Monday, July 1st 2019
AMD Ryzen 9 3950X Cinebench R15 Performance Spied
Market availability of the 16-core Ryzen 9 3950X may be far away, given its September 2019 launch, but engineering samples (ESes) of the chip seem to be already in circulation. "uzzi38" on Twitter posted this spy-shot of a 3950X ES making short work of Cinebench R15. CPU-Z recognizes the chip by its codename "Matisse," and puts out the correct CPU core and thread count, but doesn't give a name-string. It also recognizes the MSI MEG X570 GODLIKE motherboard this test is run on.
The purported Ryzen 9 3950X ES, overclocked to 5.42 GHz, scores a gargantuan 5,501 points in the multi-threaded benchmark. To put this number into perspective, at stock frequencies, a Ryzen Threadripper 2950X (same core-count, double the memory bus width), scores 3,645 points. The 3950X benefits from not just its massive overclock that's over 1 GHz higher than the stock TR-2950X, but also higher IPC, and a more consolidated memory interface. This feat goes to show that AMD's upcoming Ryzen chips love to overclock, and deliver a significantly higher single-thread performance over the previous generation.
Source:
uzzi38 (Twitter)
The purported Ryzen 9 3950X ES, overclocked to 5.42 GHz, scores a gargantuan 5,501 points in the multi-threaded benchmark. To put this number into perspective, at stock frequencies, a Ryzen Threadripper 2950X (same core-count, double the memory bus width), scores 3,645 points. The 3950X benefits from not just its massive overclock that's over 1 GHz higher than the stock TR-2950X, but also higher IPC, and a more consolidated memory interface. This feat goes to show that AMD's upcoming Ryzen chips love to overclock, and deliver a significantly higher single-thread performance over the previous generation.
38 Comments on AMD Ryzen 9 3950X Cinebench R15 Performance Spied
It's also been floating around that TSMC is using a new formulation of elements in the process that reduces resistance in the circuit pathways, which would have the direct effect of reducing the heat generated by component electrical resistance. This is a problem that Intel and Global Foundries have been unable to solve. It's not that either one can't go down to 7nm, it's that the chemistry formulations they're using can not withstand the heat naturally generated by the elements within the circuit components at the clock speeds needed to compete.
I know what a 3900x is capable of under ambient water....weve seen other leaks too using extreme cooling and similar voltage and clocks.
Your guess, what would you think that "extreme cooling" might be? LN2? Refrigerated liquid/water?
LN2 is what the articles say.
Either way, are you doing a review on your site?
1.5V or so is likely be 360mm+ water...but maxed out.
In my early days, I used ln2 on my daily drivers before I built out test benches. They are plenty valid after ln2 sessions. Surely they can die and more likely to during an ln2 sesh, but
Edit: I need to correct my info... the 4.4ghz clock I know was 1.35v under water and about tapped out thermally. That may be chilled water at 1.55v? Not sure.