AMD Announces Ryzen 6000 Mobile CPUs for Laptops: Zen3+ on 6nm with RDNA2 Graphics (anandtech.com) 20
AnandTech: The notebook market is a tough nut to crack with a single solution. People want that mix of high performance at the top, cost effectiveness at the bottom, and throughout there has to be efficiency, utility, and function. On the back of a successful ramp last year, AMD is striking the notebook market hot again in 2022 with the launch of its new Ryzen 6000 Mobile processors. These 'Rembrandt' APUs feature AMD's latest RDNA2 graphics, up to eight Zen3+ cores with enhanced power management features, and it uses TSMC's N6 manufacturing process for performance and efficiency improvements. Yesterday AMD disclosed that they would be launching the new Ryzen 6000 Mobile series today -- updated cores, better graphics, more features, all in a single monolithic package a little over 200 mm2.
There will be 10 new processors, ranging from the traditional portable 15 W and 28 W hardware, up to 35 W and 45 W plus for the high-end gaming machines. AMD is expecting 200+ premium systems in the market with Ryzen Mobile in 2022. At the heart of the design is AMD's Zen 3+ core, which affords an improvement in power management between the cores, but keeps the Zen 3 performance characteristics. The focus here is mainly to improve idle power consumption and power when using accelerators, to help extend the life of ultraportable devices -- AMD is claiming 15-40% lower power between web browsing and video streaming. There is a frequency uplift as well, with the top processors going up to 5.0 GHz. AMD is claiming up to 1.3x single thread performance for the Ryzen 7 6800U.
There will be 10 new processors, ranging from the traditional portable 15 W and 28 W hardware, up to 35 W and 45 W plus for the high-end gaming machines. AMD is expecting 200+ premium systems in the market with Ryzen Mobile in 2022. At the heart of the design is AMD's Zen 3+ core, which affords an improvement in power management between the cores, but keeps the Zen 3 performance characteristics. The focus here is mainly to improve idle power consumption and power when using accelerators, to help extend the life of ultraportable devices -- AMD is claiming 15-40% lower power between web browsing and video streaming. There is a frequency uplift as well, with the top processors going up to 5.0 GHz. AMD is claiming up to 1.3x single thread performance for the Ryzen 7 6800U.
Desktop APUs? (Score:3)
Considering the current state of GPU market, getting a beefy APU for desktop would have been nice.
Mobile APUs are nice and all, but laptop discrete GPUs are actually decently available and their pricing isn't mining-crazed like desktop GPUs.
Re:Desktop APUs? (Score:4, Informative)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Nobody gives a fuck what "architecture" it is
Re: (Score:2)
8 core GCN based CPU on desktops. Those are fundamentally 8 year old GPU cores that are utter garbage by modern standards, even before the memory constraints.
RDNA2 cores are way bigger, so they're usually only using 2 on APUs and those are only really available on mobile.
Re: (Score:2)
TFA doesn't say if it supports USB 4 and Thunderbolt. Hopefully the answer is that it does.
Re: (Score:1)
TFA doesn't say if it supports USB 4 and Thunderbolt. Hopefully the answer is that it does.
The new Rembrandt CPUs will also feature updated memory controllers, with support up to DDR5-5200 and LPDDR5-6400 – it’s worth noting that there doesn’t seem to be DDR4 support here, thus making a clean cut to the DDR5 standards. (Ian: that might mean increased cost during the first few months of these products.) AMD has also increased from PCIe 3.0 to PCIe 4.0, supporting 8x for a discrete GPU and 12x split between NVMe, SATA, and chipset. There is also native support for USB4, which all
Re: (Score:2)
And here I'm just waiting for the next gen Threadripper...
Finally something to compare M1 against (Score:2)
It's still a node behind, but at least it's getting close to parity. The 6800U will likely be around the size of M1 taking into account the node difference.
Watt for watt M1 will almost certainly still be faster for single core across the clock range, though 8 cores plus SMT has some upper range advantages compared to 4+4 big little.
Re: (Score:2)
It's not, they are basically the same node. Just like DBZ power levels, nodes are bullshit.
