PCI SIG’s hardline on compliance approval being only available at quarterly events. I work as a test house that offers PCIe testing but we don’t get much traffic there because there’s little benefit to testing twice.
Really doubt it. +1 hardware upgrades drive new PC sales if nothing else. Skipping 4.0 would mean skipping sales, and Intel rarely leaves money on the table.
Intel needs to have their chip designs finalized a decent way before manufacturing can start. A 2019 date probably means the earliest they could get it out would be 2020 with 2021 more likely. There's no way they'd risk the spec being delayed and Icelake not being releasable as a result. I wouldn't be surprised if the 2021 socket comes with PCIe5. For 2019, at most they might have a pre-release PCIe5 derived version of DMI since they control that entirely.
Very excited to see faster busses! As more things have just started plugging direct into PCIe lanes it has caused no end of headaches when you want lots of m.2, GPUs, and network in the same box. I really think that the current lack of PCIe lanes is a lot of the reason why we dont have 10gig Ethernet standard yet. It would take away too many lanes from other devices that need it. With faster lanes we can assign fewer lanes per device, and hopefully have more nice things :D
No, lackluster adoption of 10GbE has nothing to do with PCIe speed (current PCIe 3.0 x1 already manages 80% of 10GbE(!)). It has everything to do with power consumption & the shift to wlan for everything outside the server room.
- 10GbE switches are just at ~$60/10gbaset-port... - It's just been 3 month's since 10GbE consumer cards can be had for ~$100 - It's been just 2.5 years since intel managed to cram 10GbE into their low power server platform (Xeon-D) and supermicro's x10sdv mainboards have been selling really well since then... - It's been about a year since we have 10GbE SFP+ Modules - It's been about a year since WLAN-APs managed to handle more than 1Gb/s, thus becoming bottlenecked on the uplink...
in a nutshell: 10GbE is currently taking off. mostly be
Fiber and DA (Direct-Attach copper) 10GbE SFP+ modules have been available for about 10 years now. It is specifically the 10GBASE-T SFP+ modules that have only been available for about a year.
You're both looking at the subject with datacenter tinted glasses. Outside the enterprise world ethernet and RJ-45 are one and the same. SFP+, optical fiber, etc don't exist.
"In regards to the seven year delay, PCI-SIG noted that PCIe 3.0 provided sufficient bandwidth for some time before the developments and rising demands of AI compute workloads, PCIe NVMe and 3D XPoint storage, and networking speeds, particularly as 10GbE becomes more and more accessible to consumers."
Or, maybe, put another way, the PCIe standard provided sufficient bandwidth until they stopped updating it; then it didn't. If PCIe 4.0 were released in 2014 there probably wouldn't have been such a problem of PCIe becoming a performance bottleneck for the "demands of AI compute workloads, PCIe NVMe and 3D XPoint storage, and networking speeds, particularly 10GbE".
Yeah, it's really really annoying that the PCI-SIG can force people to spend money on new hardware. Such a shame we don't have any choice over whether or not to upgrade.
He wants a new processor that works with the old surrounding technology. But, because of the way large scale semiconductor design and manufacturing works, and because significant markets need these new memory, bus, and networking technologies (including the fastest growing and highest margins ones), consumers can't afford to lag behind the advancing standards (otherwise all the processors cost more because the market becomes very segmented). I am guessing Intel makes more money on their CPUs than on their chip sets. I don't think they are trying to force people to upgrade more components of their system than they might otherwise want to. They are just keeping their platform up to date. And if they didn't do that, so many people would complain "why can't I have X new technology that I want, just like Y users get" such as NVMe SSDs or DDR4 DRAM or PCIe 4.0.
But a new processor DOES work with "the old surrounding technology". Most of the people in the world do NOT upgrade their external HDs, or wifi base stations, or ethernet switches, or screens, or ..., when they buy a new PC. They replace those surrounding technology pieces when they break, and not a day sooner.
I don't understand his complaint, and I don't understand your attempt to justify it.
I think the reason you can't understand either of us is because you do not have the relevant technology in mind. The article concerns the system bus, not peripherals. So, it makes sense that op's complaint is about the system bus. Since a new system bus on the market doesn't force anyone to buy a new processor or new peripherals, I can only imagine his complaint is that he has to buy a new motherboard when he buys a new processor. The peripherals will still work regardless, as you point out. You are correct in that that partially refutes his point that they are forcing customers to upgrade, milking cash. But, in general, when one gets a new processor one will probably need a new motherboard, and perhaps new RAM.
