- Status
Noone seems to get it. I'd get one, but Until I have 2,000 saved up for my dream PC I won't get it. If you have anything less than a 4850 / Core i7 and you have a dedicated NIC I'll laugh at you.
Your benchmarks get a thumbs down and the "need" concern in the title/article is unnecessary bashing of the product. Thanks for wasting my time : concluding the other things that are obvious. Noone with an 800 dollar gaming PC budget should get one.
Your benchmarks get a thumbs down and the "need" concern in the title/article is unnecessary bashing of the product. Thanks for wasting my time : concluding the other things that are obvious. Noone with an 800 dollar gaming PC budget should get one.
bellbillsnow
Distinguished
- Jul 2, 2009
- 25
- 0
- 18,530
Cleeve,
Thanks for the response. We are just ships passing in the night I guess. I just can't keep re-explaining myself.
I was questioning that "Pings=lag". Pings send a weird empty ICMP packet. I have read in several places that this is not going to tell you much about how game play is affected. In fact, the NIC should have the least effect on Ping times because most of the travel is off your LAN. The Killer is addressing what happens in your game.
Finally, I do not think you results tell us much. I guess they work for you. From my stand point the article does not tell me much. It is kind of like someone putting up charts of census data and saying "Tide is the better cleaner" without explaining how the census tells us about cleaners. This is not benching CPU's. It is very complicated and there is nothing to indicate to me that your "tests" measure anything relevant. In fact, after too many posts, you still cannot even acknowledge what I am getting at! For example, high FPS's and low Ping times are worthless if the game engine cannot process where other characters are because that data is stuck in the network stack. But whatever, you "know" that high pings - lag, so no reason to think about it anymore.
The interesting thing here is that if your conclusions are correct than BigFoot is a fraud right? Why don't you find a lawyer to sue them?
Because your methodology does not prove that it doesn't work!
Thanks for the response. We are just ships passing in the night I guess. I just can't keep re-explaining myself.
I was questioning that "Pings=lag". Pings send a weird empty ICMP packet. I have read in several places that this is not going to tell you much about how game play is affected. In fact, the NIC should have the least effect on Ping times because most of the travel is off your LAN. The Killer is addressing what happens in your game.
Finally, I do not think you results tell us much. I guess they work for you. From my stand point the article does not tell me much. It is kind of like someone putting up charts of census data and saying "Tide is the better cleaner" without explaining how the census tells us about cleaners. This is not benching CPU's. It is very complicated and there is nothing to indicate to me that your "tests" measure anything relevant. In fact, after too many posts, you still cannot even acknowledge what I am getting at! For example, high FPS's and low Ping times are worthless if the game engine cannot process where other characters are because that data is stuck in the network stack. But whatever, you "know" that high pings - lag, so no reason to think about it anymore.
The interesting thing here is that if your conclusions are correct than BigFoot is a fraud right? Why don't you find a lawyer to sue them?
Because your methodology does not prove that it doesn't work!
- Mar 7, 2003
- 12,012
- 0
- 40,790
[citation][nom]bellbillsnow[/nom]Why don't you find a lawyer to sue them?[/citation]
My sarcasm detector is going off but it's not making sense. How has Killer Networks defrauded me? I haven't paid for one of their cards.
[citation][nom]bellbillsnow[/nom]Because your methodology does not prove that it doesn't work![/citation]
We'll have to agree to disagree on that one. I measured network latency and frame rate and detected no objective advantage over free software. From my experience, ping is a pretty good measurement of latency, as high ping rates have often accompanied lag and being kicked from servers.
I also experienced no subjective advantage.
I'm not sure what kind of test you would find satisfying.
My sarcasm detector is going off but it's not making sense. How has Killer Networks defrauded me? I haven't paid for one of their cards.
[citation][nom]bellbillsnow[/nom]Because your methodology does not prove that it doesn't work![/citation]
We'll have to agree to disagree on that one. I measured network latency and frame rate and detected no objective advantage over free software. From my experience, ping is a pretty good measurement of latency, as high ping rates have often accompanied lag and being kicked from servers.
I also experienced no subjective advantage.
I'm not sure what kind of test you would find satisfying.
bellbillsnow
Distinguished
- Jul 2, 2009
- 25
- 0
- 18,530
It is hard to believe that that Killer would do much to resolve high ping rates... most of that is on the network. And any NIC is going to get booted when ping rates are high... again most of it is on the network.
What the killer is supposed to do is reduce the fraction of a second that it takes to get the data from the NIC to the game. That has very little to do with Ping times.
