Andres Rodriguez
New Member
I am running Windows 7 and just updated to Vers. 6.0.7.4. The side panel that should display the variables does not make sense. See picture.
Andres
Andres
Variables not displayed on Vers 6.0.7.4
- Thread starter Andres Rodriguez
- Start date
i have exactly the same problem after update yesterday to 6.0.7.4.! what is wrong ? what can i do ?.. or how can i get the the older version back ?
help needed !!
Ron
Andres Rodriguez
New Member
I wonder if Technical looks at posts in this part of the forum.
Buzby
Senior Member
PE 6.0.7.4 is worse then the previous betas.
I have the same variable display problem as above, and the simulation instabilities are worse than before.
It is a real shame, as PE6 has so much going for it, but I still have to use PE5 if I need to guarantee no lock-ups while I'm working on a project.
The simulation problem seems to be centered around the updating of the screen when dynamically updating. If the running simulation does not need to scroll the screen, then it works better, but if it needs to scroll then it can't even run as fast as PE5.
I still don't believe that the problems are due to 'antique PCs', I think it's more due to Rev-Ed using the wrong toolset to build PE6.
I have the same variable display problem as above, and the simulation instabilities are worse than before.
It is a real shame, as PE6 has so much going for it, but I still have to use PE5 if I need to guarantee no lock-ups while I'm working on a project.
The simulation problem seems to be centered around the updating of the screen when dynamically updating. If the running simulation does not need to scroll the screen, then it works better, but if it needs to scroll then it can't even run as fast as PE5.
I still don't believe that the problems are due to 'antique PCs', I think it's more due to Rev-Ed using the wrong toolset to build PE6.
The variable display is simply a flowchart flag also being incorrectly used by the BASIC simulation, this will be fixed in the next release.
Please uncheck: File>Options>Flowcharts>'Hide Advanced Features within Explorer Panels'
There are no other main simulation run changes at all.
Buzby - you have made the same point numerous times in various threads, we will not be changing the development architecture.
Last edited:
Buzby
Senior Member
Most of the time I don't need to see the step-by-step execution of my code, I only need to see the results.
I do this by running the simulation as fast as possible, and use serout to the virtual terminal to see specifics.
This technique worked reasonably well on PE5, but slowly because each step was constrained to a minimum 30mS delay. On PE6 I can't even run at 50mS reliably, and no chance at all at 30mS.
My suggestion is to include some new #directives to disable the update of the GUI panels when they are not needed.
E.g. #Variables_Off / On, #Pinouts_Off / On, and especially #CodeFollowing_Off / On. Directives like these would not affect the simulation engine itself.
I know I've gone on about the Syncfusion toolset, and now accept that you are not going to change from it, but are you using it properly ?.
It does not seem right that a 3.2GHz 4Gb PC should struggle to simulate a simple language running on a tiny PIC that has hardly any resources.
I do this by running the simulation as fast as possible, and use serout to the virtual terminal to see specifics.
This technique worked reasonably well on PE5, but slowly because each step was constrained to a minimum 30mS delay. On PE6 I can't even run at 50mS reliably, and no chance at all at 30mS.
My suggestion is to include some new #directives to disable the update of the GUI panels when they are not needed.
E.g. #Variables_Off / On, #Pinouts_Off / On, and especially #CodeFollowing_Off / On. Directives like these would not affect the simulation engine itself.
I know I've gone on about the Syncfusion toolset, and now accept that you are not going to change from it, but are you using it properly ?.
It does not seem right that a 3.2GHz 4Gb PC should struggle to simulate a simple language running on a tiny PIC that has hardly any resources.
Fully agree, and we test without any problems on much older machines with much fewer resources and do not see these issues.
So that points to something else going on on your machine, maybe some time investigating why your machine is struggling would be more beneficial.
Buzby
Senior Member
I wish I knew how to !.
grim_reaper
Senior Member
I don't think it's a matter of using it properly or not; with third-party controls/toolset you're at the mercy of the developer. If Rev-Ed didn't use any third-party tools, there would be just as many (probably more) issues using the bog-standard Microsoft controls.
Fortunately, the benefits of Syncfusion massively outweigh the problems, again especially compared to MS. But if there's an issue deep down in a Syncfusion library that only affects certain machine configurations, it's something that they need to fix, not Rev-Ed.
For your peace of mind Buzby, I'm just updating to 6.0.7.4 and then I'll be testing it to death to see if I can offer any insights into why your machine might be struggling. No promises though.
Buzby
Senior Member
Hi grim_reaper,
Thanks for the interest, it looks like you have a better idea than me as to how Windows works !.
I honestly don't know how to go about finding what goes on inside my PC. I knew exactly what went on in my Apple ][, had a reasonably good idea in DOS, but complelety lost it in Windows. There's just too much to learn if it's not your day job.
