PCI-SIG Talks PCIe 4.0 at Its Developers Conference

Status
Not open for further replies.

Pinhedd

Champion
Moderator
[citation][nom]loops[/nom]I just need to find a GPU that will fill the lanes.[/citation]

Graphics cards may be the most popular PCIe add-in cards but many onboard devices also communicate with the chipset or CPU by PCIe. Onboard peripherals that are added by the motherboard manufacturer such as audio, ethernet, USB 3.0, extra SATA ports, bluetooth, wifi, etc... are often wired up using PCIe links. Newer PCIe generations will allow these devices to provide greater services without increasing the complexity of the motherboard.
 

the3dsgeek

Honorable
Apr 22, 2012
24
0
10,510
0
i don't know where are they going with this, today's graphic cards performs the same in the old generation PCIe slots, or i don't know, maybe 2-3% difference. First make the hardware that can take advantage of the available bandwidth. Its like putting a car on a 200mph race track when it can only do 100.
 

rocknrollz

Distinguished
Nov 16, 2011
750
0
19,010
24
[citation] Its like putting a car on a 200mph race track when it can only do 100.[/citation]

Exactly, but what happens when we get the tech to go up to the track? Then we gotta rebuild it all. Better to have it available now, then later.
 
G

Guest

Guest
It's not all about graphics cards. PCIe SSD's will need the bandwidth to keep pushing I/O.
 

A Bad Day

Distinguished
Nov 25, 2011
2,256
0
19,790
2
For those who say that there's no need for so much bandwidth...

PCI-E 1.0 had lots of bandwidth during its introduction, but now GPUs saturate it fairly well.

AGP had lots of bandwidth during its introduction, until it reached a point where OCing it actually improved GPU performance.

PCI original had lots of bandwidth during its introduction. However, the last GPU that used it today was a modified GTX 520.

USB 2.0 had lots of bandwidth during its introduction. But today, SSDs require a lot more than two USB 2.0 ports to avoid being bottlenecked.
 

opmopadop

Distinguished
Apr 12, 2009
277
0
18,780
0
I know 'GigaTransfers' is the correct description for PCIe transfer speed, but I miss the old tried-and-true Gigabits-per-second measurement.

http://en.wikipedia.org/wiki/Transfer_(computing)
 
G

Guest

Guest
do renember this tech would not hit the market until 2016-2017. just look at 4 years ago, saturating a pcie-gen2 seemed impossible. looking foward, high end gpus might start reaching the upper limits of pci-e gen3, (sli and crossfire also present their own issues, as often getting two x16 slots is tough) besides, more lanes will affect other areas, more bandwidth for SATA, usb3 (which someday every port will be native 3.0) thunderbolt.
 

DRosencraft

Distinguished
Aug 26, 2011
743
0
19,010
6
I understand where the3dsgeek is coming from. The problem with building stuff like this so far out into the future sometimes is that it takes what can seem like an eternity for the useful hardware to catch up. Enterprise will love this because they do virtually everything through PCI and can never get it done fast enough. But most people here are consumers, and the consumer sector has limited use for PCIe 3.0 at this point, such that even talking about 4.0 seems at least a little early, if not ridiculous. I mean, if you're an AMD fan, they don't even have a MB with 3.0 on it yet. And the benefit of 3.0 for the latest gen. of GPUs is not that terrific. As for consumer SSDs on PCIe, most don't take full advantage of 2.0, let alone touch 3.0, and are so wildly expensive that few people seem to give a care for them at all.

It is nice to have these sorts of upgrades planned out, but increasingly we are seeing that they come out farther and farther in advance of any actually useful means of implementation. So, if you're a consumer now, you end up buying hoping that a use will come along now, or you end up paying a premium, even if only slightly, for the sake of having a newer/higher standard that no one will have any meaningful use for. But, 2015 is 3 years away, and the first implementations probably wont be until 2016, so maybe things will change faster than they have been of late.
 

the3dsgeek

Honorable
Apr 22, 2012
24
0
10,510
0
If we had that, then people of this mind set would be saying "why do we have cards that can go faster when we don't have the slots?" You have to start somewhere. I remember a time when games came out with features that weren't yet supported to help make them a bit more future-proof. Innovation is never incorrect.
i get it man, but all i am saying is that they are introducing PCIe 4.0, but there isn't hardware available in the market which can take full advantage of PCIe 2.0. By the time we'll be able to take advantage of PCIe 3.0, they will introduce PCIe 5.0. Evrytime we will be paying more money for the technology we don't need "yet".
 

