Device performance under OpenWrt

The same measurement across every ported device, so one board can be compared with another on identical firmware.

Two frame sizes are published here, 60 B and 1400 B, and both as bandwidth on the cable: megabits per second counted the way the wire carries them, each frame plus its checksum, preamble and inter-frame gap (L1). Measured the same way at both sizes, every wired bar reads against the same gigabit link, so a 60 B bar and a 1400 B bar can be compared directly. Whether a given bar is AT the link's capacity or far below it is what the bar itself says, and the caption claims nothing about which is which. WiFi figures have no cable framing and are counted at the frame itself (L2). Every size the suite measures, and every qualifier each figure was measured with, stays in the dated artifact these pages are generated from.

DHCP LAN->WAN, 60 B frames

G24W 238 Mbps V2801RGW 240 Mbps X100DG 275 Mbps X111W 329 Mbps X400AXF 248 Mbps

DHCP WAN->LAN, 60 B frames

G24W 460 Mbps V2801RGW 466 Mbps X100DG 476 Mbps X111W 576 Mbps X400AXF 489 Mbps

DHCP+VLAN LAN->WAN, 60 B frames

G24W 220 Mbps V2801RGW 238 Mbps X100DG 236 Mbps X111W 300 Mbps X400AXF 336 Mbps

DHCP+VLAN WAN->LAN, 60 B frames

G24W 337 Mbps V2801RGW 430 Mbps X100DG 444 Mbps X111W 665 Mbps X400AXF 663 Mbps

PPPoE CHAP LAN->WAN, 60 B frames

G24W 214 Mbps V2801RGW 204 Mbps X100DG 208 Mbps X111W 215 Mbps X400AXF 214 Mbps

PPPoE CHAP WAN->LAN, 60 B frames

G24W 134 Mbps V2801RGW 180 Mbps X100DG 180 Mbps X111W 201 Mbps X400AXF 203 Mbps

PPPoE CHAP+VLAN LAN->WAN, 60 B frames

G24W 182 Mbps V2801RGW 210 Mbps X100DG 17.2 Mbps X111W 212 Mbps X400AXF 204 Mbps

PPPoE CHAP+VLAN WAN->LAN, 60 B frames

G24W 139 Mbps V2801RGW 152 Mbps X100DG 21.5 Mbps X111W 178 Mbps X400AXF 138 Mbps

PPPoE PAP LAN->WAN, 60 B frames

G24W 203 Mbps V2801RGW 208 Mbps X100DG 214 Mbps X111W 213 Mbps X400AXF 205 Mbps

PPPoE PAP WAN->LAN, 60 B frames

G24W 118 Mbps V2801RGW 151 Mbps X100DG 184 Mbps X111W 200 Mbps X400AXF 202 Mbps

PPPoE PAP+VLAN LAN->WAN, 60 B frames

G24W 194 Mbps V2801RGW 202 Mbps X100DG 16.9 Mbps X111W 208 Mbps X400AXF 210 Mbps

PPPoE PAP+VLAN WAN->LAN, 60 B frames

G24W 153 Mbps V2801RGW 155 Mbps X100DG 29.1 Mbps X111W 174 Mbps X400AXF 150 Mbps

DHCP LAN->WAN, 1400 B frames

G24W 999 Mbps V2801RGW 1000 Mbps X100DG 1000 Mbps X111W 1000 Mbps X400AXF 1000 Mbps

DHCP WAN->LAN, 1400 B frames

G24W 979 Mbps V2801RGW 1000 Mbps X100DG 1000 Mbps X111W 1000 Mbps X400AXF 1000 Mbps

DHCP+VLAN LAN->WAN, 1400 B frames

G24W 994 Mbps V2801RGW 1000 Mbps X100DG 1000 Mbps X111W 1000 Mbps X400AXF 1000 Mbps

DHCP+VLAN WAN->LAN, 1400 B frames

G24W 960 Mbps V2801RGW 992 Mbps X100DG 1000 Mbps X111W 1000 Mbps X400AXF 1000 Mbps

