No reason to be upset or to get angry.
I'm quite sure I know at least a little about parallel processing/programming and computing big data in synchronized threads, and I'm quite sure I'm not the only one here. I'm also sure I'm not the only one who understood that your PowerPar library is good for getting the best from a multiprocessor system used by one user at a time. It is not necessarily a good solution if you have multiple users working on one system in parallel, like on a RDP server. Besides all that, we just exchanged our thoughts about selecting one processor if needed. And I also think it's good that we made it clear that your library is not a solution for multi-user scenarios.
So, calm down.
New library PowerPar for parallel programming
Re: New library PowerPar for parallel programming
Best regards,
Tom
"Did I offend you?"
"No."
"Okay, give me a second chance."
Tom
"Did I offend you?"
"No."
"Okay, give me a second chance."
Re: New library PowerPar for parallel programming
Hay, I am not angry and upset, I think you misunderstood. I wanted to make it clear that one-processor apps and multi-user discussion is not the topic of this thread, and not proper to put PowerPar usage in that context. I didn't mean that you are quite ignorant about parallel programming, but that PowerPar is the best for that scenarios to use and you probably don't have them. If I am wrong I apologize.
Slavoljub Damnjanovic
SD-SoftDesign, Alaska Software Technology Partner
https://www.sd-softdesign.com
https://www.sd-softdesign.rs
SD-SoftDesign, Alaska Software Technology Partner
https://www.sd-softdesign.com
https://www.sd-softdesign.rs
Re: New library PowerPar for parallel programming
Hi,
I'm wondering why you are promoting your library here if we are too stupid to understand in your opinion.You don't know for better and it is better for you.
Re: New library PowerPar for parallel programming
Hi,
I wanted to introduce my work and my achievement, that is the purpose of this heading "Announcements". If you have nothing to say regarding the topic, there is no need to discuss at all. The topic here is multi-CPU app and not one-CPU app. You don't have to agree with my work, but don't compare it with something that is not relevant.
I agree with Jan on German Forum that posts in each thread should keep track the topic of the thread. If you want to speak about something that is not relevant for the topic, create a new thread for that topic.
My words "You don't know for better" means "You don't know the other way", and that is true. You work in one-CPU per app scenarios, like many others. PowerPar offers using multi-CPU scenarios per app, which is unique. So, you misunderstood and no need to be offended. I think there are no stupid members on this Forum.
I wanted to introduce my work and my achievement, that is the purpose of this heading "Announcements". If you have nothing to say regarding the topic, there is no need to discuss at all. The topic here is multi-CPU app and not one-CPU app. You don't have to agree with my work, but don't compare it with something that is not relevant.
I agree with Jan on German Forum that posts in each thread should keep track the topic of the thread. If you want to speak about something that is not relevant for the topic, create a new thread for that topic.
My words "You don't know for better" means "You don't know the other way", and that is true. You work in one-CPU per app scenarios, like many others. PowerPar offers using multi-CPU scenarios per app, which is unique. So, you misunderstood and no need to be offended. I think there are no stupid members on this Forum.
Slavoljub Damnjanovic
SD-SoftDesign, Alaska Software Technology Partner
https://www.sd-softdesign.com
https://www.sd-softdesign.rs
SD-SoftDesign, Alaska Software Technology Partner
https://www.sd-softdesign.com
https://www.sd-softdesign.rs
Re: New library PowerPar for parallel programming
Slavkodam is right, he and I want to get the most out of more CPU cores, if the processing of the task on one CPU cannot be accelerated in any way, neither by optimizing the code nor by creating another algorithm. Only if by rewriting in C language and compiling in C/C++ (this is how I had some routines made in Borland C++ even under Ca-Clipper and they were really very fast), or directly in machine code, which is not efficient nowadays. And it is also uneconomical to go the route of a more powerful CPU or a faster disk.
That's why I divide the tasks into smaller pieces and each one is processed separately on its own CPU, and this significantly shortens the total process time, if x is the number of CPUs, and h is the processing time, then it's almost h/x, not quite, but almost. Windows uses multiprocessor processing in a slightly different way, and more efficiently, especially for computer games, probably one CPU core would not slow down, but I don't know how it works.
That's why I divide the tasks into smaller pieces and each one is processed separately on its own CPU, and this significantly shortens the total process time, if x is the number of CPUs, and h is the processing time, then it's almost h/x, not quite, but almost. Windows uses multiprocessor processing in a slightly different way, and more efficiently, especially for computer games, probably one CPU core would not slow down, but I don't know how it works.