Processador AMD ZEN4 (Ryzen 7000 series)

[topspeed]

Qual seria uma boa board entre os 150/200€, para levar um 7800x3d e aguentar um futuro upgrade daqui a uns aninhos? Tenho uma MSI de momento e estou satisfeito, mas quanto às b650 não sei bem. Parece-me que a ASROCK e a Asus estão melhores esta geração não?

Aqui tem mais MB:

Evita as que estão acima dos 70ºC.

 

[JPgod]

Zen2 já deu o que tinha para dar mesmo...

Por outro lado os 3D pouco justifica do ponto de vista de performance em vários casos, Por outro lado
- não precisa de rans muito rápidas
- Baixo consumo, logo pode optar por boards mais simples
- E obviamente não consumindo mais que 50-60w em gaming, qualquer cooler tower 120mm barato aguenta e sobra.

 

[Dark Kaeser]

Supostamos disponibilidade global amanhã

 

[w@rrior]

300 paus por um 8700F parece-me um pouco puxado, um cpu que nem gráfica integrada tem.

 

[JPgod]

Sim, é um CPU à evitar...

para gaming é ruim, so tem 16MB L3, aka um 7600 vai andar às voltas disso. Para trabalhar vai precisar de placa gráfica na mesma, enfim.

Se viesse sub 250 paus ainda poderia ser uma opção e usar um gpu antigo para basic gaming machine/entry point/ter um 8 cores decentes para trabalho.

 

[Dark Kaeser]

AMD Ryzen 5 8400F vs. Intel Core i5 14400F: 230+ Benchmarks For Sub-$200 CPU Performance​

For gaming, highly threaded content creation, and related scenarios of software being highly scalable but not taking advantage of AVX-512, the Core i5 14400F tended to be the front-runner. But for HPC-like workloads, AI, and other software able to leverage AVX-512, the Ryzen 5 8400F is the obvious choice. The Ryzen 5 8400F was also pulling ahead in various single-threaded workloads like audio encoding and web browser Chrome/Firefox benchmarks.

he Ryzen 5 8400F performance at least under Linux was many times better than the Core i5 14400F, the AVX-512 support helped a great number of workloads, this Zen 4 processor was far more power efficient. The mix of hybrid P/E cores on the Core i5 14400F also caused higher run-to-run variance due to task scheduling between the cores sometimes being less than optimal.

When looking at the CPU power consumption over the span of ~22 hours benchmarking each processor, the Ryzen 5 8400F on average had a 59 Watt CPU power consumption and recorded peak of 88 Watts. The Core i5 14400F meanwhile had a 64 Watt average and a peak of 104 Watts. So for overall power efficiency, the Ryzen 5 8400F was ahead of the Core i5 14400F.

https://www.phoronix.com/review/amd-ryzen-5-8400f

 

[Dark Kaeser]

The AMD EPYC 4004 is Finally Here and Intel Xeon E Needs an Overhaul​

Other execs at AMD are probably sick of me asking over the last six years for a Ryzen-based EPYC. Finally, I can stop asking as we now have the AMD EPYC 4004 series for entry-level servers. With this launch, something we have discussed in a few pieces over the years is coming true, Intel needs to re-think its Xeon E series since it is now getting walloped in the segment.

AMD uses the “P” to show these are single-socket only SKUs as it does with the rest of the EPYC range. The X signifies 3D V-cache. We wish there was a 8 core 3D V-cache SKU as the AMD Ryzen 7800X3D has been one of our favorite SKUs to use for Ryzen-based servers. That is a glaring omission in the SKU stack.

Given that we were doing this pre-launch, we only had two SKUs, an AMD EPYC 4344P 6-core part and an AMD EPYC 4564P part. Probably the most interesting for folks would be the AMD EPYC 4464P with 12-cores at 65W TDP, the 3D V-Cache parts, and the 8-core SKUs that match Intel’s top-end Intel Xeon E-2488G. Alas, we have a 16 core part, so instead we will take a look at the per-socket performance.

AMD and Intel are closer on a per-core basis at a similar TDP. The challenge is really this. AMD can scale to 16 cores and 32 threads in a socket whereas Intel tops out at 8C/16T. That makes performance just look like they are on a different scale. On one hand, this is an unfair comparison just based on the core count. On the other hand, the parts are $606 for the Xeon E-2488 and $699 for the EPYC 4564P so they are fairly close on price. The Xeon E may use less power for fewer compute resources, but AMD has a 12 core 65W TDP part that would likely change that narrative based on what we have seen with the Ryzen parts.

AMD now has a somewhat crazy portfolio of SKUs. That is saying a lot since the last time it launched a mainstream server CPU was over 18 months ago with AMD EPYC Genoa. Intel has launched two mainstream generations in that time and be moving onto a new platform before we get a Genoa refresh. Part ofthe reason for that is that EPYC has gone from being a one CPU offering in 2020 to five different flavors in 2024 with three different sockets. This is a big reason AMD is gaining market share in servers. It is taking common components and scaling them to different market segments.

https://www.servethehome.com/the-amd-epyc-4004-is-finally-here-and-intel-xeon-e-needs-an-overhaul/

AMD EPYC 4004 Benchmarks: Outperforming Intel Xeon E-2400 With Performance, Efficiency & Value​

AMD has now solidified its positioning for entry-level servers with the introduction of the EPYC 4004 series processors. The EPYC 4004 series is derived from the Ryzen 7000 series offerings to facilitate cost conscious server options and putting the Intel Xeon E-2400 series in the crosshairs. In this review is a look at the EPYC 4004 series along with benchmarks of nearly the entire EPYC 4004 product stack compared to Intel's current top-end Xeon E-2400 series processor, the Intel Xeon E-2488 Raptor Lake.

For this round of testing the EPYC 4244P (6 cores), EPYC 4344P (8 cores), EPYC 4364P (8 cores), EPYC 4464P (12 cores), EPYC 4484X (12 cores + 3D V-Cache), EPYC 4564P (16 cores), and EPYC 4584PX (16 cores + 3D V-Cache) were all benchmarked. I am still testing the EPYC 4124P 4-core processor that due to arriving late and already testing these other processors hadn't arrived in time for launch day. The EPYC 4124P will be part of its own interesting 4-core showdown soon. The AMD EPYC 4004 line-up was compared to the flagship Intel Xeon E-2488 8-core processor. That was the only Xeon E-2400 series processor I had available and with being their highest-end Xeon E processor right now shows the best possible Raptor Lake server performance.

When taking the geometric mean of all 360+ raw performance benchmark results, the Xeon E-2488 was similar to the 6-core EPYC 4244P processor. Ouch for Intel, considering the Xeon E-2488 is an 8-core processor with a list price of $606 USD. Meanwhile the EPYC 4244P has a $229 list price. The top-end EPYC 4564P/4584PX processors were delivering around 1.81x the performance of the Xeon E-2488 flagship while costing only 1.15x the price of that Xeon Raptor Lake CPU.

Next is the AC wall power consumption for these servers. The Xeon E-2488 server had a 138 Watt average and a peak of 338 Watts, which was higher than even the EPYC 4564P server configuration peaking at 301 Watts. Again, the EPYC 4584PX is the most interesting for delivering great 16-core performance, the 3D V-Cache helping out in some workloads, and the power efficiency boost over the EPYC 4564P.

https://www.phoronix.com/review/amd-epyc-4004