AMD Ryzen CPU Core Scaling Performance

[duby229]

You talk about indoctrination yet all the arguments you give are that there is an inherent performance problem with OOP that only you understand yet it should be obvious to all. My guess is that you have no knowledge of software development and you just read some blogs and are now spreading that new gained "knowledge". Isn't that the definition of indoctrination?

Again, the way you organize code (OOP vs. procedural) is orthogonal to how you organize the data. That so called "data oriented design" can be done in both OOP and procedural or (probably) even functional languages (I'm not verry familiar with functional programming).

Are you serious? You can't possibly think what you just said make sense do you? The very concept of objects itself makes debugging code literally nearly impossble and that's -BECAUSE- if you have many thing you need to do to a block of data then that bl;ock of data needs to be slput up among mny objects. It most definitely is -not- orthogonal.

[Ansla]

Did you read my previous post?

What part of OOP says "you must have an object for each bot"? And how come it doesn't say "you must have an object for each pixel that forms each bot"?

Stop blaming OOP for personal design mistakes.

[duby229]

Did you read my previous post?



Stop blaming OOP for personal design mistakes.

I didn't blame anything. It just is what it is. And did you read my previous post? Why would you want to take a concept that makes massive amounts of overhead an absolute requirement and then exacerbate it?

[Ansla]

There is no overhead requirement, there is just good or bad design. And you can make both good and bad designs in either OOP or procedural.

[duby229]

There is no overhead requirement, there is just good or bad design. And you can make both good and bad designs in either OOP or procedural.

Obviously, but OOP makes writing good designs hard for code that outputs data. It's just not the inherently correct approach for that scenario -because- there -is- massive amounts of overhead.

[pal666]

Higher memory clocks increase performance because they increase bandwidth and reduce latency.

they come with higher timings, which increase latency

[pal666]

in object oriented programming all that exists is objects.

wrong. that is object based programming. for object oriented you need class hierarchies and for that you need relevant application domain

They can be all sorts of things, but again emphasis is always on code.

wrong again. emphasis of objects-based is encapsulation i.e. bundling of code with data

Where as with data oriented programming emphasis is always on data. What that means is you have to look at you data from the output and imagine what code could produce that output. The code then becomes subset of the data. You can make many threads and work on it in many different ways and the bottleneck is moved to cache coherency hardware on the CPU. That technology was mastered in the early 2000s with MESI and MOESI techniques.

data oriented is like use vector instead of list, which is orthogonal to object-orientedness

[pal666]

The very concept of objects itself makes debugging code literally nearly impossble and that's -BECAUSE- if you have many thing you need to do to a block of data then that bl;ock of data needs to be slput up among mny objects.

this is nonsense. are you programmer at all?

[pal666]

In procedural programming all that exists is the procedures you write. They can do whatever you write them to do, but emphasis is always on code.

this is also wrong. data is always more interesting. here is quote from 1975: Show me your flowcharts and conceal your tables, and I shall continue to be mystified. Show me your tables, and I won’t usually need your flowcharts; they’ll be obvious.

[pal666]

The example about how much the performance can be improved by "data oriented programming" just replaces a list of independent bots that actually do the same thing at the same time with a homogenous "army".

that "example" started by counting 4 icache misses from 4 calls to same method