Windows 10 vs. Eight Linux Distributions On The Threadripper 3970X

[andyprough]

That doesn't explain Tumbleweed using really similar package versions, some newer like the kernel, and having better results. Doing a quick glace at the table, I think it's because they're using GCC 9.2.0 and all the rest of the newer ones are using GCC 9.2.1...Manjaro is the odd one out with 9.2.0.

Could be. But even so, CentOS is using 8.3.1.

[sdack]

It would be great if people actually engaged their brains before commenting on here...

True, if only to stop them from making dumb remarks.

But there is also more than the instruction cache. Enabling more features means the compiler has to consider more complex code paths and cannot be as aggressive as you would like it to be, i.e. dead code removal. Larger binaries also means longer startup time just for loading the code into memory. Then keeping it in L3 and L2 cache does indeed create more cache pressure. Hot and cold sections tend to be further apart, adding to the problem. Also more features tends to pull in more shared libraries, which also need to be loaded and further fill up the caches. Some features of the compiler like stack protection alone cause noticeable performance hits all on their own. And just because it has mechanisms, which counter some of these effects doesn't mean they completely negate them, but they often only reduce these effects, and aren't magic bullets.

[pipe13]

Am I the only one who thinks that those last and first place finishers are completely irrelevant?

Probably. Last-place Manjaro might be irrelevant for those of us it's not relevant for, but Clear Linux is relevant for everybody who is not Microsoft. Because open source: Clear Linux was originally intended as a test bed to demonstrate how fast was reasonably possible. If any other distro lags "significantly" behind Clear Linux in an "important" benchmark, the devs and maintainers of the lagging distro are welcome to dive in, find out why, and fix or not as they deem appropriate.

Since then, Clear Linux has taken on some trappings of a "real" desktop distribution, so those of us not named Michael Larabel can see for ourselves. Jim Salter has a review at Ars Technica: https://arstechnica.com/gadgets/2020...lear-linux-os/

[edwaleni]

.......without the optimizations that Clear Linux creates specifically to win benchmark tests.

What exactly are you implying? Context needed, can be seen 2 ways.

[edwaleni]

Probably. Last-place Manjaro might be irrelevant for those of us it's not relevant for, but Clear Linux is relevant for everybody who is not Microsoft. Because open source: Clear Linux was originally intended as a test bed to demonstrate how fast was reasonably possible. If any other distro lags "significantly" behind Clear Linux in an "important" benchmark, the devs and maintainers of the lagging distro are welcome to dive in, find out why, and fix or not as they deem appropriate.

Since then, Clear Linux has taken on some trappings of a "real" desktop distribution, so those of us not named Michael Larabel can see for ourselves. Jim Salter has a review at Ars Technica: https://arstechnica.com/gadgets/2020...lear-linux-os/

Yes, I read same article. It should have been called "My personal exploration of running ZFS on Clear Linux" but it gave the world a good idea that anyone used to tweaking Linux to their whims may have several hurdles to jump over with Clear.

[andyprough]

Probably. Last-place Manjaro might be irrelevant for those of us it's not relevant for, but Clear Linux is relevant for everybody who is not Microsoft. Because open source: Clear Linux was originally intended as a test bed to demonstrate how fast was reasonably possible. If any other distro lags "significantly" behind Clear Linux in an "important" benchmark, the devs and maintainers of the lagging distro are welcome to dive in, find out why, and fix or not as they deem appropriate.

Since then, Clear Linux has taken on some trappings of a "real" desktop distribution, so those of us not named Michael Larabel can see for ourselves. Jim Salter has a review at Ars Technica: https://arstechnica.com/gadgets/2020...lear-linux-os/

What a horrible desktop experience to read about. Phoronix should really stop promoting Clear as anything but a way to win benchmark tests. Intel's own survey last year showed over 90% of their own developers do not use it.

[Danny3]

Really interesting to see that Ubuntu 20.04 not only that it comes with an outdated kernel, but it's also the slowest Linux distro from all of them, even slower than their previous LTS.

I wonder what is Canonical doing in 2 years of development when actually most of the hard work is done by the kernel developers and debian developers.
I wonder if Canonical tight friendship with Microsoft has anything to do with this, like an artificial limitation or not enough optimizations so Windows 10 would not look so bad in benchmarks against Ubuntu.
Anyway I think that Ubuntu 20.04 will be the most disappointing Ubuntu release ever.

[wertigon]

Really interesting to see that Ubuntu 20.04 not only that it comes with an outdated kernel, but it's also the slowest Linux distro from all of them, even slower than their previous LTS.

I wonder what is Canonical doing in 2 years of development when actually most of the hard work is done by the kernel developers and debian developers.
I wonder if Canonical tight friendship with Microsoft has anything to do with this, like an artificial limitation or not enough optimizations so Windows 10 would not look so bad in benchmarks against Ubuntu.
Anyway I think that Ubuntu 20.04 will be the most disappointing Ubuntu release ever.

Yes, it may be disappointing.

On the other hand and speaking from experience, performance is almost always addressed late in the product lifecycle (premature optimization being the root of all evil and all that), so it is quite possible it will improve by quite a bit the following two months.

[Spooktra]

Here's what I would like to know, do all distros run the cpu at the same clock speed? I am asking because with Ubuntu, depending on the governor setting I use, the clock speed will vary widely, especially during long encodes, as the processor heats up and thermal throttles.

What I'm wondering is if Clear Linux, as a rule, simply runs the cpu flat out on all cores as often as possible, where as other distros are more gentle on the cpu. This could mean that Clear would be great for short benchmarks but not good for long workloads or cpu longevity.

[schmidtbag]

Am I the only one who thinks that those last and first place finishers are completely irrelevant?

Well, depends how you look at it. Does it matter when you're watching the Olympics?
The overall performance is more important but that doesn't always tell a complete story. Windows for example performs the worst on average, but there are moments where it was #1, or at least placed better than everything other than Clear (and Clear isn't exactly a good representative of the average Linux setup). That shouldn't be ignored - that means Linux has room for improvement.