• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

bigfootnetworks is a joke

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
This is so ridiculous. They should offer interent, game servers, routers, and networking cable before they put this out.
 
There are similar products that have been around for years and exist for good reason - in highly loaded servers the network overhead is significant. For your average gamer who's using perhaps 2% of their adaptor's available bandwidth, it's a waste. Nevertheless, there are 1440 new customers per day.
 
i read a interview on the people selling this nic (note every image ive seen of the "killer nic" are all generated images i've YET to see a phyiscall prototype image of a card.) the suguested retail price is going to be around 300 dollars 275 to be more exact (somewhere near that price range).


i'd rather buy a new video card with that amound of cash :bday:
 
The cards are ready and we will see how they work. I interviewed the guy not that long ago and he is very sharp. All of those high end Intel NICs that you have been using since Win 98...he designed them.

With that said give the guy a chance, he is the expert. I can't wait to get a card in for testing...then again, I won't be paying for it either.
 
That may be the case, but does this not remain a solution in search of a problem? The size of the pipe has always been the primary limiting factor and as such games have evolved with spartan bandwidth requirements. Unless your node is a server working at almost full utilization of gigE, there really won't be many CPU cycles to save especially when cost is factored in. Perhaps this card will also run IOS, with 3DES tunnels to and from the game clients, firewall rules preventing undesired access, and have Quicken acceleration to balance my checkbook and monitor investments while I play?
 
Bigfootnetworks said:
...PCI 32-bit is capable of tremendous throughput capabilities approaching gigabit speeds in unshared PCI buses.

Highendtoys said:
...With that said give the guy a chance, he is the expert. I can't wait to get a card in for testing...then again, I won't be paying for it either.

Balderdash and poppycock...Even a neophyte in our community realizes that the bandwidth of the PCI bus is 133 MB/S. The author of the "white paper" regarding this card has more smoke and mirrors than Houdinini...
 
If this card in a gaming PC reduces the ping more than 0.1 seconds, the game is misdesigned and misprogrammed. And then not even a gold plated network card can save your playing.
 
hafa said:
Balderdash and poppycock...Even a neophyte in our community realizes that the bandwidth of the PCI bus is 133 MB/S. The author of the "white paper" regarding this card has more smoke and mirrors than Houdinini...
Gigabit = 1000 megaBITS.
Now we divide by 8 to convert to megaBYTES.
1000 megabits / 8 = 125 MB/S.
The PCI bus is capable of full gigabit speeds, as long as no other devices are sharing the bandwidth. Poppycock indeed :rolleyes:
 
Yet the advertisement is for shared busses. It's a rare day when there's more than one device on the bus and it's completely quiet. The PCI bus protocol is not particularly efficient for small data transfers, as it's pretty much a fixed cost per transfer. Larger transfers increase the efficiency but degrade performance if you're sharing the bus with other devices, especially isochronous devices. Supporting a large on-controller cache and using a TOE will reduce the amount of the smaller burst communications on the PCI bus that are common with TCP networks and reduce processor loading, but again, what does this gain in a desktop or workstation? Cards with TOEs are critical at multi-gigabit speeds, but at lower than that they're largerly expensive toys. Increasing PCI efficiency by using larger transfer sizes is a bad idea if the bus is shared with other isochronous devices (sound cards et al). It's also something of a moot point with the transition to P2P peripheral links since there's nothing shared and more than adequate bandwidth available.
 
someone who today releases a supposedly ultra high performance network card for the PCI bus and not PCI-E simply has lost all credibility for any performance claims

It's been known since the first GBitE card that the 32bit PCI bus is not enough for such a network card. Intel designed their CSI specially for GBit ethernet in their chipsets just to avoid the PCI bottleneck.
 
PCI is a known quantity and a stepping stone to PCIe and PCI-X. All undertakings in engineering are iterative efforts.
 
Back