Instale o Steam
iniciar sessão
|
idioma
简体中文 (Chinês simplificado)
繁體中文 (Chinês tradicional)
日本語 (Japonês)
한국어 (Coreano)
ไทย (Tailandês)
Български (Búlgaro)
Čeština (Tcheco)
Dansk (Dinamarquês)
Deutsch (Alemão)
English (Inglês)
Español-España (Espanhol — Espanha)
Español-Latinoamérica (Espanhol — América Latina)
Ελληνικά (Grego)
Français (Francês)
Italiano (Italiano)
Bahasa Indonesia (Indonésio)
Magyar (Húngaro)
Nederlands (Holandês)
Norsk (Norueguês)
Polski (Polonês)
Português (Portugal)
Română (Romeno)
Русский (Russo)
Suomi (Finlandês)
Svenska (Sueco)
Türkçe (Turco)
Tiếng Việt (Vietnamita)
Українська (Ucraniano)
Relatar um problema com a tradução
Regras e diretrizes para discussões
Denunciar esta mensagem
Yeah, you're right. I incorrectly read that it was unlocked, thanks for clearing that up.
Still doesn't change the fact that XFR2 makes the 2600 a better chip overall for only 10$ more, and the rest of the system would still cost less as well.
Zen 2 is expected to bring superior IPC, so fingers crossed, because it'll force Intel to make actual progress for once instead of releasing side-step upgrades.
Likely not getting DDR5 until after 2020.
As it is a FPU.
Silly me.
Because what should bother for the OP is the synthetic test with games, not the "theoretical" performance of a reduced instruction set. If you want to test FPU perf, go on but test all the FPU instructions, not just a single set (and as I said, on real tests there isn't that big distance between performance for floating point operations).
And following your premise, what you are getting on "the papers" it just cannot be visualized on the real tests, but that doesn't bother you at all for some reason.
And, btw, you're very wrong about real tests. All depend on tests. You just do not know tests where Zen sucks.
It's irrelevant either way because in all real-world applications, Zen is just fine, and superior for the price compared to similar graded builds from Intel. That's all that actually matters.
Seriously? You're going to bring supercomputers and datacentre level systems into a discussion about CONSUMER-GRADE hardware? That has no relevance here because there is a big difference between the two, and you're never going to find even a 9900K in a datacenter or supercomputer because the hardware is very different.
Also, when people talk about real-world applications, they're talking about consumers using consumer-grade hardware in applications typically used daily, and gaming, not datacenters and supercomputers and the workloads those are constantly running, nor does it have any relevance to the average user who just wants a basic PC, gaming system, workstation, or streaming system. You're just spouting information that doesn't even matter to the average user and always disputing people when they say that the 2600 is the best value CPU on the market at the moment, which it is. Even LinusTechTips and Jayztwocents covered this, and the data shows that AMD Ryzen has a better price-to-performance value in the low to mid range spectrum, with Intel only coming out on top in the very high end when you consider performance above budget entirely. Even in HPDTs with Threadripper, AMD holds strong with their performance value for the cost.
It's not rocket science unless you're rich and don't care about spending way more than you need to. Ryzen is enough for anyone from the average user to a higher end user; nobody needs to spend more on Intel when the performance differences are close enough that you won't actually notice a significant difference. That's why people chose the R5 2600 over Intel's 8th gen i5s, it was less costly up until the 9400F released in January, and even then, you can't overclock the 9400F at all while the 2600 has XFR2, PB, and potentially PBO which raises voltage limits if your motherboard supports it, allowing automatic overclocking features to push the 2600 beyond it's turbo clock of 3.9 GHz, and easily towards 4.1~4.2, allowing it to close the gap in gaming performance between itself and the 9400F, which puts even less value in Intel's favor with that CPU. It also has SMT, so the 12 threads of the R5 2600 and 2600X make it more valuable for streaming and more multi-threaded workloads that would gain benefit from more than 6 threads that the 9400F offers.
If Intel works best for the type of work you do, that's fine. You can recommend Intel based on that. But using market share against the 2600 is a weak argument.
https://www.forbes.com/sites/antonyleather/2019/04/09/amds-ryzen-5-2600-outsells-every-other-desktop-cpu-as-intel-loses-market-share/#7a13e08b356b
I do think the 9400f has very good performance at a reasonable price so long as you don't plan on streaming. In gaming it beats the 2600 in a lot of benchmarks.
But I would still wait to see the benchmarks for Zen2 before making a purchase.
Also have to consider XFR2 and pricing of a Ryzen 5 2600 build vs an Intel i5-9400F build. Generally, Ryzen is cheaper and of higher cost value.
Complete platform cost and performance between AMD and Intel for gaming is very competitive. It's kinda a wash. 3000 series won't be worse though so .. That's what you should focus on for new stuff (of course you could buy the 2600(X) system used if someone upgrade.)
I assume you can live with the i7 2600 for the small time which is left.
Ok, you take whatever I say in the way you want... Can we agree that your theoretical measure you calculated is only based on reduced instruction set? Do you understand that when I say "FPU instruction" is just a FP operation?
Regarding test, all what I can say is that for the same price (and sometimes less) Ryzen has a good value. If you want to argue that for scientific operations the Intel CPU has advantage, go on with that but understand that in a context of a Steam discussion it has no sense. I mean, is the same that I say that Ryzen is better just because it beats Intel for video compression without even asking the end user if he will be running such tasks on his cpu... give me a break.
https://www.msi.com/news/detail/4c78f7b58f4de12ed2cab9bcb9ec0ba0
Do you realize that if one of the manufacturers declares that certain models of its products are not subject to this defect, that means that it acknowledges that the defect really exist?