The problem I and tmfkam have shows as the center panel stepping through the code, but none of the other panels even drawing properly. Other problems seem to be related to low sim delays, < 50mS, and to code highlighting that requires the panel to scroll. As there are at least two users on the forum, me and tmfkam, who have exactly the same problem, it would seem that the issue is replicateable. How many prospective users exploring the market have downloaded PE6, had poor success, then binned it and gone for an A ?.
Technical's attitude is very wrong, saying it's up to the users to find out why Rev-Ed's beta software won't work on their machines.
If you really want to help me find out what is going wrong I am quite happy to send you any info you can tell me how to retrieve, or run a diagnostic, or whatever.
Cheers,
Buzby
Thanks for the interest, it looks like you have a better idea than me as to how Windows works !.
I honestly don't know how to go about finding what goes on inside my PC. I knew exactly what went on in my Apple ][, had a reasonably good idea in DOS, but complelety lost it in Windows. There's just too much to learn if it's not your day job.
The problem I and tmfkam have shows as the center panel stepping through the code, but none of the other panels even drawing properly. Other problems seem to be related to low sim delays, < 50mS, and to code highlighting that requires the panel to scroll. As there are at least two users on the forum, me and tmfkam, who have exactly the same problem, it would seem that the issue is replicateable. How many prospective users exploring the market have downloaded PE6, had poor success, then binned it and gone for an A ?.
Technical's attitude is very wrong, saying it's up to the users to find out why Rev-Ed's beta software won't work on their machines.
If you really want to help me find out what is going wrong I am quite happy to send you any info you can tell me how to retrieve, or run a diagnostic, or whatever.
Cheers,
Buzby
grim_reaper
Senior Member
Buzby,
I hope I do have a good idea how Windows works - I've been doing this for 18 years now! You're right though, unless you're brought up learning it, or spend all day doing it, learning a new modern OS is a nightmare. I have the opposite problem to you; I'll never understand Apple.
My first attempts at digging into the editor have resulted in the same kind of issues you're describing. When simulating - my current project is at a leisurely 150ms delay - the simulation controls become slow to respond. If drop the delay down to 50ms I actually get a better response; presumably because it's not busy 'waiting' for the delay period to elapse. What I think is a sticking point is using the Step function - the application focus returns to the main window or the serial terminal and so several clicks are required to keep it stepping through. This also may be due to several commands being executed in the same line (a prime example being 'Low pin1, pin2, pin3').
The most basic diagnostic tool I can suggest to you is Windows Task Manager. Even in XP it's more useful than most people realise. If you edit the displayed columns under 'Processes' and make sure memory, thread count and GDI objects are displayed, then run a simulation you can see what's changing. The most notable thing when I run my simulation is that the thread count only increases by two - one for the simulation and one for the serial terminal window, telling me that PE is not taking advantage of as many threads as it could. One thing for you to check is how your GDI object count compares to other processes you have running; it might be that your graphics card is being strained, rather than your CPU/RAM etc.
Also, I feel I should point out that this IS beta software! I think Rev-Ed are doing us a favour in letting us test it to these extremes before it's 'officially released'. Most software companies don't let beta software anywhere near their customers until their internal testers are completely happy, which might be good from a 'our software works well' viewpoint, but as for customer satisfaction I prefer that we get to review and rate PE while it's being finalised.
I hope I do have a good idea how Windows works - I've been doing this for 18 years now! You're right though, unless you're brought up learning it, or spend all day doing it, learning a new modern OS is a nightmare. I have the opposite problem to you; I'll never understand Apple.
My first attempts at digging into the editor have resulted in the same kind of issues you're describing. When simulating - my current project is at a leisurely 150ms delay - the simulation controls become slow to respond. If drop the delay down to 50ms I actually get a better response; presumably because it's not busy 'waiting' for the delay period to elapse. What I think is a sticking point is using the Step function - the application focus returns to the main window or the serial terminal and so several clicks are required to keep it stepping through. This also may be due to several commands being executed in the same line (a prime example being 'Low pin1, pin2, pin3').
The most basic diagnostic tool I can suggest to you is Windows Task Manager. Even in XP it's more useful than most people realise. If you edit the displayed columns under 'Processes' and make sure memory, thread count and GDI objects are displayed, then run a simulation you can see what's changing. The most notable thing when I run my simulation is that the thread count only increases by two - one for the simulation and one for the serial terminal window, telling me that PE is not taking advantage of as many threads as it could. One thing for you to check is how your GDI object count compares to other processes you have running; it might be that your graphics card is being strained, rather than your CPU/RAM etc.
Also, I feel I should point out that this IS beta software! I think Rev-Ed are doing us a favour in letting us test it to these extremes before it's 'officially released'. Most software companies don't let beta software anywhere near their customers until their internal testers are completely happy, which might be good from a 'our software works well' viewpoint, but as for customer satisfaction I prefer that we get to review and rate PE while it's being finalised.
