PCI-e bandwidth/capacity limitations

A forum for discussing FAH-related hardware choices and info on actual products (not speculation).

Moderator: Site Moderators

Forum rules
Please read the forum rules before posting.
ComputerGenie
Posts: 230
Joined: Mon Dec 12, 2016 4:06 am

Re: PCI-e bandwidth/capacity limitations

Post by ComputerGenie »

boristsybin wrote:...
2.2. is "premium" Termaltake riser realy good for folding?
i know about 3M risers, they are perfect, but too expensive for just risers
Save $ and get something like: http://a.co/jfI80yM
SteveWillis
Posts: 389
Joined: Fri Apr 15, 2016 12:42 am
Hardware configuration: PC 1:
Linux Mint 17.3
three gtx 1080 GPUs One on a powered header
Motherboard = [MB-AM3-AS-SB-990FXR2] qty 1 Asus Sabertooth 990FX(+59.99)
CPU = [CPU-AM3-FX-8320BR] qty 1 AMD FX 8320 Eight Core 3.5GHz(+41.99)

PC2:
Linux Mint 18
Open air case
Motherboard: ASUS Crosshair V Formula-Z AM3+ AMD 990FX SATA 6Gb/s USB 3.0 ATX AMD
AMD FD6300WMHKBOX FX-6300 6-Core Processor Black Edition with Cooler Master Hyper 212 EVO - CPU Cooler with 120mm PWM Fan
three gtx 1080,
one gtx 1080 TI on a powered header

Re: PCI-e bandwidth/capacity limitations

Post by SteveWillis »

ComputerGenie wrote:Save $ and get something like: http://a.co/jfI80yM
At least two reports of ruined motherboards from these in the reviews? No I don't think so. Quality powered risers are hard to find but you can tell these are quality just by looking at them.
http://www.ebay.com/itm/261459534757?_t ... EBIDX%3AIT
I have two with no problems. You get what you pay for, at least on a good day.

if you are looking specifically for unpowered risers I like these but they are out of stock right now.
https://smile.amazon.com/gp/product/B01 ... UTF8&psc=1
Image

1080 and 1080TI GPUs on Linux Mint
ComputerGenie
Posts: 230
Joined: Mon Dec 12, 2016 4:06 am

Re: PCI-e bandwidth/capacity limitations

Post by ComputerGenie »

SteveWillis wrote:
ComputerGenie wrote:Save $ and get something like: http://a.co/jfI80yM
At least two reports of ruined motherboards from these in the reviews? No I don't think so...
I'm not sure how much I ever trust "customer reviews" on any site that allows novices to buy and review electronics. That aside, my point was simply that shielding is overkill. Personally, I have plain old dual-ribbon 16x-16x that I clipped and added power myself, which I got for $6 per.
bruce
Posts: 20824
Joined: Thu Nov 29, 2007 10:13 pm
Location: So. Cal.

Re: PCI-e bandwidth/capacity limitations

Post by bruce »

On one system, my MB has room for two double-slot x16 devices plus a couple of single-slot x1 devices. I have two double-wide GPUs in the x16 slots and I wanted to add another GPU to use the x1 slot. I'm using one of the risers that converts a x1 PCIe slot to USB and back to x16 (something like this) which allows me to leave the GPU outside the case. I put the slowest of the 3 GPUs in that slot, assuming that it would need the lowest share of the total bandwidth. I suspect that it might fold a little faster if it was provided with a x8 or better slot, but my CPU doesn't have enough lanes for that and it's getting more than it was sitting on the spare parts shelf.

I did try some plain wired risers and had some troubles, but they'd probably work except for the fact that I needed the extra length of the USB cable to reach outside of the case.

