[Power/EMI] I discover why sometime PC become fast and low input lag and otherwise feel high input lag

[nuggify]

Also, just because (cherrypicking simplified number) ...., say 99% or 99.9% people don't have major EMI-related lag/glitches can mean 10,000 (1%) out of 1,000,000 has some kind of relatively significant EMI glitch with their computer because they're living next to high voltage transmission lines or live in a very EMI-noisy old apartment building with a malfunctioning fridge compressor in the next tenant's wall right behind the computer, or whatever. Lots of near-EMP-league EMI edge cases that punch through a lot of anti-EMI stuff. EMI is an infinitely vast universe with a spectrum wider than audio spectrum and visible light spectrum, of all kinds of signal noises (broadband, narrowband, pure noises at below-light frequencies, or random noise at above-light frequencies, over-the-air, over-the-wire, or trillions/quadrillions/quintillons/infinite other patterns more numerous than the number of atoms in this universe, etc). It's a lot of FUD of wild goose chase for red herrings, very hard to debug stuff for end users. There is little hope for a full EMI debug though one can always do their best, and increase EMI-resistance success %, but not for the entire locations of the entire population...

You keep offering this sentiment that things like high voltage power lines could cause this or something, but I am again telling you that is just not the case. This issue I (and many others) have is indeed much stranger than anything obvious like that. The consensus with literature out there is that modern PCs are highly immune to a lot of interference- so much so that it would take such a high amount of EMF or RFI that it would be completely destructive interference. Things would start breaking- and not just your computer. I have consulted with professionals about this. They do very rugged testing when they build these components. There are strict Electro Magnetic Compatibility testing that is done on components before they go to market.

The thing here is that we all suffer non destructive interference. This is why I believe it is some kind of systematic neutral conductor issue. Something very strange is going on with that grounded conductor and it is systematically allowing all the noise in our environment to cause problems. That does not mean the problem is "EMI". It means that is just a symptom of the source issue. If your signal/logic reference is well above 0 then your entire noise floor is raised very high. When this happens any noise in the environment (no matter how normal) is amplified one thousand fold or more. Those signals need to compare themselves against something so presumably if that noise messes up the reference it can cause inconsistent errors- the likes of which so many posting here have seen.

I appreciate that you allow us to discuss this here, but there is a reason people come back over and over and go in constant circles about fixing this. It is because no one is actually looking in the right place. We have all just been trying filtration techniques and although they do effect the problem, it is never a fix because it simply cannot be.

[Anonymous768119]

I can confirm that the internet connection has nothing to do with it.

I've tried 4 different ISPs and I can barely see any difference between them. I replaced every component in my PC including mice, keyboard and monitor. I have massive desync (in my feeling it's around 200-400 ms) due to data correction and/or error correction I assume. When I am watching my teammates everything looks so easy and they can react without struggling. When I am watching myself on demos I am astonished how slow my reaction is. This is like playing CS:GO on nightmare mode and very seldom there are days, when everything is working more less (yeah) ok and this is the moment when I can see how I wasted my potential because of place where I currently live. When I am browsing some Polish PC forums, noone has any idea about that and is blaming the connection. I can't find anybody who can make proper measurements in my electric installation and it makes me crazy already.

Chemical ground makes sense, because it's protecting ground from environment and reduces ground resistance but it's really uncommon in my country to use such a methods.

Guys don't spend unnecessarily your money because on 99% it's not your PC's fault. Try to run it in another location and then make a conclusion. It is really comforting that this problem has become more popular and there should be more, less radical solutions.

