All AMD ThreadRipper CPU's will have 64 PCIe Lanes • Hacker News Board

Re: All AMD ThreadRipper CPU's will have 64 PCIe Lanes

#51

Post by TazeTSchnitzel » Wed, May 31, 2017, 1:27 PM UTC

DaiPlusPlus wrote: post #36

nicktelford wrote: post #14

Earlier quoted context omitted.

I doubt we'll be seeing desktop ARM systems any time soon, especially high end workstation or enthusiast systems. The problem is that the ARM64 architecture is completely different from x86-64, which almost every desktop application is compiled for exclusively . Remember when Microsoft launched an ARM Surface? It had a special version of Windows, and almost nothing would run on it, as very few applications are compil…

> It had a special version of Windows, and almost nothing would run on it, as very few applications are compiled for Windows on ARM. That wasn't the reason. It was because Windows RT was artificially locked-down to only allow sandboxed "apps" from the Windows App Store to run. There's not much of a use-case for only glorified web-service clients on a machine with a desktop UI regardless of the ISA they're compiled fo…

Yeah, it required the Windows Store, and additionally at that time it could only have WinRT (unrelated to Windows RT; basically the old name for UWP) apps. These days you can put Win32 apps on the store. If Microsoft allowed Win32 ARM apps on the store, maybe Windows RT would have succeeded. Maybe. Software publishers probably wouldn't have wanted to lose that control.

Re: All AMD ThreadRipper CPU's will have 64 PCIe Lanes

#52

Post by TazeTSchnitzel » Wed, May 31, 2017, 1:29 PM UTC

floatboth wrote: post #12

vbezhenar wrote: post #8

Earlier quoted context omitted.

I don't fully understand. So 1 SSD need 1 lane. How many lanes need something like NVisia 1080Ti? What use of the remaining lanes? If I want to build something like 1 CPU/1 GPU/1 SSD configuration without additional peripherals, 62 lanes will be effectively wasted?

A GPU can use 16x 3.0 lanes, theoretically. In practice 8 lanes is enough, people have measured performance between 16x and 8x, it's almost the same. SSDs use 4x, if you do buy an NVMe drive that is. (In my country, these things are TWICE as expensive as SATA SSDs lol)

My current “gaming machine” is still using PCIe 1.0 and it doesn't look like it's proving much at all of a bottleneck, because why would it? The game loads most things into memory once and then the rest of the time it's just sending draw calls and maybe geometry updates.

Re: All AMD ThreadRipper CPU's will have 64 PCIe Lanes

#53

Post by TazeTSchnitzel » Wed, May 31, 2017, 1:29 PM UTC

masklinn wrote: post #2

Salient point: > all 64 lanes being enabled for all ThreadRipper SKUs. This will be broken up into 60+4: 60 lanes directly from the CPU for feeding PCIe and M.2 slots , and then another 4 lanes going to the chipset And > the 16 core processor will for most purposes be half of an Epyc processor Epyc (formerly Naples) is the server-grade arch with 8 memory channels, 32 cores and 128 PCIe lanes per socket.

I wonder if it's the same silicon as Epyc, just ThreadRipper gets the worse bins.

Or maybe there's one less chip in the package?

Re: All AMD ThreadRipper CPU's will have 64 PCIe Lanes

#54

Post by steveadoo » Wed, May 31, 2017, 1:31 PM UTC

krisroadruck wrote: post #47

nicktelford wrote: post #16

Earlier quoted context omitted.

Enthusiasts tend to be gamers, and once you have a high end machine, most games tend to be limited on disk bandwidth, which affects load times. By consumers, I meant enthusiasts, since that's who the Threadripper (X399) and Intel's Extreme Edition (X299) platforms are aimed at.

Enthusiast here - can confirm, an array of nvme's makes a huge difference over single standard SSD. I set this up last year, check the benchmark results, it's night and day. http://imgur.com/a/N4FpD note: be sure to note the scaling at the bottom. 4-500MB/s old vs 3000+ MB/s new

Do games actually load any faster? I went from a Samsung SSD(I think the 850 pro) to an Intel nvme ssd, and load times in games didn't seem much faster. I definitely notice it when booting up VS though.

Re: All AMD ThreadRipper CPU's will have 64 PCIe Lanes

#55

Post by bryanlarsen » Wed, May 31, 2017, 1:31 PM UTC

TazeTSchnitzel wrote: post #53

masklinn wrote: post #2

Salient point: > all 64 lanes being enabled for all ThreadRipper SKUs. This will be broken up into 60+4: 60 lanes directly from the CPU for feeding PCIe and M.2 slots , and then another 4 lanes going to the chipset And > the 16 core processor will for most purposes be half of an Epyc processor Epyc (formerly Naples) is the server-grade arch with 8 memory channels, 32 cores and 128 PCIe lanes per socket.

