Marrying Vega and Zen: The AMD Ryzen 5 2400G Review
by Ian Cutress on February 12, 2018 9:00 AM ESTPower Consumption
For our power consumption metrics, we use a Prime 95 blend on fixed threads to generate a strong load, and then poll the internal power registers that determine power state calculations to get the power consumption. Each processor is different in how it reports its power, which depends on the level of control the processor has: some of the more advanced CPUs, such as Ryzen, will provide per-core power numbers, while the latest Intel CPUs only give a figure for the CPUs as a whole but also include DRAM controller and uncore power consumption.
An interesting element to the power consumption on the Ryzen APUs, due to the unified power delivery subsystem in play feeding the CPU and the integrated graphics, is that the power registers only report half the power consumption when probed (e.g. when 14W, shows 7W). As of yet, we are unsure if this has a knock-on effect on how the processor adjusts its turbo modes in response to power consumption. Nonetheless, a simple scaling factor gives the following results.
Total Package: The Whole Processor
For this data, we take the values of the processor as a whole, which includes all the interconnect, memory controllers, PCIe root complexes, etc. The system is still only loading the CPU cores with minimal effect on the rest of the system, however depending on how the power is managed, some of the sub-systems still remain enabled.
At full load, the difference between the Ryzen 5 and the other Ryzen CPUs shows that the 2400G is using more of its upper margin, compared to the 1400 which is rated at the same power (note TDP is only determined at the base frequency), but the extra frequency of the 2400G means that there is extra power draw overall. Part of this is due to the Infinity Fabric, which we will see below. But what these tests also underline is that in a quad-core configuration, the Intel CPUs are still very power efficient.
Cores Only: Pure Work
For the processors that split out the data, we can look at the power consumption of the cores on their own, without any of the sub-systems, like uncore, mesh, or infinity fabric. This usually paints a different picture to the package power.
For the core only power, the Ryzen 5 2400G uses less power than the Core i3-8350K, despite the situation being reversed when considering the whole package. This means that Infinity Fabric takes a lot of power here, and the ring bus solution that Intel uses benefits from being simpler, and Intel can push more power to its individual cores.
Facebook
Twitter
RSS
177 Comments
View All Comments
sonicmerlin - Tuesday, February 13, 2018 - link
Now if only AMD had a competent GPU arch. The APU performance could be given a huge boost with Nvidia’s techdr.denton - Thursday, February 15, 2018 - link
They do. It's called Vega. Very efficient in mid- to low range and compute, and if I'm not mistaken that's where the money is. Highend gaming is just wi**ie waving for us geeks.HStewart - Tuesday, February 13, 2018 - link
Check out performance of up and coming i8809G with Vega Graphics compare to Ryzen 7http://cpu.userbenchmark.com/Compare/Intel-Core-i7...
Keep in mine this is a mobile chips - this is new mobile chips is quite powerful - I thinking of actually getting one - only big concern is compatibility with Vega chip.
haplo602 - Wednesday, February 14, 2018 - link
the i8809G is a desktop chip, 100W TDP ....hansmuff - Tuesday, February 13, 2018 - link
Any idea where I could buy the MSI B350I Pro AC? I have searched every retailer I've ever bought from and can not find the damn thing. I'm hoping it can run a 2400G out of the box, at least to update to the newest BIOS.Dragonstongue - Tuesday, February 13, 2018 - link
they REALLY should not have cut back the L3 cache SO MUCH...beyond that, truly are amazing for what they are...they should have also made a higher TDP version such as 125-160w so they could cram more cpu cores or at very least a more substantial graphics portion and not limit dGPU access to 8x pci-e (from what I have read)Graphics cards and memory are anything but low cost.
2200 IMO is "fine" for what it is, the 2400 should have had at least 4mb l3 cache (or more) then there should have been "enthusiast end" with the higher TDP versions so they could more or less ensure someone trying to do it "on a budget" really would not have to worry about getting anything less than (current) RX 570-580 or 1060-1070 level.
many cpu over the years (especially when overclocked) had a 140+w TDP, they could have and should have made many steps for their Raven Ridge and not limit them so much..IMO...they could have even had a frankenstein like one that has a 6pin pci-e connector on it to feed more direct power to the chip instead of relying on the socket alone to provide all the power needed (at least more stable power)
AM4 socket has already been up to 8 core 16 thread, and TR what 16 core 32 thread says to me the "chip size" has much more room available internally to have a bigger cpu portion and/or a far larger GPU portion, now, if they go TR4 size, TR as it is already has 1/2 of it "not used" this means they could "double up" the vega cores in it to be a very much "enthusiast grade" APU, by skimping cost on the HBM memory and relying on the system memory IMO there is a vast amount of potential performance they can capture, not to mention, properly designed, the cooling does not really become an issue (has not in the past with massive TDP cpu afterall)
anyways..really is very amazing how much potency they managed to stuff into Raven Ridge, they IMO should not have "purposefully limited it" especially on the L3 cache amount, 2mb is very limiting as far as I am concerned especially when trying to feed 4 core 8 thread at 65w TDP alojng with the gpu portion.
Either they are asking a bit much for the 2400g or, they are asking enough they just need to "tweak" a bit more quickly to make sure it is not bottlenecking itself for the $ they want for it ^.^
either way, very well done....basically above Phenom II and into Core i7 level performance with 6870+ level graphics grunt using much less power...amazing job AMD...Keep it up.
SaturnusDK - Wednesday, February 14, 2018 - link
Well done AMD. Well done.Both these APUs are extremely attractive. The R5 just screams upgradable. You get a very capable 4 core / 8 thread CPU packaged with an entry level dGPU for less than the competition charges for the CPU (with abyssmal iGPU) alone. In the current market with astronomical, even comical, dGPU prices this is a clear winner for anyone wanting to build a powerful mid-tier system but doesn't have the means to fork out ridiculous cash for higher tier dGPU now.
The R3 scream HTPC or small gaming box. A good low end CPU paired with a bare bones but still decently performing iGPU. Add MB, RAM, PSU, and HDD/SSD and you're good to go. I imagine these will sell like hot cakes in markets with less overall GDP and in the brick'n'mortar retail market.
The question is now. Is Intel ever going to produce a decent iGPU for the low end market? They've had plenty of time to do so but before Ryzen, AMD APUs just wasn't that attractive. Now though, you really have to think hard for a reason to justify buying a low end Intel CPU at all.
yhselp - Wednesday, February 14, 2018 - link
"Now with the new Ryzen APUs, AMD has risen that low-end bar again."You had to do it. I understand. And thank you.
dr.denton - Thursday, February 15, 2018 - link
<3Hifihedgehog - Wednesday, February 14, 2018 - link
I have been doing some digging and found that although current-generation AM4 motherboards lack formal HDMI 2.0 certification, just like many HDMI 1.4 cables will pass an HDMI 2.0 signal seamlessly without a hitch, the same appears to be the case for these boards whose HDMI traces and connectors may indeed be agnostic to the differences, if any. Could you do a quick test to see if HDMI 2.0 signals work for the Raven Ridge APUs on the AM4 motherboards you have access to? For further reference on the topic, see this forum thread “Raven Ridge HDMI 2.0 Compatibility — AM4 Motherboard Test Request Megathread” at SmallFormFactor.