[hardhit]

Also, just because (cherrypicking simplified number) ...., say 99% or 99.9% people don't have major EMI-related lag/glitches can mean 10,000 (1%) out of 1,000,000 has some kind of relatively significant EMI glitch with their computer because they're living next to high voltage transmission lines or live in a very EMI-noisy old apartment building with a malfunctioning fridge compressor in the next tenant's wall right behind the computer, or whatever. Lots of near-EMP-league EMI edge cases that punch through a lot of anti-EMI stuff. EMI is an infinitely vast universe with a spectrum wider than audio spectrum and visible light spectrum, of all kinds of signal noises (broadband, narrowband, pure noises at below-light frequencies, or random noise at above-light frequencies, over-the-air, over-the-wire, or trillions/quadrillions/quintillons/infinite other patterns more numerous than the number of atoms in this universe, etc). It's a lot of FUD of wild goose chase for red herrings, very hard to debug stuff for end users. There is little hope for a full EMI debug though one can always do their best, and increase EMI-resistance success %, but not for the entire locations of the entire population...

You keep offering this sentiment that things like high voltage power lines could cause this or something, but I am again telling you that is just not the case. This issue I (and many others) have is indeed much stranger than anything obvious like that. The consensus with literature out there is that modern PCs are highly immune to a lot of interference- so much so that it would take such a high amount of EMF or RFI that it would be completely destructive interference. Things would start breaking- and not just your computer. I have consulted with professionals about this. They do very rugged testing when they build these components. There are strict Electro Magnetic Compatibility testing that is done on components before they go to market.

The thing here is that we all suffer non destructive interference. This is why I believe it is some kind of systematic neutral conductor issue. Something very strange is going on with that grounded conductor and it is systematically allowing all the noise in our environment to cause problems. That does not mean the problem is "EMI". It means that is just a symptom of the source issue. If your signal/logic reference is well above 0 then your entire noise floor is raised very high. When this happens any noise in the environment (no matter how normal) is amplified one thousand fold or more. Those signals need to compare themselves against something so presumably if that noise messes up the reference it can cause inconsistent errors- the likes of which so many posting here have seen.

I appreciate that you allow us to discuss this here, but there is a reason people come back over and over and go in constant circles about fixing this. It is because no one is actually looking in the right place. We have all just been trying filtration techniques and although they do effect the problem, it is never a fix because it simply cannot be.

I assume you have tried different mixes of split core ferrite. Mix j(75) for example.. please update your further findings, it is helpful

[axaro1]

Are you the only person using internet when you play? Have you tried a router(or even a switch) with QoS supporting fq_codel?

[mello]

That is funny. I have a discord of near 70 members with this issue, and it is certainly not internet. You can fully disconnect internet and its all still there. The only person that fixed it (like fully mitigated the symptoms- we still have no idea what the source is) did so with chemical grounding. He had all the classic symptoms of desync and input lag.

How many times i need to say that ? As long as someone has these problems outside of online gaming, then obviously this has nothing to do with the internet. I do not contest that. These are obviously two separate issues. What i'm saying thought, is that some people *might* be going crazy about this whole thing and start thinking and convincing themselves that they have also problems outside of online gaming. There is a very high possibility of that being the case, due to how human brain works. As i said, i have seen in the last ~20 years, people making all kinds of ridiculous claims regarding certain bios settings, windows settings, game setting - "improving" or "fixing" things, making gaming "better", when in reality 95% of these "tips & tricks" are total and meaningless garbage. Same thing when it comes to hardware, peripherals, cables etc. Have you ever heard about the audiophile market ? Many audiophiles and "special high quality equipment" promoters make many claims about improvements in sound etc. What the reality looks like ? When you put it against an AB test, they usually fail to tell the difference and 'what is what'. The power of the human mind.

Maybe you should listen to me and consider what I have to say. That thread nicko started about net issues, he got a dedicated line (T1 connection) and still same problem. Then he started to realize it simply was not internet. He lifted ground prong to PC and the issue improved a ton (temporarily).

