Does Mining Pool Ping Really Matter?
The 120ms Myth Debunked

The Claim We Keep Hearing

In our Discord and across mining communities, we hear variations of this argument regularly:

This reasoning feels intuitive if you come from competitive gaming, where 120ms ping is genuinely terrible. But mining is a fundamentally different workload, and applying gaming logic to it leads to wrong conclusions. Let’s prove it with math.

How Mining Actually Works (The 30-Second Version)

To understand why ping barely matters, you need to understand what happens during mining. Here’s the entire cycle:

The critical insight: step 2 takes seconds to minutes. Step 4 takes milliseconds. Your miner spends 99.99% of its time computing hashes locally — your GPU doesn’t care about ping, it doesn’t wait for the network, and it doesn’t slow down because of latency.

When Does a Share Become “Stale”?

A share becomes stale only when a new block is found on the network between the moment your miner starts working on a job and the moment the pool receives your share. Here’s the timeline:

Notice that your ping is just one small part of a much larger chain. Block propagation (200-500ms) and pool processing (5-50ms) happen before your ping even enters the picture. These delays exist regardless of whether your ping is 5ms or 500ms.

The Math: Stale Share Probability

Here’s the formula that governs stale shares from latency. It’s simple division:

// Stale probability from YOUR ping alone
stale_rate = ping_one_way / block_time

// One-way latency = half the round-trip ping
ping_one_way = ping_rtt / 2

// Example: 120ms ping, 60-second block time (like Groestlcoin)
stale_rate = (0.120 / 2) / 60 = 0.060 / 60 = 0.001 = 0.1%

Interpretation: With a 120ms ping on a coin with 60-second blocks, your ping contributes approximately 0.1% stale share rate. That means out of every 1,000 shares you submit, roughly 1 will be stale due to your network latency.

The Full Comparison Table

Let’s calculate stale rates for various ping values across different block times:

What the Table Tells Us

The difference between a “perfect” 20ms ping and a “terrible” 120ms ping is 0.083%. To put that in perspective: if you mine $100 worth of crypto per month, switching from 120ms to 20ms would earn you an extra 8.3 cents. You spend more electricity leaving a light on.

Why Gaming Logic Doesn’t Apply to Mining

The “120ms is too high” belief comes from applying gaming latency standards to a completely different workload. Here’s why they’re not comparable:

The Core Difference

In gaming, the server is checking your position and actions 128 times per second. If your packet arrives 120ms late, you’re 7–15 “ticks” behind — in a shooter, that’s the difference between hitting and missing.

In mining, the pool checks your work when you submit it — which happens every few seconds at most. The only question is: “Was a new block found during the 60ms (one-way) it took for your share to travel to the pool?” Since blocks are found once every 60–600 seconds, the answer is “no” approximately 99.9% of the time.

What Actually Causes Stale Shares

If your stale rate is high, your ping is almost certainly not the reason. Here are the real culprits, ranked by impact:

Block propagation is the biggest factor and it’s completely outside your control. When someone finds a block on the other side of the planet, it takes 200–2,000ms for that information to reach your pool’s node. During that time, every miner on every pool is working on the old block — regardless of their ping.

Connection stability matters more than speed. A stable 150ms connection is vastly better than a 20ms connection that drops packets or has 500ms jitter spikes. When your connection drops a share, it must be re-transmitted, adding seconds — not milliseconds — of delay. If you’re seeing high stale rates, check for packet loss first (ping -c 100 pool.address and look for lost packets), not the raw ping number.

Static Analysis: A Year of Mining at Different Pings

Let’s model a real scenario. Assume you mine a GPU-mineable coin with a 60-second block time, earning $200/month. Here’s what different ping values cost you over a full year:

Switching from a pool with 120ms ping to one with 20ms ping saves you $2 per year. Meanwhile, a pool with a 0.5% lower fee saves you $12 per year. The fee difference is 6x more impactful than the ping difference.

What If Ping Is Really High? (1,000ms+)

Even at extreme latencies, mining doesn’t fall apart. Let’s look at the worst case:

Key observations:

• At 1,000ms (1 second) ping, you lose 0.83% of shares to stale — you still keep 99.17% of your earnings. That’s entirely workable.
• At 3,000ms (3 seconds) ping — the latency of geostationary satellite internet — you still keep 97.5% of your earnings. Mining from a rural satellite connection is totally viable.
• Even at 10,000ms (10 seconds), which would mean something is seriously broken with your connection, you’d still earn 91.7% of what a zero-latency miner earns.

Technical Deep Dive: How Stratum Handles Latency

For the technically inclined, here’s exactly what happens at the protocol level. The Stratum protocol is designed to be latency-tolerant by design:

1. Asynchronous Communication

Stratum uses TCP with JSON-RPC. Your miner doesn’t wait for a response before hashing — it fires off a share and immediately continues computing. The pool sends mining.notify when new work is available, but your miner doesn’t poll for it. This means ping doesn’t affect your hashrate at all — your GPU never pauses.

2. The Stale Window Is Tiny

A share can only go stale during the brief moment between a new block being found and your miner receiving the updated job. Let’s calculate the exact probability:

// Probability of a block arriving during any given time window
// Follows Poisson distribution: P(block in t) = 1 - e^(-t/T)

T = 60 seconds (average block time)
t = 0.060 seconds (one-way latency for 120ms RTT)

P(stale) = 1 - e^(-0.060/60)
         = 1 - e^(-0.001)
         = 1 - 0.9990005
         = 0.0009995
         = 0.09995%

// For comparison, at 20ms ping:
P(stale) = 1 - e^(-0.010/60) = 0.01667%

// Difference: 0.09995% - 0.01667% = 0.08328%
// That's less than one-tenth of one percent

3. Your Hashrate Is Not Affected

This is the most misunderstood point. Some miners believe high ping “lowers their hashrate.” It does not. Your GPU computes hashes at its clock speed — it has no idea what your ping is. Whether the pool is 5ms or 500ms away, your reported hashrate in the mining software will be identical. The pool’s effective hashrate estimate may fluctuate slightly due to share submission timing, but this is a display issue, not an actual hashrate reduction. Read more about how hashrate reporting works.

4. Share Difficulty Absorbs Latency

Pools set share difficulty based on your hashrate to target roughly 1 share every few seconds. At higher difficulty, each share represents more work. Whether that share takes 60ms or 200ms to reach the pool doesn’t change the work it represents. The pool credits you for the computational work, not the delivery speed.

Putting It In Perspective: Ping vs Everything Else

Let’s compare the yearly impact of various factors for a miner earning $200/month on a coin with 60-second blocks:

Obsessing over 100ms of ping while ignoring GPU tuning, fee structures, or hardware efficiency is like adjusting the rearview mirror angle to make your car go faster.

When Should You Actually Worry About Latency?

Ping is not completely irrelevant. Here are the rare cases where it genuinely matters:

Practical Advice: What to Actually Optimize

Instead of chasing lower ping, here’s what actually moves the needle on your mining earnings:

1. Tune your GPU/CPU: Proper overclocking and power limiting can improve efficiency by 20–40% — that’s 200x more impactful than fixing your ping.
2. Choose the right payout scheme: PPS vs PPLNS can differ by 2–5% depending on your mining patterns.
3. Compare pool fees: A 1% fee difference is worth 12x more than 100ms of ping.
4. Use a wired connection: WiFi adds jitter and packet loss — not because of ping, but because of dropped packets that cause reconnections.
5. Keep mining software updated: Newer versions handle job switching faster, reducing the miner-side stale window.
6. Mine the right coin: Check which coins are most profitable for your hardware — this matters infinitely more than ping.