Slashdot Mirror

It's hard to tell from the article, but Google is obviously running the game on PC's in the backend.

No, it runs systems with x86 processors but with custom AMD GPUs. I'm not sure what your definition of a PC is but this wouldn't fit most.

Then you will be overjoyed to know that the Radeon VII draws roughly the same power as the RTX 2080. Further, it is likely to underclock very well, dropping into a sweeter spot of the 7nm power curve. At which point it will be not only a very powerful card, but a cool and quiet one too. Whereas in stock configuration, this reference card is known to be about twice as noisy as a typical 2080. That would be my only serious issue, and for that reason I will wait for the OEM cards to land, which they surely will because AMD has, by accident or design, left room for OEMs to differentiate with superior cooling solutions.

I guess this GPU is going to be a solid seller after the dust settles. It's already a hit in the Linux world. AMD will probably be selling as many as they can make for quite some time.

That is considering it purely as a gaming card. But with twice the memory, and 50% faster memory than the RTX 2080 it's not just a gaming card. And 16MB gives it a lot of longevity, maybe an uncomfortable fact for Nvidia, who prefers its products to go obsolete as quickly as possible.

They are describing an option to toggle onscreen display that shows information about the graphics card while it's running a game. Mesa is a graphics library and it's homogenized computing (Vulkan) driver is known as Gallium3D. This is an example of the HUD option that Intel is working on replicating.

This is why you don't have an ideological/emotional tie in to a computer operating system. It's why I dual boot my desktop Windows & Ubuntu and why I also have a Mac, I can just choose the best tool for the job and don't have any need to make silly, false generalizations to justify using one or another because sometimes the best choice is macOS, sometimes it's Linux and sometimes it's Windows.

https://www.phoronix.com/scan.php?page=article&item=intel-win10-mesa121&num=4
https://www.phoronix.com/scan.php?page=article&item=windows-linux-pascal&num=5
https://www.phoronix.com/scan.php?page=article&item=amdgpu-win10-fury&num=6

This is why you don't have an ideological/emotional tie in to a computer operating system. It's why I dual boot my desktop Windows & Ubuntu and why I also have a Mac, I can just choose the best tool for the job and don't have any need to make silly, false generalizations to justify using one or another because sometimes the best choice is macOS, sometimes it's Linux and sometimes it's Windows.

https://www.phoronix.com/scan.php?page=article&item=intel-win10-mesa121&num=4
https://www.phoronix.com/scan.php?page=article&item=windows-linux-pascal&num=5
https://www.phoronix.com/scan.php?page=article&item=amdgpu-win10-fury&num=6

This is why you don't have an ideological/emotional tie in to a computer operating system. It's why I dual boot my desktop Windows & Ubuntu and why I also have a Mac, I can just choose the best tool for the job and don't have any need to make silly, false generalizations to justify using one or another because sometimes the best choice is macOS, sometimes it's Linux and sometimes it's Windows.

https://www.phoronix.com/scan.php?page=article&item=intel-win10-mesa121&num=4
https://www.phoronix.com/scan.php?page=article&item=windows-linux-pascal&num=5
https://www.phoronix.com/scan.php?page=article&item=amdgpu-win10-fury&num=6

This isn't about removing x86 32 bit support as far as I can see, it's about removing 32 bit support for 64-bit processors using the x86_64 branch, it's niche and only started appearing in 2012:

details

Your link may provide some links to interesting stories at Phoronix beyond what's posted in this Slashdot story: 1, 2, and 3. However, it would have been better to link directly to the Phoronix stories instead of your comments. Details are lacking in the Phoronix stories, such as why XArray was rejected for inclusion in the 4.19 branch. It is good to see improvements in the Linux kernel that result in cleaner and more secure code, like removing VLAs.

One criticism of Linux versus *BSD is that the *BSD kernel code is much cleaner and better written than many portions of the Linux kernel. Although it's good to see new features added to the Linux kernel, it's also very good to see efforts like this to clean up the existing code. For those who would say that *BSD code cleanliness comes at the expense of supporting newer hardware and features, there are also a lot more contributors (especially corporate contributors) to Linux than for the *BSDs.

