SphinUX OS Claims To Be ~150% Faster Than GNU/Linux
Phoronix: SphinUX OS Claims To Be ~150% Faster Than GNU/Linux
SphinUX OS is an open-source POSIX-compatible operating system developed under the GPLv3 and running the Egyptian LSX Kernel Architecture. This open-source operating system claims to be much faster than Linux and that its memory usage can even be 3x less! This is an operating system with some of the most wildest performance claims we have ever seen...
If it supports only POSIX and not the range of APIs Linux supports - it's only interesting to folks who already use fringe OSes like BeOS, DOS etc.
I'm pretty sure we're presented with the usual situation: the size of the OS is generally inversely proportional to its speed. When Linux was much smaller there were also astonishing benchmark results, when it matured to support a much wider array of APIs, drivers and hw it did so generally at the expense of speed.
I guess since it's perhaps easiest, I'll start with the English language and their use thereof. "The kernel proved to process 100% faster than the Linux kernel, memory consumption is 300% less, and I/O is 50% better than the Linux kernel." I'm not sure how you evaluate "100% faster", but one way might be to look at the "sys" time and take it ALL away? As for "300% less" memory consumption, does that mean that if the Linux kernel uses a bit over 16M (according to System.map?) of my 8G machine, the SphinUX kernel will make it seem like that machine has 8G + 32M once it's loaded? Any time improvements get close to or above 100%, so does my skepticism. The I/O may be the only reasonable one here, depending on the pixie dust they've used to accelerate disk seek times and rotational latency.
That is of course before we get to "supports "99% of modern and legacy hardware", highly-scalable, and is a secure computing environment." To begin with, getting hardware penetration that high on an immature (speaking only of time, here) OS seems highly unlikely - even getting a Windows driver shim to work perfectly would have a hard time with that number. Then it's fast AND secure??
Finally, looking at that diagram it's got a lot of layers, and layers don't generally equal speed, unless they're technical abstractions only, and don't equate to real code. Call/return overhead would be a killer.
Extraordinary claims call for extraordinary evidence.
If it really is that much faster then I think the only honorable thing would be for the whole linux kernel team to commit seppuku. They spewed so much bullshit about their godlike programming skills and the linux speed that there will be no way out for them....
When it comes to the topic of "spewing bullshit", you of all people probably shouldn't throw rocks in a glass house...
I find all their claims a little hard to believe, but we'll leave it up to Michael to help prove everything. Personally, I don't see it as being particularly hard to outperform linux, as it's a very versatile and general purpose OS - in fact, it's more general purpose than just about any OS considering you'll find it running in devices like watches, phones, routers, robots, kitchens appliances, vehicles, and obviously, computers. I'm not aware of any other OS that can be found in as many places like that, except maybe free-BSD.
It's a hoax. The entire OS is probably some kind of malware that hijacks your computer for a bitcoin/litecoin mining botnet.
Think about it, it'd be brilliant: it's so much harder to hijack Linux/BSD user's computers, for various reasons, and who are the people who spend lots of money for high-end hardware? Geeks. Geeks, who are harder to infect with malware by regular means, but give them something new to play with - a new kernel, a new OS that claims to be the best thing since sliced bread - either way you're hooked: you either think it's amazing or disbelieve it and set out to try it to disprove their claims. Then bam, you get a rootkit and your high-end hardware is used to make money for some shady guy in Egypt.