The killer gaming nic :laugh;

[Goosemaster]

Originally posted by: BD2003

Originally posted by: Goosemaster

Originally posted by: BD2003

Originally posted by: spidey07
Goose,

Trying to explain networking to PC guys is an effort in futility. They just don't get it.

I'll let some smart person do the math...

1) Calculate the serialiazation on a common 100 Base-T network.
2) Calculate the the propogation delay on the length of the cable
3) Calculate the hold time/delay of a typical broadband modem.
4) Calculate the serialzation plus latency to the next hop.
5) repeat until you reach the destination.
6) You have now determined end-to-end latency, from the switching delay in ideal situations, with no buffering.

Then prove to me that your NIC can control this. Then do it for the return path.

Until I see pure sniffer analysis of this so called netwok card i will continue to debunk it as "bullcrap"

And you're still trying to burn down the strawman. I haven't seen anyone here claim that actually believe the unbelieveable, that it will reduce lag when theres no other network traffic present.

You're absolutely right, as you said, under *ideal* conditions. Which aren't always present. When you've got a 500k up/down bittorrent stream choking your connection, it's far from ideal. If it can prioritize the gaming packets, then clearly there will be less lag, under *those conditions*. Can it ACTUALLY do that? I don't know, I dont have one...but the theory is sound. I haven't even read the marketing...lags per second sounds ridiculous.

And regardless of how miniscule the CPU usage of your standard NIC is, it does not take a mathematician to figure out that even 1% more FPS is still a benefit. An absurd one, but not too far from what you'd get by tightening your memory timings.

I know you all want to be right, so badly. And you are, on the points you are making. But that isnt the entire story. It's horrifically overpriced, compared to the admittedly minimal theoretical benefits.

Any true "gamer" isn't going to purposely take resources away from their comp if they have to. Does this card magically process the bitorrent files for you (perhaps, but its current iteration? NO!) Does it magically take care of disk access when your copy of bf2 is setting up the 1.3GB of files that are going to flood the ram?

So many are defending this thing based on specualtion, and frankly, that is just wrong. Worst of all, it seems that this company has done everything they can to make logical folk who care what the hell crawls out of their wallet, bleed from their rectum.

Their jargon, their whitepaper, their website, their shills.

We've seen it before. Sure we are basing our assumptions on specualtion, but the difference is that we are basign it on knowledge and past experiences as well.

I'm not claiming it can generate any performance benefit but offload the *miniscule* CPU load that your average NIC will generate. Keep throwing up strawmen.

I'm not even defending it. I don't think ANYONE should buy it unless they have money burning a hole in their pocket, and even then, it will still likely be better spent literally being burnt.

An honest question: Say youre transferring a file over the network, that is completely saturating your ethernet connection. Ignore any other bottlenecks such as disk or memory I/O. Ignore even the context of gaming. Assume you want to make a skype call, or surf the net, or any random internet usage.

Under those conditions of the file transfer saturating your connection, wouldn't your internet access be slowed down? Isnt there any sort of prioritization scheme that would let your more crucial packets through when you need them?

switches have QoS. Concievably, one could say that having data prioritized before it leaves the machine sounds quite nice, and that consumer switches don't have QoS and those people need to crack open a TCP/IP book before they start arguing that fact. I would defintiely rely on people like spidey, cmets, ramirez, scottmac, and the like to wield legit answers to such questions.

That said, one also needs to be realistic. What average gamer would be saturating a gigabit connection while palying game In lien with the facts, one also has to be realistic about the human element. These are the types of people who turn off SERVICES to get better performance. These are the type of people who would havea little NAS implementatiosn in there house. They would NOT be uploading @ 70MB/s from their machines, if not out of F.E.A.R that the massive transfers are taxign their HDs. gigabit is very fast....faster than msot people's storage solutions.

 

[Baked]

I'm gonna get 2 to fill up the PCI slots on my mobo. Gotta love the giant Killer K logo. That alone is worth $280.

 

[BD2003]

Originally posted by: Goosemaster

Originally posted by: BD2003

Originally posted by: Goosemaster

Originally posted by: BD2003

Originally posted by: spidey07
Goose,

Trying to explain networking to PC guys is an effort in futility. They just don't get it.

I'll let some smart person do the math...