I have said previously what would be satisfying: a blind experiment with pro gamers on $600 machines (killer ones would be $730). Have them play with them online and in a lan.
Or two machines side-by-side playing the same game at the same time.
Again, I am not saying that it can be tested... just that I don't you did.
What the killer is supposed to do is reduce the fraction of a second that it takes to get the data from the NIC to the game. That has very little to do with Ping times.
I have said previously what would be satisfying: a blind experiment with pro gamers on $600 machines (killer ones would be $730). Have them play with them online and in a lan.
Or two machines side-by-side playing the same game at the same time.
Again, I am not saying that it can be tested... just that I don't you did.
To the Aussie posters...
Australian and New Zealand Modems tend to have lower latency, because we are so far away the high end ones are built pretty well... Netcomm NB9W + Telstra = 240ms-260ms to the US if you set up QOS for each game. Was the same as a $1500 Cisco modem in lab tests. NIC to NB9W = 0ms so I don't see how a NIC could make this faster.
Australian and New Zealand Modems tend to have lower latency, because we are so far away the high end ones are built pretty well... Netcomm NB9W + Telstra = 240ms-260ms to the US if you set up QOS for each game. Was the same as a $1500 Cisco modem in lab tests. NIC to NB9W = 0ms so I don't see how a NIC could make this faster.
- Mar 7, 2003
- 12,012
- 0
- 40,790
[citation][nom]bellbillsnow[/nom]It is hard to believe that that Killer would do much to resolve high ping rates... [/citation]
From the Killer Xeno Pro reviewer's guide:
"There are no settings to adjust here - log into a server with a low ping value and a high number of gamers. After logging in, note the ping (latency) reported by the game"
You see, I followed Killer's suggested test procedure. Measuring the ping is exactly how they claim to measure latency. So if you believe their claims, then you believe that ping measures latency.
If you don't believe their claims, then there is no data whatsoever suggesting that the Killer Xeno Pro enhances network performance...
From the Killer Xeno Pro reviewer's guide:
"There are no settings to adjust here - log into a server with a low ping value and a high number of gamers. After logging in, note the ping (latency) reported by the game"
You see, I followed Killer's suggested test procedure. Measuring the ping is exactly how they claim to measure latency. So if you believe their claims, then you believe that ping measures latency.
If you don't believe their claims, then there is no data whatsoever suggesting that the Killer Xeno Pro enhances network performance...
bellbillsnow
Distinguished
- Jul 2, 2009
- 25
- 0
- 18,530
[citation][nom]Cleeve[/nom]From the Killer Xeno Pro reviewer's guide:"There are no settings to adjust here - log into a server with a low ping value and a high number of gamers. After logging in, note the ping (latency) reported by the game"You see, I followed Killer's suggested test procedure. Measuring the ping is exactly how they claim to measure latency. So if you believe their claims, then you believe that ping measures latency.If you don't believe their claims, then there is no data whatsoever suggesting that the Killer Xeno Pro enhances network performance...[/citation]
Nice piece of logic work there. I take from this that the "Experts" at Toms hardware have nothing to contribute to the conversation beyond validating what hardware manufacturer's claim. I will keep that in mind as I read future reviews.
Nice piece of logic work there. I take from this that the "Experts" at Toms hardware have nothing to contribute to the conversation beyond validating what hardware manufacturer's claim. I will keep that in mind as I read future reviews.
- Mar 7, 2003
- 12,012
- 0
- 40,790
Nice. First, you question whether ping is a valid method of measuring latency for the Killer Xeno Pro. Then, when you find out that's how the manufacturer measures latency, you lash out at me for - what, trying to see whether or not their claims are valid?
What did you want us to do, ignore their claims of lower latencies? Did you want us to benchmark how well it works as a graphics adapter? Maybe we should have measured how well it works as a paperweight?
You complain that we use ping to measure latency when pings are a latency measurement, and you complain that we try to validate the manufacturer's claims. Let me explain something that might have gone over your head: we review PC hardware. It's our job to see if the product does what the manufacturer claims it does, and to facilitate this we tend to use the proper tool for the job.
You may be surprised to find that you're alone in your crusade to convince people that a latency measurement (ping) isn't adequate to measure latency...
What did you want us to do, ignore their claims of lower latencies? Did you want us to benchmark how well it works as a graphics adapter? Maybe we should have measured how well it works as a paperweight?
