Intel G41

[Man-wai Chang]

Is its PCI-e 1x slot using PCI-e 2.0 speed?

--
@~@ You have the right to remain silence.
/ v \ Simplicity is Beauty! May the Force and farces be with you!
/( _ )\ (Fedora 15 i686) Linux 3.0.4
^ ^ 22:56:01 up 5 days 2:09 0 users load average: 0.09 0.05 0.06
ä¸å€Ÿè²¸! ä¸è©é¨™! ä¸æ´äº¤! ä¸æ‰“交! ä¸æ‰“劫! ä¸è‡ªæ®º! è«‹è€ƒæ…®ç¶œæ´ (CSSA):
http://www.swd.gov.hk/tc/index/site_pubsvc/page_socsecu/sub_addressesa

 

[VanguardLH]

Is its PCI-e 1x slot using PCI-e 2.0 speed?

G41 Express is a chipset, not a motherboard. You actually have the
chipset separate of a motherboard implementation?

Some info here:
http://www.intel.com/content/www/us/en/chipsets/mainstream-chipsets/g41-express-chipset.html

More info:
http://www.intel.com/p/en_US/support/highlights/graphics/g41

--
@~@ You have the right to remain silence.
/ v \ Simplicity is Beauty! May the Force and farces be with you!
/( _ )\ (Fedora 15 i686) Linux 3.0.4
^ ^ 22:56:01 up 5 days 2:09 0 users load average: 0.09 0.05 0.06
ä¸å€Ÿè²¸! ä¸è©é¨™! ä¸æ´äº¤! ä¸æ‰“交! ä¸æ‰“劫! ä¸è‡ªæ®º! è«‹è€ƒæ…®ç¶œæ´ (CSSA):
http://www.swd.gov.hk/tc/index/site_pubsvc/page_socsecu/sub_addressesa

11% body (43 characters)
89% signature (336 characters)

The noise from the off-topic *fluff* SPAM signature swamps out the
question in the body.

 

[Paul]

G41 Express is a chipset, not a motherboard. You actually have the
chipset separate of a motherboard implementation?

Some info here:
http://www.intel.com/content/www/us/en/chipsets/mainstream-chipsets/g41-express-chipset.html

More info:
http://www.intel.com/p/en_US/support/highlights/graphics/g41

And my interpretation of the information in the diagram, is *all*
PCI Express lanes on that chipset, are 1.1 at 250MB/sec bidirectional.
Even the graphics slot is version 1.1 and 250MB/sec bidirectional
per lane times 16 lanes.

So, for example, if you install a USB3 PCI Express x1 card with a
NEC USB3 controller on it, that will be limited in theoretical top
speed by the PCI Express slot. That's because it expects a Rev.2
slot, and a Rev1.1 slot would run at half of the desired speed.
The card would still work though. And the only time that matters,
is if you're connecting a BlackMagic video capture box to a USB3
card, which has a hard requirement of needing peak transfer speed.
For many USB3 hard disk enclosures, they can't go faster than about
200MB/sec anyway, even if you have a Rev.2 interface for your USB3 chip.
The USB3 disk enclosure wouldn't particularly care about the details.
There is a BlackMagic box that does care (this one needs peak USB3
transfer rates). So the USB3 add-in card in the computer, hosting
this, should have a real Rev.2 x1 slot.

http://p2ptalk.org/wp-content/uploads/2010/04/271f53d278341996.jpg.jpg

Paul

 

[Man-wai Chang]

So, for example, if you install a USB3 PCI Express x1 card with a

NEC USB3 controller on it, that will be limited in theoretical top
speed by the PCI Express slot. That's because it expects a Rev.2
slot, and a Rev1.1 slot would run at half of the desired speed.

I am glad I didn't buy any Intel Gxx motherboard. The PCIe slots of my
Asus M3N78 PRO are all running at PCIe 2.x speed....

--
@~@ You have the right to remain silence.
/ v \ Simplicity is Beauty! May the Force and farces be with you!
/( _ )\ (Fedora 15 i686) Linux 3.0.4
^ ^ 22:56:01 up 5 days 2:09 0 users load average: 0.09 0.05 0.06
ä¸å€Ÿè²¸! ä¸è©é¨™! ä¸æ´äº¤! ä¸æ‰“交! ä¸æ‰“劫! ä¸è‡ªæ®º! è«‹è€ƒæ…®ç¶œæ´ (CSSA):
http://www.swd.gov.hk/tc/index/site_pubsvc/page_socsecu/sub_addressesa

 

[Paul]

I am glad I didn't buy any Intel Gxx motherboard. The PCIe slots of my
Asus M3N78 PRO are all running at PCIe 2.x speed....

Yup.

http://techreport.com/r.x/nvidia-geforce-8300/block.jpg

One other issue with Intel, is the block diagram may show
add-in slots with Rev.2 capability, but the design lacks
enough low-jitter clock signals to make it happen. In which case,
depending on how that was fixed, the slot reverts to Rev.1.1
operating speed. (I.e. If they route a high jitter clock to
the slot, it's no longer a good candidate for Rev.2 500MB/sec.)
Jitter, is the cycle-by-cycle displacement of the clock, from
the ideal position, and Revision 2 apparently requires that
the clock signal be of a higher quality. (I don't have a copy
of the actual spec, so I can't verify the requirements.)

http://www.edn.com/article/459246-PCIe_clock_distribution_in_embedded_systems.php

"For 2.5-Gbps operation, the limit is 86-psec p-p phase jitter for a
sample set of 106 samples.

The 5-Gbps operational limit is 3.1-psec-rms jitter. However, to operate at
5 Gbps, a transceiver first negotiates at 2.5 Gbps and then moves up to the
higher rate if both ends can do so. That is, if the system supports any 5-Gbps
links, then the reference clock must meet both jitter specifications."

And some motherboards, the chipset doesn't have enough low-jitter clocks,
even though the chipset diagram shows all the slots are Rev.2!

Paul