1) Calculate the serialiazation on a common 100 Base-T network.
2) Calculate the the propogation delay on the length of the cable
3) Calculate the hold time/delay of a typical broadband modem.
4) Calculate the serialzation plus latency to the next hop.
5) repeat until you reach the destination.
6) You have now determined end-to-end latency, from the switching delay in ideal situations, with no buffering.

Then prove to me that your NIC can control this. Then do it for the return path.

Until I see pure sniffer analysis of this so called netwok card i will continue to debunk it as "bullcrap"

And you're still trying to burn down the strawman. I haven't seen anyone here claim that actually believe the unbelieveable, that it will reduce lag when theres no other network traffic present.

You're absolutely right, as you said, under *ideal* conditions. Which aren't always present. When you've got a 500k up/down bittorrent stream choking your connection, it's far from ideal. If it can prioritize the gaming packets, then clearly there will be less lag, under *those conditions*. Can it ACTUALLY do that? I don't know, I dont have one...but the theory is sound. I haven't even read the marketing...lags per second sounds ridiculous.

And regardless of how miniscule the CPU usage of your standard NIC is, it does not take a mathematician to figure out that even 1% more FPS is still a benefit. An absurd one, but not too far from what you'd get by tightening your memory timings.

I know you all want to be right, so badly. And you are, on the points you are making. But that isnt the entire story. It's horrifically overpriced, compared to the admittedly minimal theoretical benefits.

Any true "gamer" isn't going to purposely take resources away from their comp if they have to. Does this card magically process the bitorrent files for you (perhaps, but its current iteration? NO!) Does it magically take care of disk access when your copy of bf2 is setting up the 1.3GB of files that are going to flood the ram?

So many are defending this thing based on specualtion, and frankly, that is just wrong. Worst of all, it seems that this company has done everything they can to make logical folk who care what the hell crawls out of their wallet, bleed from their rectum.

Their jargon, their whitepaper, their website, their shills.

We've seen it before. Sure we are basing our assumptions on specualtion, but the difference is that we are basign it on knowledge and past experiences as well.

I'm not claiming it can generate any performance benefit but offload the *miniscule* CPU load that your average NIC will generate. Keep throwing up strawmen.

I'm not even defending it. I don't think ANYONE should buy it unless they have money burning a hole in their pocket, and even then, it will still likely be better spent literally being burnt.

An honest question: Say youre transferring a file over the network, that is completely saturating your ethernet connection. Ignore any other bottlenecks such as disk or memory I/O. Ignore even the context of gaming. Assume you want to make a skype call, or surf the net, or any random internet usage.

Under those conditions of the file transfer saturating your connection, wouldn't your internet access be slowed down? Isnt there any sort of prioritization scheme that would let your more crucial packets through when you need them?

switches have QoS. Concievably, one could say that having data prioritized before it leaves the machine sounds quite nice, and that consumer switches don't have QoS and those people need to crack open a TCP/IP book before they start arguing that fact. I would defintiely rely on people like spidey, cmets, ramirez, scottmac, and the like to wield legit answers to such questions.

That said, one also needs to be realistic. What average gamer would be saturating a gigabit connection while palying game In lien with the facts, one also has to be realistic about the human element. These are the types of people who turn off SERVICES to get better performance. These are the type of people who would havea little NAS implementatiosn in there house. They would NOT be uploading @ 70MB/s from their machines, if not out of F.E.A.R that the massive transfers are taxign their HDs. gigabit is very fast....faster than msot people's storage solutions.

If one were being realistic, they wouldn't even be considering this card. Obviously we have no control over what the ISP sends down, only what we send up. But thats half of the battle when we're talking about pings, when those packets are competing with other less important ones. I believe NVIDIA was introducing a similar tech with their new nforce chipset. Same idea with the "hardware firewall" as well. I forget which company has it, but there was even a router that actually has its own bitorrent client + storage.


If someone wants to pull rank to address the actual issue of upload prioritization on a taxed connection, and not just feed their ego about how they KNOW they're right, then I'm all ears. But all I see is the same point about ideal end to end bandwidth being addressed over and over, and calls for realism when I'm only interested if there is a *theoretical* benefit.

I'm just curious as to why the theory behind it seems to be offending people and getting them all upset over a NIC. I still can't see how there is "*********ZERO******* net benefit".

By the same vein, is the "D-Link gaming router" pure BS too? It has many similar claims.