I still have another x1 slot to try later.
boristsybin
Posts: 50
Joined: Mon Jan 16, 2017 11:40 am
Hardware configuration: 4x1080Ti + 2x1050Ti
Location: Russia, Moscow

Re: PCI-e bandwidth/capacity limitations

Post by boristsybin »

ComputerGenie wrote:Save $ and get something like: http://a.co/jfI80yM
already waiting for a couple of similar cheap risers, and i also ask about them
boristsybin wrote:2.1. is cheap non-shielded pci-e x16-x16 or pci-e x16-x4 20cm riser enough to provide stable folding at pci-e 3.0 standard? if not, does handmade aluminium foil shield helps?
Image
ryoungblood
Posts: 39
Joined: Wed Jan 18, 2017 10:13 pm
Location: Boulder, CO

Re: PCI-e bandwidth/capacity limitations

Post by ryoungblood »

FYI, those unshielded risers have a chance to drop your PPD. There is an ongoing discussion of this in the Overclockers.com Folding Forum. The EZDIY risers from Amazon are great. They have smaller ones that are in stock.
Image
ComputerGenie
Posts: 230
Joined: Mon Dec 12, 2016 4:06 am

Re: PCI-e bandwidth/capacity limitations

Post by ComputerGenie »

ryoungblood wrote:FYI, those unshielded risers have a chance to drop your PPD...
I had my GTX 1080s on an M5A99FX PRO (purchased for reasons unrelated to folding) and there was no discernible difference between with unshielded risers (even when laying directly on each other) and directly in the slots; so, I guess ymmv. :egeek:
foldy
Posts: 2040
Joined: Sat Dec 01, 2012 3:43 pm
Hardware configuration: Folding@Home Client 7.6.13 (1 GPU slots)
Windows 7 64bit
Intel Core i5 2500k@4Ghz
Nvidia gtx 1080ti driver 441

Re: PCI-e bandwidth/capacity limitations

Post by foldy »

I guess it depends on slot speed, gen3 x16 with risers reducing bandwidth by half doesn't hurt. But the same on gen3 x4 bandwidth and PPD goes down.
SteveWillis
Posts: 389
Joined: Fri Apr 15, 2016 12:42 am
Hardware configuration: PC 1:
Linux Mint 17.3
three gtx 1080 GPUs One on a powered header
Motherboard = [MB-AM3-AS-SB-990FXR2] qty 1 Asus Sabertooth 990FX(+59.99)
CPU = [CPU-AM3-FX-8320BR] qty 1 AMD FX 8320 Eight Core 3.5GHz(+41.99)

PC2:
Linux Mint 18
Open air case
Motherboard: ASUS Crosshair V Formula-Z AM3+ AMD 990FX SATA 6Gb/s USB 3.0 ATX AMD
AMD FD6300WMHKBOX FX-6300 6-Core Processor Black Edition with Cooler Master Hyper 212 EVO - CPU Cooler with 120mm PWM Fan
three gtx 1080,
one gtx 1080 TI on a powered header

Re: PCI-e bandwidth/capacity limitations

Post by SteveWillis »

My apologies as I imagine this has been answered before but I can't find it. With a 1080 TI what, if anything, do you lose with pcie V2 vs V3?
Image

1080 and 1080TI GPUs on Linux Mint
Nathan_P
Posts: 1164
Joined: Wed Apr 01, 2009 9:22 pm
Hardware configuration: Asus Z8NA D6C, 2 x5670@3.2 Ghz, , 12gb Ram, GTX 980ti, AX650 PSU, win 10 (daily use)

Asus Z87 WS, Xeon E3-1230L v3, 8gb ram, KFA GTX 1080, EVGA 750ti , AX760 PSU, Mint 18.2 OS

Not currently folding
Asus Z9PE- D8 WS, 2 E5-2665@2.3 Ghz, 16Gb 1.35v Ram, Ubuntu (Fold only)
Asus Z9PA, 2 Ivy 12 core, 16gb Ram, H folding appliance (fold only)
Location: Jersey, Channel islands