I wonder if it's the same silicon as Epyc, just ThreadRipper gets the worse bins. Or maybe there's one less chip in the package?

ThreadRipper has 2 CPUs in its MCM, and Epyc has 4.

Re: All AMD ThreadRipper CPU's will have 64 PCIe Lanes

#56

Post by redtuesday » Wed, May 31, 2017, 1:32 PM UTC

Quequau wrote: post #29

johansch wrote: post #11

Earlier quoted context omitted.

I've long been thinking that it would make sense for e.g. US+EU to fund AMD just in order to get some actual competition going.

Some years ago I read a moderately well fleshed out proposal to create a European semiconductor consortium as a sort of an Airbus for CPUs. My recollection is that, while it was somewhat flawed, it was an idea with significant merit.

They should do the same for software companies that produce software for the governments, with the requirement to open source the software.

Re: All AMD ThreadRipper CPU's will have 64 PCIe Lanes

#57

Post by redtuesday » Wed, May 31, 2017, 1:33 PM UTC

znpy wrote: post #42

Competition is supposed to bring prices down, yet Intel released i9 processor for an insane prize (nearly 2000$). If AMD manages to ship similar performance (not necessarily higher) at a lower price... Sooner or later Intel will have to slash its prices. Looking forward to such times.

Intel lowered their prices with the release of the 18 core for 2000$. Before AMD became competitive again you had to pay 1700$ for only 10 cores or go for a 3000$ or 4000$ 18 core Xeon.

EDIT - of course there are also Xeons between 10 and 18 cores, but it should show that you had to pay much more to get 18 cores, or only slightly less for much less cores.

Re: All AMD ThreadRipper CPU's will have 64 PCIe Lanes

#58

Post by BatFastard » Wed, May 31, 2017, 1:33 PM UTC

wlesieutre wrote: post #39

nicktelford wrote: post #14

Earlier quoted context omitted.

I doubt we'll be seeing desktop ARM systems any time soon, especially high end workstation or enthusiast systems. The problem is that the ARM64 architecture is completely different from x86-64, which almost every desktop application is compiled for exclusively . Remember when Microsoft launched an ARM Surface? It had a special version of Windows, and almost nothing would run on it, as very few applications are compil…

Funny you say that, MS just announced they're taking another stab at it. https://www.theverge.com/2017/5/31/15711334/microsoft-window...

Interesting, was watching the video on the new MS Surface Laptop at the end, I saw 4 gig of Ram and shut down the page...

Re: All AMD ThreadRipper CPU's will have 64 PCIe Lanes

#59

Post by drbawb » Wed, May 31, 2017, 1:34 PM UTC

masklinn wrote: post #2

Salient point: > all 64 lanes being enabled for all ThreadRipper SKUs. This will be broken up into 60+4: 60 lanes directly from the CPU for feeding PCIe and M.2 slots , and then another 4 lanes going to the chipset And > the 16 core processor will for most purposes be half of an Epyc processor Epyc (formerly Naples) is the server-grade arch with 8 memory channels, 32 cores and 128 PCIe lanes per socket.

>32 cores >128 PCIe lanes

Well shit... what am I going to do with this pair of useless Xeon E5s I just bought?

Re: All AMD ThreadRipper CPU's will have 64 PCIe Lanes

#60

Post by redial » Wed, May 31, 2017, 1:35 PM UTC

nicktelford wrote: post #14

xbmcuser wrote: post #6

I was of the opinion that Intel will only face real competition when arm reaches parity with desktop class cpu. Nice to see AMD rise back up again. Hope Intel is not able to strong arm the oems like last time when AMD had design wins but Intel using anti competitive practice stalled its growth till it was able to catch-up and later go ahead.

I doubt we'll be seeing desktop ARM systems any time soon, especially high end workstation or enthusiast systems. The problem is that the ARM64 architecture is completely different from x86-64, which almost every desktop application is compiled for exclusively . Remember when Microsoft launched an ARM Surface? It had a special version of Windows, and almost nothing would run on it, as very few applications are compil…

Isn't the main problem with ARM the lack of a standard hardware architecture that supports of-the-shelf pc components?

Every motherboard has it's own size, it's own closed bios supporting only some kinds of memory (if at all, due to on die memory of most soc), which is fine for embedded use but leaves a lot to be desired if you try to use it as a "desktop pc" replacement.