You complain that we use ping to measure latency when pings are a latency measurement, and you complain that we try to validate the manufacturer's claims. Let me explain something that might have gone over your head: we review PC hardware. It's our job to see if the product does what the manufacturer claims it does, and to facilitate this we tend to use the proper tool for the job.
You may be surprised to find that you're alone in your crusade to convince people that a latency measurement (ping) isn't adequate to measure latency...
bellbillsnow
Distinguished
- Jul 2, 2009
- 25
- 0
- 18,530
I just thought you could think for yourself instead of acting as if the only thing in the world is the Mfg's claims. Moreover, since y'all have the big fancy hardware expert website, I thought perhaps you could justify using ping as a measure of latency. So far, all you have been able to say is "that is what everyone else uses".
And I don't know why you cannot follow my posts, but I have suggested that I am not convinced that you are measuring the right things. I want you to see if it reduces lag in games. It shouldn't be that hard to do some actual research and talk to someone other than Bigfoot about how a few games use the internet and see if it makes a difference. From my point of view, your tests do not test what you claim. In fact, several posters have made some interesting points that could be explored.
Now if you just want to leave your review at "the Xeno does live up to the Mfg's claim", that's fine. Nice work. I thought you would want to press further and try to understand if there is a benefit that is not captured by ping times. Finally, it would be really interesting if you tried to figure out why companies like EVGA and Alienware would sell a product that does not do what it claims.
In the end, you argue that card is as good as QoS software. That claim does not seem to be supported by your data. Your data does not contradict it, but it is insufficient.And when I raise issues with your methods, your retort is "well the Mfg says blah blah". When it seems you could concede that a margin of error remains (which, ironically, you do in your review).
So here is a simple test that would satisfy me. Get three machines. Two with Killer Xenos (one in app mode and one in game mode, both in the DMZ running FNA Forewall) and one without (people on several sites are calling for an intel pro card, matters not to me). Have three people play on the same server of the same game. Measure ping times and record the game play. Then have the players rotate machines (blindly) and comment on each. This should at the very least show that if there is a benefit, then it is not discernible.
Now you can say that test is too complicated and too expensive. That's fine, but that is what it takes to make the claims you are trying to make.
And I don't know why you cannot follow my posts, but I have suggested that I am not convinced that you are measuring the right things. I want you to see if it reduces lag in games. It shouldn't be that hard to do some actual research and talk to someone other than Bigfoot about how a few games use the internet and see if it makes a difference. From my point of view, your tests do not test what you claim. In fact, several posters have made some interesting points that could be explored.
Now if you just want to leave your review at "the Xeno does live up to the Mfg's claim", that's fine. Nice work. I thought you would want to press further and try to understand if there is a benefit that is not captured by ping times. Finally, it would be really interesting if you tried to figure out why companies like EVGA and Alienware would sell a product that does not do what it claims.
In the end, you argue that card is as good as QoS software. That claim does not seem to be supported by your data. Your data does not contradict it, but it is insufficient.And when I raise issues with your methods, your retort is "well the Mfg says blah blah". When it seems you could concede that a margin of error remains (which, ironically, you do in your review).
So here is a simple test that would satisfy me. Get three machines. Two with Killer Xenos (one in app mode and one in game mode, both in the DMZ running FNA Forewall) and one without (people on several sites are calling for an intel pro card, matters not to me). Have three people play on the same server of the same game. Measure ping times and record the game play. Then have the players rotate machines (blindly) and comment on each. This should at the very least show that if there is a benefit, then it is not discernible.
Now you can say that test is too complicated and too expensive. That's fine, but that is what it takes to make the claims you are trying to make.
bellbillsnow
Distinguished
- Jul 2, 2009
- 25
- 0
- 18,530
I just wanted to add... the claims you make in your summary go beyond what the Mfg says... that is why I think you need to be able to justify your claims outside of the reference of the Mfg.
Anyway, before this whole thing got so nasty, I was trying to push for the fact that there is still fertile ground to test networks and gaming... routers, cards, connections, etc.
Anyway, before this whole thing got so nasty, I was trying to push for the fact that there is still fertile ground to test networks and gaming... routers, cards, connections, etc.
bellbillsnow
Distinguished
- Jul 2, 2009
- 25
- 0
- 18,530
Specifically - you claim the money is better spent on a gaming router... where is the evidence for that? Or that a 4850 would help more? In a CPU bound machine, the Killer would probably make a bigger difference than a graphics card.
bellbillsnow
Distinguished
- Jul 2, 2009
- 25
- 0
- 18,530
One last gripe - the minimum ping would be more interesting than the average... assuming that the Ping is a useful measure.