 

[Goosemaster]

Originally posted by: BD2003

Originally posted by: Goosemaster

Originally posted by: BD2003

Originally posted by: Goosemaster

Originally posted by: BD2003

Originally posted by: spidey07
Goose,

Trying to explain networking to PC guys is an effort in futility. They just don't get it.

I'll let some smart person do the math...

1) Calculate the serialiazation on a common 100 Base-T network.
2) Calculate the the propogation delay on the length of the cable
3) Calculate the hold time/delay of a typical broadband modem.
4) Calculate the serialzation plus latency to the next hop.
5) repeat until you reach the destination.
6) You have now determined end-to-end latency, from the switching delay in ideal situations, with no buffering.

Then prove to me that your NIC can control this. Then do it for the return path.

Until I see pure sniffer analysis of this so called netwok card i will continue to debunk it as "bullcrap"

And you're still trying to burn down the strawman. I haven't seen anyone here claim that actually believe the unbelieveable, that it will reduce lag when theres no other network traffic present.

You're absolutely right, as you said, under *ideal* conditions. Which aren't always present. When you've got a 500k up/down bittorrent stream choking your connection, it's far from ideal. If it can prioritize the gaming packets, then clearly there will be less lag, under *those conditions*. Can it ACTUALLY do that? I don't know, I dont have one...but the theory is sound. I haven't even read the marketing...lags per second sounds ridiculous.

And regardless of how miniscule the CPU usage of your standard NIC is, it does not take a mathematician to figure out that even 1% more FPS is still a benefit. An absurd one, but not too far from what you'd get by tightening your memory timings.

I know you all want to be right, so badly. And you are, on the points you are making. But that isnt the entire story. It's horrifically overpriced, compared to the admittedly minimal theoretical benefits.

Any true "gamer" isn't going to purposely take resources away from their comp if they have to. Does this card magically process the bitorrent files for you (perhaps, but its current iteration? NO!) Does it magically take care of disk access when your copy of bf2 is setting up the 1.3GB of files that are going to flood the ram?

So many are defending this thing based on specualtion, and frankly, that is just wrong. Worst of all, it seems that this company has done everything they can to make logical folk who care what the hell crawls out of their wallet, bleed from their rectum.

Their jargon, their whitepaper, their website, their shills.

We've seen it before. Sure we are basing our assumptions on specualtion, but the difference is that we are basign it on knowledge and past experiences as well.

I'm not claiming it can generate any performance benefit but offload the *miniscule* CPU load that your average NIC will generate. Keep throwing up strawmen.

I'm not even defending it. I don't think ANYONE should buy it unless they have money burning a hole in their pocket, and even then, it will still likely be better spent literally being burnt.

An honest question: Say youre transferring a file over the network, that is completely saturating your ethernet connection. Ignore any other bottlenecks such as disk or memory I/O. Ignore even the context of gaming. Assume you want to make a skype call, or surf the net, or any random internet usage.

Under those conditions of the file transfer saturating your connection, wouldn't your internet access be slowed down? Isnt there any sort of prioritization scheme that would let your more crucial packets through when you need them?

switches have QoS. Concievably, one could say that having data prioritized before it leaves the machine sounds quite nice, and that consumer switches don't have QoS and those people need to crack open a TCP/IP book before they start arguing that fact. I would defintiely rely on people like spidey, cmets, ramirez, scottmac, and the like to wield legit answers to such questions.

That said, one also needs to be realistic. What average gamer would be saturating a gigabit connection while palying game In lien with the facts, one also has to be realistic about the human element. These are the types of people who turn off SERVICES to get better performance. These are the type of people who would havea little NAS implementatiosn in there house. They would NOT be uploading @ 70MB/s from their machines, if not out of F.E.A.R that the massive transfers are taxign their HDs. gigabit is very fast....faster than msot people's storage solutions.

If one were being realistic, they wouldn't even be considering this card. Obviously we have no control over what the ISP sends down, only what we send up. But thats half of the battle when we're talking about pings, when those packets are competing with other less important ones. I believe NVIDIA was introducing a similar tech with their new nforce chipset. Same idea with the "hardware firewall" as well. I forget which company has it, but there was even a router that actually has its own bitorrent client + storage.