Re: PCI-e bandwidth/capacity limitations

Post by Nathan_P »

at x16 maybe 1-2% depending on the project.
Image
SteveWillis
Posts: 389
Joined: Fri Apr 15, 2016 12:42 am
Hardware configuration: PC 1:
Linux Mint 17.3
three gtx 1080 GPUs One on a powered header
Motherboard = [MB-AM3-AS-SB-990FXR2] qty 1 Asus Sabertooth 990FX(+59.99)
CPU = [CPU-AM3-FX-8320BR] qty 1 AMD FX 8320 Eight Core 3.5GHz(+41.99)

PC2:
Linux Mint 18
Open air case
Motherboard: ASUS Crosshair V Formula-Z AM3+ AMD 990FX SATA 6Gb/s USB 3.0 ATX AMD
AMD FD6300WMHKBOX FX-6300 6-Core Processor Black Edition with Cooler Master Hyper 212 EVO - CPU Cooler with 120mm PWM Fan
three gtx 1080,
one gtx 1080 TI on a powered header

Re: PCI-e bandwidth/capacity limitations

Post by SteveWillis »

Nathan_P wrote:at x16 maybe 1-2% depending on the project.
Thanks Nathan_P!
Image

1080 and 1080TI GPUs on Linux Mint
Sven
Posts: 71
Joined: Fri Nov 01, 2013 8:12 pm

Mining/F@H only Graphiccards

Post by Sven »

These Graphiccards are specially made for mining.

Eventually there also interesting for Folding@Home, depending on the pricepoint.

https://www.techpowerup.com/234113/nvid ... their-gpus
Sven
Posts: 71
Joined: Fri Nov 01, 2013 8:12 pm

Re: Mining/F@H only Graphiccards

Post by Sven »

And that's how one look like:

https://videocardz.com/newz/first-look- ... ng-station
Aurum
Posts: 292
Joined: Sat Oct 03, 2015 3:15 pm
Location: The Great Basin

Re: Mining/F@H only Graphiccards

Post by Aurum »

Horrible idea!!! Miners will take these and overvolt and overclock them and burn them up. Never buy one used as it will be near death. I've yet to see they offer any benefit for folding.
In Science We Trust Image
Aurum
Posts: 292
Joined: Sat Oct 03, 2015 3:15 pm
Location: The Great Basin

Re: PCI-e bandwidth/capacity limitations

Post by Aurum »

SteveWillis wrote:
ComputerGenie wrote:Save $ and get something like: http://a.co/jfI80yM
At least two reports of ruined motherboards from these in the reviews? No I don't think so. Quality powered risers are hard to find but you can tell these are quality just by looking at them.
http://www.ebay.com/itm/261459534757?_t ... EBIDX%3AIT
I have two with no problems. You get what you pay for, at least on a good day.

if you are looking specifically for unpowered risers I like these but they are out of stock right now.
https://smile.amazon.com/gp/product/B01 ... UTF8&psc=1
I've fried two motherboards using a new powered riser. I didn't realize the riser was the problem and tried it on a second mb.
Too bad there's not a tester.
The blue tape is just there to hide the terrible Chinese workmanship. The cables are soldered off center and at an angle to connector plug. From what I've seen there appears to be no QA or inspection process whatsoever, just pack'em and ship'em.
When I get a new unpowered riser I lay it flat on my workbench and peel off the silly blue tape. I inspect both sides with a magnifying glass for shorted neighbors and broken connections, if any then trash. If not, I squeeze on a thick coat of Locite Vinyl, Fabric & Plastic transparent adhesive and leave it sit for a day to cure. Then I test it on an old rig.
Bending a riser at the ribbon cable connector joint is very likely to break solder joints or even the wire. The ribbon cables are way too stiff for this function and not meant to be bent.
Last edited by Aurum on Tue Jun 13, 2017 1:27 pm, edited 1 time in total.
In Science We Trust Image
Post Reply