By the way, this is the third time you've posted this comment. Even if your comment is on-topic, this is the third time you've posted the comment in this article. You have a tendency of doing this. Posting the same comment more than once makes any reports, by definition, redundant. The links are actually interesting, but you deserve your -1.

Your link may provide some links to interesting stories at Phoronix beyond what's posted in this Slashdot story: 1, 2, and 3. However, it would have been better to link directly to the Phoronix stories instead of your comments. Details are lacking in the Phoronix stories, such as why XArray was rejected for inclusion in the 4.19 branch. It is good to see improvements in the Linux kernel that result in cleaner and more secure code, like removing VLAs.

One criticism of Linux versus *BSD is that the *BSD kernel code is much cleaner and better written than many portions of the Linux kernel. Although it's good to see new features added to the Linux kernel, it's also very good to see efforts like this to clean up the existing code. For those who would say that *BSD code cleanliness comes at the expense of supporting newer hardware and features, there are also a lot more contributors (especially corporate contributors) to Linux than for the *BSDs.

By the way, this is the third time you've posted this comment. Even if your comment is on-topic, this is the third time you've posted the comment in this article. You have a tendency of doing this. Posting the same comment more than once makes any reports, by definition, redundant. The links are actually interesting, but you deserve your -1.

Your link may provide some links to interesting stories at Phoronix beyond what's posted in this Slashdot story: 1, 2, and 3. However, it would have been better to link directly to the Phoronix stories instead of your comments. Details are lacking in the Phoronix stories, such as why XArray was rejected for inclusion in the 4.19 branch. It is good to see improvements in the Linux kernel that result in cleaner and more secure code, like removing VLAs.

One criticism of Linux versus *BSD is that the *BSD kernel code is much cleaner and better written than many portions of the Linux kernel. Although it's good to see new features added to the Linux kernel, it's also very good to see efforts like this to clean up the existing code. For those who would say that *BSD code cleanliness comes at the expense of supporting newer hardware and features, there are also a lot more contributors (especially corporate contributors) to Linux than for the *BSDs.

By the way, this is the third time you've posted this comment. Even if your comment is on-topic, this is the third time you've posted the comment in this article. You have a tendency of doing this. Posting the same comment more than once makes any reports, by definition, redundant. The links are actually interesting, but you deserve your -1.

See subject & some VERY COOL 'optimizations' that hopefully make it into the kernel + usermode space like XArrays vs. RADIX trees or "Trie" https://www.phoronix.com/scan.... & spin lock micro-optimizations https://www.phoronix.com/scan.... AFTER I've seen VLA = GONE https://www.phoronix.com/scan.... + Restartable Sequences System Call Is Living Up To Its Performance Claims https://www.phoronix.com/scan....

* LITTLE THINGS mean a LOT (there's nothing BIGGER) - They start to add up into tangible speed/performance & stability benefits...

APK

P.S.=> Lastly, as an aside: I'm really, Really, REALLY starting to LIKE Mr. Larabel's site PHORONIX... apk

See subject & some VERY COOL 'optimizations' that hopefully make it into the kernel + usermode space like XArrays vs. RADIX trees or "Trie" https://www.phoronix.com/scan.... & spin lock micro-optimizations https://www.phoronix.com/scan.... AFTER I've seen VLA = GONE https://www.phoronix.com/scan.... + Restartable Sequences System Call Is Living Up To Its Performance Claims https://www.phoronix.com/scan....

* LITTLE THINGS mean a LOT (there's nothing BIGGER) - They start to add up into tangible speed/performance & stability benefits...

APK

P.S.=> Lastly, as an aside: I'm really, Really, REALLY starting to LIKE Mr. Larabel's site PHORONIX... apk

See subject & some VERY COOL 'optimizations' that hopefully make it into the kernel + usermode space like XArrays vs. RADIX trees or "Trie" https://www.phoronix.com/scan.... & spin lock micro-optimizations https://www.phoronix.com/scan.... AFTER I've seen VLA = GONE https://www.phoronix.com/scan.... + Restartable Sequences System Call Is Living Up To Its Performance Claims https://www.phoronix.com/scan....

