I would like to purchase the a PCI-E ethernet card with the RTL8111H chipset you've mentioned, however, your findings seem very anecdotal without much evidence.6yToFindTheAnswer wrote: ↑16 Nov 2022, 22:48Thank you guys for your interest on this topic, i'm reading everything Very interesting to see that so many people have the same problem as me; with exactly the same network chip (intel i219-v)... There are already at least 10 people only on this post claiming having the same issues.
Yes it is. Ever since i bought this mobo in 2017 nothing changed (same issue ever since with all the different drivers and windows versions, etc).
Yes, yes, yes, totally agree with you here. Found a friend with a msi Z390 having the i219-v chip, so i asked him (without any previous explanation) to unplug his ethernet cable and see if there is a difference in mouse feeling... guess what? Yep he was losing his mind, as i personally felt on my rig, he felt the same on his; huge difference in mouse movementsChief Blur Buster wrote: ↑07 Nov 2022, 21:42There are indeed crappy on-board Ethernet implementations, and I've sometimes found a PCIe Ethernet card to produce better results.
Other times, the Ethernet port on a motherboard is really good.
It's very motherboard dependant.
I bought a bunch of PCie to ethernet card and one USB-C to ethernet here are my findings :
1st card : chip : Realtek 8168 : Very bad result overall. Mostly having huge ping spikes CONSTANTLY without much load making online gaming impossible. Didn't go further with this one.
2nd card : chip : Realtek 8111H : Now this one is a very very good surprise, it was super stable and seems like it handles small packets very well making it a superb chip for competitive online gaming. I got 2ms less ping overall on every servers on it was a VERY stable card.
3rd card : chip : Broadcom 5751 : Because my router have a Broadcom chip, i was curious to buy a PCIe to Ethernet with a chip from Broadcom. This chipset was very weird : it is insanely stable in downloads (minimal jitter) and as soon as you upload something it starts to jitter (8ms +) this was not present on every other chip tested so it is an isolated issue specifically to this chip. When you shoot in csgo or valorant there is a very noticeable delay between the click and the hits on the bodies (like if you play with 150ms ping)... So it is not a good chip because as soon as you start shooting (upload) it starts to slightly jitter. Also it handles big packets very well but small packets makes it jitter (wtf???)
4th is a USB-C to Ethernet : chip : REALTEK 8153 . I don't know what dark magic went into this one but as soon as i started testing my ping was 3ms lower on every server and it never, lowest amounts of jitter in download / uploads between all the cards. Never seen such a stable chip. My download speed was 2mbp/s slower than on the other cards but i will take it everyday for the stability it provides.
In conclusion : Yes a PCIe to ethernet card solve my issue. As soon as i plug my ethernet cable to the integrated chip in my motherboard (Intel i219-v) my mouse movements are slower (like if the DPI decrease) and i can even feel it on desktop. With all that has been said before (heavy feeling in game, bad hitboxes, getting pre-shots, etc ) So having a PCIe card solve all those issue. Also you have to buy a bunch like me and test them because there are some differences between them, some are far better than others. They are not equivalent.
Also don't believe what you read on Reddit; people are writing eevrywhere that intel chips are far better and years ahead of realtek ones (not true at all) Seems like people like to repeat what they heard without having anything to back their claims... People seems to forget that Intel doesn't have such a clear and bright past in network chips history as people tend to make you believe...
yes indeed, if you don't know, there was a huge lawsuit against Intel a few years ago for their PUMA 6 and PUMA 7 chips that were using in millions of routers.
in summary : https://approvedmodemlist.com/intel-pum ... t-defects/
Those were not firmware related issues but it was the chip in itself, bad conception.
Z200, Z300 and Z400 series motherboard were mostly using intel chips... don't you find weird that with newer AMD/Intel mobo generations all the new ones that comes out are using Realtek network chips (like 80% of the past 2 gens and even newer ones) even the high ends ones. Such a weird U-turn that mobo manufacturers have made, yet 0 public announcement but there are reasons for sure...
Here are an example : most of the new B550 mobos that came out use a Realtek 8111H chip (yes the same one i tested) which is a great chip, one of the best i tested. But you know what? this chip was released in 2011 almost 12 years old and it still beats the newer poorly made intel ones....
Could you provide us with a more detailed explanation on:
- How are your ethernet settings are setup,
- What drivers you're using,
- Does the chipset support MSI and MSI-X (the latter being better),
- What your DPC latency is like (using xperf from Windows ADK, not LatencyMon) before and after
I cannot seem to find any specific recommendations on which chipsets to avoid (this seems to be easier to find out than to find which are ok), but there have definitely been multiple reports on newer i2xx chipsets not feeling OK for people.
I myself have a Atheros AR8151 onboard ethernet chipset, which is bringing me problems in form of DPC latency spikes (according to LatencyMon, I have not had time to test it with xperf) thus giving me unstable polling rate (or perhaps some other Windows issue is not allowing me to have 1000.5 and 999.5 polls all the time without spikes), which is what got me into the rabbit hole of ethernet drivers and chipsets
The adapter supports MSI https://github.com/djdallmann/GamingPCS ... /README.md according to the command used in this link, but not MSI-X, thus not allowing me to utilise RSS properly. I have a very 'old' CPU so any CPU % saved can impact my experience with high polling rate mice (2kHz+), which is what got me into this whole rabbit hole.
To clarify, the spikes are very small and very likely imperceptible in the grand scheme of things, however I'd like to just add my thoughts here as reference.
It could be possible that earlier Intel chipsets (specifically found in the "Intel(R) Gigabit CT Desktop Adapter" as per link above) don't suffer from this input lag issue as mentioned by the OP and that drivers on newer cards are playing tricks on people (most of this stuff is very hard to discern for individuals, as seen by the A/B input lag thread where most scored ~10ms lag only)
Of course, I have no ways to confirm anything, but I'd just like to give my thoughts on all of this while also mentioning my experience going down the hole of analysing ndis.sys (networking in Windows) DPC spikes.