There is no single reliable electricity bill for running a 24/7 YouTube stream from a Synology NAS with Docker and FFmpeg. The cost depends on the NAS’s average wall-power draw while doing your particular job and the electricity tariff that applies to you.
Measure the complete NAS at the wall during a representative stream, convert its average watts into kilowatt-hours for your chosen period, and multiply by your price per kilowatt-hour. Synology’s published DS923+ figures can provide context, but they do not measure FFmpeg streaming draw.
Why there is no single NAS streaming bill
A NAS model name alone cannot tell you what a stream will cost. Two owners may use the same model yet have different drive counts, drive activity, video settings, network traffic and background tasks. Their actual average wall draw can therefore differ.
Docker does not come with a fixed electricity figure. It is a way of running the software; the relevant quantity is the power used by the whole NAS system while it runs the container and carries out the stream. Include the drives and the rest of the NAS, not just a CPU estimate or a container metric.
FFmpeg demand depends on what it has to do. Passing through a video is different from decoding, filtering, resizing or encoding it. Resolution, frame rate, filters and encoder settings all affect the work. If encoding is involved, hardware support may or may not be usable in your model and container configuration. The available documentation does not provide a benchmark for an unspecified Synology running your particular stream.
There is also no tariff in the question. Electricity rates vary by location and, in some places, by plan or time. A bill calculated using a guessed rate would look precise without being personal to you. Keep the measured wattage and your applicable tariff separate until you have both.
The same caution applies to stream design. If your source files, bitrate or loop behaviour change, the workload may change too. For a starting point on the video side, see the bitrate settings checklist for prerecorded streams; it helps you define the stream you are measuring rather than substituting a generic load assumption.
Measure average wall draw during the workload
The useful input is average wall draw while the intended stream is running. A plug-in electricity usage monitor can measure a NAS plugged into a suitable outlet. Check the monitor’s instructions and rating, and make sure it is appropriate for your electrical supply. The measurement should include the NAS power supply and its installed drives, as used in normal operation.
Set up the same job you expect to run continuously: the same video or playlist, FFmpeg command, Docker configuration, resolution, frame rate, filters, encoding path and network use. Let the system settle into its ordinary behaviour before recording readings. A brief idle reading or a momentary peak does not represent the average workload.
Measure across a representative period that includes ordinary changes in the stream, such as loop transitions and any routine background activity you expect to keep enabled. Record the monitor’s energy reading or take repeated power readings often enough to understand how the load varies. If the monitor reports cumulative energy rather than average watts, note both the energy and elapsed time; average watts can be derived from that interval. Do not confuse an instantaneous maximum with a period average.
A fair measurement should also reflect your actual storage configuration. If you intend to run with four drives, measuring an empty NAS or a one-drive test setup will not account for the installed system. Likewise, do not disable normal services just to produce a lower reading if you will leave them enabled for the real channel.
For a useful record, write down the model, installed drives, stream settings, measurement start and end, average wall watts if available, and anything unusual during the interval. If the figure changes substantially from one representative period to another, repeat the measurement and use a result that reflects normal use. A single unusual event should not become the assumed year-round average.
The FFmpeg documentation describes its options and capabilities, but documentation is not an electricity test of your container on your NAS. Check the FFmpeg documentation when validating what a command does, and measure the resulting system rather than treating command settings as a substitute for wall-power data.
Calculate continuous-use energy and cost
Once you have an average in watts, the conversion is straightforward. Multiply watts by the number of hours, then divide by 1,000 to get kilowatt-hours (kWh). For an always-on schedule, use 24 hours per day, 24 × 30 hours for a 30-day month, or 24 × 365 hours for a 365-day year.
| Period | Energy calculation from average watts | Cost calculation |
|---|---|---|
| One day | watts × 24 ÷ 1,000 = kWh per day | kWh per day × tariff per kWh |
| 30-day month | watts × 24 × 30 ÷ 1,000 = kWh per 30 days | kWh per 30 days × tariff per kWh |
| 365-day year | watts × 24 × 365 ÷ 1,000 = kWh per 365 days | kWh per 365 days × tariff per kWh |
These are unit conversions, not test results. For example, if you assume a measured average of 40 W purely to demonstrate the arithmetic, daily energy would be 40 × 24 ÷ 1,000 = 0.96 kWh. That assumed wattage is not a Synology specification or a claim about what FFmpeg will draw. To get cost, multiply the resulting kWh by your own tariff, using the currency and unit price shown on your electricity plan or bill.
If your tariff has different rates at different times, a single price per kWh may not represent the bill. In that case, calculate energy for the relevant periods separately and apply the corresponding rates, or use the tariff’s billing method. Fixed charges and taxes may also appear on an electricity bill; the simple multiplication estimates energy charges only unless your applicable rate already includes those items.
Do not overstate precision. If your meter only gives a rough average, report a rounded estimate. Recalculate when you change drives, stream settings or other meaningful parts of the workload. For an operational checklist beyond the arithmetic, this guide to running a continuous FFmpeg YouTube stream covers stream behaviour that can affect what you should test.
