Live data from Hacker News

AMD Rome Second Generation EPYC Review: 2x 64-core Benchmarked

anandtech.com

121–130 of 208 posts

Re: AMD Rome Second Generation EPYC Review: 2x 64-core Benchmarked

#121

The thing is you can't even get Ryzen inside a Dell or Lenovo machine - so for businesses it's kind of hard to get onboard with they hype. So intel is going to keep up market share until AMD can get into some prebuilt systems.

What ? Lenovo is selling Ryzen based desktop by the truck load. I've bought more than a dozen this year already.

Dell is in that game too.

What AMD need for consumer computers is to get in the laptop oem build.

Re: AMD Rome Second Generation EPYC Review: 2x 64-core Benchmarked

#122
post #95

Earlier quoted context omitted.

Then check out the discounted deal and mail them you'd like to pay extra

Good Morning Sir, I hope this letter finds you in good health. Since I've seen that you are offering AMD 3900X at a discount, I'd like to inform you that I am not like those pleebs and would therefore like you to pay in full MSRP. Please let me know where I can send the goats. I have the honor to be your obedient customer. N. Pleeb

Funny enough, I had two buyers saying pretty much this. "I'll pay 49" on a £35 item. Why? Because apparently that's "what it's worth". I appreciate the thought, but it just looks insanely suspicious...

Re: AMD Rome Second Generation EPYC Review: 2x 64-core Benchmarked

#123

Earlier quoted context omitted.

>a lot of the chiplets sold as 3900X are not fast enough to hit the clocks advertised for 3900X Do you have some sort of source for this claim?

every forum pretty much. If you have been interested in getting one, and following along with the launch this is not a controversial statement. It's not 100% sure its the chips fault though, bios issues are still running rampant nearly 5 weeks later, and each new bios is changing performance significantly. It will take a while before everyone knows exactly where they stand.

Forums are not reputable sources unless it is insiders leaking info that can be verified through other means.

Re: AMD Rome Second Generation EPYC Review: 2x 64-core Benchmarked

#124

Earlier quoted context omitted.

every forum pretty much. If you have been interested in getting one, and following along with the launch this is not a controversial statement. It's not 100% sure its the chips fault though, bios issues are still running rampant nearly 5 weeks later, and each new bios is changing performance significantly. It will take a while before everyone knows exactly where they stand.

Forums are not reputable sources unless it is insiders leaking info that can be verified through other means.

They might not be definitive sources, but they can definitely be the canary that triggers an investigation by somebody more authoritative.

As it stands, this mine has some dead canaries.

Re: AMD Rome Second Generation EPYC Review: 2x 64-core Benchmarked

#125
post #26

Earlier quoted context omitted.

wow, I don't think I have ever seen such a dominant benchmark victory across the board... for 40% less dollars.

At the end of the Phoronix review, you'll find that performance per dollar is 4x - 8x better with AMD. Assuming Intel slashes prices, I still dont' know if they can match it.

It also beats Intel on performance / watts according to the last page.

Seems like a slam dunk from AMD.

Re: AMD Rome Second Generation EPYC Review: 2x 64-core Benchmarked

#126
post #25
post #2

> For those with little time: at the high end with socketed x86 CPUs, AMD offers you up to 50 to 100% higher performance while offering at a 40% lower price. Unless you go for the low end server CPUs, there is no contest: AMD offers much better performance for a much lower price than Intel, with more memory channels and over 2x the number of PCIe lanes. These are also PCIe 4.0 lanes. What if you want to more than 2 T…

FWIW Intel still has an edge over AMD for speciality use cases, I would like to see some benchmarks that use Intel's MKL library to AMD (OpenBLAS).

You are correct at that, and that's because for over 10 years, especially in academia / HPC / Sci Comp, AMD was non existent. If they get the edge long enough for the HPC community to care, maybe an alternative will be present. Otherwise, if AMD just overpowers intel in hardware / clocks so much, then it might not be an issue.

In addition, some specific software (eg Android emulator on Windows) doesn't exist for AMD CPUs.

Re: AMD Rome Second Generation EPYC Review: 2x 64-core Benchmarked

#127
That AMD prices this aggressively to me indicates that they expect intel to slash prices hard too. If the graphics card releases is anything to go by, AMD might even cut further after intel shows its hand. I know intel have good margins for server parts but it’s also their bread & butter - what would it mean to intel’s bottom line if they had to cut prices (not just to select whales but msrp) by enough to compete here, say 40%?

Re: AMD Rome Second Generation EPYC Review: 2x 64-core Benchmarked

#128

Earlier quoted context omitted.

Physical space is almost never a constraint in data centers. It's power and cooling that are almost always the limit. If AMD's chips offer better performance/watt, that would make them attractive. The capital cost of purchasing a server tends to be a smaller number than the cost of the power to run and cool it over its lifetime. However, what also matters is vendor support. Lots of companies have contracts for pricin…

Less physical space = less cooling requirements

No that doesn't make any sense. If you can fit two 200W systems in the same space as a single big 200W system then using less physical space increases the cooling requirements and the power density might become high enough to warrant exotic cooling solutions like immersion cooling.

Re: AMD Rome Second Generation EPYC Review: 2x 64-core Benchmarked

#129
post #83

Earlier quoted context omitted.

OSs are not most software and are not designed around the instruction set changing underneath them during normal operation. Why would they be? You can't physically swap out a processor while the system is booted and you can't swap a virtual processor either. It works fine if you shut everything down and reboot the system, but that is often undesirable. The whole point of the feature is that the VM can be migrated aro…

>You can't physically swap out a processor while the system is booted and you can't swap a virtual processor either. I think you can do that on multisocket systems: https://www.kernel.org/doc/html/v4.14/core-api/cpu_hotplug.h...

If the hardware supports it, yes, but only with basically the same type of processor.

Re: AMD Rome Second Generation EPYC Review: 2x 64-core Benchmarked

#130
post #25

Earlier quoted context omitted.

FWIW Intel still has an edge over AMD for speciality use cases, I would like to see some benchmarks that use Intel's MKL library to AMD (OpenBLAS).

You are correct at that, and that's because for over 10 years, especially in academia / HPC / Sci Comp, AMD was non existent. If they get the edge long enough for the HPC community to care, maybe an alternative will be present. Otherwise, if AMD just overpowers intel in hardware / clocks so much, then it might not be an issue. In addition, some specific software (eg Android emulator on Windows) doesn't exist for AMD…

The Android emulator can run on AMD CPUs under Windows 10 as of a year or so ago, actually. It now only requires Intel's HAXM on older versions of Windows and I don't think Zen 2 or recent Intel chips officially support those versions of Windows anymore.
Post reply on HN