Programming language ray tracing benchmarks project

#1

Post by jblindsay » Thu, Apr 25, 2019, 5:34 PM UTC

Programming language ray tracing benchmarks project
github.com

Re: Programming language ray tracing benchmarks project

#2

Post by ilitirit » Thu, Apr 25, 2019, 6:09 PM UTC

Interesting that Nim is slightly faster than C it considering that it compiles down to C.

Re: Programming language ray tracing benchmarks project

#3

Post by azhenley » Thu, Apr 25, 2019, 6:09 PM UTC

This is awesome! More good press for Nim.

There is a big variation in performance, some of which I find surprising. Do you know what exactly causes some languages to be so slow (e.g., small objects being created and garbage collected frequently)?

Re: Programming language ray tracing benchmarks project

#4

Post by kyberias » Thu, Apr 25, 2019, 6:13 PM UTC

I don't think the C# time is representative. I suspect Mono is really slow here. I just ran it with VS 2015 in 1 min 24 sec.

Re: Programming language ray tracing benchmarks project

#5

Post by ygra » Thu, Apr 25, 2019, 6:21 PM UTC

kyberias wrote: post #4

I don't think the C# time is representative. I suspect Mono is really slow here. I just ran it with VS 2015 in 1 min 24 sec.

For some workloads Mono seems awfully slow. The compiler I'm maintaining at work takes about twice as long on Mono on Windows and about four times as long as Mono on Linux compared to NET. I guess NET Core would be comparable or faster than .NET Framework and similar on both platforms.

Re: Programming language ray tracing benchmarks project

#6

Post by once-in-a-while » Thu, Apr 25, 2019, 6:29 PM UTC

A number crunching benchmark? Seriously?

Any more experienced language user could trim down the execution time by applying known rules for any given language.

Sorry, but this is not serious!

Re: Programming language ray tracing benchmarks project

#7

Post by kev6168 » Thu, Apr 25, 2019, 6:30 PM UTC

I am surprised PyPy has such a huge lead over Python.

    $ time python pyrb.py 

    real    348m35.965s
    user    345m51.776s
    sys     0m22.880s

    $ time pypy pyrb.py

    real    14m2.406s
    user    13m55.292s
    sys     0m1.416s

Re: Programming language ray tracing benchmarks project

#8

Post by alkonaut » Thu, Apr 25, 2019, 6:33 PM UTC

kyberias wrote: post #4

I don't think the C# time is representative. I suspect Mono is really slow here. I just ran it with VS 2015 in 1 min 24 sec.

The C# implementation looks flawed (uses reference types for vectors etc). Using value types and .NET Core should give a much better result than that. Will try to remember doing a PR.

Re: Programming language ray tracing benchmarks project

#9

Post by kyberias » Thu, Apr 25, 2019, 6:36 PM UTC

alkonaut wrote: post #8

kyberias wrote: post #4

I don't think the C# time is representative. I suspect Mono is really slow here. I just ran it with VS 2015 in 1 min 24 sec.

The C# implementation looks flawed (uses reference types for vectors etc). Using value types and .NET Core should give a much better result than that. Will try to remember doing a PR.

Tested this on the same machine with Ubuntu for Windows + old .NET Core 1.0. 3 min 20 seconds.

Re: Programming language ray tracing benchmarks project

#10

Post by FartyMcFarter » Thu, Apr 25, 2019, 6:37 PM UTC

kev6168 wrote: post #7

I am surprised PyPy has such a huge lead over Python. $ time python pyrb.py real 348m35.965s user 345m51.776s sys 0m22.880s $ time pypy pyrb.py real 14m2.406s user 13m55.292s sys 0m1.416s

10-100x speedup seems normal for CPython vs well compiled code. I don't have experience with PyPy though, so in this specific case I'm not sure.