PPPoE CHAP LAN->WAN, 1400 B frames

G24W 601 Mbps V2801RGW 1000 Mbps X100DG 999 Mbps X111W 999 Mbps X400AXF 1000 Mbps

PPPoE CHAP WAN->LAN, 1400 B frames

G24W 408 Mbps V2801RGW 815 Mbps X100DG 812 Mbps X111W 826 Mbps X400AXF 840 Mbps

PPPoE CHAP+VLAN LAN->WAN, 1400 B frames

G24W 931 Mbps V2801RGW 1000 Mbps X100DG 282 Mbps X111W 1000 Mbps X400AXF 710 Mbps

PPPoE CHAP+VLAN WAN->LAN, 1400 B frames

G24W 645 Mbps V2801RGW 786 Mbps X100DG 154 Mbps X111W 769 Mbps X400AXF 678 Mbps

PPPoE PAP LAN->WAN, 1400 B frames

G24W 485 Mbps V2801RGW 1000 Mbps X100DG 1000 Mbps X111W 999 Mbps X400AXF 1000 Mbps

PPPoE PAP WAN->LAN, 1400 B frames

G24W 513 Mbps V2801RGW 779 Mbps X100DG 817 Mbps X111W 844 Mbps X400AXF 835 Mbps

PPPoE PAP+VLAN LAN->WAN, 1400 B frames

G24W 1000 Mbps V2801RGW 1000 Mbps X100DG 265 Mbps X111W 1000 Mbps X400AXF 1000 Mbps

PPPoE PAP+VLAN WAN->LAN, 1400 B frames

G24W 784 Mbps V2801RGW 793 Mbps X100DG 108 Mbps X111W 759 Mbps X400AXF 777 Mbps

WiFi

Each device shows its highest valid measured OpenWrt throughput across WiFi bands for the same direction, WAN service and frame size. Each bar names its own operating point -- band, width, standard and cipher; the channel is recorded per measurement in the dated artifact, and is left off the bar because at one band, width, standard and cipher two channels negotiate the same thing -- beside its figure, because bars on one chart may have been taken at different points; where they differ, the gap between them includes that difference and no percentage is claimed. These are end-to-end figures over the device's own WAN, not an isolated radio measurement, and the client, the channel and the air are in them too. A device with no WiFi does not appear here at all.

WiFi -> WAN -- DHCP, 60 B frames

G24W 19.6 Mbps 2g/20MHz/HT/CCMP V2801RGW 12.4 Mbps 2g/20MHz/HT/CCMP X111W 11.1 Mbps 2g/20MHz/HT/CCMP X400AXF 58.1 Mbps 5g/80MHz/HE/CCMP

WiFi -> WAN -- DHCP+VLAN, 60 B frames

G24W 9.36 Mbps 2g/40MHz/HT/CCMP V2801RGW 14.6 Mbps 2g/20MHz/HT/CCMP X111W 13.9 Mbps 2g/40MHz/HT/CCMP X400AXF 86.2 Mbps 5g/80MHz/HE/CCMP

WiFi -> WAN -- PPPoE CHAP, 60 B frames

G24W 9.38 Mbps 2g/40MHz/HT/CCMP V2801RGW 11.8 Mbps 2g/20MHz/HT/CCMP X111W 9.74 Mbps 2g/20MHz/HT/CCMP X400AXF 48.9 Mbps 5g/80MHz/HE/CCMP

WiFi -> WAN -- PPPoE CHAP+VLAN, 60 B frames

G24W 6.65 Mbps 2g/40MHz/HT/CCMP V2801RGW 5.12 Mbps 2g/20MHz/HT/CCMP X111W 10.8 Mbps 2g/20MHz/HT/CCMP X400AXF 82.6 Mbps 5g/80MHz/HE/CCMP

WiFi -> WAN -- PPPoE PAP, 60 B frames

G24W 9.12 Mbps 2g/40MHz/HT/CCMP V2801RGW 3.94 Mbps 2g/20MHz/HT/CCMP X111W 10.7 Mbps 2g/20MHz/HT/CCMP X400AXF 49.9 Mbps 5g/80MHz/HE/CCMP

