How Playing on Wi-Fi Hurts Your Game

Megalith

24-bit/48kHz
Staff member
Joined
Aug 20, 2006
Messages
13,000
Here is a piece on how you’re doing it wrong if you’re gaming on Wi-Fi. I bet you already knew that, but the numbers may be interesting to some of you. I was just considering a Powerline adapter instead of running a cable through my attic, but I think I will just bear the trouble and stick with the latter.

While hard to argue against the advantages of a direct wired connection, there are umpteen reasons why a connection other than Ethernet gets used. Drilling holes to run cables—while tempting—may not be practical, especially in older, multilevel homes, or when renting. And let’s be honest, most of my friends don't even know which end of a drill to use, let alone how to pull cable. With advances in both Wi-Fi and Powerline networking, we decided to seek to answer if a wired connection was truly the best for gaming, and why. In addition, for those times that gaming could only be done on Wi-Fi or Powerline networking, which was the better choice with current gear.
 
shrug. new house 2 years ago, AC, 5ghz same room as my router with my gaming rig and ps4/x1 and never have an issue with latency.
 
I ran my line through a vent to the basement over to a shaft to the 2nd floor and under the carpet to the wall where my desk is. less then 50ft once it is pulled tight.
 
It can be real pain to run cable. My house has the switch in the basement (where I hang out most of the time) - the bedrooms are on the second floor. I still haven't ran a cable to the second floor yet - I would need to cut a few holes in the dry wall, which leads to patching, and then painting.
This past winter, I put in an Ubiquiti AP on the first floor of my house (not too hard to run a cable from the basement to the first floor)- almost dead center in the middle. Since then, I have 5ghz to the second floor - it works good enough for me. (We have an Amazon Fire and an Apple TV upstairs - my kids sometimes use tablets or their computers. They aren't really gaming - well, Roblox...
 
If you really want to do it wrong just game over WiFi with your VPN on.
I still have ptsd from the comments.:spitoutdummy:
 
This is all bullshit.

1) A 3ms ping isn't going to be noticeable to any gamer.

2) Using a USB wifi adapter? We didn't they try a quality pcie adapter?

3) Not convinced their router of choice is the very best. Why not try a goosed up Netgear Nighthawk running kong's build with higher amplification?

4) The remarks about conjestion don't mean anything unless there's actually conjestion. Which for many is a non-issue. :rolleyes:
 
why is this even news ? yes a wired connection always = faster if ya have a crap USB dongle
 
This is clearly a case of "If I could, I would, but I can't, so I won't."

This reminds me of Network 101... if you have a wired network, then there is active "collision detection" verses WiFi which, by it's very nature, uses Collision Avoidance. Collision detection isn't that big of a thing now since network hubs are more likely to show up on Antiques Roadshow rather than ebay, and a 10/100/1000 eight port switch for around $25. On the other hand, you probably have one access point, and multiple devices accessing that access point at that same time. Thus, wired is preferred over wireless.

However, outside of the classroom, reality sets in. WiFi ends up being "good enough" for most things, and unless you are paying the mortgage with your gaming winnings, you live with WiFi.
 
LOL, you guys don't realize that you won't feel the lag on your end, but the people you play with online will see you lag, teleport, and move erratically in the game, even on 5Ghz. To you, it all feels normal. I had to record a buddy of mine playing on wifi to show him what it was like on our end for him to realize even though it was smooth on his end, that there was an actual issue with him playing over wifi.

The main reason you don't feel the lag, is you receive more date from the server than you send to the server, so when you lag over wifi, your download isn't noticeably effected so it appears smooth on your end. But because you are sending much less data, and your position in your upload, when you lag, the server does not receive the packet, and other players don't get your true location: hence.. the lag other's see from you.
 
Last edited:
If you have cat5 and don't use it over WiFi ... you're lazy and cheap. Lazy because, yeah, they do make flat cat5 that you can tact and run anywhere and it's cheap. $12 dollars maybe for 50' maybe even 75 for that amount.
 
2) Using a USB wifi adapter? We didn't they try a quality pcie adapter?