The memory controllers are integrated into the processor die. PCIe controllers are integrated into the processing die. I believe it's not easy to get the new technology to work with the old pin layout. It certainly reduces the flexibility in the engineering to try. Most of Intel's customers move to a new platform when they buy a new processor, because they rely on various parts of the improved ecosystem, or because they buy new computers each time and don't upgrade. The number of people truly interested in buying a new processor and keeping their old motherboard is low. Probably mostly hobbyists/enthusiasts. It doesn't make sense for Intel to change their engineering work flow for a small group of customers.
3840 x 2160 pixels x 30 bits/pixel x 120 fps = ~30 Gbps, which would be PCIe 5.0 1x or PCIe 3.0 4x. But then, we don't really connect these devices through PCIe. Meanwhile, transfer rate really doesn't mean much when discussing GPUs, so if you were talking about rendering the scene that's entirely different.
I noticed in the AMD Carizzo review that the chip has 8 PCIe lanes citing power efficiency due to its mobile origins. If I've understood this article correctly PCIe 4.0 should be able to deliver the same bandwidth with half the lanes, so it sounds like we could see efficiency gains here.
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
26 Comments
Back to Article
GTRagnarok - Thursday, October 26, 2017 - link
Typo in the title >_>Nate Oh - Thursday, October 26, 2017 - link
Fixed, thank you!willis936 - Thursday, October 26, 2017 - link
PCI SIG’s hardline on compliance approval being only available at quarterly events. I work as a test house that offers PCIe testing but we don’t get much traffic there because there’s little benefit to testing twice.DanNeely - Thursday, October 26, 2017 - link
No indication of if Intel's going to have it in Icelake, or if we'll end up having to wait a few years for their next major hardware refresh?CaedenV - Thursday, October 26, 2017 - link
And with that PCIe5 spec looking to be finalized in 2019... I could see Intel skipping this one.saratoga4 - Thursday, October 26, 2017 - link
Really doubt it. +1 hardware upgrades drive new PC sales if nothing else. Skipping 4.0 would mean skipping sales, and Intel rarely leaves money on the table.beginner99 - Friday, October 27, 2017 - link
Not to mention that upgrading at least DMI to 4.0 would be a huge benefit.DanNeely - Friday, October 27, 2017 - link
Intel needs to have their chip designs finalized a decent way before manufacturing can start. A 2019 date probably means the earliest they could get it out would be 2020 with 2021 more likely. There's no way they'd risk the spec being delayed and Icelake not being releasable as a result. I wouldn't be surprised if the 2021 socket comes with PCIe5. For 2019, at most they might have a pre-release PCIe5 derived version of DMI since they control that entirely.CaedenV - Thursday, October 26, 2017 - link
Very excited to see faster busses! As more things have just started plugging direct into PCIe lanes it has caused no end of headaches when you want lots of m.2, GPUs, and network in the same box. I really think that the current lack of PCIe lanes is a lot of the reason why we dont have 10gig Ethernet standard yet. It would take away too many lanes from other devices that need it. With faster lanes we can assign fewer lanes per device, and hopefully have more nice things :Dbernstein - Thursday, October 26, 2017 - link
No, lackluster adoption of 10GbE has nothing to do with PCIe speed (current PCIe 3.0 x1 already manages 80% of 10GbE(!)).It has everything to do with power consumption & the shift to wlan for everything outside the server room.
- 10GbE switches are just at ~$60/10gbaset-port...
- It's just been 3 month's since 10GbE consumer cards can be had for ~$100
- It's been just 2.5 years since intel managed to cram 10GbE into their low power server platform (Xeon-D) and supermicro's x10sdv mainboards have been selling really well since then...
- It's been about a year since we have 10GbE SFP+ Modules
- It's been about a year since WLAN-APs managed to handle more than 1Gb/s, thus becoming bottlenecked on the uplink...
in a nutshell: 10GbE is currently taking off. mostly be
bernstein - Thursday, October 26, 2017 - link
cause it's getting cheap (because power consumption has finally come down)cdillon - Thursday, October 26, 2017 - link
Fiber and DA (Direct-Attach copper) 10GbE SFP+ modules have been available for about 10 years now. It is specifically the 10GBASE-T SFP+ modules that have only been available for about a year.Notmyusualid - Friday, October 27, 2017 - link
@ cdillonYes, I was thinking, wait, what? 10GE SFPs/XFPs have been around for half my working lifetime...
