[USG Flex 500H + NWA130BE] - Different speedtest value between LAN and Wifi
All Replies
-
Hi @Maverick87 I hope that what I'm writing would be considered as a pursuit to the best possible analysis, not some tech banter.
Obviously, there is always an adapter with own driver (USB-A/USB-C/Thunderbolt/PCI-Express), it's always an adapter internal or external to the PC. Also, there is always a driver that "translate" the information into human-readable. So can be normal.
Yes and no.
Data is put into subsequent "envelopes", and any "envelope" can change the theoretical maximal speed achieved. In these times, PCIe is the best "envelope" available to consumer devices, but still is not always the powerful enough-one for deliver full duplex 2.5GBe, even if PCIE 3.0x is usually "enough". It's not guarantee that any connection between adapters is delivering "enough" bandwidth to max-out 2.5GBe, expecially if it's connected throught other interfaces like USB or Thunderbolt, but it do not ends here.
USB and Thunderbolt slow down PCIe, so if the ethernet chip is more "CPU dependant" than other, the overhead for transferring such data could be also increased; Realtek are nice network adapter, delivers stable and reliable connection, usually at good performances but… there are better chips, with more computational power and that requires less CPU "attention" to work properly.Gigabit Ethernet is quite stale network interface, roughly 1999. However, I was able to seen maxed-out the performance of these interface (via SMB) only through a "heavily" cached NAS on one side and a SATA SSD on the other one.
Therefore.
Before taking measures with speedtesters, Iperf on the same network segment is far more useful to understand what you're achieving from your network adapter. It's harder obviously compare wireless lan with LAN because also the AP plays a major role into data exchange.
Last but not least, i'm a "wire guy", whenever I cannot live without Wireless lan. It's simply better, becaue have less latency, less throughtput issues and it's connected to the switch, which can usually handly way more datatransfer than a bajillion of APs.0 -
Hi @mMontana,
I completely agree with you regarding the "wire guy" (I had my home renovated two years ago, and the first thing I told the contractor was that they had to ensure I had Cat6A cabling running to 15 RJ45 outlets throughout the house).
In any case, it’s clear that speed test results are lower than Iperf results, just as it’s clear that hardware plays a crucial role.
The problem is, I can't figure out how it's possible — given identical interface configurations — to get two distinct speeds (differing by even a few hundred Mbit) between ports P2 and P4.
I also need to understand why I get lower speeds using LAN compared to Wi-Fi.That said, I’m well aware that higher LAN throughput requires more CPU power to manage; however, my tests were all conducted using Realtek adapters on a system featuring a 12-core ARM 64-bit Snapdragon X processor, 64GB of RAM, and an NVMe SSD.
0
Categories
- All Categories
- 442 Beta Program
- 3.1K Nebula
- 241 Nebula Ideas
- 6.8K Security
- 755 USG FLEX H Series
- 380 Security Ideas
- 1.8K Switch
- 87 Switch Ideas
- 1.5K Wireless
- 58 Wireless Ideas
- 7.2K Consumer Product
- 321 Service & License
- 512 News and Release
- 99 Security Advisories
- 31 Education Center
- 10 [Campaign] Zyxel Network Detective
- 5.3K FAQ
- 34 Documents
- 90 About Community
- 119 Security Highlight
Guru Member
Master Member