Use DS923+ figures only as reference points
Synology’s DS923+ product page lists 35.51 watts for “Access” and 11.52 watts for “HDD Hibernation”. These are manufacturer figures for that named model and stated conditions, not measured FFmpeg encode-and-stream draw. They should not be treated as the power use of every Synology NAS or as the answer to your running-cost question.
The two conditions themselves illustrate why context matters: an access state and drive hibernation are not interchangeable descriptions of a NAS doing a continuous video job. FFmpeg may be reading files, processing media and sending a stream while the system and drives behave differently from the conditions behind a published reference figure. The product page also cautions that specifications can vary with configuration, deployment and active operations.
You can use these figures to orient yourself to a manufacturer’s stated model data, provided you keep the labels attached. Do not multiply either one by continuous hours and present the result as your stream’s bill. That calculation would turn a contextual reference into a workload claim the figures do not establish.
For primary-source context, see Synology’s DS923+ product specification page. The published hardware specification document is also available from Synology; it describes the DS923+ hardware, not a benchmark of your Docker and FFmpeg stream.
Account for model, drives and workload differences
A different Synology model may have different components and operating characteristics. Even the DS923+ specification document describes only that model’s hardware configuration. Its processor, memory and other component details should not be generalised to a different DiskStation, and none alone yields a reliable wattage for a particular stream.
Drive population and activity matter because your measured system includes the installed disks. Capacity, number of drives, access patterns and whether drives remain active can all distinguish one setup from another. Include routine NAS services and the network equipment only if your measurement boundary includes them; for the NAS’s own cost, measure the NAS itself and keep separate devices out of the reading.
Then consider the media workload. A simple loop may be mostly file reading and network transfer, while a job that transcodes or applies filters asks more of the processor. Hardware encoding is not automatic merely because a NAS has a video-capable component. The container build, device access and software support need to be checked for the particular setup. SynoCommunity’s FFmpeg FAQ discusses device access in its community packaging context; it is not a guarantee that every NAS or container can use hardware encoding.
YouTube’s setup instructions use a server URL and stream key in an encoder. Keep the stream configuration stable while measuring, and check current YouTube guidance for your account and workflow. YouTube Help says streams under 12 hours are automatically archived, so do not assume a single continuous 24-hour event will be automatically archived as one stream. If recording or archiving behaviour affects your workload, test the actual operating pattern you intend to use.
If maintaining your own NAS means watching CPU use, device permissions and reconnect behaviour through the night, account for that practical work as well as electricity. StreamNeo addresses the specific need to keep the computer switched off after the video and YouTube stream key are set up, with monitoring and automatic restart if the broadcast drops; it is YouTube-only, so it is not a replacement for a NAS workflow that must serve other destinations or local services.
For a stream that drops and needs recovery, the FFmpeg reconnect troubleshooting guide is relevant to testing reliability alongside power. Reliability and cost are separate questions: a lower measured draw does not show whether the stream remains healthy, and a successful test does not establish a long-term electricity bill.
Report tariff and measurement period with a result
A result is useful only when someone can understand what it describes. State the NAS model and drive configuration, the average wall draw measured during the stream, the period over which you measured it, the tariff and currency per kWh, and whether the final figure is for a day, month or year. If the tariff is variable, say how you applied it.
For example, a report might say: “Measured average wall draw was [your measured watts] W during a representative [your measurement period] stream on a [model] with [drive configuration]. At [your tariff and currency] per kWh, the calculated energy charge is [your calculated amount] for [stated period].” Fill the brackets with your own readings and bill information. This is a reporting template, not a supplied result or an implied Synology measurement.
Keep the source of each input clear. The watts should come from your measurement, not the DS923+ access or hibernation references; the tariff should come from the applicable electricity rate, not an assumed national average. If you have not measured the workload or do not know which tariff applies, report the formula and say what is missing rather than publishing an exact bill.
You can also make the result more useful by noting changes that would invalidate it: replacing drives, changing the encoding path, adding filters, or switching to a different stream schedule. That gives you a reason to measure again rather than treating one number as universal. If you are comparing configurations, compare their measured averages under the same workload and tariff basis.
Before committing, compare the operating options on the pricing page. When the file and channel are ready, start free — 24-hour trial, no card.
FAQ
Can I calculate my cost from Synology’s DS923+ watt figures?
Not as the cost of your FFmpeg stream. Synology lists 35.51 W for Access and 11.52 W for HDD Hibernation, but those are model-specific reference figures, not a measurement of this workload. Measure average wall draw during your own stream and use your applicable tariff.
Does Docker add a fixed amount to the electricity bill?
No fixed Docker cost can be applied to every NAS. Measure the complete system while it runs the intended container and stream, since model, drives and the work FFmpeg performs affect the result. Container monitoring may help explain workload, but wall draw is the input for electricity cost.
What information do I need for a personalised estimate?
You need the NAS model and drive configuration, average wall watts during a representative stream, the measurement period, and the electricity tariff in its currency per kWh. If you do not yet have the measured average or applicable tariff, you can calculate the formula but not a defensible exact bill.
Will YouTube automatically archive a continuous 24-hour stream?
YouTube Help says streams under 12 hours are automatically archived. Do not assume that a single 24-hour event will be archived as one stream; check current YouTube guidance and plan your own recording or event schedule if an archive matters.