3) Not convinced their router of choice is the very best. Why not try a goosed up Netgear Nighthawk running kong's build with higher amplification?
Author was being lazy with testing (used a laptop)
I assume they had it around (it's their "Best Gaming Router pick in the budget category")

If you have cat5 and don't use it over WiFi ... you're lazy and cheap. Lazy because, yeah, they do make flat cat5 that you can tact and run anywhere and it's cheap. $12 dollars maybe for 50' maybe even 75 for that amount.
Not everybody wants to drill/patch walls, or have cables running around the hallways?


Not sure how much I trust the author with goodies like this:
"This is a built in ping test that runs in MS-DOS, and can be found in Windows 10"
 
My reflexes are slower than they were 20 years ago and I don't play shooters much anymore so I don't really give a shit either way. If I ever get into e-sports(ahahahaaa) I will worry about 5-30ms of extra latency.
 
Wi-Fi sucks, is there anyone who doesn't realize that it adds latency?

I wonder if some of the other comments in this thread are from people who don't have very good internet access, because if you're looking at a 200ms ping anyway, a 100ms spike probably isn't noticeable. If you have fiber to the home it's more noticeable.
 
Author was being lazy with testing (used a laptop)
I assume they had it around (it's their "Best Gaming Router pick in the budget category")

I run a solid pcei wifi card and a Nighthawk (Kong)... just tested the wireless ping and got 1ms each and every time. So there you have it.

This story isn't even worth posting on Hardocp. Lame.
 
Wired is always better. When I bought my house 20 years ago, it needed a lot of work, so I took the opportunity to run Cat 5 to most the rooms. The short cat 5 runs have no problem supporting GB Ethernet :D

I have one part of the house where there was no easy way to run a cable. Tried AC extenders, but most seem to have a problem with occasionally dropping the connection.
Ended up using a wireless bridge that uses the 5Ghz band and has multiple GB ports on it. Rock solid.
 
3ms on wifi

oh noes :ROFLMAO:

I haven't indulged into the article, so does he have/use Mu-MiMo?

my mainboard has a direct wifi antenna connection which I'm using

it has MU-MiMo as well as my router
not concerned

3ms is so low it might be overhead from using USB

3ms is so low we should talk about using 144hz for our display instead of 60 first and how we achieve such high frame rates in games if one is so competitive


if anything should've tested power line adapters

because I used some and I could make them repeatedly/reliably overheat, which caused them to shutdown, by just using an Amazon fire tv(wired from power line) and 2 cell phones browsing (it provided wifi as well)

terrible
 
I went the powerline route since drilling holes in concrete isn't exactly my thing. It's way more reliable than wi-fi but latency is about 10ms.

I've been hosting dedicated Quake Live servers on a spare machine and my friends all around the city have had better pings than me.
 
^ Change out your router and wifi card. Will be rock solid if configured correctly (on the right channels and such).
 
As an Amazon Associate, HardForum may earn from qualifying purchases.
I run dual wireless routers (one for outside) - 2.4 GHz and 5 GHz bands for both - and powerline ethernet (I have run two different types). I also run gbit wired down to my server room. I game over all of them, either directly or with in-home streaming. I also have an NVIDIA SHIELD TV, etc. In my experience the only truly consistent experience is wired. I often get interference on the 2.4 GHz bands (bluetooth, microwaves, neighbors, whatever) which is why I doubled up on 5 GHz for inside and out. Nevertheless, I often had buffering issues with 5 GHz (even at places getting 100-150 mbit), which for some games caused inconsistent spikes. It's not helped by the issue that not all devices handle the same band of 5 GHz the same way; many of my streaming devices hate higher bands, etc. That leaves the powerline ethernet - which is good, but is also inconsistent. Bandwidth can vary from 30 to 60 mbit on that even on my new wiring, which means I might have decent play for 30 minutes and then some spike on the powerline can suddenly cause the stream to drop.

For any fast-paced game you really need wired. I manage with wireless on anything less rigorous, but multiplayer games? Wired. Powerline ethernet is pretty good these days and I trust it more than wireless but even so, I don't use it for anything serious.
 
