Make no mistake – the Ryzen 5 and up CPUs in the 3000 series aren't just incrementally improved – they are legends in the making.
This is a companion discussion topic for the original entry at https://level1techs.com/article/ryzen-3000-navi-megathread
7 Likes
Good luck on the stock price @PhaseLockedLoop @anon46267848 . I wonder what it will look like tomorrow.
Techspot/Hardware Unboxed managed to kill their 3900X with overclocking…
It's finally time to review AMD's new 3rd-gen Ryzen processors. The Ryzen 9 3900X is a 12-core, 24-thread processor with a massive 64MB L3 cache. It costs $500, placing it in direct competition with the Core i9-9900K. Then the Ryzen 7 3700X costs...
On an even more disappointing note, we somehow managed to end the life of our 3900X sample at this stage of the review. We don’t recall exactly what settings were applied, but we know we hadn’t manually adjusted voltages yet. We believe after testing the 4.3 GHz overclock with auto voltage, we increased the LLC to see what impact that had on temperatures and during our first CB20 pass the system crashed and reset, and never booted up again.
The CPU now gets stuck at code 07 after microcode. We tried running the chip on a different X570 board among other tests before sadly declaring it dead. AMD says no other reviews had managed to kill their 3900X so we’re just special or unlucky, your pick. AMD has sent us a replacement but in the meantime we were unable to test the 3900X on B450 boards, or include it in the IPC test with a few cores disabled in each chiplet.
1 Like
From just the early benchmarks, it appears the 3700x, which is $329, is only like 4.5% slower then a near $500 9900k. And a 3900x basically performs like a Xeon with its multicore and does great with streaming.
You can 1t oc a 3700x to match a 9900k at least mine can but pbo and auto oc were not (currently) to my liking. Check out the longish written article.
1 Like
Phoronix is reporting that newer distros like Manjaro, Fedora 31 and Ubuntu 19.04 will simply not boot Zen 2.
Yet oddly, 18.04.2 LTS boots just fine.
https://www.phoronix.com/scan.php?page=article&item=ryzen-3700x-3900x-linux&num=2
Should I pick up a 3900x or wait for 3950x reviews in september? My use case is light gaming, heavy handbrake usage.
the 3900x has all the cache the 3950x will have. The 3950x is formidable but on am4 I think there will be somewhat diminishing returns. $500 is a killer deal, too.
3 Likes
3950X is probably for full time streamers that want to use x264 on slow on the same system. (and video editors that need extra CPU)
3900X would probably be better for those at the casual level of streaming, or as a dedicated 2nd streaming system.
Any chance for a sudo lspci -vvv from somebody with a 5700 XT, I’m curious to see if it’s still plagued by lack of actual PCI reset support.
2 Likes
Argone
July 7, 2019, 1:43pm
13
What about handbrake uhd encoding/compression. With a 2700x an entire movie takes 12hrs to encode.
More cores helps there as well if you need to transcode.
1 Like
Argone
July 7, 2019, 1:45pm
15
All I play is runescape and much older games. Mainly I transcode.
So did I, see megathread and other post on the forum
Fedora 30 (and Pop!_OS 19.0 and probably any linux based on 5.x kernels) won’t boot on Ryzen 3000?! What the heck?
It’s true. AMD is aware of the issue and it is being worked on. In the slides from the E3 event, changes to virtualization and new instructions were mentioned. Not a lot is known about this yet, but I have been working on it as I have time.
In a pinch, you can boot older kernels like 4.14 or use a distro like Ubuntu 18.04.2 LTS or Debian 9.9; those seem to work fine.
What’s the error?
When booting Fedora 30, you get a lot of error messages from SystemD. Same with Pop 19.04. Pop is especially useful for easily manipulating kernel parameters like maxcpus and noacpi and disabling rdrand but I haven’t found what actually works. When I change the init line, I get a kernel panic and something about offline CPUs. Last time I saw that error, it was just a bad ACPI table for Linux but… things are different now.
Here’s the messed up thing
So if you install Windows on bare metal, which is fine, then install VirtualBox, then try to boot from Fedora 30 it dies exactly the same way In a VM.
What? How on earth is that possible?
Virtualbox is using the SVM/hardware extensions. I am not sure why this would be. Does it make sense to you?
Disabling NPT support in Virtualbox will allow the system to boot normally.
Any known work arounds?
I installed Debian 9.9 so I could experiment. It worked fine. I updated my kernel to 5.1.14 and that worked fine. When I rolled back to 5.0.9 it still worked fine BUT only for the first boot (!?!?!?!?!).
It seems like to me that the system was left in a state by the newer kernel (or the older kernel!?) that would allow it to boot. I am not sure what this means, but it doesn’t seem good.
AMD is aware of the issue and investigating. I am really hoping this is a quick kernel tweak or firmware update. I will be surprised if it is a kernel bug, unless it is one that AMD introduced themselves a while back a la the encrypted-virtual-memo…
Yeah, 'tis a shame that the bleeding edge gets shut out this time.
They want to ensure the AM4 socket actually backs it’s word about upgradability.
1 Like
Here’s something that Looking Glass users might not like about Ryzen 3000.
Write bandwidth in AIDA64 is HALF of that of typical processors.