- Mar 7, 2003
- 12,012
- 0
- 40,790
Yes, this is nastier than it needs to be. So I'm happy to kick it down a notch.
Bellbillsnow, I guess what it comes down to is this: I believe ping is an accurate measure of latency, and you do not. These beliefs are our prerogative, and there is certainly room on this earth for us to disagree.
I am also of the belief that blind, subjective tests prove *nothing* unless an objective test has already established something to prove or disprove. It is my belief that sticking 3 people on three different machines is pointless when an objective measurement shows us no measurable difference.
Once again, you certainly have the right to disagree.
But since we're on opposite sides of the fence here, I don't have a lot of reason to go out and perform another test that I think is useless - especially when yours is the lone voice out of all the comments that calls for more investigation.
If a good measure of folks were demanding more, I might oblige the readership regardless of whether or not I personally think it's a waste of time. I have done so in the past when enough readers have suggested as much. But frankly with you being the only person asking for more, I find it hard to justify.
So if you want to start a grassroots campaign to get more forum users to vouch for a blind subjective tests, go to town. If a good number of forum users with a few years behind their belt are as interested as you in this test, I may consider giving it a try.
But as it stands, I have no reservations whatsoever with the testing or conclusions.
Bellbillsnow, I guess what it comes down to is this: I believe ping is an accurate measure of latency, and you do not. These beliefs are our prerogative, and there is certainly room on this earth for us to disagree.
I am also of the belief that blind, subjective tests prove *nothing* unless an objective test has already established something to prove or disprove. It is my belief that sticking 3 people on three different machines is pointless when an objective measurement shows us no measurable difference.
Once again, you certainly have the right to disagree.
But since we're on opposite sides of the fence here, I don't have a lot of reason to go out and perform another test that I think is useless - especially when yours is the lone voice out of all the comments that calls for more investigation.
If a good measure of folks were demanding more, I might oblige the readership regardless of whether or not I personally think it's a waste of time. I have done so in the past when enough readers have suggested as much. But frankly with you being the only person asking for more, I find it hard to justify.
So if you want to start a grassroots campaign to get more forum users to vouch for a blind subjective tests, go to town. If a good number of forum users with a few years behind their belt are as interested as you in this test, I may consider giving it a try.
But as it stands, I have no reservations whatsoever with the testing or conclusions.
bellbillsnow
Distinguished
- Jul 2, 2009
- 25
- 0
- 18,530
I think your summary undercuts the complexity and subtlety of the comments I have made. There is more going on than Ping times, but whatever. The mob has decided this product is a fraud, so there is little chance of anyone else challenging the wisdom of the group.
Seriously, have you read the comments? I am not the only one calling for further investigation - I am just the only one challenging your conclusions. Several people have asked for the card to be compared to an Intel Pro amongst other things.
Seriously, have you read the comments? I am not the only one calling for further investigation - I am just the only one challenging your conclusions. Several people have asked for the card to be compared to an Intel Pro amongst other things.
- Mar 7, 2003
- 12,012
- 0
- 40,790
How would comparing vs. an Intel pro instead of integrated help the Xeno when the Xeno already equals integrated in testing?
Once again, not interested in pouring more time into hardware that I'm more than satisfied I tested thoroughly.
I can understand your position, I simply disagree with it.
Once again, not interested in pouring more time into hardware that I'm more than satisfied I tested thoroughly.
I can understand your position, I simply disagree with it.
bellbillsnow
Distinguished
- Jul 2, 2009
- 25
- 0
- 18,530
I didn't say it made sense - I just said that I am not the only one calling for more testing. My standard is more about if I am convinced, you are the one who said you had a democratic standard.
Also, have you seen this...
http://www.bigfootnetworks.com/killer-nic-and-the-tab-key/
I just found it. Any response?
Also, have you seen this...
http://www.bigfootnetworks.com/killer-nic-and-the-tab-key/
I just found it. Any response?
- Mar 7, 2003
- 12,012
- 0
- 40,790
Yes, the bigfoot guys mentioned this issue with certain titles. I followed their procedures and even suggested game titles to get around this issue.
Remember that in the games we tested, the method we used did show differenes when multiple downloads occured, so we did measure a benefit to the Killer Xeno. We simply matched that benefit with freeware.
And I will correct you; I never once used the term democratic standard. I said I might oblige the readership if a good measuere of folks are demanding more. By my reckoning that requisite hasn't been met. Maybe by yours it has, but you and I disagree on a lot of things so that's no great surprise.
