Windows 11, 5200DDR5, 170Watt power limit, background processes like steam. Its just 7/24 condition.
It's also not just about your hardware. For example, Gnumeric outperforms MS Excel.
And if you want to write a document of more than 30 pages, LaTeX is much faster than MS Word.
You can sometimes also save a lot of time by using the fastest software for your task versus the slowest software.
Many people usually use cycles from Blender but what I can learn you is that LuxCore is often faster.
In very simple scenes Blender can be faster. At least that is what is frequently claimed.
LuxCoreRender is excellent at creating caustics, which means if you have a scene with a lot of transmission materials like glass, vodka or any fluids it can make a huge difference in realism.
By contrast, Cycles (Blender) does not have good caustics support.
CPU rendering can produce more accurate and consistent results than GPU rendering, as it can handle more complex effects and algorithms with less noise and artifacts.
If you don't care about quality then it's best to use a GPU for rendering.
But even then, there are still situations where CPU rendering is sometimes faster than GPU rendering.
Your gpu is limited by the amount of vram it has. Your cpu is limited by the ram it has.
If your scene doesn't fit in your vram, then your gpu can't render it.
Some scenes are rendered faster by a CPU than by a GPU, even if your GPU has enough VRAM.
Furthermore, you can also argue that the programmers of LuxCore and Blender were limited by their skills and knowledge of programming languages. C++ is a programming language that scales poorly to multi-core processors. The larger an app gets and the more cores an app uses, the harder C++ is usually outperformed by Java and Haskell. That's a general law that applies in most situations. So if someone comes up with the idea to write a 3D rendering software in a programming language most suitable for 3D rendering then your CPU is normally going to be able to perform a lot better.
Personally, I think Lisp would also make more sense as a scripting language for Blender and FreeCAD. Since AutoLISP has been the standard for AutoCAD for over 30 years and is very suitable for x, y, z, and 3D data. It does not really make sense that there is no option offered to use a Lisp variant similar to AutoLISP in these two apps. Software could be made much faster and more user-friendly than how it currently is. There is actually extremely little thought put into all the basic stuff.
AutoLISP has such a strong following that other computer-aided design (CAD) application vendors add it to their products. Bricscad, IntelliCAD, DraftSight and others have AutoLISP functionality, so that AutoLISP users can consider using them as an alternative to AutoCAD.
Haskell is very competitive with C, and on some benchmarks, it is faster. How is that possible? With all that Haskell does on top of the raw C code, how can it possibly be faster? In this episode, I talk about two advantages of Haskell that can make it faster than C.
ericnormand.me
That is another way that Haskell benefits over C. That if the problem is complicated enough or big enough where a linked list...The difference between in-access complexity and Big O notation complexity, between a linked list and a tree, boom, it's a big enough difference. Haskell's going to win.
It is really hard to write threads in C compared to Java. In Java, it's very easy to write threads and to start new threads. What do you do? What do you do? Who won?
I say that Java people won. That's the whole point of Java, is that it makes those kind of things easy. It makes threads on cross-platform really easy. It makes garbage collection cross-platform really easy.
I think that the same thing is true of Haskell.
If it makes writing a data structure that gives you an advantage over a linked list, let's say, easy. That's what you get. That's why it's faster.
Apache Cassandra has a more structured data storage system than MongoDB.
So it's not a 100% fair comparison because the Cassandra database is heavily structured which makes things more efficient.
But you do see here a Java database that is many times faster than the most popular NoSQL database (and written in C++).
Here you can see that a low effort Haskell HTTP server outperforms NginX with only two workers.
Haskell isn't just faster than C on benchmarks, benchmarks don't mean anything really. It's even better than that: programs written in Haskell, actual tangible useful software I personally run every day on my PC, works faster than a similar program in C, and not only that, but in this case we're talking about the suckless project program, whose whole claim to fame is that it doesn't ship features that would actually be useful for like 3 people, so it's as bare bones as it gets. But wait, there's more: while dwm only had the most basic window manager functionality, the Xmonad build I use is packed to the brim with features, like program select menu, minimizable windows, workspace name displayed on screen and all that good stuff, and it actually was faster like this, which is solely responsible for my borderline worship of Haskell in particular.
The last time I checked out with gnome-desktop I had an ugly lose of about 30/60fps and DWM dropped 10fps compared with Xmonad. 
