Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!


jsg, the "Server Review King": can you trust him? Is YABS misleading you? - Page 10
New on LowEndTalk? Please Register and read our Community Rules.

All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.

jsg, the "Server Review King": can you trust him? Is YABS misleading you?

1567810

Comments

  • The US government utilises vpsbench to test all DOD systems and supercomputers in all their black sites & official facilities.

  • @Ahfaiahkid said:
    The US government utilises vpsbench to test all DOD systems and supercomputers in all their black sites & official facilities.

    Is it also used for testing and benchmark of the computers responsible for launching thermonuclear weapons?

  • raindog308raindog308 Administrator, Veteran

    @default said:

    @Ahfaiahkid said:
    The US government utilises vpsbench to test all DOD systems and supercomputers in all their black sites & official facilities.

    Is it also used for testing and benchmark of the computers responsible for launching thermonuclear weapons?

    Doubt it unless @jsg has extended it to run on 1960s-era mainframe custom operating systems.

  • @jsg said: Your fio parameters lead to a benchmark that works totally different from mine

    Apart from the "resumed" messages because of fio spawning processes, let's see the syscalls otherwise and how very dissimilar they are:

    fio in sync (sync + fdatasync=1 with 4m block sizes, similar to vpsbench):

    vpsbench:

    BTW, if you were thinking if the 4m size helps it get any closer to vpsbench, here's what fio had to report.

    WRITE: bw=129MiB/s (135MB/s), 129MiB/s-129MiB/s (135MB/s-135MB/s), io=4096MiB (4295MB), run=31732-31732msec
    

    @jsg said: You trust the yabs/fio results for one simple reason

    Because it is open source, and accurate, yours isn't.

    @jsg said: vpsbench v2 closed source -> evil!!! geekbench closed source -> no problem, that's OK, hey they give you a PDF.

    Pot, meet kettle.

    @jsg said: One does not need the source code and/or debugging symbols for [l|s]trace.

    Correct. It's only that you really lack reading comprehension, otherwise you'd see that I've been stracing your binaries right from Aug 25th and on this thread, no source given.

    @jsg said: Yes, regularly.

    Might want to get yourself checked for selective amnesia, seems like you had forgotten about the additional complexity and indirection those features in Nim add.

    @jsg said: I'll be generous and respond to at least some of the nonsense you spread.

    Yeah, because you took a big bet on spreading bullshit without evidence and get all erect for a dick measuring contest. Instead, you got your dick cut off and handed right back to you.

  • jsgjsg Member, Resident Benchmarker

    @raindog308 said:

    @default said:

    @Ahfaiahkid said:
    The US government utilises vpsbench to test all DOD systems and supercomputers in all their black sites & official facilities.

    Is it also used for testing and benchmark of the computers responsible for launching thermonuclear weapons?

    Doubt it unless @jsg has extended it to run on 1960s-era mainframe custom operating systems.

    Nuh, I didn't. But that doesn't mean much because thanks to a certain "researcher" we now can assume that all OSs are the same anyway.

    @stevewatson301

    Yay, hurray, thanks for the good news! Since the newest results of your "dissertation" we developers can largely stop working, because we have learned that all programs doing a 'lseek, write, fdatasync, nanosleep' cycle are basically equal and the same.
    There's just one problem: the very same "researcher" tells us in the same post that they are quite different. But don't worry.
    I got mine cut off and you won? Yeah right ... in your "logic" that's probably even a valid conclusion. ("Mom, I killed him, I won! I proved that two things are the same yet drastically different").

  • bulbasaurbulbasaur Member
    edited September 2021

    @jsg said: we developers can largely stop working

    You've already stopped working for the most part and have become a keyboard warrior.

    @jsg said: I got mine cut off and you won?

    Absolutely, because you seem to be unable to come up with a coherent answer since your lack of reading comprehension and knowledge about organization of Nim binaries was pointed out.

  • @jsg said:

    @raindog308 said:

    @default said:

    @Ahfaiahkid said:
    The US government utilises vpsbench to test all DOD systems and supercomputers in all their black sites & official facilities.

    Is it also used for testing and benchmark of the computers responsible for launching thermonuclear weapons?

    Doubt it unless @jsg has extended it to run on 1960s-era mainframe custom operating systems.

    Nuh, I didn't. But that doesn't mean much because thanks to a certain "researcher" we now can assume that all OSs are the same anyway.

    @stevewatson301

    Yay, hurray, thanks for the good news! Since the newest results of your "dissertation" we developers can largely stop working, -redacted-

    Please stop using "We" when it is only "You".

    Thanked by 2adly Andrews
  • jsgjsg Member, Resident Benchmarker

    @stevewatson301 said:

    @jsg said: we developers can largely stop working

    You've already stopped working for the most part and have become a keyboard warrior.