WiFi -> WAN -- PPPoE PAP+VLAN, 60 B frames

G24W 10.1 Mbps 2g/40MHz/HT/CCMP V2801RGW 1.52 Mbps 2g/20MHz/HT/CCMP X111W 10.6 Mbps 2g/20MHz/HT/CCMP X400AXF 47.7 Mbps 5g/80MHz/HE/CCMP

WAN -> WiFi -- DHCP, 60 B frames

V2801RGW 7.42 Mbps 2g/20MHz/HT/CCMP X111W 7.98 Mbps 2g/20MHz/HT/CCMP X400AXF 106 Mbps 5g/80MHz/HE/CCMP

WAN -> WiFi -- DHCP+VLAN, 60 B frames

V2801RGW 2.17 Mbps 2g/20MHz/HT/CCMP X111W 3.09 Mbps 2g/40MHz/HT/CCMP X400AXF 183 Mbps 5g/80MHz/HE/CCMP

WAN -> WiFi -- PPPoE CHAP, 60 B frames

V2801RGW 2.17 Mbps 2g/20MHz/HT/CCMP X111W 3.52 Mbps 2g/40MHz/HT/CCMP X400AXF 53.3 Mbps 2g/40MHz/HE/CCMP

WAN -> WiFi -- PPPoE CHAP+VLAN, 60 B frames

V2801RGW 2.17 Mbps 2g/20MHz/HT/CCMP X111W 2.10 Mbps 2g/40MHz/HT/CCMP X400AXF 88.4 Mbps 5g/80MHz/HE/CCMP

WAN -> WiFi -- PPPoE PAP, 60 B frames

V2801RGW 7.14 Mbps 2g/20MHz/HT/CCMP X111W 2.03 Mbps 2g/40MHz/HT/CCMP X400AXF 118 Mbps 5g/80MHz/HE/CCMP

WAN -> WiFi -- PPPoE PAP+VLAN, 60 B frames

V2801RGW 2.24 Mbps 2g/20MHz/HT/CCMP X111W 3.34 Mbps 2g/40MHz/HT/CCMP X400AXF 117 Mbps 5g/80MHz/HE/CCMP

WiFi -> WAN -- DHCP, 1400 B frames

G24W 14.1 Mbps 2g/20MHz/HT/CCMP V2801RGW 43.0 Mbps 2g/20MHz/HT/CCMP X111W 92.4 Mbps 2g/20MHz/HT/CCMP X400AXF 689 Mbps 5g/80MHz/HE/CCMP

WiFi -> WAN -- DHCP+VLAN, 1400 B frames

G24W 43.3 Mbps 2g/20MHz/HT/CCMP V2801RGW 76.1 Mbps 2g/20MHz/HT/CCMP X111W 100 Mbps 2g/40MHz/HT/CCMP X400AXF 85.5 Mbps 2g/40MHz/HE/CCMP

WiFi -> WAN -- PPPoE CHAP, 1400 B frames

G24W 102 Mbps 2g/40MHz/HT/CCMP V2801RGW 81.3 Mbps 2g/20MHz/HT/CCMP X111W 101 Mbps 2g/20MHz/HT/CCMP X400AXF 184 Mbps 5g/80MHz/HE/CCMP

WiFi -> WAN -- PPPoE CHAP+VLAN, 1400 B frames

G24W 132 Mbps 2g/40MHz/HT/CCMP V2801RGW 79.7 Mbps 2g/20MHz/HT/CCMP X111W 60.4 Mbps 2g/40MHz/HT/CCMP X400AXF 258 Mbps 5g/80MHz/HE/CCMP

WiFi -> WAN -- PPPoE PAP, 1400 B frames

G24W 41.0 Mbps 2g/20MHz/HT/CCMP V2801RGW 84.5 Mbps 2g/20MHz/HT/CCMP X111W 102 Mbps 2g/20MHz/HT/CCMP X400AXF 160 Mbps 5g/80MHz/HE/CCMP

