AMD Ryzen 9 3950X Review • Hacker News Board

Re: AMD Ryzen 9 3950X Review

#141

Post by gameswithgo » Thu, Nov 14, 2019, 6:12 PM UTC

MrGilbert wrote: post #20

neogodless wrote: post #12

This isn't the most groundbreaking release - it's not the first 16-core chip you can buy (edited), nor the first 7nm. No new clockspeed records were set. Still, $750 for all that power! Any professionals shopping for this, or waiting for 24 or 32-core Threadripper? Anyone trying to upgrade on an older motherboard, or are you getting a matching X570 to ensure maximum boost and PCI-e bandwidth?

I did some math for myself and realized that I can skip the 4 extra cores, and go with the Ryzen 9 3900X (530€ vs 820€). That's 44.16 € / Core (3900X) vs 51.25 € / Core (3950X) where I live. Yes, the base clock is 200 Mhz lower, but for my use case (homeserver with gaming vm and minecraft server etc), I need cores that I can assign. Im not ready to pay 290€ für an extra 4 cores yet.

the 3900x base clock is actually higher.

Re: AMD Ryzen 9 3950X Review

#142

Post by velox_io » Thu, Nov 14, 2019, 6:12 PM UTC

simplyinfinity wrote: post #86

mmastrac wrote: post #74

Earlier quoted context omitted.

I believe that some of the YouTube channels have started including compilation benchmarks. GamersNexus [1] and Linus Tech Tips [2] both do! [1] https://www.gamersnexus.net/guides/3460-new-cpu-bench-method... [2] https://youtu.be/stM2CPF9YAY?t=4m49s

In addition, Level1Techs is more developer/sysadmin oriented than consumer oriented : https://www.youtube.com/channel/UC4w1YQAJMWOz4qtxinq55LQ

Likewise (I 2nd L1Techs btw). Having 16 cores/ 32 threads will greatly aid testing server applications. Making multi-threading & locking issues much more apparent (Amdahl's law) than they would on a slightly older CPU with only 4-6 cores (a i7 6700k in my case).

Re: AMD Ryzen 9 3950X Review

#143

Post by yifanlu » Thu, Nov 14, 2019, 6:13 PM UTC

davidy123 wrote: post #44

Slight meta, and part of a larger 'rant,' but when are we going to get away from reviews that might as well have been printed on glossy pages in PC Magazine in 1986? Anandtech has been at it since the 90s and hasn't changed their format at all. This is a serious decades-long stagnation of the web. The graphs should be dynamic (able to choose what scenarios and components to compare, and able to search within them), a…

> and user contributed

There's been a lot of debate in the enthusiasts community, but reviewers believe that user benchmarks don't have much value. There's so much variation in software, cooling, RAM speed, GPU speed, etc. Even misconfigurations like different running background apps can skew the results.

However, for a processor that's been in the market for a while, I think userbenchmarks is a good site to look at the aggregate data. Their rankings were recently updated to disfavor AMD chips, so don't take those too seriously. But for head-to-head comparisons of processors with a lot of users, you can get a good idea of how much faster a processor is.

However, I disagree that review sites should consider "user data" because 1) these are new processors and people who read these reviews are usually early adopters who want to make a buying decision and 2) the testing setup and methodology is a time consuming and scientific process and shouldn't be discounted by just asking random people to run an app.

Re: AMD Ryzen 9 3950X Review

#144

Post by Matthias247 » Thu, Nov 14, 2019, 6:13 PM UTC

ibobev wrote: post #72

As a developer I am keen to see some compilation benchmarks. Unfortunately those kinds of benchmarks are almost never included in such reviews. Instead there are many gaming benchmarks which purpose is not exactly clear for me, after obviously gaming is not the primary target market for R9 3950X.

As already said by others, Phoronix has some compilation benchmarks.

I have recently bought a 3900X. Some things about compilation I would note is that it heavily depends on your programming language and tooling. And that you would note is that you get most benefits on clean builds of big projects, which scale very well per core. On incremental builds or smaller projects it is however not uncommon to see It's all nice and fast, but on day to day use you likely won't see a 100% speed increase compared to a 3600.

Re: AMD Ryzen 9 3950X Review

#145

Post by stamps » Thu, Nov 14, 2019, 6:14 PM UTC

fock wrote: post #96

Can anyone comment on the IOMMU-grouping on typical boards? From what I've just googled, the CPU "lanes" seem to support ACS so it could indeed be working to replace my IGP+GPU-for-the-gaming-VN-Haswell system with a dual-GPU-Ryzen one (contrary to what I believed previously). Have been eyeing TR for that reason, but as I don't really need this amount of I/O (and cores), I might be well served by AM4.

