• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

AMD launches Zen+ 12nm Ryzen and X470 motherboards

Page 8 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
How are these guys even publishing stuffs like this? They got their sample from someone else so no NDA?
I understand the need to get the review out, but it such a weird thing, publishing data but withholding something because of not wanting to show everything but already disclosing anyhow. Then using a Low end board for a high end part? That too in print when you want to get first coverage ?
Also that choice of games...
I cannot fully understand this Canard PC.
They're veterans in this industry, they've got contacts to get hardware early. Sometimes obscenely early, like when they had a sample of Sledgehammer a year early.

They didn't release a lot because they respect other publications that ARE under NDA. They're leaving much to be investigated for them when NDA lifts.
 
Some info about the CanardPC article from this thread.


Latencies should be really good, especially if these chips support 3466+ Mhz more widely. Precision Boost 2.0 also seems very nice (especially when considering that the benches were run on a A320 chipset:
hfvtPmC.jpg

Hmm wasn't ryzen max turbo xfr using 2 core ?
 
Some info about the CanardPC article from this thread

So, you increase tdp by 10% and you use a process that's supposed to bring 10% improvement, and you wind up delivering < 10% on both the top and the bottom? Huh. Well, clearly not where I would have put the bar for release, but it's nice that new Ryzen buyers will get better ddr4 support, lower access latencies, and a boost mode that works properly. Unclear why they didn't let this one bake longer, though.
 
So with Precision Boost Overdrive and XFR 2.0 Enhanced we could see 4.5 GHz SC boost, when not thermal constrained?
I would like to know this too.

(I have a screen-grab of HWMonitor, after playing Tekken 7 PC for an hour, that shows my Core Speed Max at over 4.50Ghz for all four cores, on my 2400G. Base clock multiplier was increased to 39.0 multiplier. Didn't change any of the specific CPU settings, so maybe "Turbo" of whatever flavor AMD calls it was still active?)
 
"Reliability" whatever it means in practice.

AMD, nVidia and most likely Intel as well implement reliability calculations in their boost algorithms.
This allows them to reach the highest possible boost speeds in light loads. The silicon is able to withstand significantly higher voltage in low load conditions (i.e. partial utilization, low current and temperature).
If the same voltage levels used at the highest boost speeds would be applied during the full load as well, there would most likely be permanent damage to the silicon.
 
They're veterans in this industry, they've got contacts to get hardware early. Sometimes obscenely early, like when they had a sample of Sledgehammer a year early.

They didn't release a lot because they respect other publications that ARE under NDA. They're leaving much to be investigated for them when NDA lifts.

I seriously doubt this. AMD would not give 1 company a sneak peak without giving them all a sneak peak. Also, AMD would have provided an X470 motherboard and a complete kit for testing. Either they pieced things together from the leaked benchmarks or they managed to get their hands on an engineering sample. We'll find out soon enough. The actual NDA isn't until April.
 
I seriously doubt this. AMD would not give 1 company a sneak peak without giving them all a sneak peak. Also, AMD would have provided an X470 motherboard and a complete kit for testing. Either they pieced things together from the leaked benchmarks or they managed to get their hands on an engineering sample. We'll find out soon enough. The actual NDA isn't until April.

Huh, quote CatMerc "they've got contacts in the industry" - no word about AMD handing out samples. computerbase.de also wrote they got CPUs but not from AMD. Seems CPUs leak, but different news outlets have different standards for respecting NDAs.
 
I would like to know this too.

(I have a screen-grab of HWMonitor, after playing Tekken 7 PC for an hour, that shows my Core Speed Max at over 4.50Ghz for all four cores, on my 2400G. Base clock multiplier was increased to 39.0 multiplier. Didn't change any of the specific CPU settings, so maybe "Turbo" of whatever flavor AMD calls it was still active?)

That is almost certainly the infamous "sleep bug" which came back after Raven Ridge was introduced.
https://www.youtube.com/watch?v=nTdgB63xGX8
 
Huh, quote CatMerc "they've got contacts in the industry" - no word about AMD handing out samples. computerbase.de also wrote they got CPUs but not from AMD. Seems CPUs leak, but different news outlets have different standards for respecting NDAs.

Also, seems CB had something to say on the Canard PC for their article

In their opinion they said CPC wouldn't have had the final BIOS and the 2700X would have been depreived of the two new boost mechanisms.

CB has an X470 board.
 