WiFi -> WAN -- PPPoE PAP+VLAN, 1400 B frames

G24W 10.3 Mbps 2g/40MHz/HT/CCMP V2801RGW 82.0 Mbps 2g/20MHz/HT/CCMP X111W 100 Mbps 2g/20MHz/HT/CCMP X400AXF 225 Mbps 5g/80MHz/HE/CCMP

WAN -> WiFi -- DHCP, 1400 B frames

G24W 6.00 Mbps 2g/20MHz/HT/CCMP V2801RGW 80.1 Mbps 2g/20MHz/HT/CCMP X111W 69.8 Mbps 2g/20MHz/HT/CCMP X400AXF 579 Mbps 5g/80MHz/HE/CCMP

WAN -> WiFi -- DHCP+VLAN, 1400 B frames

V2801RGW 37.6 Mbps 2g/20MHz/HT/CCMP X111W 37.5 Mbps 2g/40MHz/HT/CCMP X400AXF 336 Mbps 5g/80MHz/HE/CCMP

WAN -> WiFi -- PPPoE CHAP, 1400 B frames

V2801RGW 39.9 Mbps 2g/20MHz/HT/CCMP X111W 39.3 Mbps 2g/40MHz/HT/CCMP X400AXF 277 Mbps 5g/80MHz/HE/CCMP

WAN -> WiFi -- PPPoE CHAP+VLAN, 1400 B frames

V2801RGW 37.3 Mbps 2g/20MHz/HT/CCMP X111W 34.8 Mbps 2g/40MHz/HT/CCMP X400AXF 381 Mbps 5g/80MHz/HE/CCMP

WAN -> WiFi -- PPPoE PAP, 1400 B frames

V2801RGW 27.8 Mbps 2g/20MHz/HT/CCMP X111W 39.0 Mbps 2g/40MHz/HT/CCMP X400AXF 359 Mbps 5g/80MHz/HE/CCMP

WAN -> WiFi -- PPPoE PAP+VLAN, 1400 B frames

V2801RGW 27.8 Mbps 2g/20MHz/HT/CCMP X111W 33.8 Mbps 2g/40MHz/HT/CCMP X400AXF 280 Mbps 5g/80MHz/HE/CCMP

WiFi -> WAN -- TCP max

G24W 56.9 Mbps 2g/20MHz/HT/CCMP V2801RGW 49.8 Mbps 2g/20MHz/HT/CCMP X111W 93.8 Mbps 2g/20MHz/HT/CCMP X400AXF 160 Mbps 5g/80MHz/HE/CCMP

WAN -> WiFi -- TCP max

G24W 31.1 Mbps 2g/20MHz/HT/CCMP V2801RGW 33.7 Mbps 2g/20MHz/HT/CCMP X111W 77.7 Mbps 2g/20MHz/HT/CCMP X400AXF 181 Mbps 5g/80MHz/HE/CCMP

WiFi latency

Round-trip delay over the air, at the 95th percentile -- the tail a user actually notices, not the average. Lower is better on these charts, the opposite of every chart above, so the bars are drawn in a different colour and each states the direction. Each bar names its own operating point. The idle chart is the floor; the loaded ones show what happens to it while another device is using the same radio, which is the case a shared medium makes worst.

WiFi -> WAN, idle

G24W 4.02 ms 2g/20MHz/HT/CCMP V2801RGW 31.3 ms 2g/20MHz/HT/CCMP X111W 11.6 ms 2g/20MHz/HT/CCMP X400AXF 38.2 ms 5g/80MHz/HE/CCMP lower is better: these bars are milliseconds of delay, so a SHORTER bar is the better device

WiFi -> WAN, a second device loading the WAN over the cable (downstream)

V2801RGW 27.6 ms 2g/20MHz/HT/CCMP X111W 6.12 ms 2g/20MHz/HT/CCMP X400AXF 40.5 ms 5g/80MHz/HE/CCMP lower is better: these bars are milliseconds of delay, so a SHORTER bar is the better device