* LITTLE THINGS mean a LOT (there's nothing BIGGER) - They start to add up into tangible speed/performance & stability benefits...

APK

P.S.=> Lastly, as an aside: I'm really, Really, REALLY starting to LIKE Mr. Larabel's site PHORONIX... apk

See subject & some VERY COOL 'optimizations' that hopefully make it into the kernel + usermode space like XArrays vs. RADIX trees or "Trie" https://www.phoronix.com/scan.... & spin lock micro-optimizations https://www.phoronix.com/scan.... AFTER I've seen VLA = GONE https://www.phoronix.com/scan.... + Restartable Sequences System Call Is Living Up To Its Performance Claims https://www.phoronix.com/scan....

* LITTLE THINGS mean a LOT (there's nothing BIGGER) - They start to add up into tangible speed/performance & stability benefits...

APK

P.S.=> Lastly, as an aside: I'm really, Really, REALLY starting to LIKE Mr. Larabel's site PHORONIX... apk

Github has security issues "Git Users Should Get To Updating Due To An Arbitrary Code Execution Vulnerability" https://www.phoronix.com/scan.... on top of being part of an infector in Kodi Apk pointed out.

This would have been a more helpful link.

2. Give up games and

In recent news : Valve is integrating Wine capability into their Steam linux client in order to handle exactly *this* specific problem.

Over time, the problem will get lesser.
(Valve indeed needs it, if they want SteamOS to be anything more than a glorified remote streamcasting device and to be instead worthy of a good SteamBox)

(And until then, my extremely subjective suggestion would be to try picking up an out-door hobby and/or a significant other : both could be healthier way to spend leisure time :-P )

a whole host of creativity / specialist software that isn't available for Linux.

Depends on your reliance on specific software.
For some users, a combo of VirtualBox and/or Wine might fill the gap to run *those few applications* while at the same time constricting the mess that is Microsoft Windows to a very small danger level.
(I am lucky enough that this happens to be my case. Might not be everyone's though)

Phoronix seems to have disregarded that part and published some benchmarks anyway. https://www.phoronix.com/scan....

Read the license. It does not restrict the publication of benchmarks. It restricts the OS vendors from publishing benchmarks directly. Not cool on Intel's part. But no where does the license prevent anyone from running benchmarks. That would be impossible to control and completely impossible to enforce.

Phoronix seems to have disregarded that part and published some benchmarks anyway. https://www.phoronix.com/scan....

Superior performance of Intel?

Here, watch this: 10 EPIC games with VULKAN support (PC 2018). There is a good reason why it is easier to name the game engines that do not support Vulkan than those that do. What reason? Better performance. Those engines that currently do not support Vulkan, do support DX12 or Metal, basically the same thing. So 100% of major engines are already moved to Vulkan or similar. For most of those, their Vulkan render path gives the best performance.

Now, somebody desperate to justify their pet theory that Vulkan is not the way forward for 3D rendering might point to some cases where Vulkan (or DX12) actually slows the game down. This is because the initial implementation of Vulkan is sometimes just a translation layer for OpenGL, which is single-threaded, so the translation is single-threaded too. On top of that you have the translation overhead. Result: no improvement, or actually a regression. However, that was then, this is now. (...we continue to see much lower CPU utilization when using Vulkan rather than OpenGL.)

Vulkan improves performance in several ways: 1) distributes rendering across multiple cores 2) lifts state out of the inner render loops 3) shortens the path for submitting render data 4) supports fine grained control over caching of render assets. (Not an exhaustive list.) None of this is a secret; all of it is known to anybody who could call themselves a game developer. You are very obviously not one of those. But do us all a favor and please stop shitting on the internet with your ignorance. If you are confused then do a little research.