Had seen that argument million times already. Someone changes something, things magically improve, but after some time (same or next day usually) things get worse once again. I have been throught the exact same things in the past, making all kinds of both, dumb and smart, changes, feeling an improvement but eventually problem came back with full force. This happens because this problem fluctuates in severity on its own, which is very deceitful. Even if you do nothing, the problem goes away, comes back on its own and changes in severity without any interventions. Also, i'm also unable to know what issues other people are having and perceiving. So, we might be talking about the same, similar or completely different things.

There is something causing this that is electrical in nature and many people suffer it. I strongly urge you to try some of the tests we have spoken about just to see for yourself.

I can make this problem go away in 15 seconds, by switching the internet from one to another. Same server, same people, same map, everything is the same, an near immediate disconnection & reconnetion (15 seconds difference) to the server. One connection (Fibre 1 Gb/s both ways, ~10ms) = problem is constant, only the severity of it changes all the time, second connection (mobile internet, don't know the speed, most likley the usual ~20-80 Mb/s, ping ~25ms) = problem is literally non existent. I'm sorry, but you can't beat this with anything you do, say or hypothesize.

[nuggify]

These are obviously two separate issues. What i'm saying thought, is that some people *might* be going crazy about this whole thing and start thinking and convincing themselves that they have also problems outside of online gaming. There is a very high possibility of that being the case, due to how human brain works. As i said, i have seen in the last ~20 years, people making all kinds of ridiculous claims regarding certain bios settings, windows settings, game setting - "improving" or "fixing" things, making gaming "better", when in reality 95% of these "tips & tricks" are total and meaningless garbage. Same thing when it comes to hardware, peripherals, cables etc. Have you ever heard about the audiophile market ? Many audiophiles and "special high quality equipment" promoters make many claims about improvements in sound etc. What the reality looks like ? When you put it against an AB test, they usually fail to tell the difference and 'what is what'. The power of the human mind.

In regards to all the ridiculous claims I think I have already posited the idea that it could be they have this very problem I have. For me BIOS changes, Windows settings and all of that stuff does cause system responsiveness to change (no it is not placebo as it is reproducible although it varies) but it is temporary. So your anecdotal case study with how people are experiencing placebo with these things is explained from my point of view by people who suffer some degree of this issue. I am fully aware of the audiophile market. And I am also aware of the absolute complexity of this problem (and power problems in general), having observed it myself and I think there may be some merit to my argument here. These things may only seem illogical because we do not know the exact interaction that is taking place here- cause and effect (we are only seeing the effect side). In regards to the audiophile market I will give you an example. Many power conditioners and filters are sold to improve the sound, and many, many folks say they have tried the device in question with no observable difference to sound quality. But did you ever consider that those people do not benefit from noise reduction on their homes power grid? Nearly everyone has a different situation and setup. Some will benefit and some will not. The real question to ask is what are the actual specs and components of the power conditioning in the first place and is it targeting the problems you may have with your power quality. My point being user experiences differ vastly (especially in regard to analogue signals and power quality) and are highly dependent on the specific situation. Now our problem is a much stranger version of this as we are having non destructive problems on digital devices (that run on DC no less).

Had seen that argument million times already. Someone changes something, things magically improve, but after some time (same or next day usually) things get worse once again. I have been throught the exact same things in the past, making all kinds of both, dumb and smart, changes, feeling an improvement but eventually problem came back with full force. This happens because this problem fluctuates in severity on its own, which is very deceitful. Even if you do nothing, the problem goes away, comes back on its own and changes in severity without any interventions. Also, i'm also unable to know what issues other people are having and perceiving. So, we might be talking about the same, similar or completely different things.

So you have observed this phenomenon yourself and yet you still say others are experiencing placebo with it? What kind of logic is that.

I can make this problem go away in 15 seconds, by switching the internet from one to another. Same server, same people, same map, everything is the same, an near immediate disconnection & reconnetion (15 seconds difference) to the server. One connection (Fibre 1 Gb/s both ways, ~10ms) = problem is constant, only the severity of it changes all the time, second connection (mobile internet, don't know the speed, most likley the usual ~20-80 Mb/s, ping ~25ms) = problem is literally non existent. I'm sorry, but you can't beat this with anything you do, say or hypothesize.