If someone wants to pull rank to address the actual issue of upload prioritization on a taxed connection, and not just feed their ego about how they KNOW they're right, then I'm all ears. But all I see is the same point being argued over and over, and calls for realism when I'm only interested if there is a *theoretical* benefit.

I'm just curious as to why the theory behind it seems to be offending people and getting them all upset over a NIC. I still can't see how there is "*********ZERO******* net benefit".

By the same vein, is the "D-Link gaming router" pure BS too? It has many similar claims.

no it is not. while it includes a lot of marketing, it is legit. Quality of Service (QoS) and overall traffic managment are the foundations for well-run networks, and will it incldues some rudimentary forms of it, they are very very effective given the type of data that passes through them. These devices are what matter most, assumgin that your WAN link is in good shape.


Basically, your 10/100/1000 links are serial, so 1 by one, packets travel in series. Even so, check your ping times to your rotuer when you are "saturating your connection."

<---currently downloading a file off of his server @ ~ 84Mb/s, which hoenstly, is n't that fast

Still, at least for 10/100, using an integrated controller, I am pinging the living hell out of everything and am getting <1ms pings.

While windows networking is responsible for the 38% cpu usage (a64 3000+), I won't give this card my ear at ALL until we get some legit numbers.

 

[EyeMWing]

Word on the street is that it's either a viral marketing experiment, or a RTL-8160 in a fancy suit.

If it does indeed actually exist, I'll attempt to requisition some funds and reverse-engineer it.

 

[Goosemaster]

Originally posted by: EyeMWing
Word on the street is that it's either a viral marketing experiment, or a RTL-8160 in a fancy suit.

If it does indeed actually exist, I'll attempt to requisition some funds and reverse-engineer it.

Jsut read up on those realteks...interesting.



http://www.csie.nctu.edu.tw/~cfliu/work/8650.htm ...neat stuff.


http://www.realtek.com.tw/products/products1-2.aspx?modelid=2006036

 

[SampSon]

BD2003 and others: I'm not sure what exactly any of you are trying to argue. You're arguing in very broad, simplistic, psuedo-theoretic manner that uses a lot of words to say very little.

I'm not sure if you read what anybody else wrote about them but..... you wanna buy some monster cables?

 

[EyeMWing]

Originally posted by: Goosemaster

Originally posted by: EyeMWing
Word on the street is that it's either a viral marketing experiment, or a RTL-8160 in a fancy suit.

If it does indeed actually exist, I'll attempt to requisition some funds and reverse-engineer it.

Jsut read up on those realteks...itneresting.

Features are eerily similar, eh?

The 8160 is a gigabit version of the 8139E, which introduced the coprocessor concept. Every other feature of this NIC is doable in drivers.

Realtek fell out of favor when the gigabit transition happened, because Marvell chipsets are a few cents cheaper, so everybody has Marvell on their motherboards instead of Realtek, these days.

 

[Goosemaster]

Originally posted by: EyeMWing

Originally posted by: Goosemaster

Originally posted by: EyeMWing
Word on the street is that it's either a viral marketing experiment, or a RTL-8160 in a fancy suit.

If it does indeed actually exist, I'll attempt to requisition some funds and reverse-engineer it.

Jsut read up on those realteks...itneresting.

Features are eerily similar, eh?

The 8160 is a gigabit version of the 8139E, which introduced the coprocessor concept. Every other feature of this NIC is doable in drivers.

Realtek fell out of favor when the gigabit transition happened, because Marvell chipsets are a few cents cheaper, so everybody has Marvell on their motherboards instead of Realtek, these days.

only thing that seems weird is the processor speed. Maybe they are pullign the 400mhz out of their ass sicne this is a risc chip?




as for the info on realtek/marvell...thanks learned somethig new today

What other routing-capable chipsets are there out there? Any specific products that have come out of them?


<--has a soekris and is imrpessed with it to this da...50Mbps is a lot ...too bad that even with a vpn accelerator the vpn traffic is a paltry 1-3Mbps

 

[spidey07]

Originally posted by: BD2003
And regardless of how miniscule the CPU usage of your standard NIC is, it does not take a mathematician to figure out that even 1% more FPS is still a benefit. An absurd one, but not too far from what you'd get by tightening your memory timings.

I know you all want to be right, so badly. And you are, on the points you are making. But that isnt the entire story. It's horrifically overpriced, compared to the admittedly minimal theoretical benefits.