Here, watch this: 10 EPIC games with VULKAN support (PC 2018). There is a good reason why it is easier to name the game engines that do not support Vulkan than those that do. What reason? Better performance. Those engines that currently do not support Vulkan, do support DX12 or Metal, basically the same thing. So 100% of major engines are already moved to Vulkan or similar. For most of those, their Vulkan render path gives the best performance.

Now, somebody desperate to justify their pet theory that Vulkan is not the way forward for 3D rendering might point to some cases where Vulkan (or DX12) actually slows the game down. This is because the initial implementation of Vulkan is sometimes just a translation layer for OpenGL, which is single-threaded, so the translation is single-threaded too. On top of that you have the translation overhead. Result: no improvement, or actually a regression. However, that was then, this is now. (...we continue to see much lower CPU utilization when using Vulkan rather than OpenGL.)

Vulkan improves performance in several ways: 1) distributes rendering across multiple cores 2) lifts state out of the inner render loops 3) shortens the path for submitting render data 4) supports fine grained control over caching of render assets. (Not an exhaustive list.) None of this is a secret; all of it is known to anybody who could call themselves a game developer. You are very obviously not one of those. But do us all a favor and please stop shitting on the internet with your ignorance. If you are confused then do a little research.

IIRC, Devs. were QUITE whining about the state of OpenGL on Apple's platforms.

Entirely Apple's doing, a transparently obvious of the whole sleazy plan.

Yeah, they had this plan in motion for YEARS. Riiiight.

Apparently, Apple's intentionally broken OpenGL support

you'd better wrap another layer on that tinfoil hat.

You do realize you are an idiot, right?

  will be fixed by MoltenVK. MoltenVK is the new reality that means no developer needs to waste resources on Metal to target Mac. Initial Vulkan Performance On macOS With Dota 2 Is Looking Very Good

For once we agree!

So everyone needs to just STFU.

IIRC, Devs. were QUITE whining about the state of OpenGL on Apple's platforms.

Entirely Apple's doing, a transparently obvious of the whole sleazy plan. Apparently, Apple's intentionally broken OpenGL support will be fixed by MoltenVK. MoltenVK is the new reality that means no developer needs to waste resources on Metal to target Mac. Initial Vulkan Performance On macOS With Dota 2 Is Looking Very Good

Like? I know There are a few graphics and video application than can bring the ILP needed to saturate an intel core, though hyper-threading tends to have less than a 1% penalty there. https://www.phoronix.com/scan..... The other kind of workload that doesn't benefit are can only utilize a few threads, in which case an i3 or i5 with 4 cores and an equal clock would perform just as well on the workload. The reason I bough the sylake i7 over the i5 was precisely hyperthreading to speed up compiling and the virtualization extensions.

Also, if your code is that performance-critical, an interpreted language is not a good idea.

Yup, try convincing ESR about that.

Even using your out of the ass numbers of 0.5% of computers (really using just Windows 7 sales (300 million), which is a smaller number), and you're talking 1.5 million computers. You don't think 1.5 million computers as part of a botnet would be a big thing? Or replacing 1.5 million computers early would be a big thing?

Most of those have already been replaced and sent off to the metal recyclers. Congratulations, you managed to offer even dumber numbers than the GP.

Of course, the real lesson here is to cut Windows products out of your environment as much as possible. Any company that believes it can just mandate substantial changes to your business so you can keep accepting patches to fix ITS bugs is a trainwreck.

"Linux Set To Shed Nearly 500k Lines Of Code By Dropping Old CPUs"
https://www.phoronix.com/scan....

Hyperthreading v. hypervisors is a really difficult and long topic to talk about. There's a lot of information and performance comparisons on the net and in the end it boils down to the type of work that you're doing.

https://medium.com/data-design...
https://medium.com/data-design...
https://medium.com/data-design...
https://www.phoronix.com/scan....
https://blogs.vmware.com/apps/...
https://blog.heroix.com/blog/s...

Also, last time I checked OpenBSD is not widely used as a virtualization platform.

Of course, the drivers will probably be binary blobs.

