I tested it with Kazarus (26 ms, with B4J (19,9 ms) with Java (14ms) with C++ (11ms) with Python (35 ms), with Ruby (16 sec). What you want to tell me? All of them are optimizing? No. If I am optimizing the Java Code in runner then I will have around 6ms while the Java is than threading. It is not optimized on all sites. Sorry. I know what I am doing.
It is like you said but the difference is not that big and I think it is comparing to Xojo SuperUltraMegaHyperFast
I just have to say: there is the adding to the outer variable in Java what makes it impossible to optimize without a runner I have to define in Sourcecode what I have not done like everybody can see. Also C++ has the same Cose structure without optimizations you can take Clion and do it
Yes, they could all be optimising most of the work out. Even the setting of the variable ‘b’ could be optimised out.
I asked Norman to optimize the code for me, fair enough? I am taking the counter from Java out and I am taking Variable declaration from both out of the loop so we have same conditions. Is that acceptable for you as fairly compared?
Optimized Java: 6,6 ms!
@npalardy found trhe Workaround. At the beginning of the code I have to write
#If debugBuild = false
#Pragma BackgroundTasks False
#Pragma BoundsChecking False
#Pragma BreakOnExceptions False
#Pragma NilObjectChecking False
#Pragma StackOverflowChecking False
#EndIf
The code runs then in 72 ms. Fast enough.
1 Like
The Xojo Inc. cage should be kicked over and the birds inside served up with curry and rice. Its precisely everybody’s willingness for going along to get along, some not wanting to damage “relationships” and still others who regard any effort no matter how many participants or to what end, as being absolutely futile and keeps this downward trajectory.
If every person on this site had banded together and taken some (any) collective action to hold them to account, it still would have been better than logging back in months later and see that absolutely nothing has changed and it has now gotten worse.
EVERY COMMERCIAL BUSINESS is accountable to its customers and must eventually be brought into line with the needs of its patrons. There are no exceptions. They deliver, or they are buried in customer complaints and starve to death. The only reason that Xojo Inc. is ignoring its customers is because the customers (for the most part, and there are exceptions) are apathetic, defeatist and spineless.
I know, you all have moved to other and better platforms, and nobody cares about this company or product, and I understand that to ever cause the situation to change is absolutely and utterly futile, so please… carry on.
1 Like
See, we will carry on because there is no other way and especially not a way back. We need to work and we need to get our Jobs done. To get a Job done the tool needs to be professional, if there are questions about that, ask your plumper, he will tell you why tools need to be professional. The same situation you have with Software Development.
Therefore I was pressing on Xojo, inc until Geoff Perlman wrote to me personally and later in public again that Xojo 2019R3.2 users will still have Support for their Web 1.0 and that there still can be a development and a software lifecycle. What shall I say: the last days showed the evidence for that what I was becoming sure and aware when I was reading first time from Geoff that there will be Support for Web 1.0 that this will never be happened. The Support stopped immediately with delivering 2020R1.0.
The Company was never telling trues. I am not shure if the CEO believed what he wrote when he wrote it or if he was writing it as a wrong information to calm down the users of Web 1.0 while their development environment was switchet off and kicked of from the maintenance from that moment on. In my believing he was in the knowledge that he will never fulfill what he promised but for that I have no evidences so it is only a meaning.
Xojo was never telling how to make the compiled files faster. What I wrote a few lines higher was knowledge from @npalardy and not officially from any documentation at all. There are in a compiled App all the Debug Stuffs still enabled. Therefore compiled Apps are not faster. Because I was not really sure I spoke yesterday on Discord with Norman about this facts.
Xojo was never promoting this to its customers. So most people developing with Xojo have this problem. On their own documentation site they wrote nothing about the situation that this Checkings, Tasks and Services are always on and costing tons of performance. That’s why I was doing the demonstration yesterday. Xojo is blaming slow even with Pragmas cause it ias slower than Java, Lazarus, C++. It could be as fast as them. And it could be published or sattled down as Option Menu when building Apps. Nothing like this is happened.
So within the fight of xojo, inc against their own users, against their paying customers we have a small but detailled problem. Xojo is settling down on the fact that users don’t have courage to speak on the forum. There are many people with the same meaning like me. But if they would speak, they would be banned. Speaking trues on xojo.com can violate their forum rules so you are out. In my case for 1000 years like Norman also.
For me is clear in this situation that I will not stop to fight because the users have rights. That is one problem which shows that we need rules for businesses like that and for Software vendors publishing programming languages.
Another side of the deal is the company itself. They have 5 or 6 programmers. 3 to 5 people arouind for Marketing, Sales and so on. And as you can see, they are a really small company at all cause xojo has a small list of employees.At the same Moment they have many construction areas: Android, Web 2.0, Desktop, IOS, Lin/Win/Mac with the need of Crosscompiling and many errors and Bugfixes which are not done until now.