I see dumb people everywhere. They don't even know that they are dumb.

Have you figured out that latency yet?

 

[Goosemaster]

seeing as how I am studying, I probably should get more into embedded pcs than m0n0wall Whiel I will probably setup a freeSAN box up in the future, all of this is childplay compared to this stuff


Whiel I could setup the serial controller so that I can install freebsd myself onto the soekris and screw around, it is my gateway rotuer so right now I can't

 

[Goosemaster]

Originally posted by: spidey07

Originally posted by: BD2003
And regardless of how miniscule the CPU usage of your standard NIC is, it does not take a mathematician to figure out that even 1% more FPS is still a benefit. An absurd one, but not too far from what you'd get by tightening your memory timings.

I know you all want to be right, so badly. And you are, on the points you are making. But that isnt the entire story. It's horrifically overpriced, compared to the admittedly minimal theoretical benefits.

I see dumb people everywhere. They don't even know that they are dumb.

Have you figured out that latency yet?

I thought you were goign to bed.....

or was that jack?

 

[spidey07]

Originally posted by: BD2003
If one were being realistic, they wouldn't even be considering this card. Obviously we have no control over what the ISP sends down, only what we send up. But thats half of the battle when we're talking about pings, when those packets are competing with other less important ones. I believe NVIDIA was introducing a similar tech with their new nforce chipset. Same idea with the "hardware firewall" as well. I forget which company has it, but there was even a router that actually has its own bitorrent client + storage.


If someone wants to pull rank to address the actual issue of upload prioritization on a taxed connection, and not just feed their ego about how they KNOW they're right, then I'm all ears. But all I see is the same point about ideal end to end bandwidth being addressed over and over, and calls for realism when I'm only interested if there is a *theoretical* benefit.

I'm just curious as to why the theory behind it seems to be offending people and getting them all upset over a NIC. I still can't see how there is "*********ZERO******* net benefit".

By the same vein, is the "D-Link gaming router" pure BS too? It has many similar claims.

If one were to be realistic one would have a fargin clue.

Just stop. stop it. stop.

 

[EyeMWing]

Originally posted by: Goosemaster

Originally posted by: EyeMWing

Originally posted by: Goosemaster

Originally posted by: EyeMWing
Word on the street is that it's either a viral marketing experiment, or a RTL-8160 in a fancy suit.

If it does indeed actually exist, I'll attempt to requisition some funds and reverse-engineer it.

Jsut read up on those realteks...itneresting.

Features are eerily similar, eh?

The 8160 is a gigabit version of the 8139E, which introduced the coprocessor concept. Every other feature of this NIC is doable in drivers.

Realtek fell out of favor when the gigabit transition happened, because Marvell chipsets are a few cents cheaper, so everybody has Marvell on their motherboards instead of Realtek, these days.

only thing that seems weird is the processor speed. Maybe they are pullign the 400mhz out of their ass sicne this is a risc chip?




as for the info on realtek/marvell...thanks learned somethig new today

What other routing-capable chipsets are there out there? Any specific products that have come out of them?


<--has a soekris and is imrpessed with it to this da...50Mbps is a lot ...too bad that even with a vpn accelerator the vpn traffic is a paltry 1-3Mbps

You're looking at the wrong Realtek chips, those are the 8650-series, not the 8160-series.

The 8650 are a routing solution, integrating a NIC chipset (RTL8139 in that case), a CPU (MIPS), and a switching solution into the same chip. You'll find one (or a similar, Broadcom model, mostly) in EVERY consumer router on the market

The 8160 are an actual NIC chipset, and that's what we're likely to find if we take apart this "Killer NIC"

And as for "400MHz!!!!", all you need to do to be able to legitimately claim that is to put an oscillator on the board somewhere.

 

[Goosemaster]

Originally posted by: EyeMWing

Originally posted by: Goosemaster

Originally posted by: EyeMWing

Originally posted by: Goosemaster

Originally posted by: EyeMWing
Word on the street is that it's either a viral marketing experiment, or a RTL-8160 in a fancy suit.

If it does indeed actually exist, I'll attempt to requisition some funds and reverse-engineer it.

Jsut read up on those realteks...itneresting.

Features are eerily similar, eh?

The 8160 is a gigabit version of the 8139E, which introduced the coprocessor concept. Every other feature of this NIC is doable in drivers.

