New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Debate time: What is more important? Low ping OR Low Hops?
DataIdeas-Josh
Member, Patron Provider
in General
Whats your take?
I know everyone will say "Both".
Explain your reasoning.
Comments
Low ping is important for me while I am playing FPS games.
I have actually added more paths to my network by using proxies and VPNs so I can get a lower ping while connecting to another region's game server.
low ping
hops can be hidden / doesn't matter
Low ping. Latency is matter..
low ping
I would have to say low ping, hops although for those who are more knowledgable would notice more hops likely 90% of people wouldn't even notice more or less hops, but they would notice ping whether they just see it as lag.
less hops - better. Latency related issue (so called ping) complex stuff.
Usually game devs using udp for that stuff and avoid tcp as much as they can, or building their own queue system for delivering packets. For what?
For solving key problems:
I've seen few apps that did awesome job related to latency and client-side prediction of what is going on on server-side for some online rpg game. Results pretty awesome.
So why less hops important more, than latency?
Ping it's about icmp, ping do say nothing about tcp/ip connection at all.
Do not say anything about different packet losses or fragmentation of packets.
Or throughput of tcp, router configurations (border ones). And path from server to client routing.
Usually less hops means almost always that a provider is great. I mean if you have server in one country, while you're in another. More hopes - more chances that some hope in the middle of route will be broken or damaged (which is non rare thing). And can provoke issues in game. And hard to trace and fix it, if it's related to tier1, or tier2 providers.
For example Equinix in Warsaw still for maybe 6 years for sure can't fix extremely weird routing bugs. They simply do not care.
So latency is important, less is better. But hopes in my opinion better. I.e. less hopes - better.
AS200690 has no more than 2 hops between any of our locations.
It simply means the network is interconnected via L2 links.
Each L2 link may traverse 50 Ethernet switches, but you won't see them in MTR.
Low Latency, hops don't matter.
As long you they don't really add latency, for each jump.
Like my route to Tokyo, with usually 5-6 hops, and shiny 150ms.
God, I love Jesus for inventing route bending.
Latency matters over hops, 'cause lots of factors combined, including hops, leads to the latency.
low ping high throughput
Isn't that question of reliability and dependency? The more hops you have - the more likely you will have problems in the middle? Fewer hops is better. Lattency is relevant in a very specific cases.
Low ping. Hops aren't necessarily meaningful (even though less hops can make you imagine that the route is shorter than it really is... can be some kind of marketing trick!).
Latency matter no one looks for hops anyways unless a network geek
This thread needs a poll, but the answer is ping.
Low price
low latency
hops are hops they dont do anything
Latency is all that matters for performance when everything goes right. But each hop could be an additional source of nuisance.
Neither.
E.g. I've got a connection with low hops (5) and low latency (15ms) that is the worst ever. Throughput is like bits per second and a 40kb file needs 2 to 4 seconds while on a 190ms, 20 hops connection (Australia-Europe) it needs 200ms. And the provider peers with the ISP directly (see low hops).
Also got a connection where icmp is ok but on the tcp level you've got hellish packet loss.
So both numbers are ok for a fast check and I would vote for latency but the real life experience is a whole lot different.
So I'll second this:
Latency over whatever, you cannot control the upstream providers paths, usualy. :-)
Issue is kinda hand / hand.
More hops can keep the signal strong per say.
However it does add more possibilities that something can happen.
Although begs the question is hops being hidden?
!ping me
I'm skipping over reading everyone's (probably very well thought out) replies but I just want to say that because of the fact that hops are harder to control I would say latency. (You won't be able to control the hops a customer takes within their ISPs network to get to yours for example).
More hops can also just be caused by a spine-leaf architecture at the datacentre (where aggregation and access switches are doing routing). So even though you could argue there's more points of failure in the path, in reality, the network operator has recognised the potential for failure and mitigated it by deploying their network in such a way that it is not only redundant but also reasonable (you wouldn't want redundancy at the expense of having half your switchports occupied for any given access switch would you?).