch it worked when i tested the same with (somewhat Modified (shames on self)) version of scratch it bugged .hwnwilson and Alec_2 have u also modified the source code of scratch , some other bugs due to bad source code modification

1. Mouse Position been shown in Presentation Mode (that * me off to reinstall scratch)
2.Cant enter in Preview Mode.

Alec_2 wrote:

Thanks for this.  I've run your version - but this exibits the same problem.  I'll try an un-install and re-install of Scratch.  (Thinks - could there be a silent failure of disk write(s), e.g. due to insufficent permissions in running the script?)

You should be VERY careful about performing a test looking for an exact match of a value (variable = 42) particularly in a language that doesn't support the integer data type (such as Scratch).  You may be seeing a case of round-off error here, where the value obtained by incrementing the variables is very close to 42 (may even print out as 42) but is slightly different due to the internal representation of the variable.

A much safer approach would be to have a test like

    variable > 41
    or
    ABS(variable - 42) < 0.001
    or
    Round(variable) = 42

Maybe the last one is the best for your purposes.

Hmm thats a bit new kinda problem and yeas Paddle2see has the best answer.

Alec_2 wrote:

Bill, did you mean that I post via "Share this project" online?  Or something else?  I could add scratch blocks below, but hopefully this exchange, in my original post, and particulaly floatingmagictree's reply, give the details.

Yes, that's what I meant, as this is a totally different player environment.
I got the script working, as did a few other people, so if you upload it and post a link then we can all see for ourselves exactly what subtle nuisance is causing it.

I'm using winXP SP3, scratch 1.3.1 using the 1.3.0 EXE and a system with 2 Gigs of RAM and a dual core D820 model.

But I don't think that any of those matter.

Solved.  <blush>  Thanks to Amos of the Scratch Team </blush>

I was using the string 'all_sprites_count', rather than the value of that variable, in evaluating the loop termination condition(s). I.e. using the text literal rather than the block itself. ;-(