Realtek fell out of favor when the gigabit transition happened, because Marvell chipsets are a few cents cheaper, so everybody has Marvell on their motherboards instead of Realtek, these days.

only thing that seems weird is the processor speed. Maybe they are pullign the 400mhz out of their ass sicne this is a risc chip?




as for the info on realtek/marvell...thanks learned somethig new today

What other routing-capable chipsets are there out there? Any specific products that have come out of them?


<--has a soekris and is imrpessed with it to this da...50Mbps is a lot ...too bad that even with a vpn accelerator the vpn traffic is a paltry 1-3Mbps

You're looking at the wrong Realtek chips, those are the 8650-series, not the 8160-series.

The 8650 are a routing solution, integrating a NIC chipset (RTL8139 in that case), a CPU (MIPS), and a switching solution into the same chip. You'll find one (or a similar, Broadcom model, mostly) in EVERY consumer router on the market

The 8160 are an actual NIC chipset, and that's what we're likely to find if we take apart this "Killer NIC"

And as for "400MHz!!!!", all you need to do to be able to legitimately claim that is to put an oscillator on the board somewhere.

oh....wow, that is bad. I thought they had a mini-rotuer thign goign on there.




Also, how much do you think the Marvell Libertas AP-32 would go for and what languages would I have to know to do anything with it? It looks like the controller is ready to go out of the box in terms of having/not havng to add any harware immediately to do soemthign with it.


<--only knows a pathetic amount of c++ and unix shell scripting




edit: why in the hell did I have to pick THIS thread to post a serious question?! :laugh:

 

[EyeMWing]

Originally posted by: Goosemaster

Originally posted by: EyeMWing

Originally posted by: Goosemaster

Originally posted by: EyeMWing

Originally posted by: Goosemaster

Originally posted by: EyeMWing
Word on the street is that it's either a viral marketing experiment, or a RTL-8160 in a fancy suit.

If it does indeed actually exist, I'll attempt to requisition some funds and reverse-engineer it.

Jsut read up on those realteks...itneresting.

Features are eerily similar, eh?

The 8160 is a gigabit version of the 8139E, which introduced the coprocessor concept. Every other feature of this NIC is doable in drivers.

Realtek fell out of favor when the gigabit transition happened, because Marvell chipsets are a few cents cheaper, so everybody has Marvell on their motherboards instead of Realtek, these days.

only thing that seems weird is the processor speed. Maybe they are pullign the 400mhz out of their ass sicne this is a risc chip?




as for the info on realtek/marvell...thanks learned somethig new today

What other routing-capable chipsets are there out there? Any specific products that have come out of them?


<--has a soekris and is imrpessed with it to this da...50Mbps is a lot ...too bad that even with a vpn accelerator the vpn traffic is a paltry 1-3Mbps

You're looking at the wrong Realtek chips, those are the 8650-series, not the 8160-series.

The 8650 are a routing solution, integrating a NIC chipset (RTL8139 in that case), a CPU (MIPS), and a switching solution into the same chip. You'll find one (or a similar, Broadcom model, mostly) in EVERY consumer router on the market

The 8160 are an actual NIC chipset, and that's what we're likely to find if we take apart this "Killer NIC"

And as for "400MHz!!!!", all you need to do to be able to legitimately claim that is to put an oscillator on the board somewhere.

oh....wow, that is bad. I thought they had a mini-rotuer thign goign on there.




Also, how much do you think the Marvell Libertas AP-32 would go for and what languages would I have to know to do anything with it? It looks like the controller is ready to go out of the box in terms of having/not havng to add any harware immediately to do soemthign with it.


<--only knows a pathetic amount of c++ and unix shell scripting




edit: why in the hell did I have to pick THIS thread to post a serious question?! :laugh:

If you're looking for a fun embedded system for n00bs, grab an older-revision Linksys WRT54g. They use various Broadcom chipsets, and can run a BSD varient called OpenWRT (the new revisions - V4 and later, I think - run VXWorks, and OpenWRT hasn't yet been ported to that hardware, since the vxworks implementation is closed source and thus it's harder to exploit it to flash the damn thing)

 

[Rubycon]

This thread is worth a million dollars + now.

That's the environmental fine for dumping snake oil into the sea. Where the hell is Loke?