This is great, I am truly happy for you (no sarcasm here). Have you gotten any closer to explaining what is wrong with the connections from the ISP compared to the telecom (4G network)? A few people tried dedicated lines, with self purported "network issues/desync etc" and it did nothing. Can you explain how this might be the case? Those dedicated lines are business grade, very costly and I highly doubt there is any throttling or congestion issues happening (hence its called a dedicated line). I also believe anything like that would be very measurable, and yet you have not been able to provide any statistics that reflect the issues you experience- just like the folks that suffer the electrical related stuff.

I personally think our problems are more similar than you might think. When I stream video to my laptop/phone or other wireless devices the video clarity is extremely poor at times (this varies). The colors are insanely washed out, there is heavy screen tearing, video hitching/buffering at random times, and often audio/video become de synchronized. I also get incredibly random disconnects to my network (the modem never reboots) and overall poor network performance with incredibly varied consistency). This only happens on my ISP provided connection. When on data (4G LTE Verizon) the video once again becomes very smooth with none of the symptoms I explained above. All of these devices have heavy input latency (even when on battery) which is unaffected by switching to Mobile LTE connection. My connection speeds are very good (ping, download, upload and jitter), modem signal and power is extremely clean (as far as the ISP techs, and my electrician are concerned) and I own commercial grade networking equipment (tried 7 different routers, and 3 modems). Also replaced all of the coaxial cables and connections on my property. Had ISP techs out multiple times. Nothing adds up.

I am not trying to beat your solution, I am merely searching for solutions that are practical and illuminate the real cause of these problems. Similarly you cannot beat the fact that when I disconnect my Online UPS from the mains AC input lag is gone and my system runs as intended. In your case the source is from coming straight from that ISP connection whereas in my case that network connection is just another factor in the whole of the equation.

[nuggify]

I assume you have tried different mixes of split core ferrite. Mix j(75) for example.. please update your further findings, it is helpful

Hey thanks for asking, yes I have tried a variety of ferrite and toroid cores. They both have a reproducible effect on the problems/symptoms. However it is not a fix in any way. Placing them on data signal cables makes the symptoms worst very quickly. On power cables it helps for a bit but seems to stop working and in some cases makes it worst as well. The fact that these have an effect on the problems clearly points to some high frequency currents traveling on these conductors (the cables themselves).

[Anonymous768119]

Are you the only person using internet when you play? Have you tried a router(or even a switch) with QoS supporting fq_codel?

QoS makes no difference. It has nothing to do with the internet.

[mello]

In regards to all the ridiculous claims I think I have already posited the idea that it could be they have this very problem I have. For me BIOS changes, Windows settings and all of that stuff does cause system responsiveness to change (no it is not placebo as it is reproducible although it varies) but it is temporary.

Yes, especially the settings that can't possibility affect, in any way, shape or form, the things one is trying to fix or improve. These tips & tricks, optimization guides etc. are literally the best. As i said, i have seen this insanity before in a gaming community. It was happening 20+ years ago and it is still going to this very day. Are you aware that majority of pros literally do nothing to their systems and they are clueless when it comes to bios/windows settings and they simply don't waste their time with this ? They just buy hardware, turn on the PC, connect mouse/keyboard/headphones and play with everything basically on default settings. Same with most casuals, who play on random PCs, laptops, default peripherals and settings. Do you want to know why that works for many of them ? Because the INTERNET performance is the single biggest factor when playing online, nothing else even comes near close to it.

In regards to the audiophile market I will give you an example. Many power conditioners and filters are sold to improve the sound, and many, many folks say they have tried the device in question with no observable difference to sound quality. But did you ever consider that those people do not benefit from noise reduction on their homes power grid? Nearly everyone has a different situation and setup. Some will benefit and some will not. The real question to ask is what are the actual specs and components of the power conditioning in the first place and is it targeting the problems you may have with your power quality.