The problem an it is thet they are not aware that it is to much. If the amount of burning fires is growing to big, the entire surface is burning faster than you can say No. and in this situation the pressure comes from the market that they have to bring every day something new so they can catch the citizens. The professional programmers with their needs are left on the side: they have no chance to work with.
1 Like
The Pragmas are documented:
http://docs.xojo.com/Pragma_Directives
Stating that this is “debug stuff” is not 100% correct. The majority of them are switching off runtime safety checks (which incur an overhead) and disable the yielding to other threads.
So, is that so? Nice that YOU say it because exactly that is the point: when and if the Switches are off you are working outside any assistance. Exactly that means: they where never telling how to make it faster.
Well, as @s7g2vp2 pointed, those are DOCUMENTED and the documentation clearly states: " Pragma Directives, Used to suspend and/or override actions to improve code execution times"
Exactly that mens: They are telling how to make it faster.
But they are not telling what you are exactly doing and which functionality will after setting eventually not work correctly anymore. A knowledge they have but not telling within the documentation. Also the compiler optimization. Sometimes not usable. So you can say what you want, that should have been better documented especially in view on the functionality to other elements. You may say it is under control of the programmer. Let’s speak about the compiler optimization. You will possibly not be able to tell me that this is really a full documented functionality. Looking on other documentations I see that it is not.
There is a pretty good description of what each pragma does.
I dont know why you are fighting for xojo here. First: it is described on one Page in a documentation where you need to know what you are searching for. I was asking many developers and they had no Idea about it. Guess why. And I did a try yesterday where many people looked on but nobody was in the knowledge how to do it. Seamlessly I did the same before seven month on xojo.com forum where also nobody was in the knowledge how. Only a few developers where thinking about it. Think about: you have a running program needs 58 seconds for a task you can do within milliseconds. Nobody could give that answer what tells me: if I check the usability of the documentation following to the standards I can tell you: no way, to less people finding out how. You byself didn’t know about it. Means: even your programs are becoming slower and could be faster.
So again: for what are you fighting. Anyhow: I would prefere to have your NAME instead of S7g2vp4. Thanks.
It is not “fighting for” any one, it is just stating the facts.
It is documented what is needed, use agressive for slow build but fast executions, and viceversa. There a hole series of compiler related post by xojo, one of the chapter is about optimization. It says what compiler they use and if anyone want to know in deep what the compiller does internally, it is free to review the compiler documentation.
The documentation is there, what you want, a home visit to tell you all the secrets and functions of the tool?
When someone mentioned a possible cause/solutions you inemediatly jump screaming, I KNOW WHAT I´M DOING Even if it was obvios that you dont, I bet it killed anyone interest in helping you.
Otherwise, someone could explained to you what the compiler optimization is and how using it your example code could be executed so fast that is actually not measurable in ms.
Anyway, here is yet another thread hijacked by you. 
2 Likes
I’m not fighting for Xojo. I’m just stating facts.
Maybe not everyone knows about the Pragmas but i’m pretty sure they aren’t some sacred information. I did actually know about them and have possibly been using them for over 10 years.
Why is knowing my name important?
1 Like
So first: when optimiztation is set to maximum often programs are not working anymore
There is no documentation about cause the compiling time is not interesting for me but the optimization.
If you believe it is documented what is needed, not for me, I need more informations.
And sorry: you said: you don’t know what you are doing. But I did know. That is the result when compiled with standard Xojo. Without any settings. Even so I programmed all the other Languages. So same conditions everybody. Optimization helps. But it is not that it is working with switched on controllings so fast. Java does. That is the difference.
What are you fighting for here? Honestly, with all due understanding for your problems. You scream months in the forum, understandable, but the tone makes the music. After all, that’s exactly what you expect from others. You’ve been writing for a long time that you left Xojo, so why are you still fighting this battle and mobilizing against the company and insulting Xojo and their users on other platforms as “A******”? That’s not the nicest thing to do either.
You want freedom of speech, so it would be only fair if you also accept the opinion and commitment of other users here in this forum without making fun of them.
I can say I’m pretty sure he isnt doing NEW development in Xojo
But legacy devices etc are a different matter and he has to support those
At least thats what I have gleaned from here & other places when we’ve spoken
1 Like
Documented is one thing
They are and have been for some time
But they are also not well exposed or easy to find
Search the wiki for “speed optimization” and they dont get shown in the results
There are sections about the code profiler and even notes in there about how to manually speed up code by limiting calculations etc
But nothing about the pragmas
Its not hard to see how someone could not know about these