Are there any ARM vendors actually providing full driver sources? nVidia made noises about Tegra GPU sources (outright stating that Tegra was not encumbered like GeForce was, and they were capable of such a release) but... ah yes, thank you google. Wow, nVidia released Tegra driver sources in February. Alas, all Tegra SoCs are allegedly vulnerable to both MELTDOWN and SPECTRE... all three variants, too. Whee! nVidia has put out a bulletin about these vulnerabilities, but note that there is not a whisper about Tegra anywhere in that document.

Is it presumptive to assume that their attempt to pretend Tegra doesn't exist effectively means that Tegra no longer exists? RIP my TF201.

is a little bit too difficult to parse.

Here's a few human readable sources:

https://kernelnewbies.org/Linu...

https://www.phoronix.com/scan....

German: https://www.heise.de/ct/artike...

Russian: https://www.opennet.ru/opennew...

• List of things created by LP & KS
• PulseAudio/PolypAudio - Linux sound - to replace ALSA or OSS 4.0 - complete, took about 4 years to finish
• Avahi - Linux ZeroConf - complete (abandoned?)
• systemd - Linux init system - to replace Upstart, initd, and openrc - WIP
• journald - Linux logging system - replaces syslog-ng and rsyslog. Replaces text logs with new, improved binary format - WIP
• logind - Linux user authentication - replaced ConsoleKit because it was unmaintained - WIP
• networkd - Linux networking - replaces isc-dhcp-server, dhclient, dhcpcd, ifconfig, etc... because the previous code was too slow or unmaintained
• tempfiles - Linux temp file cleanup (this was needed?)
• timedated - Linux time & date functionality, including network time - because the code was 'old'
• udev - Linux device support. At one time a useful, standalone project, this as been absorbed into systemd. Cannot work without it.
• gummiboot/systemd-boot - Linux bootstrap support. Gummiboot will (eventually) replace Grub2/LILO/eLILO/PLoP and other boot managers. Supports secure boot (assuming kernel is signed with the appropriate key), and booting to encrypted LVM2 file systems
• nspawn - Linux chroot support. Replaces chroot - because it was insecure; perhaps idea taken from BSD jail
• machinectl - Linux user masquerading / privilege escalation utility - replaces su because "'su' is a broken concept". Perhaps he hadn't heard of sudo, then?
• systemd-firewall - Linux Firewall - to replace iptables, ufw, firewalld, etc...

In my experience, LP and KS are great at getting code out the door. Revisiting code to fix bugs - less so. I'm not the only person that thinks like this. Even if your log files are getting filled up with Avahi warning messages about IP address blah sent an invalid request Y, ignored' messages. Which, of course, is fixed in the next version, instead of patched in the current version. Therefore, this patch was not backported to CentOS 6.

things like Power4, Blackfin, M32R etc.

Which are all newer than the motorola 68000 in the powerbook 100.

"Newer" has nothing to do with "less obsolete". There is a link in TFA, describing eight dropped architectures as "without active users". POWER4 was dropped because POWER4 support is broken since two years already and no fix or rework available (and apparently no urge to fix it).

So yes, a newer architecture can be obsolete when an older one is still alive and kicking.

Very recent? That link is to 2012.

In any case your assertion is wrong.

It's also the proprietary NVidia driver, not the open source driver that most Linux users would be using in their dist.

The default nVidia driver in SteamOS is nvidia, not nouveau. Also nVidia does not produce the open source nouveau driver so when you say "NVidia / AMD push out frequent Windows driver updates that optimize performance or fix issues for specific games" obviously the driver in question is the proprietary nVidia driver and not the open source one and your assertion about that driver is wrong as they do indeed optimize performance and fix issues for specific games on Linux.

Not to mention that anybody concerned with gaming on Linux is going to be using the proprietary nVidia driver and not the nouveau driver because of its far superior performance.

Besides that, if your complaint is actually about the open source driver then there's certainly no reason that nouveau developers couldn't work closely with game developers to optimize their drivers for specific games.