Re: Finally something to compare M1 against (Score:2)
Up to 5 vs up to 15 EUV layers, at the very least N6 is much cheaper.
waiting for full speed pci-4 (nvme) with gfx (Score:2)
I'm using a ryzen 5 5600g and the desktop gfx are quite good (simple videos, code devel use for dual screen, moving windows fast even over vnc/network).
recently blew a motherboard and bought a new one, with B550 chip. nice silent chipset (no fan) and does support pci4.
only problem: the gfx versions of the cpus have limited pci4 speed ;( so even if I buy pci4 nvme and do all the right things, it wont ever run faster than v3 speeds.
standalone gfx on regular 5000 series cpu, yes, you can get pci4 nvme speeds
Re: (Score:2)
Re: (Score:3)
gigabyte b550m ds3h
seems a lower cost version meant for 'overseas market' (I guess ble and wifi are not on the mobo so locals can get their own with possibly different rf characteristics?). rarely do I see wifi/ble removed from boards.
the X version does have a fan but I see no value in the X chipset. the b550 is decent enough for now and yes, its fanless.
ryzen 5600g is tdp of 65w and so that's also fanless territory with a heatpipe HTPC case.
I've had a fanless i7 that dates back at least 6 years ago, with
Re:waiting for full speed pci-4 (nvme) with gfx (Score:5, Informative)
Once you understand this, you realize that fast sequential speeds are pretty meaningless (unless you're regularly working with huge files, like copying large videos back and forth). They're already so fast that the time to read or write a large file is over before you can notice it. Consider reading a 250 MB file.
Notice how every time the MB/s figure doubles, the time saved is half the previous jump. In other words, the bigger MB/s becomes, the less difference it makes in wait time. Unless you're regularly working with multi-GB files, you're not gonna notice the difference between 4 GB/s and 8 GB/s. Both are going to virtually be indistinguishable from instantaneous. Heck, the vast majority of the time people can't even tell the difference between a SATA and NVMe SSD [youtu.be] in their system.
The more mathematically astute among you may also recognize that this is a converging series. So the 1 sec of time you save in the first halving - switching from a HDD to a SATA 2 SSD - is equal to the time you'd save by switching from a SATA 2 SSD to instantaneously fast storage. i.e. Even if you switched from the SATA 2 SSD to something with a million TB/s speeds, it will never reduce the wait time by more than the 1 sec you saved switching from a HDD to SATA 2.
So the fastest speed (sequential read/writes) doesn't really matter. What actually matters is making sure the slowest speed is fast. For SSDs, that's 4k (small file) read/writes. Imagine you're giving a task which requires reading 1 GB of sequential data, and 200 MB of 4k data. Which do you think will complete it faster? a NVMe SSD with 4 GB/s sequential speeds and 30 MB/s 4k speeds, or a SATA SSD with 500 MB/s sequential speeds and 45 MB/s 4k speeds? Well, there's 5x as much sequential data to read, and the NVMe drive is 8x faster at sequential reads, while only being 1.5 slower at 4k reads. So obviously the NVMe SSD will be faster, right?
Surprise! the SATA SSD is faster. That's because the bigger MB/s becomes, the less difference it makes. You'll notice that despite there being only 1/5 as much 4k data, both drives spent more time (a lot more for the NVMe SSD) working on the 4k data. That's because (repeat one more time), the bigger MB/s becomes, the less difference it makes. And it's the small MB/s operations which consume the most time.
tl;dr - If you want a fast SSD, the #1 spec you should be looking at are its 4k speeds. You want to make its slowest operation as fast as possible, which for SSDs are the 4k speeds. The sequential speeds only matter if you're gonna work with lots of big files. If you're only gonna sling a large video around about once a month, you can just ignore the sequential speeds entirely.
Re: (Score:2)
I cant think of a single thing a spinning disk does faster than even the slowest first gen nvme pcie drive.
[biden] "come on, man!" [/biden]
Re: (Score:2)
I remember the speed improvement moving from an older ATA-33 hard disk to the larger cache, ATA-66, 80-lines ribbon connector. Copying single MP3 files was so fast that it didn't even "invoke" the copying files dialog in Windows (copying two mp3 files did show it though).