I actually question how he got 3ms... I do remember back in the day on G you had a substantial ping that might be 7+ ms, but that isn't the case anymore. At home and at work I can routinely ping my AP with 1ms response times. With so many access points offering concurrent dual band support, you should be able to set it up so that your high priority devices are on a different channel than the low priority. (2.4ghz for low priority, 5ghz for high)

I'd definitely still wire in if possible for your main gaming computer, or if you plan on playing FPS games for those what if scenarios. But I don't think you are going to be at a big disadvantage if you have quality wifi. At this time, those who have fiber probably have a bigger advantage than those on cable, simply because of jitter. That and a DOCSIS 3.0 connection is 6ms to the first hop where fiber is <1ms.
 
The article does not indicate typical performance users will see, and is misleading. Reference: I've been designing wireless communications systems for > 20 years.

In a typical home with reasonable connection strength (S/N) and modest numbers of devices, wireless will normally have latencies which are indistinguishable from wired.
In incredibly crowded areas, latency may increase due to collision avoidance. Even then, it's going to be incredibly unlikely it has any impact on your gaming experience at all.

Run a wire if you can, but that would be mostly due to bandwidth improvement and interference immunity. WiFi is absolutely fine.
 
In a typical home with reasonable connection strength (S/N) and modest numbers of devices, wireless will normally have latencies which are indistinguishable from wired.

I concur with this. I have one segment running over powerline ethernet to a wireless router and even through both my latency is negligible.

Just to clarify to others, my point in my other post had nothing to do with latency - I was talking about consistency, which to me means a solid 60 FPS over a full play session. I have one machine that is both wired and has 5 GHz 1x1 built-in, and I also have a 5 GHz USB dongle 2x2. I've tested all 3 including over the powerline (total of 4 potential pathways - actually more since I also have two wireless routers over powerline) and people who say they're equal for something like The Division are just wrong. Unless you're suggesting I need a router that doubles as a home protection weapon with its octo-antennae and airways clearer than Mar-a-Lago to get there.
 
Last edited:
It's disappointing to see some disregard for performance here. Not just this thread too, for example we see people accepting adaptive sync instead of asking for stable performance from engines and maps. So sure, not everyone can get a wire plugged in. And now with client side hit detection, there's a sentiment among players that having a slower internet connection is better because you can win gunfights that you should lose, since the server allows you to lag behind the simulation and still accepts your shots. Devs gotta set the netcode for the average user.

As example, I've seen in the playerunknown battlegrounds thread that people complain about lag. Is this the fault of the devs writing trash netcode or something they've just had to accommodate due to net connections of their players?
 
For most it is not a consideration due to the convenience of Powerline adapters but mains wiring is both a transmittor and reciever of signals, furthermore PLAs can and do add noise-hum to the mains (and so also broadcasts) and some electronics can be sensitive to this.
Yeah does not stop my friend from using PLAs due to the convenience :)

CHeers
 
I've had bad experience with powerlines going from different circuits. Which is a shame, because where i'd need the wired, is on a different circuit from the room with the equipment!
 
And now with client side hit detection, there's a sentiment among players that having a slower internet connection is better because you can win gunfights that you should lose, since the server allows you to lag behind the simulation and still accepts your shots. Devs gotta set the netcode for the average user.

You're playing the wrong games. Probably on the wrong platform, too. Which games are you referring to?
 
I still don't trust wifi.

I have two Unifi 5 Ghz pro access points in my house, but they are only used by phones tablets and our one laptop.

Everything stationary gets hard wires gigabit Ethernet, no matter how difficult it is to run wires there.

I will never put up with a wifi compromise. I still facepalm every time I see a MiniITX board with on board wifi.

I don't even do much multiplayer gaming anymore.
 
I will never deny, for now, wired is still preferable, but in reality, with a high end router and good adapter, there is absolutely no issue gaming on WiFi. Unless you just have no idea what you are doing. Do what's convenient for you.
 
Power line adapters work great... been using them in my house, and we have installed in customer homes out in the field. Speed is great, will work all over the house, even on diff phase, only issue is sometimes a ballast or led driver can cause disconnections if on the same circuit. We had a customer with Neon signs... caused problems with power line, so Wifi was a better option. I run my Steam link box in my living room over power line adapter.. its just like sitting at my desk, smooth as butter.
 
they have gotten pretty good nowadays. was playing around with this set the other day

https://www.amazon.com/gp/product/B01H74VKZU/ref=oh_aui_detailpage_o05_s00?ie=UTF8&psc=1

and in my crappy old wired house was getting a solid 70Mb transfer from one end of the house to the other.

i do use an older set to connect my irrigation system in the garage to the network, and it works well.