I'm very much looking forward to installing this. I recently put together a nice Machine Learning / linux workstation / build machine.... https://pcpartpicker.com/list/... And Linux pre 4.14 just flubbed pretty bad with the processor... https://www.phoronix.com/scan.... I got things working somewhat smoothly with Manjaro linux, but getting the CUDA support working was a total hack (currently GCC 6.3 is all they support, not 6.4, much less 7.3 and the arch linux "fix" is very much an admitted dirty hack), and getting Caffe 2 to compile right was turning into more work than it was worth ... http://docs.nvidia.com/cuda/cu... ... So yes, these problems are largely the fault of Intel and NVIDIA clutching their proprietary pearls, but I am looking forward to running a well supported and stable version of linux that can support and be optimized to the latest hardware that came out 6 months ago.

My now-ancient ASUS G50VT included ExpressGate. Based on Splashtop, burned into the BIOS ROM, manageable. Rudimentary Firefox browser, email client, Skype, and obviously hard to update. But it ran independently of any OS installed on storage.

Splashtop is now done, but it was also used by ASUS on some motherboards, and then endured obscurity, competition, and finally turned into something else.

It did work. It was pretty minimal, and could have been cool. And it certainly is possible today, even in BIOS, with flexibility and update capabilities, but somehow I don't see any of this on the market.

The obvious solution would be to embed ChromeOS or something similar, fairly lightweight and useful. This could let you keep your primary OS invisible.

Cost?

Linux on Windows is an emulation layer

No, WSL is a native implementation of the Linux kernel interface. It is not an emulator.

The major performance issues that remain are with I/O. Otherwise it's actually pretty good, in some cases equal to or even slightly better than bare-metal Ubuntu performance.

https://www.phoronix.com/scan....

1) Summary indicates there are people who are annoyed. No actual links to annoyance.
2) Summary indicates it's quoting a blog post. No blog post linked, just the rules page.
3) The rules page has been around since 2015: https://archive.fo/https://www... - not that "new code of conduct" means that the writer intended to convey it was brand new, but certainly it will be interpreted that way by a lot of folks.
4) FreeBSD had some sort of discussion around it when it came to be in 2015: https://www.phoronix.com/scan.... and it looks like there was some actual internal stuff for project participants that occurred but again, nothing really happened
5) This type of code of conduct isn't really crazy in the OSS world by a simple search. For example, https://www.contributor-covena... shows a plethora of OSS projects that participate and is based on similar principles. Big names OSS participants include Eclipse, Spring, Atom,
6) Microsoft has code of conduct that touches on similar issues: https://opensource.microsoft.c...
7) Github has a guide actively encouraging codes of conduct within communities: https://opensource.guide/code-... and pointing to other OSS projects that have them: https://www.djangoproject.com/...

If you look at FreeBSD's code of conduct in context it really seems like they're late to the party, which may just be a formality (the community norms might already be enforcing these types of rules anyway) or a dramatic change, but there's no way to actually get that from the summary at all.

ATI has become a subsidiary of AMD, and their quality has dropped accordingly...

Unless you care about open source graphics in which case ati/amd has never been this good. And vulcan support is likely to be even better than opengl.
https://mesamatrix.net/
Meanwhile Nvidia makes running open source drivers harder.
https://www.phoronix.com/scan.php?page=news_item&px=Nouveau-XDC2017

Despite it's touted simplicity, Wayland lags behind X functionality in both network awareness and driver support, as well as still a slight lag in performance despite its purported closeness to the hardware compared to X. Am I misunderstanding something?

This is what happened after China acquired AMD license to produce x64 chips in China, and acquired VIA's x86 license which VIA got from acquiring Cyrix.

The CPU license pool is cracked opened. Soon CPUs in China will be 1/4 the price of Intel/AMD but has better performance.

https://www.reddit.com/r/hardw...

Zhaoxin launched KX-5000 quad/octa-core x86 processors on Dec 28, 2017 in Shanghai, China: image, report, translation.

Zhaoxin revealed KX-6000 & KX-7000 roadmap: image, report, translation.