Burodsx

Distinguished
May 31, 2009
250
0
18,780
0
[citation][nom]the3dsgeek[/nom]First make the hardware that can take advantage of the available bandwidth.[/citation]

No... No... and No. You want to have the interface first. You saturate the market with the new capability. Otherwise what would happen is companies would delay new products because they want to sell to the mainstream market. Think about it, would you produce a costly product that only 5% of the market could use? And keep in mind just because 5% can use it doesn't mean they'll buy it... Lastly consider how long you'll have to keep the product on the shelves before it would sell.
 

murdoc

Distinguished
Aug 5, 2008
356
0
18,780
0
[citation][nom]A Bad Day[/nom]For those who say that there's no need for so much bandwidth...PCI-E 1.0 had lots of bandwidth during its introduction, but now GPUs saturate it fairly well.AGP had lots of bandwidth during its introduction, until it reached a point where OCing it actually improved GPU performance.PCI original had lots of bandwidth during its introduction. However, the last GPU that used it today was a modified GTX 520.USB 2.0 had lots of bandwidth during its introduction. But today, SSDs require a lot more than two USB 2.0 ports to avoid being bottlenecked.[/citation]

Good point! You've identified a pattern. The PC vendors have sold you on the notion of "the future". You are buying tech that you don't need today, but for "tomorrow" and you are always buying tech "for tomorrow" when you are living in the present. What a way to get us on a continuous spending cycle!
 
G

Guest

Guest
I design PCI Express hardware. Everything I've designed is used for the enterprise market. Petabyte storage arrays. 10+ gigabit networking. These products cost over $100,000. Many of these products and companies could have used PCI Express 4 yesterday. This isn't for the average consumer or even the average Tom's Hardware reader, at least not yet.
 

bitbike

Honorable
Jul 13, 2012
1
0
10,510
0
[citation][nom]the3dsgeek[/nom]Its like putting a car on a 200mph race track when it can only do 100.[/citation]

Look at where the OCZ RevoDrive and other technology like it are heading. SSDs have nearly saturated the SATA III interface. PCI Express interface is the fastest way to move data at the moment without creating an entirely new standard/interface. Building up the infrastructure first is only logical.
 

headscratcher

Distinguished
Dec 2, 2009
129
0
18,680
0
[citation][nom]opmopadop[/nom]I know 'GigaTransfers' is the correct description for PCIe transfer speed, but I miss the old tried-and-true Gigabits-per-second measurement.http://en.wikipedia.org/wiki/Transfer_(computing)[/citation]
Perhaps because Gigabits implies data and not all that is transferred on PCIe is user data. The encoding includes framing bits. There are a variety of link specific house-keeping type packets that are never seen by the user. Even packets that have useable data payloads have start and stop packets, ecrc packets, and header information. So the actual bit rate the user will get will never be the nominal transfer rate.
 

A Bad Day

Distinguished
Nov 25, 2011
2,256
0
19,790
2
[citation][nom]murdoc[/nom]Good point! You've identified a pattern. The PC vendors have sold you on the notion of "the future". You are buying tech that you don't need today, but for "tomorrow" and you are always buying tech "for tomorrow" when you are living in the present. What a way to get us on a continuous spending cycle![/citation]

You're welcome to stay with your Pentium II, GeForce 2 MX, 64-128 MB of RAM, and 20 GB hard drive.
 
[citation][nom]DRosencraft[/nom]I understand where the3dsgeek is coming from. The problem with building stuff like this so far out into the future sometimes is that it takes what can seem like an eternity for the useful hardware to catch up. Enterprise will love this because they do virtually everything through PCI and can never get it done fast enough. But most people here are consumers, and the consumer sector has limited use for PCIe 3.0 at this point, such that even talking about 4.0 seems at least a little early, if not ridiculous. I mean, if you're an AMD fan, they don't even have a MB with 3.0 on it yet. And the benefit of 3.0 for the latest gen. of GPUs is not that terrific. As for consumer SSDs on PCIe, most don't take full advantage of 2.0, let alone touch 3.0, and are so wildly expensive that few people seem to give a care for them at all. It is nice to have these sorts of upgrades planned out, but increasingly we are seeing that they come out farther and farther in advance of any actually useful means of implementation. So, if you're a consumer now, you end up buying hoping that a use will come along now, or you end up paying a premium, even if only slightly, for the sake of having a newer/higher standard that no one will have any meaningful use for. But, 2015 is 3 years away, and the first implementations probably wont be until 2016, so maybe things will change faster than they have been of late.[/citation]