Last edited:
So, you increase tdp by 10% and you use a process that's supposed to bring 10% improvement, and you wind up delivering < 10% on both the top and the bottom? Huh. Well, clearly not where I would have put the bar for release, but it's nice that new Ryzen buyers will get better ddr4 support, lower access latencies, and a boost mode that works properly. Unclear why they didn't let this one bake longer, though.

Everything is as announced, at 3.9GHz frequency is considerably higher than the 1800X on a frequency/voltage POV, the 1800X is in a quasi quadratic slope at those frequencies, if you look at the curves posted by TheStilt those 300MHz required 30% more power with the previous process.
 
So, you increase tdp by 10% and you use a process that's supposed to bring 10% improvement, and you wind up delivering < 10% on both the top and the bottom? Huh. Well, clearly not where I would have put the bar for release, but it's nice that new Ryzen buyers will get better ddr4 support, lower access latencies, and a boost mode that works properly. Unclear why they didn't let this one bake longer, though.

10.53% increase in TDP

2.78% increase in base clock and 8.75% increase in boost clock

That's not bad.

The increase in boost clock helps in single thread performance where Ryzen is particularly weak
 
Good progress from AMD, the way they are going, a chip of theirs may end up in my hot little hands in a few years time when I next upgrade.
 
Anyway 3,9GHz looks great out of the box with 3466MT/s DDR4.

I really don't expect it to overclock that much. If 95% of them can get to 4.3 with good thermals then they will sell very well, but I'm thinking it will top out at 4.2 for most people. Perhaps even 4.1

I think it will be like the 4770k and 4790k. A more hefty factory overclock but the real world overclocking isn't much different. Of course the real winner will be the 2-4 core performance out of the box.
 
Any word on the X470 lineup?
ASROCK X470 Falal1ty Gaming ITX/AC
ASROCK X470 Falal1ty Gaming K4
ASROCK X470 Falal1ty Professional Gaming
ASROCK X470 Master SLI
ASROCK X470 Master SLI/AC
ASROCK X470 Taichi Ultimate
ASUS X470-F ROG STRIX GAMING
ASUS X470-I ROG STRIX GAMING
ASUS X470 PRIME PRO
ASUS X470 ROG CROSSHAIR VII HERO (WI-FI AC)
ASUS X470 ROG CROSSHAIR VII HERO
ASUS X470 TUF PLUS GAMING
GIGABYTE X470 AORUS Gaming 7 WIFI
MSI X470 GAMING M7 AC
MSI X470 GAMING PLUS
MSI X470 GAMING PRO
 
Good improvement, but doesn't quite look good enough to beat a hypothetical 8-core mainstream part from Intel. Although I suppose we'll likely see a 2800X at that point, it feels like AMD could have been more aggressive here. Part of that may be due to being pin limited on AM4, and bound to the promise to keep that platform relevant. A 12-core triple-channel part, for example, would have been difficult for Intel to counter and could have added significant momentum, but also wouldn't work on the current socket infrastructure.
 
Good improvement, but doesn't quite look good enough to beat a hypothetical 8-core mainstream part from Intel. Although I suppose we'll likely see a 2800X at that point, it feels like AMD could have been more aggressive here. Part of that may be due to being pin limited on AM4, and bound to the promise to keep that platform relevant. A 12-core triple-channel part, for example, would have been difficult for Intel to counter and could have added significant momentum, but also wouldn't work on the current socket infrastructure.

I am glad that we finally moved beyond 4 cores for the mainstream, but 8 cores is plenty right now for the mainstream.

Throwing a gazillion cores isn't the answer.

AMD needs to improve the IPC and hopefully get GlobalFoundries to improve the process for higher clock speeds.
 
ASROCK X470 Falal1ty Gaming ITX/AC
ASROCK X470 Falal1ty Gaming K4
ASROCK X470 Falal1ty Professional Gaming
ASROCK X470 Master SLI
ASROCK X470 Master SLI/AC
ASROCK X470 Taichi Ultimate
ASUS X470-F ROG STRIX GAMING
ASUS X470-I ROG STRIX GAMING
ASUS X470 PRIME PRO
ASUS X470 ROG CROSSHAIR VII HERO (WI-FI AC)
ASUS X470 ROG CROSSHAIR VII HERO
ASUS X470 TUF PLUS GAMING
GIGABYTE X470 AORUS Gaming 7 WIFI
MSI X470 GAMING M7 AC
MSI X470 GAMING PLUS
MSI X470 GAMING PRO
Thanks, but it would be nice to have more details such as actual features, number of sata ports, number of ethernet ports, number of m.2 slots, and pcie slot layout.
 
Back
Top