I wonder how ExitLag works.
If ExitLag
uses the algorithm I described here, then I understand why ExitLag improves esports scores.
There's lots of placebo, but there's a counter-intuitive way for a gateway-in-middle to improve gaming experiences by smoothing ping jitter
TL;DR: Lag conditioning algorithms / lag dejittering algorithms that converts random [10..30ms] latency jitter into a consistent [30ms] latency) which eliminates 50-75% of games' "random predictive errors" (hits that don't register, misses that register).
For this to be fair, lag-conditioning ideally should be done symmetrically (both in the incoming and outgoing direction), as some of the intentional asymmetric lag tricks can sometimes trigger gameserver-side anticheat etc.
Lag conditioning REALLY helps (adding intentional lag to dejitter), it's real, and it makes gaming more fun, but it must be done properly, ethically, and fully kosher without cheat sheninigians.
This also helps level the playing field and equalizes FTTH lag with DSL lag too, and if done fairly between all gamers on the same server, it creates WAY fewer predictive errors (false positive/false negative hit registration). But Some FTTH users resent the fact that some game engines handicap FTTH users (e.g. some games perform more reliably at 30ms lag instead of 10ms lag or 100ms lag, as an example), so this is always a controversial discussion topic in some other circles. It's because the game netcode is tuned to a ideal median lag, creating a handicap effect for LPB (Low Ping B*stards).
But the best of all worlds is good lag conditioning. Sometimes good lag conditioning is built into the games but more often it is not, and you want a third party gateway (gaming VPN with built-in lag conditioning) to do the lag conditioning. Turn a lag-jittery ISP connection into a perfect jitter-free slightly-higher-lag connection, that feels as if it's lower lag, because all the predictive errors disappear (etc)
This is why I tell my forum members to try "VPN/Tunnel Provider Roulette" in your particular country/location with your particular ISP
Switching Locations Situation
Random Example Country/Location #6 with ISP #3
1. VPN Provider "A" - Connection becomes worse and game is unplayable
2. VPN Provider "B" - Almost no difference
3. VPN Provider "C" -
Massively better esports performance
4. VPN Provider "D" - Slightly worse.
Random Example Country/Location #22 with ISP #2
1. VPN Provider "A" -
Massively better esports performance
2. VPN Provider "B" - Slightly worse.
3. VPN Provider "C" -
Massively better esports performance
4. VPN Provider "D" - Connection becomes worse and game is unplayable
Random Example Country/Location #7 with ISP #1
1. VPN Provider "A" - Almost no difference
2. VPN Provider "B" - Slightly worse.
3. VPN Provider "C" - Connection becomes worse and game is unplayable
4. VPN Provider "D" - Much worse
5. Nothing improves in this country
Random Example Country/Location #3 with ISP #6
1. VPN Provider "A" - Almost no difference
2. VPN Provider "B" - Slightly worse.
3. VPN Provider "C" - Connection becomes worse and game is unplayable
4. VPN Provider "D" -
Massively better esports performance
Switching ISPs in Same Location While Trying All VPN/Tunnels
Random Example Country/Location #6 with one ISP #3
1. VPN Provider "A" - Connection becomes worse and game is unplayable
2. VPN Provider "B" - Almost no difference
3. VPN Provider "C" -
Massively better esports performance
4. VPN Provider "D" - Slightly worse.
Random Example Country/Location #6 (SAME COUNTRY) with DIFFERENT ISP #7
1. VPN Provider "A" - Almost no difference
2. VPN Provider "B" - Much worse
3. VPN Provider "C" - Slightly worse.
4. VPN Provider "D" - Connection becomes worse and game is unplayable
5. No VPN provider seems to improve things on this ISP
Random Example Country/Location #6 (SAME COUNTRY) with DIFFERENT ISP #9
1. VPN Provider "A" - Almost no difference
2. VPN Provider "B" -
Massively better esports performance
3. VPN Provider "C" - Connection becomes worse and game is unplayable
4. VPN Provider "D" - Slightly worse.
Similar samples have have happened; with all the people reporting better gaming through a specific VPN (but only in one location), or better gaming through a specific ISP (but only in one location with a specific VPN).
So you definitely have lots of claims and outliers that doesn't help someone else much (because they're on a different ISP and different location). Which is why forums skream "PLACEBOZ" and just call it a day -- understandable -- but this is real science, buddy.
The famous chicken and egg problem, the Catch-22 problem, of making sure claims are not fake is difficult: You can't perform a scientific test with the same variables as someone else -- becuase you're not sitting at their chair (ISP-specific, location-specific, VPN-specific combo that works). You can't make "2+2=4" if you change only one of the three numbers (2+3=4, or 2+2+5) -- you need all variables the same (same ISP, same location, same routing, same city, same neighborhood, even same DSLAM, even same fiber gateway box, and even maybe in the same building, etc -- just to merely confirm their claim, because of the way Internet Routing works.
When doing ISP/VPN/Tunnel Provider "Random Combos" Roulette...
...YMMV!
While ExitLag is not a VPN, they are indeed a Tunnel Provider, and Tunnel Providers, can actually do some lag-conditioning behaviors that are definitely DEFINITELY NOT PLACEBO (at least in certain countries, even if it is placebo in YOUR location).
Your Country "A" tunnel provider placebo may scientifically not be placebo in Country "B".
This happens because of internet routing differences (different ISP, different location, different backbone, different country, etc)
Both VPNs and Tunnels modify the Internet routing between you and your game server, which can affect the latency jitter. You will usually have not much control over your first hop(s) of ISP routing, but beyond the first hop(s), a VPN or tunnel can dramatically reroute your gaming to a different (better) routing or (better) backbone, allowing better gaming. Concurrenttly, also possibly doing also double duty such as latency dejittering algorithms on the fly (like the one I described).
Some VPN providers seems to defacto provide nice (intentional and/or unintentional) latency conditioning behaviors. Intentional includes algorithms that
I described here, while unintentional includes simple sheer network routing that is more stable through them (VPN/tunnel) than bypassing them (pure ISP chosen routing).
. I'm thankful in a way because he always used to think it was just me moaning, but now he understands.
