Rocksolid Light

Welcome to RetroBBS

mail  files  register  newsreader  groups  login

Message-ID:  

"We don't have to protect the environment -- the Second Coming is at hand." -- James Watt


computers / alt.comp.hardware.pc-homebuilt / Re: AMD Ryzen 7950X / AM5 Socket / X670E Motherboards, first impression based on reviewer/youtube data.

SubjectAuthor
o Re: AMD Ryzen 7950X / AM5 Socket / X670E Motherboards, firstPaul

1
Re: AMD Ryzen 7950X / AM5 Socket / X670E Motherboards, first impression based on reviewer/youtube data.

<thbdjr$1g82$1@gioia.aioe.org>

  copy mid

https://www.rocksolidbbs.com/computers/article-flat.php?id=919&group=alt.comp.hardware.pc-homebuilt#919

  copy link   Newsgroups: alt.comp.hardware.pc-homebuilt
Path: i2pn2.org!i2pn.org!aioe.org!YpWFAxXpXJOozLnmFNLutA.user.46.165.242.75.POSTED!not-for-mail
From: nospam@needed.invalid (Paul)
Newsgroups: alt.comp.hardware.pc-homebuilt
Subject: Re: AMD Ryzen 7950X / AM5 Socket / X670E Motherboards, first
impression based on reviewer/youtube data.
Date: Sun, 2 Oct 2022 03:11:56 -0400
Organization: Aioe.org NNTP Server
Message-ID: <thbdjr$1g82$1@gioia.aioe.org>
References: <67b778e3-290d-4310-92ed-f91ef378e608n@googlegroups.com>
<a5259b45-3cd4-446a-a973-7c38d2fcb81en@googlegroups.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 7bit
Injection-Info: gioia.aioe.org; logging-data="49410"; posting-host="YpWFAxXpXJOozLnmFNLutA.user.gioia.aioe.org"; mail-complaints-to="abuse@aioe.org";
User-Agent: Ratcatcher/2.0.0.25 (Windows/20130802)
Content-Language: en-US
X-Notice: Filtered by postfilter v. 0.9.2
 by: Paul - Sun, 2 Oct 2022 07:11 UTC

On 10/1/2022 11:56 PM, Skybuck Flying wrote:
> One thing not mention yet, cause I kinda got interrupted by the call of nature shall we say is:
>
> X. Micro stutter ? The top model, 7950x has slightly different core clock speeds per compute complex, it has two of them. 100 Megahertz difference.
>
> Not sure if this will affect the smoothness of gaming...
>
> Bye,
> Skybuck.
>

There has been timer handoff software in OSes for quite
some time now. It is intended to handle unlocked cores
running at independent speeds.

Usually, the BIOS will have a setting to lock cores,
which may reduce performance a bit. If you want to run
locked, there is usually a BIOS setting for it. For example,
you could turn off Turbo, and lock cores. That would be
used, if you had any concern about the problem you describe.

It does not normally harm process handoff by the scheduler.
If SuperPI moves from Core 1 to Core 3, the only impact of
doing that, is some cache coherency traffic for anything that
might need it in the cache. In a CCX, the cache coherency
traffic is on internal wiring. And at pretty high speed too.
If SuperPI is "counting the seconds of execution", it
does not gain or lose any seconds, as it moves from core
to core.

How time pieces in an OS work, that can be described in
some detail on VirtualBox or VMWare web pages. The people
who write hosting software, are quite proud of the knowledge
of the time pieces in the OS, and they can explain all these
details to your satisfaction.

And the hosting softwares, didn't get there on the first day.
At first, the Linux boot sequence was reporting all sorts
of bad stuff related to clocks. But that's been fixed over
the years. The state of VirtualBox today, the overall experience,
is not as good as it used to be. I have to switch to VirtualBox 5
more often than I would like to. As VirtualBox 6 has issues
(trouble installing Windows XP when checking Windows XP issues).

Paul

1
server_pubkey.txt

rocksolid light 0.9.81
clearnet tor