Correct. But have you considered the fact that when it is all put into the AB test, people fail to hear the difference and "see" what is what ? Also, changes in sound does not equal an "improvement" or making it "better". This is also one of the biggest misconceptions there. Just because it sounds (or feels) different, it does not mean that the change in eq has caused the worsening or the improvement of the sound. It just as well, might be just... different.

So you have observed this phenomenon yourself and yet you still say others are experiencing placebo with it? What kind of logic is that.

Now, this is the problem i was talking about. I think i was pretty clear what i meant, and yet you still don't get it. If you do nothing, nothing at all, the problem still fluctuates in severity, disappears for short periods and comes back on its own. So, why do you think that making ANY CHANGES, ANY AT ALL, makes any difference or improves things, even for a short while ? Why do you insist on that, when people are making any kind changes to their hardware, software, settings etc., that it actually changes or improves anything ? Think about it from the psychological perspective and how human brain works. This is a collective insanity and people are simply fooling themselves and they don't even know it.

This is great, I am truly happy for you (no sarcasm here).

No need to be happy as mobile internet (the kind i have an access to) is not suitable for gaming 24/7, because when lags appear (the kind of lags that were present on old internet connections that were available in ~1995-2005), then you can forget about gaming. It simply becomes unplayable & frustrating, but... in a very different, old fashioned way ;>

Have you gotten any closer to explaining what is wrong with the connections from the ISP compared to the telecom (4G network)?

The same thing as with any other connection. The configuration / devices / software used within the network to control the traffic. The easiest explanation is that certain ISP's are using specific configurations / devices / software that directly interferes with gaming UDP packets. It is most likely advertised as a benefit / feature by the manufacturers, programmers & IT specialists to the entire network and its stability. The fact that it destroys online gaming performance is a side "feature" that no one is suspecting or is aware of. So, what you end up with (quality of the internet connection) is literally based on luck, place you live in and what your ISP's are using in their infrastructure. With the very specific way this issue occurs and manifests itself in online games it becomes obvious that is it man-made artifact.

I suspect that this problem is directly tied to the upstream channel. So, something happens to the UDP packets, they are either being throttled, delayed or delivered out of order. Based on what is actually happening in games (how the problem manifests itself) this makes perfect sense. Very recently i have been testing this whole thing for months in CS 1.6 (same private server, same community of people playing) and have discovered (last 48h) that this problem is at least partially related to what is called a 'prediction errors' and smoothing. When you disable prediction error smoothing (via certain cvars) or use the default (very low) value, then you get what many people are experiencing, a huge variability in gaming performance (if your internet is shitty). Very good or perfect internet connections will do fine with no smoothing or with the default setting, but others will experience enemy players being incorrectly presented on their screens. This results in instant deaths & enemies seeing you before you can see them. This also most likely contributes to the hit registration & many other issues. But, guess what ? When you increase prediction error smoothing (most likely beyond what game / netcode developers ever intended to be used) to a very high value, it magically fixes enemy positions instantly (!) (good enough, but not as good as on my mobile connection or on LAN obviously), but in exchange it creates some weird screen view jumping effect (bearable) & and you can no longer watch other player POVs because the view is no longer smooth. This can't be considered as a fix, but it certainly improves few crucial aspects (like correct enemy model positioning) when it comes to gaming online. So, now the question becomes... why changing the lag compensation algorithm cvars (prediction error smoothing) improves & partially fixes these issues instantly ? According to the all available scientific knowledge around networking, my Fibre 1Gb/s should be perfect, right ? Ping ~10, 0 loss, minimal jitter, no measurable lags, a dream not achievable for the most people around the world. According to the all conventional knowledge this problem should not exist... and yet, putting in an aggressive lag compensation algorithm in place, partially fixes this issue !? Wait.. how that can possibly even happen if i have NO MEASURABLE LAGS OR ERRORS using all the conventional means & analytical tools ? Well, this suggests a major flaw in internet connection design and how errors are being reported & can be measured on the client side. And before anyone says that experts, engineers, programmers & IT networking specialists know what they are doing and some random guy on the internet obviously doesn't... well, this wouldn't be the first time when great minds hadn't foreseen certain issues occuring in their technologies & implementations. This would also explain, why no one can fix this issue and why everyone is chasing their own tails with no end in sight. People are completely disregarding the internet as the source of this issue from the very start ! An occam's razor principle. The truth is literally staring everyone in the face and people simply can't see it.