WiFi -> WAN, the same station loading its own link (downstream)

V2801RGW 102 ms 2g/20MHz/HT/CCMP X111W 40.1 ms 2g/20MHz/HT/CCMP lower is better: these bars are milliseconds of delay, so a SHORTER bar is the better device

WiFi -> WAN, a SECOND WiFi device loading the air (upstream)

V2801RGW 61.6 ms 2g/20MHz/HT/CCMP X111W 108 ms 2g/20MHz/HT/CCMP X400AXF 640 ms 5g/80MHz/HE/CCMP lower is better: these bars are milliseconds of delay, so a SHORTER bar is the better device

WiFi -> WAN, a SECOND WiFi device loading the air (downstream)

G24W 1194 ms 2g/20MHz/HT/CCMP V2801RGW 166 ms 2g/20MHz/HT/CCMP X111W 96.3 ms 2g/20MHz/HT/CCMP X400AXF 281 ms 5g/80MHz/HE/CCMP lower is better: these bars are milliseconds of delay, so a SHORTER bar is the better device

4 WiFi latency point(s) on G24W, X400AXF carry no figure that can be placed beside another device's: measured, but the producer recorded no figure at this point; the cell CONTRADICTS ITSELF as a device figure: it records limited_by=host, and the lab host is the INSTRUMENT -- a many-core desktop measuring a sub-gigabit CPE. The producer's own verdict is that the repair is the generator, so this number describes the bench and would be read as the device's. It stays in the dated artifact with every field it was measured with.; the point is BLOCKED on this device -- nothing was measured, so there is no figure to place on an axis

WiFi stability

Lower is better for loss and delay; higher is better for the download and upload rates.

Round trips from the first WiFi client to the WAN-side lab target. Idle download and upload are that client's own maximum TCP rate on the same band and cipher. Under load, a second client on the same AP downloads or uploads, or the first client itself downloads; the rate that load achieved is shown beside the round trips measured under it. Loss is packet loss; p95 is delay in milliseconds. A dash means that metric is unavailable; n/a marks the direction a load row does not run.

Idle

DeviceBandDownloadUploadLossp95
G24W2.4 GHz31.1 Mbps56.9 Mbps7.10%4.02 ms
V2801RGW2.4 GHz33.7 Mbps49.8 Mbps0.00%31.3 ms
X111W2.4 GHz77.7 Mbps93.8 Mbps0.00%11.6 ms
X400AXF5 GHz181 Mbps160 Mbps0.00%38.2 ms

Another client downloading or uploading

DeviceBandDownloadUploadLossp95
G24W2.4 GHz0.6 Mbpsn/a63.4%1194 ms
G24W2.4 GHz—Not captured——
V2801RGW2.4 GHz28.6 Mbpsn/a0.00%166 ms
V2801RGW2.4 GHzn/a26.8 Mbps0.1%61.6 ms
X111W2.4 GHz27.6 Mbpsn/a0.00%96.3 ms
X111W2.4 GHzn/a54.7 Mbps0.00%108 ms
X400AXF5 GHz217 Mbpsn/a0.00%281 ms
X400AXF5 GHzn/a225 Mbps0.2%640 ms

The client itself downloading

DeviceBandDownloadUploadLossp95
G24W2.4 GHzNot captured———
V2801RGW2.4 GHz32.5 Mbpsn/a0.00%102 ms
X111W2.4 GHz25.8 Mbpsn/a0.00%40.1 ms
X400AXF5 GHzNot captured———

GPON protocol processing

Microseconds each device's own CPU spends per PLOAM and OMCI message: a real OLT's activation and provisioning, frozen, replayed with no OLT, the same bytes on every device -- so the figures compare across devices and do not move when the OLT changes. Lower is better. The chart is the OMCI median; for scale, an OLT asks about one OMCI message every 20 000 us.