:laugh:

 

[chuckywang]

An elite gamer feature, PingThrottle allows gamers to restore the sense of fair play when there is a server proximity advantage. The gamer with the unfair advantage can dial the ping back up, so that they can win without worrying about being the LPB (low Ping bastard). Also, elite gamers can use this feature to handicap themselves for training purposes. This feature, if activated by the user, utilizes hardware to add in up to a maximum 40ms of Ping, and cannot be used to cheat in an online game.

I'm surprised they didn't spell it as "l33t" in the official product description.

 

[BD2003]

Originally posted by: spidey07

Originally posted by: BD2003
And regardless of how miniscule the CPU usage of your standard NIC is, it does not take a mathematician to figure out that even 1% more FPS is still a benefit. An absurd one, but not too far from what you'd get by tightening your memory timings.

I know you all want to be right, so badly. And you are, on the points you are making. But that isnt the entire story. It's horrifically overpriced, compared to the admittedly minimal theoretical benefits.

I see dumb people everywhere. They don't even know that they are dumb.

Have you figured out that latency yet?

I've known that from the start, and never said otherwise. Keep feeding your ego. You're the one coming off as an asshole.

 

[yllus]

Amusing product, and I can't really complain about another programming interface to muck around in. That could be fun.

If this NIC improves pings by 1ms, I'll be impressed. But it's worse than claiming that running Winamp on today's processors is slowing you down - the CPU time required approaches zero, and what is offloaded still does require processing time, just on a different (and slower) processor. So where exactly are you saving anything?

 

[darkamulets]

How much you wanna bet you rip off the heatsink and it's a Realtek chip?

 

[flexy]

it strikes me as odd that on a place like AT (where i assume a high number of computer-literates) is such an excitement over
"a new device/NIC which runs off it's own OS and has network prioritizing capabilities"

Such a device is called a fvcking ROUTER (see other comments in GHW)...and for $70 you can get a
"device which runs its own Os [Linux], has QoS/traffic shaping, firewall, NAT *and* an USB port...eg. a wrtg54G router and flash it with dd-wrt.

In GHW i already let out my disbelief over that BS how a NIC (or a router for that matter can (in any way) influence the whole internet/tcp network-topology... eg. point A (me) to a server with XYZ internet nodes/switches in between.

If i ping to a game sever and get, say, 100ms or higher ....then the LAST place i'd look for the high ping's reason is *my* PC or my NIC

Despite the fact that such a miracle NIC of course cannot do anything regarding ping-times eg. to the other players.
Much use (just theretically assuming) that it would reduce my ping...other people would still have a high ping, because they dont use the miracle NIC and/or arre located in swasiland

Also...funny idea to market this for gamers "because it reduces pings" and then having a feature to have a P2P/BT client running on it.
If i were to look at my pings i for sure would not load the net with some additional BT traffic while i am gaming.

 

[SZLiao214]

Anandtech finally reviewed it hehe.

 

[DaveatBigfoot]

Dave from Bigfoot Networks here. We wanted to reach out to comments and forums around the Internet, address some of the issues being discussed, and be available for any questions you may have.

I worked with Gary while he was writing this review. We have a tremendous amount of respect for him and Anandtech.com. I'd be liar if I didn't admit that we were disappointed with the performance and experience that the Anandtech review reflects. We welcomed the "Pepsi Challenge", and appreciated the real-world approach taken.

While the performance numbers reported were lower than what our customers report, and what we see internally, we thought one of the best testimonials for the Killer was the blind test where a the Killer was added to gamers PC without his knowledge, and he thought there was a new video card or more RAM in the system. Truly, that is what the Killer is all about...smoother, faster gaming...less lag, better performance.

Back when this review was written, we did have some issues with our drivers. I believe each and every issue manifested itself during Anandtech's testing. It was very unfortunate and not anticipated. Bypassing the windows network stack and putting a Linux computer on a PCI slot is a bit tricky. We aren't using that as an excuse, just stating it as a fact. Our latest software suite addresses all the issues that are referenced in this review.

We have also recently released IPtables firewall for the Killer NIC. Many more FNApps are on the way, and with time the Killer's value will increase. A rarity in the hardware world.

We sincerely hope, at some point, Anandtech will give the Killer another shot. We firmly stand by our product and believe it holds tremendous value for online gamers.

I am also happy to answer any questions you may have about the Killer, so fire away!