If you have multiple commands on a single line you will need to step multiple times. This is intentional, the compiler works command by command, so if you choose to put multiple command on a single line you will need to click 'step' for each command on that line.
Remember you can also use spacebar or down arrow.
grim_reaper
Senior Member
First of all, I haven't touched XP in years now so I'm rusty to say the least. The low thread count (think of threads like program slots in an 'AXE) means PE5 wasn't built with multiple CPUs in mind, which is fair enough and to be expected. Your thread count on Win 7 is the same as mine, basically showing PE6 is running exactly the same way on both your machine and mine; again, to be expected.
The GDI is the Graphics Device Interface. Basically every little box, bar, shape, textbox, bit of text, etc is drawn as a 'graphics object', so obviously the bigger the number the more graphical stuff the applications drawing and/or showing. Lower is generally better, but then you don't want a blank screen two lines of text on these days, do you?
It's not the quantity you're looking for as such, it's the change in those numbers from the usual. Now that you have a rough idea of what the numbers are normally, stress your machine(s) and see what changes.For example, running 10 web browsers all showing different YouTube pages should show you a massive increase in CPU usage and GDI objects - i.e. playing videos takes processing power and screen display time. The memory usage shouldn't be too bad, because the videos are streaming from the internet.
On the other hand, if you were running a PE6 simulation and the memory usage doubled each time, that would indicate PE6 is using too much memory and not freeing it properly. [Just examples, I'm sure Rev-Ed do all this basic testing every time they rebuild the application!].
So, if you can figure out what's being strained on your system when you have the problems you've described, you can narrow it down. If the CPU, threads, memory or GDI objects get excessive then the processor, RAM or graphics card respectively could be struggling. If the I/O Reads/Writes climb a lot then your hard drive is struggling and so on.
I think that's enough gumpf for one post, let us know if you can diagnose anything!
Buzby
Senior Member
I think we might be getting somewhere !.
On my XP machine I can open 7 separate Firefox browsers and play 7 Youtube videos at once.
The results look like this :
On my Win7 machine I can't play any Youtube videos at all !.
The video starts, then a few seconds later it is replaced by a very helpful Youtube message saying 'An error occurred, please try later'.
So it looks like there is some problem with the Win7 box, but it must be very specific, it doesn't seem to affect anything else. I wrote a bit of VBA to scroll an Excel sheet, and that can do nearly a hundred full-screen scrolls a second.
I will try to get the Win7 box working better.
On my XP machine I can open 7 separate Firefox browsers and play 7 Youtube videos at once.
The results look like this :
On my Win7 machine I can't play any Youtube videos at all !.
The video starts, then a few seconds later it is replaced by a very helpful Youtube message saying 'An error occurred, please try later'.
So it looks like there is some problem with the Win7 box, but it must be very specific, it doesn't seem to affect anything else. I wrote a bit of VBA to scroll an Excel sheet, and that can do nearly a hundred full-screen scrolls a second.
I will try to get the Win7 box working better.
grim_reaper
Senior Member
Checked for updates to Adobe Flash? And there's an update to Oracle's Java that was going around recently too.
God only knows what stuff YouTube uses to play videos though...
God only knows what stuff YouTube uses to play videos though...
Buzby
Senior Member
I've taken the graphics card out of the Win7 PC and reverted to the onboard graphics.
Still no success with Youtube, so I'm not even bothering to try PE6.
Back to my antique XP 1.8GHz 2Gb, and comparing PE 6.0.7.5 with PE 5.5.6
The attached code was used to test the syntax check times.
PE 6 takes about 4 seconds, PE 5 takes less than 1 second.
Is this as expected, or do I need a 7.2GHz machine to make PE 6 run as fast as PE 5 ?
Still no success with Youtube, so I'm not even bothering to try PE6.
Back to my antique XP 1.8GHz 2Gb, and comparing PE 6.0.7.5 with PE 5.5.6
The attached code was used to test the syntax check times.
PE 6 takes about 4 seconds, PE 5 takes less than 1 second.
Is this as expected, or do I need a 7.2GHz machine to make PE 6 run as fast as PE 5 ?
grim_reaper
Senior Member
Hmm... that code takes about 2 or 3 seconds on mine (seems to vary, probably because I'm playing music and have Visual Studio running as well).
Good luck finding a processor running at over 3.6GHz for less than the price of a small house!
Good luck finding a processor running at over 3.6GHz for less than the price of a small house!
grim_reaper
Senior Member
If they're different applications, it's extremely unlikely they're performing exactly the same operation. I'm assuming you're talking about PE5 vs PE6?
Because of the differences in technologies, there will be differences in the way things work in the murky depths.
Because of the differences in technologies, there will be differences in the way things work in the murky depths.
grim_reaper
Senior Member
You're talking about Microsoft here.