ROI ASIC

ROI ASIC / One S21, two useful records

Your S21 changed. What did the change actually do?

You have changed a profile on one S21. The dashboard shows a different number. Before that number becomes a decision, we need two records that describe the same miner under comparable conditions. We will help you collect them, spot an unfair comparison and take your own inputs into our calculator

A → BOne device. A recorded baseline. A measured change
This is a worksheet for your own S21 pilot, not a published hardware test. Measurement fields start empty. We do not fill them with the model catalogue's figures or promise a particular result
01

Give the pilot one device and one question

Use the exact model on the device. This page is for Antminer S21. Our current S21 catalogue lists an AML build; confirm your own board and use the catalogue for the current file choice

Write down your local device identifier, installed build and profile. Then choose one question: does this profile reduce measured power, increase useful output, or improve the calculated result at your tariff? Change one thing at a time so you can follow the answer

Confirm the control board
02

Save A before you create B

A is your baseline from this device. Before changing it, capture the profile, measurement window, hashrate source and measured power

Keep TH/s and W tied to the same operating interval. Label hashrate as a device or pool reading and as instantaneous or averaged

Measure power at the same point with the same instrument in both runs. If A uses a dashboard estimate and B a wall measurement, they need that difference resolved. If the meter includes other equipment, record the boundary

03

Measure B without losing the context

Record B's build or profile and observation window. Keep the hashrate source, power boundary and surrounding conditions comparable to A. Note material changes in cooling, intake conditions, pool or other settings

Mark startup, adjustment periods and restarts separately. Record the window you actually used; repeat it if unstable readings or changed conditions prevent a useful conclusion. Errors and gaps belong beside the performance figures

04

Decide whether the pair is ready to compare

Before calculating money, read the two records together. Same S21? Comparable windows? TH/s and W from the same run? Clear source and fee basis?

A missing answer gives you a specific next action: measure again or resolve that gap. A complete pair gives you an observation on this device, under the recorded conditions. Use it to plan the next step

A short check before the next step
CheckReady to compareIf it differs
DeviceThe same S21Separate records from different devices
Window and sourceTH/s and W describe the same run and comparable windowsRepeat the missing measurement
PowerThe same instrument and measurement pointUse a common boundary
EventsErrors and restarts are recordedResolve the unknown before deciding
05

Keep the two records together

Fill the worksheet while the readings are in front of you. It creates a short text record to copy for your notes or a question to us. Your entries stay on this page. Copy the record to keep it

Keep a pool figure that already excludes the developer share labelled as a pool observation. The calculator needs hashrate on a basis before the fee it will apply

S21 before-and-after record

Enter what you know. Missing fields remain marked as not recorded

Your records, with no prefilled performance values
What to recordAB
Build and profile
Observation start and end
Hashrate, TH/s
Hashrate source and fee basis
Measured power, W
Measurement point and instrument
Applicable Dev Fee, %
Errors, restarts and gaps

06

Now take your pair to the calculator

Enter your recorded TH/s and W in A and B, each profile's applicable fee and a shared gross revenue basis before that fee. Add your electricity tariff and start with one miner

The calculator assumes 24 operating hours per day and constant inputs. Assess different downtime or extra costs separately

Use the comparison to choose the next action: further observation, another comparable run or a specific support question. The current S21 catalogue leads to your firmware choice and the existing installation route

FAQ

Questions about the pilot

Can I use the S21 catalogue figures as my A result?

Use them to identify the product context, not as a measurement from your device. A should describe the S21 and operating conditions you are actually comparing

Does every S21 need the same observation time?

This worksheet does not set a universal time. Record the actual windows and relevant changes. If transient behaviour or different conditions prevent a useful comparison, gather another comparable record

Does a successful pilot prove the result for the whole fleet?

It proves an observation on the device and conditions recorded. Use a limited follow-up on comparable equipment before treating the result as a fleet scenario

Use your S21 readings, not someone else's headline

Take the recorded pair into our calculator, compare at your tariff and keep the reason for the next step beside the result