Potentially overlooked problem that can be mistaken for EMI

Everything about latency. Tips, testing methods, mouse lag, display lag, game engine lag, network lag, whole input lag chain, VSYNC OFF vs VSYNC ON, and more! Input Lag Articles on Blur Busters.
Post Reply
Eonds
Posts: 262
Joined: 29 Oct 2020, 10:34

Potentially overlooked problem that can be mistaken for EMI

Post by Eonds » 11 May 2022, 00:59

I just realized that my display port cable is getting heated up dramatically by my GPU exhaust. I think you guys should keep an eye out for that. I think this could help some people. I'm currently testing this and it honestly could of been my MAIN issue.

woodyfly
Posts: 91
Joined: 03 Jul 2020, 07:53

Re: Potentially overlooked problem that can be mistaken for EMI

Post by woodyfly » 14 May 2022, 19:31

theres no way lol

Thatweirdinputlag
Posts: 305
Joined: 27 Aug 2021, 14:09

Re: Potentially overlooked problem that can be mistaken for EMI

Post by Thatweirdinputlag » 15 May 2022, 02:04

Could be, since the operating temperature of the cable itself is up to 50c "what a simple search on google said". But in general, most of the people having the "issues" are FPS players playing games like CSGO or Valorant on low settings, when their GPU barely hits 50-55c in any given session. So this specific case might be only on your side if you put heavy loads on your GPU constantly.

But even if your GPU is hitting 80c constantly, thats the die itself, there will be multiple stages where heat loss would happen while being transferred. I.E the GPU cooler in general will have a lower temperature, and the air running through that cooler would even have a lower temperature than the cooler itself "Physics". So by the time the air gets exhausted from the back, yes it will warm up the cable but to a lesser degree, I'd imagine the inner parts of the cable barely reaching 40c. But thats just me speculating!

Do you have a small fan to point at your DPI cable for a test?
Rog Strix Z79i - Intel 13700K - 4090 OC ROG Strix - 7200 Trident G.Skill - 1TB SK Hynix Platinum P41 - 1000W ATX3.0 Asus Tuf - 34'' Odyssey OLED G8 - FinalMouse Tenz S/Pulsar Xlite V2 Mini - Wooting 60HE - Sennheiser HD 560s - Shure SM7b - GoXLR Mini

Anonymous768119

Re: Potentially overlooked problem that can be mistaken for EMI

Post by Anonymous768119 » 17 May 2022, 05:10

Can't see the correlation. I get much better CSGO performance in summer (no A/C) and still don't know why. Can be lesser internet traffic congestion or some factor in atmospheric parameters like humidity (rather doubtful).

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: Potentially overlooked problem that can be mistaken for EMI

Post by Chief Blur Buster » 18 May 2022, 20:26

woodyfly wrote:
14 May 2022, 19:31
theres no way lol
It's very hard to trace these kinds of things. Extra heat can improve/worsen different computers, but most of it will be retransmit-error-correction latency changes caused by temperature changes.

DisplayPort uses Reed-Solomon Forward Error Correction (FEC) but FEC can be enabled/disabled on DisplayPort. FEC was introduced in DisplayPort 1.4 which can have slightly more latency than non-FEC DisplayPort 1.2 for the same cable bandwidth. It's possible to dynamically enable/disable FEC on DisplayPort in certain circumstances, creating a very slight lag change. However, heating would not be the culprit.

In theory, longshot though, thermal expansion increasing the S/N ratio of the cable to trigger an enabling of FEC in theory, but this is unlikely here. Also, changing the bitrate of a DisplayPort cable (different number of gigabits per second) can change the latency too, e.g. badly made or high-EMI situations on DisplayPort 2.0 cables getting too many errors and then transceivers falling back to a DisplayPort 1.4 mode, etc. So one computer bootup may have some weird extra DisplayPort lag that a different computer bootup does not have, because the cables automatically trained/sync'd itself to a specific bitrate/FEC/etc upon startup handshake.

See DisplayPort FEC Technical Information

More likely, higher heat is affecting a different problem -- Random example is there could be a ~1% chance of a overheated NVIDIA GDDR6 memory chips that is running extra repeat-reads and increasing latency. Who knows -- it's really hard to trace. It can be any other component that is near its S/N margin -- anything that use retroactive / repeat / retransmit error correction is prone to EMI lag caused by temperature changes too.

Silver lining, extra heat can speed up certain types of electronics -- by unusual mechanisms such as pushing certain resonant-frequency-type EMI regions (e.g. two different chips interfering with each other at colder temperatures but not at warmer temperatures). (Spread spectrum is designed to fixes resonant-EMI types of issues too).

Another example is thermal expansion -- warmer tempertures caused contacts to thermally expand and become tighter and more solid with less EMI -- it might be just one PCIe pin or one CPU pin or one port pin becoming tighter in its socket -- thousands of pins, who knows? Causing error-correction latencies to drop. Etc, etc.

There are million different dominoes that can tip, and it's hard to tell what the hell happened -- could be the Internet too -- Rheoretically, could easily be a DSL line performing better because the phones box out on the curb had drier, tighter, wires during warm weather.

Almost impossible to trace these sort of things without spending a huge painful truckload of hours of troubleshooting time.

Doubt it is the cable, but it's not impossible, given automatic error correction algorithm changes can cause latency changes...
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

Post Reply