Ethernet Frame Efficiency Calculator

Payload bytes:

Enter the useful payload size before Ethernet headers, FCS, tags, padding, preamble, or inter-frame gap.

bytes

Link rate:

Use the Ethernet line rate for the path you want to model.

Frame profile:

Pick untagged, 802.1Q, QinQ, or a custom tag count.

VLAN tags:

Set the number of 4-byte tags included between source MAC and EtherType.

tags

Include preamble, SFD, and inter-frame gap

Adds 20 wire-time bytes per frame: 7-byte preamble, 1-byte start delimiter, and 12-byte inter-frame gap.

Target payload rate:

Leave at 0 to skip target-rate planning, or enter a desired goodput target.

Curve max payload:

Use 1500 for standard MTU analysis or 9000 for jumbo comparison.

bytes

Metric	Value	Detail	Copy
{{ row.metric }}	{{ row.value }}	{{ row.detail }}

Check	Status	Action	Reason	Copy
{{ row.check }}	{{ row.status }}	{{ row.action }}	{{ row.reason }}

Payload	Wire bytes	Efficiency	Goodput	Copy
{{ row.payloadLabel }}	{{ row.wireBytesLabel }}	{{ row.efficiencyLabel }}	{{ row.goodputLabel }}

{{ formattedJson }}

Embed:

Customize

Include current inputs

Size

Advanced

Width

Height

Aspect ratio

Max height

Collapsible embed

Allow fullscreen

Referrer policy

Sandbox tokens

Ethernet capacity is advertised as a line rate, but the useful work moves in repeated frame units. Every transmitted frame carries payload bytes plus addressing fields, type information, optional tags, an error-check trailer, and sometimes the physical byte time before and after the frame. Ethernet frame efficiency is the share of that repeated byte pattern that is actually payload.

The same link can look very different under large and small frames. A 1500-byte payload spreads fixed Ethernet overhead across enough data that the loss is modest. A short control packet, storage acknowledgment, or tunnel fragment still needs a valid frame envelope, so overhead and packet-rate pressure can dominate long before the nominal line rate looks exhausted.

Frame efficiency is most useful when a capacity question mixes bandwidth and packet rate. Firewalls, switches, capture appliances, service-provider handoffs, storage fabrics, and overlay-heavy networks can hit forwarding or inspection limits even when a 1 Gbps, 10 Gbps, or faster link has theoretical headroom. The efficiency ratio separates Layer 2 byte cost from higher-layer headers, application behavior, retransmissions, shaping, and device performance.

Common Ethernet frame efficiency terms
Term	Plain meaning	Why it changes efficiency
Payload	The useful bytes being carried inside the Ethernet frame.	It is the numerator of the efficiency ratio.
MAC overhead	Destination address, source address, type or length field, optional tags, and frame check sequence.	Most of it repeats once per frame.
Preamble and IFG	Physical-layer byte time before and after the frame.	It consumes line time even though captures often do not show it.
VLAN tag	A 4-byte field inserted for VLAN or service tagging.	Each tag adds bytes to every frame.

Payload size also connects to maximum transmission unit (MTU) choices, but the terms are not interchangeable. MTU usually names the largest Layer 3 packet a link can carry without fragmentation, while Ethernet wire accounting includes the Layer 2 fields around that payload and, for line-rate planning, the physical spacing that packet captures may hide.

The number is a Layer 2 byte-accounting estimate. It does not prove real path throughput, and it does not include IP, TCP, UDP, VXLAN, IPsec, MACsec, TLS records, flow control, congestion, or retransmission. It is most useful when the same accounting scope is used for every comparison.

How to Use This Tool

Start by matching the frame shape to the path you want to model, then read the percentage together with the byte ledger and goodput values.

Enter Payload bytes. Use 1500 for a normal Ethernet payload, 9000 for a jumbo-frame comparison, 46 for the usual minimum payload, or a smaller value when you want to see how padding behaves.
Set Link rate to the nominal line rate you want to translate into goodput, frame rate, and overhead bandwidth.
Choose the closest Frame profile. Untagged, 802.1Q tagged, QinQ, and MAC-frame-only profiles set the tag count and wire-time scope for common cases.
Adjust VLAN tags if the selected profile does not match the path. Each tag adds 4 bytes per frame.
Keep Include preamble, SFD, and inter-frame gap enabled for line-utilization planning. Turn it off only when you intentionally want MAC-frame-only byte accounting.
Use the advanced target rate when you need to know whether a desired payload goodput fits the selected line rate and frame shape.
Use the efficiency curve and curve data to compare several payload sizes without changing the main input repeatedly.

Interpreting Results

Payload efficiency is the main result. It is the entered payload divided by the total modeled byte count for one transmitted frame. Payload goodput applies that same ratio to the selected line rate, while Overhead bandwidth shows the remaining line capacity consumed by selected Ethernet overhead.

Total wire bytes is the clearest place to troubleshoot a surprising result. If preamble, start frame delimiter, and inter-frame gap are included, the denominator includes 20 byte times that ordinary packet captures may not display. If the payload is very small, the ledger may also show padding added to satisfy the minimum MAC frame size.

Result labels used by the Ethernet frame efficiency calculator
Result cue	Boundary	How to read it
High efficiency	At least 94%	The selected payload is large enough that Ethernet overhead is a small share of the line.
Normal overhead	85% to under 94%	The frame shape is usable, but overhead is visible in capacity planning.
Overhead heavy	Below 85%	Small frames, extra tags, or a narrow accounting scope may dominate the result.
Target fits	Payload goodput is at or above the target	The modeled line rate can carry the entered target before higher-layer costs are considered.
Target short	Payload goodput is below the target	Raise line rate, increase payload size, remove avoidable overhead, or revisit the target.