Remember that in the games we tested, the method we used did show differenes when multiple downloads occured, so we did measure a benefit to the Killer Xeno. We simply matched that benefit with freeware.
And I will correct you; I never once used the term democratic standard. I said I might oblige the readership if a good measuere of folks are demanding more. By my reckoning that requisite hasn't been met. Maybe by yours it has, but you and I disagree on a lot of things so that's no great surprise.
bellbillsnow
Distinguished
- Jul 2, 2009
- 25
- 0
- 18,530
I don't know Cleeve. You are are having a really hard time following this conversation. Is there someone at the office that can help you interpret my posts? You accused me of being the only person that wanted further investigation and it is simply not true. As for the number of people, I already agreed that there is not going to be a "good measure" and you got me - I am the one that used the term Democratic Standard.
Personally, my credibility in Tom's Hardware has plummeted- this is subject is just over your head. You have consistently failed to acknowledge things that I have said and been all to keen to follow any side track from the main issues. I sent you a link explaining that Ping happens over TCP (or ICMP) and the killer optimizes UDP. You say that you looked for game titles to get around this. How? That makes zero sense. If you had said, I looked for a way to ping UDP, that would make sense. But if you mean you looked for games that use TCP rather UDP, then you just further proved your ignorance since the Killer optimizes UDP.
And in all fairness, Bigfoot has also lost credibility. They are putting out all this info about FPS's and Ping times when their own site contradicts those as standards. Sadly, this project was always going to require a herculean PR effort and between the confusion they cause on testing and the buggy drivers - they cannot have much left to stand on. Oh well, I like my K1. Too bad it probably won't see win7 drivers.
Personally, my credibility in Tom's Hardware has plummeted- this is subject is just over your head. You have consistently failed to acknowledge things that I have said and been all to keen to follow any side track from the main issues. I sent you a link explaining that Ping happens over TCP (or ICMP) and the killer optimizes UDP. You say that you looked for game titles to get around this. How? That makes zero sense. If you had said, I looked for a way to ping UDP, that would make sense. But if you mean you looked for games that use TCP rather UDP, then you just further proved your ignorance since the Killer optimizes UDP.
And in all fairness, Bigfoot has also lost credibility. They are putting out all this info about FPS's and Ping times when their own site contradicts those as standards. Sadly, this project was always going to require a herculean PR effort and between the confusion they cause on testing and the buggy drivers - they cannot have much left to stand on. Oh well, I like my K1. Too bad it probably won't see win7 drivers.
bellbillsnow
Distinguished
- Jul 2, 2009
- 25
- 0
- 18,530
bellbillsnow
Distinguished
- Jul 2, 2009
- 25
- 0
- 18,530
You really proved that my opinion doesn't matter by responding to my posts for the past five days. Seriously, are you a teenager?
plonkaclunk
Distinguished
- Feb 27, 2007
- 19
- 0
- 18,510
Is this Killer card just performing QoS?
Some routers now have QoS built in that can be configured.
If your router was configured with QoS why would we need to spend the money on one of these network cards?
Obviously if your switching infrastructure is not QoS capable this may be a solution, but this just looks to be a cheap way of fixing poor infrastructure.
Some routers now have QoS built in that can be configured.
If your router was configured with QoS why would we need to spend the money on one of these network cards?
Obviously if your switching infrastructure is not QoS capable this may be a solution, but this just looks to be a cheap way of fixing poor infrastructure.
- Mar 7, 2003
- 12,012
- 0
- 40,790
[citation][nom]bellbillsnow[/nom]You really proved that my opinion doesn't matter by responding to my posts for the past five days. [/citation]
A few days ago it mattered. Then you posted rudely, so now I'm just having a little fun at your expense.
If you actually mattered, I'd probably consider retesting the card as you suggested...
A few days ago it mattered. Then you posted rudely, so now I'm just having a little fun at your expense.
If you actually mattered, I'd probably consider retesting the card as you suggested...
- Status
TRENDING THREADS
-
-
News Intel CEO Pat Gelsinger retires, effective immediately — also steps down from BOD, two co-CEOs step in- Started by Admin
- Replies: 124
-
Question I'm looking for a temporary new board after mine died ?
- Started by chucky9
- Replies: 3
-
Discussion What's your favourite video game you've been playing?- Started by amdfangirl
- Replies: 4K
Latest posts
-
-
-
Question Are there any compatibility issues I'm missing with this build- Latest: Vanquisher71
-
Facebook
Twitter
Instagram