DanNeely - Friday, October 27, 2017 - link
You're both looking at the subject with datacenter tinted glasses. Outside the enterprise world ethernet and RJ-45 are one and the same. SFP+, optical fiber, etc don't exist.Yojimbo - Friday, October 27, 2017 - link
"In regards to the seven year delay, PCI-SIG noted that PCIe 3.0 provided sufficient bandwidth for some time before the developments and rising demands of AI compute workloads, PCIe NVMe and 3D XPoint storage, and networking speeds, particularly as 10GbE becomes more and more accessible to consumers."Or, maybe, put another way, the PCIe standard provided sufficient bandwidth until they stopped updating it; then it didn't. If PCIe 4.0 were released in 2014 there probably wouldn't have been such a problem of PCIe becoming a performance bottleneck for the "demands of AI compute workloads, PCIe NVMe and 3D XPoint storage, and networking speeds, particularly 10GbE".
damianrobertjones - Friday, October 27, 2017 - link
"Every 3 years..."Which means that a new product hits every 3 years, ensuring that the cycle of upgrading continues milking customers of their cash.
mkaibear - Friday, October 27, 2017 - link
Yeah, it's really really annoying that the PCI-SIG can force people to spend money on new hardware. Such a shame we don't have any choice over whether or not to upgrade.Yojimbo - Friday, October 27, 2017 - link
He wants a new processor that works with the old surrounding technology. But, because of the way large scale semiconductor design and manufacturing works, and because significant markets need these new memory, bus, and networking technologies (including the fastest growing and highest margins ones), consumers can't afford to lag behind the advancing standards (otherwise all the processors cost more because the market becomes very segmented). I am guessing Intel makes more money on their CPUs than on their chip sets. I don't think they are trying to force people to upgrade more components of their system than they might otherwise want to. They are just keeping their platform up to date. And if they didn't do that, so many people would complain "why can't I have X new technology that I want, just like Y users get" such as NVMe SSDs or DDR4 DRAM or PCIe 4.0.name99 - Saturday, October 28, 2017 - link
But a new processor DOES work with "the old surrounding technology".Most of the people in the world do NOT upgrade their external HDs, or wifi base stations, or ethernet switches, or screens, or ..., when they buy a new PC.
They replace those surrounding technology pieces when they break, and not a day sooner.
I don't understand his complaint, and I don't understand your attempt to justify it.
Yojimbo - Sunday, October 29, 2017 - link
I think the reason you can't understand either of us is because you do not have the relevant technology in mind. The article concerns the system bus, not peripherals. So, it makes sense that op's complaint is about the system bus. Since a new system bus on the market doesn't force anyone to buy a new processor or new peripherals, I can only imagine his complaint is that he has to buy a new motherboard when he buys a new processor. The peripherals will still work regardless, as you point out. You are correct in that that partially refutes his point that they are forcing customers to upgrade, milking cash. But, in general, when one gets a new processor one will probably need a new motherboard, and perhaps new RAM.The memory controllers are integrated into the processor die. PCIe controllers are integrated into the processing die. I believe it's not easy to get the new technology to work with the old pin layout. It certainly reduces the flexibility in the engineering to try. Most of Intel's customers move to a new platform when they buy a new processor, because they rely on various parts of the improved ecosystem, or because they buy new computers each time and don't upgrade. The number of people truly interested in buying a new processor and keeping their old motherboard is low. Probably mostly hobbyists/enthusiasts. It doesn't make sense for Intel to change their engineering work flow for a small group of customers.
willis936 - Friday, October 27, 2017 - link
Heaven forbid that my next replacement machine is also an upgrade. Unless you mean that electronics should live forever.Sahrin - Friday, October 27, 2017 - link
Infinity Fabric 2.0 here we come!BubbaJoe TBoneMalone - Friday, October 27, 2017 - link
Does PCIe 5.0 have the bandwidth for 4k HDR10+ VR at 120fps?mkozakewich - Monday, October 30, 2017 - link
3840 x 2160 pixels x 30 bits/pixel x 120 fps = ~30 Gbps, which would be PCIe 5.0 1x or PCIe 3.0 4x. But then, we don't really connect these devices through PCIe. Meanwhile, transfer rate really doesn't mean much when discussing GPUs, so if you were talking about rendering the scene that's entirely different.skavi - Sunday, October 29, 2017 - link
I'm most hyped for Thunderbolt docks with less compromise.jimjamjamie - Wednesday, November 1, 2017 - link
I noticed in the AMD Carizzo review that the chip has 8 PCIe lanes citing power efficiency due to its mobile origins. If I've understood this article correctly PCIe 4.0 should be able to deliver the same bandwidth with half the lanes, so it sounds like we could see efficiency gains here.