    @jsg said: I got mine cut off and you won?

    Absolutely, because you seem to be unable to come up with a coherent answer since your lack of reading comprehension and knowledge about organization of Nim binaries was pointed out.

    Bad news for you: The weird constructs in your head != reality. Sorry.

  • @jsg said:

    @stevewatson301 said:

    @jsg said: we developers can largely stop working

    You've already stopped working for the most part and have become a keyboard warrior.

    @jsg said: I got mine cut off and you won?

    Absolutely, because you seem to be unable to come up with a coherent answer since your lack of reading comprehension and knowledge about organization of Nim binaries was pointed out.

    The weird constructs in your head != reality

    Seems like the way vpsbench was designed.

  • jsgjsg Member, Resident Benchmarker

    @stevewatson301 said:

    @jsg said:

    @stevewatson301 said:

    @jsg said: we developers can largely stop working

    You've already stopped working for the most part and have become a keyboard warrior.

    @jsg said: I got mine cut off and you won?

    Absolutely, because you seem to be unable to come up with a coherent answer since your lack of reading comprehension and knowledge about organization of Nim binaries was pointed out.

    The weird constructs in your head != reality

    Seems like the way vpsbench was designed.

    Yeah, yeah, and you are more handsome than me too *yawn

  • Who’d have guessed this would turn into a love story.

  • @adly said:
    Who’d have guessed this would turn into a love story.

  • jenkkijenkki Member
    edited September 2021

    Just allocate a new shiny tag as suggested to @stevewatson301 for fill up his egoism and set his TUN/TAP to ,,Never been enabled,,

  • defaultdefault Veteran
    edited September 2021

    @jenkki said:
    Just allocate a new shiny tag to @stevewatson301 for fill up his egoism and set his TUN/TAP status to ,,Never been enabled,,

    Any idea what title he/she would like?

  • @default said: Any idea what title he/she would like?

    Look at the page 7 of this pulp fiction. :smiley:

  • @stevewatson301 said:

    So, mods willing, I'd definitely like a "Troll" tag.

    @raindog308 - help him out please.

  • @default said: @raindog308 - help him out please.

    Troll are too short. There a more poetic suggestions. Tag must be a long like a song.

    Egoism has too much empty space for fill up.

  • @jenkki said:

    @default said: @raindog308 - help him out please.

    Troll are too short. There a more poetic suggestions. Tag must be a long like a song.

    Egoism has too much empty space for fill up.

    Then @jsg deserves that tag more than @stevewatson301

  • ArkasArkas Moderator
    edited September 2021

    @jenkki said: Troll are too short. There a more poetic suggestions. Tag must be a long like a song.

    "Imaginary Debunker"??

  • raindog308raindog308 Administrator, Veteran

    @adly said: Who’d have guessed this would turn into a love story.

    Better than Twilight.

  • @raindog308 said:

    @adly said: Who’d have guessed this would turn into a love story.

    Better than Twilight.

    Anything is better than Twilight.

    Thanked by 2Thundas raindog308
  • ,,MJJ and Chaolin certified representative,,

  • ,,Unexpected tart fart tuner,,

  • User jenkki added to ignore list.

  • @jsg said:
    Nuh, I didn't. But that doesn't mean much because thanks to a certain "researcher" we now can assume that all OSs are the same anyway.

    What? That's flipped. You're the one insisting on testing on Freebsd instead of Debian/Ubuntu/Centos(the OS's of interest to majority of LET) despite being pointed to Phoronix benchmarks specifically showing different performance between kernels and distros ages ago? Also, in one of the first posts of this thread, the freebsd vs Linux results were posted and shown to be significantly different.

  • After reading 10 pages... All I can do is laugh.. but it's a sad laugh...

    Thanked by 1Arkas
  • It's 2021, are monarchies even hip anymore? I blame the, "Server Review King," title for this thread.

  • @yoursunny said:

    There is possibly malicious code in the pseudo "benchmark".
    Not only is the results untrustworthy, but also your server may be compromised.
    Stay away from vpsbench.

    And no, you can't discover malicious code by strace.
    In a sophisticated attack, malicious code is triggered only when it's running on the intended victim.

    Sooooooo, Malware until proven otherwise.
    Also, I am not a real software developer.

  • @TimboJones said: User jenkki added to ignore list.

    You can ignore only your ejaculation.

  • raindog308raindog308 Administrator, Veteran

    @jenkki said: You can ignore only your ejaculation.

    If there's one thing that always commands my undivided attention in this world, that's it.

    (to be clear: mine, not @TimboJones's)

    Thanked by 3adly skorous TimboJones
Sign In or Register to comment.