Other reports: golem.de, pcgameshardware.de, bitsandchips.it, phoronix

KX-5000:

Full SOC design (integrated southbridge)
28nm process by HLMC, 2.1 billion transistors
4-core / 8-core SKUs, no SMT
2.0-2.2GHz base clock, 2.4GHz max turbo
IMC supports dual channel DDR4-2400
PCIe 3.0 lanes
iGPU
integrated audio codec
ZX-200 I/O extension (chipset): SATA3.0, USB 3.1 Gen2, Gigabit Ethernet
OEM: Lenovo desktop M6200

KX-6000: 16nm tick-tock

KX-7000: new uArch, DDR5, PCIe 4.0

Related info:

About VIA & Zhaoxin: wikipedia and wikichip.

KX-5000 preview: image, report

KaiXian KX-5000 series was listed in PCI-SIG integrators list on Nov 10, 2017.

Zhaoxin KaiXian KX-5640 in SiSoftware database.

Zhaoxin ZX-C, KX-5000 series on exhibition on Nov 21, 2017 in Ukraine: report, translation.

KX-5000 CPU arch: block diagram, report, translation.

Compilers are evolving to produce hard to exploit binaries.

This comes with a 20-30% tax for many workloads.

Except it didn't. All benchmarks point to a 5-10% worst case hit, and an unmeasurable hit in pretty much all desktop / user facing workloads. Despite all the initial reports I've yet to see any benchmark, Windows, Linux, server loads, office applications, gaming, databases, or whatever get into the double digits.

Here's just some top google results:
http://www.guru3d.com/articles...
https://www.techspot.com/artic...

And here's some Linux ones on KPTI:
https://www.phoronix.com/scan....

You can get passively cooled 1030s

https://www.phoronix.com/scan....

...at least when mainboard makers support the option in UEFI.

https://www.phoronix.com/scan....

Phoronix did a few benchmarks on games

Of course, each game is a different animal, but it looks like there may be reason to hope that games aren't heavily affected.

Since they've done another round, I'll point to the Phoronix dev's actual testing where they're showing 10-15% under actual benchmarks in configurations that look like more like production to me. I don't see ANYONE able to corroborate even the lower 30% numbers people are throwing around without using fully synthetic benchmarks that do nothing but make system calls, let alone the 60% I've seen used in a couple of places (and here!).

I got no clue what they're on about - that's not even what the Phoronix devs are saying - See HERE, where they're showing real world, non-synthetic stuff at between 10% and 20%. They're saying they can produce some really bad numbers with synthetic benchmarks on extremely fast SSDs, which is about the worst possible setup for this bug, but under more normal conditions they're not seeing a large hit. They're seeing nothing on some older systems with spinning disks where waiting for the disk covers the additional overhead.

I like how they've weaseled out of the whole fiasco (why didn't /. post a link to the original press release?):

"Contrary to some reports, any performance impacts are workload-dependent, and, for the average computer user, should not be significant and will be mitigated over time".

I'm not sure I can read between the lines properly but I guess new revisions of Coffee Lake/Kaby Lake/SkyLake(X) CPUs are coming and they will contain a hardware fix (though it still seems highly unlikely considering how difficult it's to deploy a new hardware design - however unlike other fabless companies, like AMD/NVIDIA/ARM/etc Intel has everything under control). After all they've known about this issue for almost half a year.

Meanwhile as for consumer workloads they are correct. Two German websites have already tested a Windows build with a fix and found very little performance losses.

Phoronix has also run a number of tests on Linux and found out that only few (mostly artificial) tasks are seriously affected.

Intel home users may sleep well. As for enterprise customers no one has run virtualization tests yet though - that's what truly important for large deployments (clouds).

From what I'm reading, it's cause the code is still in development so they basically have it turned on for everything. They plan on fixing that soon.

https://www.phoronix.com/scan....

https://www.phoronix.com/scan....

From what I'm reading, it's cause the code is still in development so they basically have it turned on for everything. They plan on fixing that soon.

https://www.phoronix.com/scan....

https://www.phoronix.com/scan....