I use this one as well. Compared it to a Netgear MIMO and it crushed the Netgear by about 20Mb on the same outlets. It has been rock solid. Those spikes can easily be diagnosed and eliminated. I am going to say that the reviewer could have put in ten minutes of effort to figure out where those latency spikes were coming from.
 
As an Amazon Associate, HardForum may earn from qualifying purchases.
Power line adapters work great... been using them in my house, and we have installed in customer homes out in the field. Speed is great, will work all over the house, even on diff phase, only issue is sometimes a ballast or led driver can cause disconnections if on the same circuit. We had a customer with Neon signs... caused problems with power line, so Wifi was a better option. I run my Steam link box in my living room over power line adapter.. its just like sitting at my desk, smooth as butter.
They do work great, but unlike wifi and ethernet, they add a fixed latency on top of any latency spikes. This is because the power line adapters only transmit during the crossover portion of the AC signal, when voltage is near zero. So if you have 60hz power, there are only 120 transitions per second, so when it wants to transmit, it has to wait for one of those transitions. That alone adds an average of about 5ms latency, and if you're running a protocol that requires acks, that really adds up. A coax adapter may work better since they simply use a frequency spectrum not used by your TV/Satellite and doesn't need to wait for the voltage transitions, though those also have their own issues and aren't as well developed/mature.
 
I've been using MoCA 2.0 and it has been great. Used existing Coax runs throughout the house.
 
Last edited:
I've been using MoCA 2.0 and it has been great. Used existing Coax runs throughout the house.
Glad to hear that, they've been pretty pricey so I've held off on those for a couple years, I'm interested in using a pair of Moca to gigabit switches, and bridging our upstairs TV room and the computer room downstairs, a bit pricey but beats the Ethernet cord snaking along the stairwell if it works well.
 
They do work great, but unlike wifi and ethernet, they add a fixed latency on top of any latency spikes. This is because the power line adapters only transmit during the crossover portion of the AC signal, when voltage is near zero. So if you have 60hz power, there are only 120 transitions per second, so when it wants to transmit, it has to wait for one of those transitions. That alone adds an average of about 5ms latency, and if you're running a protocol that requires acks, that really adds up. A coax adapter may work better since they simply use a frequency spectrum not used by your TV/Satellite and doesn't need to wait for the voltage transitions, though those also have their own issues and aren't as well developed/mature.

Well, not going to argue your data, but all i can do is give my experience in the home and out in the field. Have had zero complaints from people, and as i said, i can game on my steam link without any noticeable lag. However, when using wifi with my steam link, i get lag.
 
LOL, you guys don't realize that you won't feel the lag on your end, but the people you play with online will see you lag, teleport, and move erratically in the game, even on 5Ghz. To you, it all feels normal. I had to record a buddy of mine playing on wifi to show him what it was like on our end for him to realize even though it was smooth on his end, that there was an actual issue with him playing over wifi.

The main reason you don't feel the lag, is you receive more date from the server than you send to the server, so when you lag over wifi, your download isn't noticeably effected so it appears smooth on your end. But because you are sending much less data, and your position in your upload, when you lag, the server does not receive the packet, and other players don't get your true location: hence.. the lag other's see from you.

No, just no.

The local connection has nothing to do with this if the transit time to the server is the same, the method matters little.

They also picked the worst possible options for WiFi, that being a USB dongle, they then went a step even further to stack the deck even more by picking the "FAR" location they do for their router tests that only the best routers reach, meaning loss of signal, dropped packets and much more sensitive to interference. They really went out of their way to get a 3ms average difference in ping times in a single test, the other tests were 1ms.

From the testing:

"For the wireless connection, we ran tests on both the 2.4 GHz, and the 5 GHz frequencies. In general, on most setups, the 5 GHz will have more throughput, and less interference. However, the range of 5 GHz is less, as it goes through walls and obstructions less well. Testing was done at our standard “Far” location that we use on our Best Routers and Best Wireless Adapters Guides."
 
Back
Top