Compare runs only when the accounting scope is the same. A MAC-frame-only result will always look more efficient than full wire-time accounting because it excludes the repeated preamble/SFD and inter-frame gap byte times.

Technical Details

An Ethernet MAC frame begins with destination and source MAC addresses and a type or length field. A frame check sequence closes the MAC frame, and VLAN tags can be inserted into the header area when the path uses 802.1Q tagging or stacked service tags. These fields repeat once per frame, so their effect depends strongly on payload size.

Wire-time accounting adds bytes that sit outside the MAC frame but still consume transmission time: 7 bytes of preamble, a 1-byte start frame delimiter, and 12 byte times of inter-frame gap. Capacity planning usually includes those 20 byte times because the physical link cannot use them for another frame's payload.

Formula Core

The model uses integer byte counts. Payload and tag values are clamped to the supported ranges before the ratio is calculated, and display rounding is applied only after the underlying values are computed.

\begin{array}{lll} H & = & 14 + 4 T \\ M & = & 64 + 4 T \\ pad & = & \max (0, M - (H + P + 4)) \\ wire & = & H + P + 4 + pad + G \\ efficiency & = & \frac{P}{wire} \\ goodput & = & line rate * efficiency \end{array}

Formula symbols and byte ranges
Symbol or value	Meaning	Modeled range or amount
`P`	Useful payload bytes	1 to 9600 bytes
`T`	VLAN tag count	0 to 8 tags
`H`	MAC header plus tag bytes	14 bytes plus 4 bytes per tag
`M`	Minimum modeled MAC frame envelope	64 bytes plus tag bytes
FCS	Frame check sequence	4 bytes
`G`	Preamble/SFD plus inter-frame gap	20 bytes when included, otherwise 0

For a 1500-byte payload with one VLAN tag and full wire-time accounting, the header is 18 bytes, the frame check sequence is 4 bytes, padding is 0, and the wire-time addition is 20 bytes. The total is 1542 bytes, so efficiency is 1500 / 1542, or about 97.28%.

Frame rate is the selected line rate divided by total wire bits per frame. Required wire rate for a target payload rate is the target divided by the current efficiency, so small-frame overhead raises the required line rate even when the payload target itself has not changed.

Limitations and Privacy Notes

The calculation is deterministic byte accounting, not a live network test. It does not send packets, inspect interfaces, or verify what a switch, router, host stack, or provider circuit is doing. Use packet captures, interface counters, and traffic generators when you need operational proof.

Higher-layer costs are outside the Ethernet result. IP and TCP or UDP headers, VXLAN, GRE, IPsec, PPPoE, TLS records, storage protocol framing, retransmissions, pause frames, shaping, and application chunking can all reduce real application throughput beyond the Layer 2 estimate shown here.

The entered values are ordinary numeric settings processed in the browser page. They are useful for planning examples and comparisons, but they should not be treated as a measurement of any private or production network unless the same values are backed by separate evidence from that network.

Worked Examples

Ethernet frame efficiency worked examples
Scenario	Modeled bytes	Efficiency result	Practical reading
1500-byte payload, one 802.1Q tag, full wire time	1542 total wire bytes	97.28%	About 972.76 Mbps of payload goodput on a 1 Gbps line.
46-byte untagged payload, full wire time	84 total wire bytes	54.76%	Small-frame packet rate dominates, even before higher-layer headers.
9000-byte payload, two tags, full wire time	9046 total wire bytes	99.49%	Jumbo payloads make stacked tags a small share of the frame.
1500-byte untagged payload, MAC frame only	1518 MAC frame bytes	98.81%	The result is higher because preamble/SFD and IFG are excluded.

If the target payload rate is 900 Mbps in the tagged 1500-byte example, the required wire rate is 925.20 Mbps. That target fits on a 1 Gbps line in this Layer 2 model, but there is not much room left for higher-layer overhead or operational loss.

FAQ

Why is payload goodput lower than line rate?

Line rate counts every selected byte time. Payload goodput removes the share consumed by Ethernet header bytes, tag bytes, FCS, padding, and optional preamble/SFD plus inter-frame gap.

Should preamble, SFD, and inter-frame gap be included?

Include them for line-utilization, packet-rate, and capacity planning. Exclude them only when comparing MAC frame sizes without physical-layer spacing.

Why can payload below 46 bytes be entered?

Very small values make padding visible. The payload is still modeled inside a valid minimum MAC frame, so padding fills the gap before efficiency is calculated.

What changes when QinQ is selected?

QinQ models two tags, adding 8 bytes per frame. The effect is easy to see on small payloads and much less important on jumbo payloads.

Can the result predict an application transfer speed?

Only as a Layer 2 starting point. Application transfers also depend on higher-layer headers, protocol behavior, host performance, congestion, retransmissions, storage speed, and test method.

Glossary

Payload: The useful data bytes counted in the efficiency numerator.
Goodput: The useful payload data rate after selected Ethernet overhead is removed from line rate.
MAC frame: The Ethernet frame from destination MAC address through FCS, excluding preamble, start delimiter, and inter-frame gap.
FCS: Frame check sequence, the 4-byte trailer used to detect frame errors.
SFD: Start frame delimiter, the 1-byte marker after the preamble.
IFG: Inter-frame gap, the idle byte time reserved between Ethernet frames.
802.1Q: The common VLAN tagging standard that inserts a 4-byte tag into an Ethernet frame.
QinQ: A double-tagged Ethernet pattern used for stacked VLAN or service-provider Layer 2 paths.

References

IEEE 802.3 Ethernet Working Group, IEEE.
IEEE 802.3as Ethernet Frame Expansion overview, IEEE 802.3.
IEEE 802.1Q Virtual LANs, IEEE 802.1.
How to change MTU with ip link, Simplified Guide.