I don't have the exact grouping on hand, but I'm able to run a Ryzen 2700x (8 core) with a 2080 (Windows VM GPU) and wx4100 (manjaro host GPU) all on a GIGABYTE X470 AORUS GAMING 5.

USB passthrough with an Startech USB PCI-E card was a bit problematic as the group was not isolated. Turned out the 2080 has a USB controller for the type C (I'm told for VR).

Re: AMD Ryzen 9 3950X Review

#146

Post by davidy123 » Thu, Nov 14, 2019, 6:17 PM UTC

wtallis wrote: post #93

davidy123 wrote: post #44

Slight meta, and part of a larger 'rant,' but when are we going to get away from reviews that might as well have been printed on glossy pages in PC Magazine in 1986? Anandtech has been at it since the 90s and hasn't changed their format at all. This is a serious decades-long stagnation of the web. The graphs should be dynamic (able to choose what scenarios and components to compare, and able to search within them), a…

> Instead we get feeble excuses like "it doesn't make sense to compare a two-year-old-generation to this one," well yes it does if I'm considering an upgrade. You don't get that excuse from AnandTech. We do our best to keep a long history of benchmark data for users to peruse: https://www.anandtech.com/bench/CPU-2019/2224 The main limiting factor on how far back our benchmark database goes is software updates. When w…

That's great to see, thanks. I thought I had asked about it some time ago to no avail. Still, if the interest is moving the front forward, making it available as cc-by would be nice, so people could use and contribute their own analysis around it.

Re: AMD Ryzen 9 3950X Review

#147

Post by tachion » Thu, Nov 14, 2019, 6:22 PM UTC

HHad3 wrote: post #55

th-miracle-257 wrote: post #33

Why did Apple not release their new MBP 16 with the 3950X? [1] [1] https://news.ycombinator.com/item?id=21523780

In addition to the points already mentioned, a good bit of Apple's software depends on Intel-proprietary features, e.g. QuickSync for video encoding. So far, no macOS system ever shipped with an AMD CPU, and porting the OS and user-mode application stack is not trivial.

You will find people using Hackintosh machines with AMD cpu's just fine (as much as fine it is possible for a Hackintosh, that is).

Re: AMD Ryzen 9 3950X Review

#148

Post by paulmd » Thu, Nov 14, 2019, 6:22 PM UTC

Someone1234 wrote: post #34

zamubafoo wrote: post #19

I find it interesting how desktop CPUs are essentially coming down to two enthusiast markets, developers/content creators/workstations and gaming. While the gaming market is (usually) seeking the highest top single core clock speed with respect to CPUs, it also relies on other expensive hardware. Meanwhile the dev/content creator/workstation market is much better served by these multi-core behemoths. Intel really has…

I agree with your overall synopsis of the status quo. We might however see gaming shift more to multi-core workloads as the next gen of consoles are highly likely to contain 8 core (16T) AMD CPUs based on Zen2 (and a lot of games are made for the lowest common denominator). This shouldn't be confused with the 8 core APUs found in current gen. Next gen will have an 8 core CPU and a NAVI 12+ GPU.

To be clear it's a shift from 8C8T to 8C16T (8C with SMT). The main change here is a massive (~3x) improvement in per-core performance.

It's not actually a shift to (significantly) more threading but actually a shift to more per-core performance, like desktop PCs currently use. Consoles are getting desktop-like processors for the first time.

This is creating quite a bit of anxiety in the desktop gaming community since high framerates depend on desktop processors being able to significantly outperform consoles on a per-core basis, and that will no longer be the case.

Re: AMD Ryzen 9 3950X Review

#149

Post by davidy123 » Thu, Nov 14, 2019, 6:24 PM UTC

dageshi wrote: post #67

davidy123 wrote: post #44

Slight meta, and part of a larger 'rant,' but when are we going to get away from reviews that might as well have been printed on glossy pages in PC Magazine in 1986? Anandtech has been at it since the 90s and hasn't changed their format at all. This is a serious decades-long stagnation of the web. The graphs should be dynamic (able to choose what scenarios and components to compare, and able to search within them), a…

They don't do it because it won't lead to any additional page views but will require extra effort.

This is probably the real answer. No profit innit. Having been through a few revolutions, there are advantages and challenges we won't know until we get there. Open it up so the users are encouraged to participate, rather than just using their computers to decide what new gaming system to build, with "participation" a ten comment per page system designed to support throwaways.

Re: AMD Ryzen 9 3950X Review

#150

Post by moonbug » Thu, Nov 14, 2019, 6:24 PM UTC

starpilot wrote: post #136

mrandish wrote: post #23

I noticed yesterday's articles about Intel's upcoming unusually large release of security mitigations. Under serious competitive threat for the first time in a while, I'm curious if Intel may have slowed the release of some mitigations to land after this round of comparison benchmarks.

Intel is part of the deep state.

If that's so, they're helping Make AMD Great Again.