Read more.It debuted alongside the new and improved Killer Intelligence Engine at Computex 2019.
Read more.It debuted alongside the new and improved Killer Intelligence Engine at Computex 2019.
I wonder if that includes itself?Meanwhile, GameFast protects gamers from applications and processes that unnecessarily consume CPU and memory resources that should be devoted to keeping games playing fast and smooth.
On a serious note, have there actually been any demonstrations that show Killer to be anything other than snake oil?
Isn't it about time faster cards game out. Intel are already making up to 40gb ones. I know this countries broadband speeds are anything close to even 1gb currently but if you're on a good lan faster speed are possible these days.
if it is that good and works Intel or someone would have bought it by now
What does it matter now if men believe or no?
What is to come will come. And soon you too will stand aside,
To murmur in pity that my words were true
(Cassandra, in Agamemnon by Aeschylus)
To see the wizard one must look behind the curtain ....
The cards exist but the interconnects and the bandwidth available from the CPU/lanes aren't capable enough to justify. With AMDs PCIe 4.0 and whatever Intel goes with next may allow more breathing room for these but also the cabling is ridiculously expensive.
Sure you can get second hand/off brand crapola but to get a 5m QSFP+ cable for "reasonable" prices, but manufacturers don't look at second hand/off brand when designing their products.
I use Mellanox 40GbE between my servers and primary desktop using IBM second hand datacentre cards that they were literally throwing away but these costs hundreds of pounds new...even for a single port card!
That is one of the reasons why. Plus, look how long it took gigabit to saturate a home environment, it'll be the same to 5GbE or even fully capable home routers that can go above gigabit.
That's just the techonology quagmire of consumer products sadly.
Snake oil in 2004 with that add on card with the giant K on it.
Snake oil later when it came back as a built in thing in motherboards with its software bundle.
Still nothing changes. Just give me realtek or intel ffs, i've had zero problems with either when gaming on windows or linux.
"Oh man if only I had better ethernet implementation I could've made that headshot" - No gamer ever
It does make you wonder how the business has survived as long as it has. Gamers can be a gullible bunch, but even so it's been around for a long time in spite of reviews, criticism, etc.
The difficulty with that statement is you can guarantee a vast majority of gamers have no idea how the traffic leaves their PC and gets to the remote server.
This card is snake oil and give me an Intel i211 anyday because once the network packets leave the NIC, any fancy stuff you paid for to optimise it leaving your PC means squat in your home network and means less than squat which it moves through the ISPs etc.
It's a complete gimmick and any performance gains from this card would be minimal and are likely because the current status of their windows installation is just a little overloaded. So it literally is akin to Game mode in Windows...
Yup: https://techreport.com/review/29144/...ght-years-on/3
The yawn for me is that it's yet another gigabit controller, probably near identical off-the-shelf silicon, but smaller/cheaper, and only really comes with driver updates. That's more snake-oily to me than performance doubts. If they still rocked a Freescale PowerQUICC NPU like the original Killer NICs K1 and M1, and it doesn't have enough processing and/or RAM chops for new software they want to add to the firmware, then I'd understand punting out a new gigabit NIC. Obviously CPUs aren't lacking in processing horsepower.
That doesn't show it's not snake oil though, those charts are in microseconds, i.e. well below the level that would be perceptible to a gamer, and totally eclipsed by any real WAN network. Heck even jitter will be substantially higher than that for the majority of connections!
Furthermore they're only comparing it to one other NIC? What sort of a review is that? It would have been far more useful to compare at least a couple and show it most likely blending into the crowd.
That 32b TCP one is very suspicious - they're connected through a network and to a NIC in a server and the difference in the Intel NIC makes up the majority of the RTT? Logic dictates the server's NIC performs much better in that (still irrelevant) test.
There are currently 1 users browsing this thread. (0 members and 1 guests)