And going a little further with this, this would mean that the idea of competitive gaming online is nothing more but pure and meaningless garbage. If your "skill" and majority of results in online games are literally decided by the quality of your internet connection, this means that most of the results and "success" is already pre-determined, regardless of your true skill & capabilities in games (mostly FPS games). In fact, this is exactly what i see when playing online with and against others or by watching well known or famous streamers. If you have perfect internet connection you can achieve much, much greater results, than you would have achieved otherwise, especially if others also had the same kind of connection you have. This is why you can see weak, average or slightly above average players getting insane kills, highlight reels, good results while having insanely good hit registration. They are able to do this, because other players, even if they are better, do not have good internet connections. This unfortunately creates an illusion of who is actually good, skilled, elite or at the pro level. Everyone who is at least slightly affected by this problem is basically skill capped to a certain degree by the quality of their own internet connection. And all of this comes from someone who was actually playing at the true pro level long time ago.

A few people tried dedicated lines, with self purported "network issues/desync etc" and it did nothing. Can you explain how this might be the case? Those dedicated lines are business grade, very costly and I highly doubt there is any throttling or congestion issues happening (hence its called a dedicated line). I also believe anything like that would be very measurable, and yet you have not been able to provide any statistics that reflect the issues you experience- just like the folks that suffer the electrical related stuff.

Everything i said might still apply to any internet connection. The root cause of the problem is still unknown, but everything points out not to random & natural occurrence but to a man-made artifacts / intervention. If it happens also on dedicates lines, this would suggest it is directly tied to configuration / devices / software used by the ISP's, as i have explained above. An "out of the box" feature of sorts.

I am not trying to beat your solution, I am merely searching for solutions that are practical and illuminate the real cause of these problems. Similarly you cannot beat the fact that when I disconnect my Online UPS from the mains AC input lag is gone and my system runs as intended. In your case the source is from coming straight from that ISP connection whereas in my case that network connection is just another factor in the whole of the equation.

Tried an Online UPS the same way you did, no change. And i'm still quite convinced that most people who think that what they have are power / EMI issues, what they actually have is shitty internet. They are just fooling themselves with all the mind tricks that our brains have at their disposal. A collective insanity in year 2020 around corona issue should be good enough evidence of that. But once again, i'm not saying that power / EMI issues are not real for some people, but these issues absolutely need to occur outside of online gaming too, to be considered real.

[Chief Blur Buster]

And i'm still quite convinced that most people who think that what they have power / EMI issues, what they actually have is shitty internet. They are just fooling themselves with all the mind tricks that our brains have at their disposal. A collective insanity in year 2020 around corona issue should be good enough evidence of that. But once again, i'm not saying that power / EMI issues are not real for some people, but these issues absolutely need to occur outside of online gaming too, to be considered real.

I totally agree with you on this. That's why I remind people about the wild goose chase full of red herrings.

That said, we've allowed EMI discussions to continue; in the hopes that improved offline troubleshooting techniques are gradually discovered over the longer term. There so many dominoes conspiring against the refresh rate race to retina refresh rates.

Alas, I can only do so much in this specific topic territory, given the hard-to-troubleshoot nature of this type of problem. These discussions do tend to go in circles.