OCZ has two PCIe 3.0 enterprise PCIe SSDs that are very fast. One breaks 7.2GB/s and the other breaks 6GB/s, the faster one has 12TB of flash and the slower one has 16TB. Considering that many (perhaps most?) servers use PCIe x8, not x16, so these SSDs could be severely bottle-necked if they were limited to PCIe 2.0 instead of 3.0, their version of PCIe. There are also PCIe 3.0 RAID cards, networking cards, and more in addition to the fact that more keep coming out. Furthermore, as others have stated, new PCIe versions are also paramount for improvements in motherboard hardware. More bandwidth for motherboard makers and chipset makers to work with helps them give us better hardware and connectivity with the motherboards.

Beyond that, PCIe 3.0 can help cards such as the Radeon 7970 somewhat when in CF with only PCIe x8 (or worse) per card when PCIe 2.0 is compared to it.

[citation][nom]the3dsgeek[/nom]i get it man, but all i am saying is that they are introducing PCIe 4.0, but there isn't hardware available in the market which can take full advantage of PCIe 2.0. By the time we'll be able to take advantage of PCIe 3.0, they will introduce PCIe 5.0. Evrytime we will be paying more money for the technology we don't need "yet".[/citation]

There are many cards that saturate PCIe 2.0 and that's a fact that you should have looked up before making such a statement. Beyond that, PCIe needs to constantly improve to make sure that it doesn't become a bottle-neck during its lifetime. The whole point of being a step ahead is to not get caught with consequences related to getting behind.
 
[citation][nom]murdoc[/nom]Good point! You've identified a pattern. The PC vendors have sold you on the notion of "the future". You are buying tech that you don't need today, but for "tomorrow" and you are always buying tech "for tomorrow" when you are living in the present. What a way to get us on a continuous spending cycle![/citation]

If someone spends money to get a new system just for a new PCIe version when by the time they can actually use it, it's time to replace that computer anyway, then they're an idiot.

[citation][nom]bitbike[/nom]Look at where the OCZ RevoDrive and other technology like it are heading. SSDs have nearly saturated the SATA III interface. PCI Express interface is the fastest way to move data at the moment without creating an entirely new standard/interface. Building up the infrastructure first is only logical.[/citation]

SSDs could saturate the SATA3 interface before it even launched. You don't see SATA3 SSDs hitting 6600MB/s because it's not reasonably possible to max out an interface completely because there is always overhead. The SATA controller, the drive's SATA interface, and more all stop even a drive capable of multiple GB/s from hitting 600MB/s on SATA3 just like it couldn't max out any other interface that is slower than it despite the fact that they are slower than it.
 
[citation][nom]bobj213[/nom]I design PCI Express hardware. Everything I've designed is used for the enterprise market. Petabyte storage arrays. 10+ gigabit networking. These products cost over $100,000. Many of these products and companies could have used PCI Express 4 yesterday. This isn't for the average consumer or even the average Tom's Hardware reader, at least not yet.[/citation]

An excellent point. Just because technology is coming out now doesn't mean that it's meant for the consumer market yet. The enterprise market might need it in order to accomplish something, so it gets worked on. Consumers can moan and groan for no good reason about it all they want, but the world isn't revolving around them and they're not even the reason for that technology coming out. We won't make use of it until our lesser hardware catches up to a point where it is necessary to continue improving. The consumers might get the next PCIe version before they need it to make sure that they have it when they need it, but it isn't being created specifically for them and it most certainly isn't being forced down their throats.

[citation][nom]opmopadop[/nom]I know 'GigaTransfers' is the correct description for PCIe transfer speed, but I miss the old tried-and-true Gigabits-per-second measurement.http://en.wikipedia.org/wiki/Transfer_(computing)[/citation]

PCIe 2.0 = 500MB/s effective data transfer rate aka 4GT/s effective transfer rate.
PCIe 2.0 = 625MB/s true transfer rate aka 5GB/s true transfer rate

Same goes for SATA, USB, and many other interfaces. PCIe 3.0 uses much more efficient signaling, so this isn't significant for PCIe 3.0 which instead of an 8/10 bit encoding (25% extra overhead), it is 128/130 bit and that's only like just under 1% extra overhead.
 
Status
Not open for further replies.

ASK THE COMMUNITY