AMD Ryzen 9 3950X: the 16 cores will arrive for the enthusiastic platform

The race of the cores in the enthusiastic platform range will reach its peak with the announced and new Ryzen 9 3950X, a processor that will house all the features of the two chiplets in full and that contrary to what you might think, will be the fastest in the range.

The slide is very clear about what we can expect from this new CPU, as it will arrive with 16 cores and 32 threads at a base frequency of 3.5 GHz and a Boost of 4.7 GHz. It will have an amazing 72 MB of total cache (adding both L1, L2 and L3) and a TDP of 105 watts.

As we already know, this TDP is measured from the base frequency, where AMD has provided this CPU with only 3.5 GHz, being the slowest of the six microprocessors presented, but where it manages to maintain the 105 watts of 3800X and 3900X.

The fastest standard CPU in the entire Ryzen 3000 range

Remember that the AM4 platform, according to AMD, can accommodate up to 140 watts in its standard configuration, so questions have already arisen in some sectors will all manufacturers support this CPU in all models of their boards? Will this affect a supposed throttling in more basic models?

The doubts are logical, especially considering how they have put aside the A320 chipset with dubious arguments. We will see if finally the support is maintained or includes some restriction, although many rumors state that it could arrive through the 7 nm+, so the consumption should be more measured than the Ryzen 3000 at 7 nm, allowing its adaptation and support as one more CPU.

At the moment there are no confirmed dates or prices.

AMD RX 5700 XT: confirmed features

Like the slide of the Ryzen 9 3950X, the one presented gets the same validity and confirmation as the first one, where AMD has surprised with the data shown.

And it is that we speak of 40 CU with a base frequency of 1605 MHz, that will ascend up to 1755 MHz in its Game Clock mode and that finally will have a Boost of 1905 MHz, reason why in this section it seems to have been a step ahead of those of Lisa Su.

If the numbers do not fail and the hybrid RDNA architecture maintains the same proportion, we would be talking about 2560 Shaders, which would give about 9.75 TFLOPS of performance in FP32.

Although this is not a totally definite measure to assess its performance and much less with a change of architecture in between, this data would place it behind a RX Vega 56, which does not fit with what is shown and said by AMD in terms of performance.

Performance per key watt in this new series

Specifically, Lisa Su announced that it was 10% faster in the game shown than the RTX 2070, where assuming that in general terms it matches, we are talking about a difference of 25-30% against the RX Vega 56.

The explanation can come from the confirmed data, where AMD revealed that RDNA would be 1.25X faster per clock, so for the first time in a long time the data in TFLOPS would not indicate a performance close but lower than the real.

Finally, it confirms what we already filtered: it will use 8 GB of GDDR6, so in this aspect it will be equal to NVIDIA and as we saw in its PCB it will arrive with 256 bits of bus, where AMD will also return to the turbine dissipater, something that some users are not liking due to the reputation they have under their GPUs.

	Extensive Product Selection ● Over 300,000 products ● 20 different categories ● 15 local warehosues ● Multiple top brands		Convenient Payment ● Global payment options: Visa, MasterCard, American Express ● PayPal, Western Union and bank transfer are accepted ● Boleto Bancario via Ebanx (for Brazil)
	Prompt Shipping ● Unregistered air mail ● Registered air mail ● Priority line ● Expedited shipping		Dedicated After-sales Service ● 45 day money back guarantee ● 365 day free repair warranty ● 7 day Dead on Arrival guarantee (DOA)