G24W 9.0 us V2801RGW 17.3 us X100DG 35.8 us X111W 17.3 us X400AXF 3.0 us lower is better: these bars are microseconds of CPU per message, so a SHORTER bar is the better device
DevicePLOAM p50PLOAM p95OMCI p50OMCI p95
G24W2.6 us2.9 us9.0 us14.7 us
V2801RGW8.0 us8.0 us17.3 us25.9 us
X100DG16.0 us16.3 us35.8 us53.8 us
X111W7.7 us8.0 us17.3 us25.9 us
X400AXF0.2 us0.2 us3.0 us5.4 us

Device resources

One static program on each device's own CPU: the CPUs online, the memory the kernel manages, a table-driven CRC-32 on one core (the byte-wise code a protocol stack runs), memcpy across buffers larger than the caches (at most a quarter of the free memory, so the 32 MB devices run it too), and the lab host streaming into the device over ssh (cipher and CPU together). Higher is better.

CPUs online

G24W 2 V2801RGW 1 X100DG 1 X111W 1 X400AXF 2

Memory the kernel manages (MemTotal)

G24W 111.5 MiB V2801RGW 52.7 MiB X100DG 20.0 MiB X111W 117.0 MiB X400AXF 462.5 MiB

CPU: table-driven CRC-32 on one core

G24W 81.8 MB/s V2801RGW 64.8 MB/s X100DG 31.8 MB/s X111W 64.6 MB/s X400AXF 148.3 MB/s

Memory: memcpy past the caches

G24W 182 MB/s V2801RGW 162 MB/s X100DG 141 MB/s X111W 215 MB/s X400AXF 1967 MB/s

Management link: ssh pipe into the device

G24W 76.8 Mbps V2801RGW 48.2 Mbps X100DG 26.3 Mbps X111W 49.4 Mbps X400AXF 166.3 Mbps

CPU core and compiler fixes

DeviceCPU coreISACore-specific compiler flags
G24WRealtek Luna - MIPS interAptiv core (RTL9607C, RTL9603CVD)mips_24kcnone
V2801RGWRealtek Luna - RLX Taroko core (RTL9602C, RTL9603C)mips_mips32-mno-branch-likely -mips32 -mtune=mips32 -fno-schedule-insns2 -mfix-rlx
X100DGRealtek Luna - RLX Taroko core (RTL9602C, RTL9603C)mips_mips32-mno-branch-likely -mips32 -mtune=mips32 -fno-schedule-insns2 -mfix-rlx
X111WRealtek Luna - RLX Taroko core (RTL9602C, RTL9603C)mips_mips32-mno-branch-likely -mips32 -mtune=mips32 -fno-schedule-insns2 -mfix-rlx
X400AXFRealtek RTL9607F Cortex-A55aarch64_genericnone

Among the WiFi points published here, 48 further WiFi point(s) on G24W, V2801RGW, X111W, X400AXF carry no figure that can be placed beside another device's: the point is BLOCKED on this device -- nothing was measured, so there is no figure to place on an axis

WiFi signal: ONU -> client

Signal the lab WiFi client receives from each ONU under OpenWrt, the strongest reading per device and band. Higher is better: -15 dBm is a stronger signal than -24 dBm. dBm is already a logarithmic unit (every 3 dB doubles the power), so each bar is the dB the signal stands above -100 dBm, about the noise floor of a 20 MHz channel: a longer bar is a stronger signal. The antennas sit within 1 m of the ONUs: this is not a range figure.

2.4 GHz

G24W -15.0 dBm V2801RGW -23.0 dBm X111W -21.0 dBm X400AXF -24.0 dBm

5 GHz

X400AXF -21.0 dBm

these devices were not heard by the same instrument in the same state (2 different receiving radios; client transmit powers that were not the same; at least one reading whose client power was not recorded, which is not the same as matching), so the figures are printed but NOT compared: a received level is the emission minus the path and the receiver, so the difference between two such columns could be the instrument's rather than the devices'.

The receiving client or its settings were not recorded for some readings; a controlled comparison is not established.