For a typical 1080p H.264 encode in HandBrake, start around RF 22, then check a short, representative section before encoding the full video. It is a practical starting point, not a YouTube requirement or a promise of a particular file size or visual result.
There is no single best value for every video. Resolution, source quality, motion and texture, encoder choice, and your preference for quality over file size all matter; YouTube publishes upload guidance in other terms, including bitrate and format.
What RF means in HandBrake
HandBrake calls its constant-quality control RF, short for Rate Factor. You may also see CRF, or Constant Rate Factor, in x264 terminology. Both names refer here to a way of asking an encoder to aim for a selected visual-quality level rather than a fixed bitrate.
In constant-quality mode, HandBrake adjusts the bitrate to meet the quality target. A lower RF value generally asks for higher quality and uses more data; a higher value allows more compression and tends to make a smaller output. The encoder makes those decisions as it handles the footage, so RF does not tell you in advance exactly how large the finished file will be.
The scale is not linear. A one-point change is not a fixed percentage change in size or quality, and the result depends on the clip. A smooth talking-head shot and a scene with fast movement, fine leaves, film grain, or animated gradients can behave differently at the same setting.
RF also does not control every part of the export. Resolution, frame rate, encoder, preset, audio settings, container, and colour handling remain separate choices. If you change several of these while comparing RF values, it becomes difficult to tell which change affected the picture or file.
A practical 1080p starting point
For ordinary 1080p material encoded with x264 in HandBrake, RF 22 is a useful first try. HandBrake documents an x264/x265 range of RF 20–24 for 1080p, and its broader x264 guidance puts HD near RF 22, plus or minus one. These are software recommendations, not results from a test of your particular footage.
Use RF 22 as a first pass, then inspect the output at the size at which viewers are likely to watch it. Look closely at the sections that contain small text, hair, patterned fabric, tree branches, moving water, fast camera movement, shadows, or smooth colour transitions. Compression damage often shows up there before it is obvious in a simple static scene.
If detail or motion looks unacceptably soft, try RF 20 or 21 on the same sample. That asks x264 to preserve more quality, with a likely trade-off in bitrate and output size. If the sample still looks good at RF 22 and a smaller upload matters more, try RF 23 or 24. These are sensible trial adjustments within HandBrake’s documented range, not guaranteed outcomes.
Keep the encoder and other settings fixed while comparing. If you switch from x264 to another encoder, change the preset, or resize the video at the same time, you are no longer testing only RF. A fair comparison means using the same source frames and playback conditions for each short encode.
For a pre-recorded playlist, the encoded file is only one part of the workflow. The video must also be prepared for repeat playback and fit the intended channel format; the guide to setting up a 24/7 Bollywood instrumental playlist stream covers the broader channel setup. Make the picture-quality decision on a representative file before building the complete playlist around it.
RF ranges by resolution
HandBrake’s documented x264/x265 ranges give you a starting bracket, not an automatic conversion from one resolution to another. Use the range for the output resolution you intend to upload, then inspect a sample from your own source.
| Output resolution | HandBrake x264/x265 RF range | How to use it |
|---|---|---|
| 480p or 576p | 18–22 | Start within the SD range and check whether small details survive at the intended viewing size. |
| 720p | 19–23 | Use this bracket for HD at 720p; do not assume a 1080p value gives the same trade-off. |
| 1080p | 20–24 | RF 22 is a reasonable first sample for typical x264 material. |
| 2160p (4K) | 22–28 | Test actual footage; the broader range does not imply equivalence with 1080p RF values. |
These ranges come from HandBrake’s quality adjustment documentation. Its separate constant-quality guidance suggests approximately RF 20, plus or minus one, for SD x264 and RF 22, plus or minus one, for HD x264. Check the documentation for your installed HandBrake version, since interface labels and preset details can change.
The ranges overlap, but an RF number is not a resolution-independent quality guarantee. An RF 22 encode at 720p and one at 2160p are not interchangeable just because the control shows the same number. The encoder is working with different amounts of image detail and different output dimensions.
If you are unsure which resolution to choose, begin with the dimensions and frame rate of the source rather than upscaling just to meet a higher number. The target should suit your footage and channel. A low-resolution original will not gain real detail simply because it is resized to 1080p or 4K before encoding.
Source quality and content change the choice
Start by checking the original file, not just the HandBrake preview. Compression cannot restore detail already lost to a low-quality download, an earlier encode, a soft focus, or noisy capture. If the source already has blockiness or ringing around edges, another encode may make those flaws more visible.
Content matters because the encoder has a different job in each shot. A static devotional image with a slow dissolve may compress differently from a concert recording with moving lights, smoke, and camera movement. Lofi visuals with subtle grain or a slowly shifting gradient can also reveal banding or texture loss, even if little is moving across the frame.
Pay attention to the parts that matter to your channel’s viewing experience. For a lecture, that may be subtitles and small writing on a board. For a bhajan video, it may be faces, instrument detail, and animated backgrounds. For a local news loop, check scrolling text, maps, and ticker edges. These are examples of where to look, not a claim that one category always needs a particular RF.
The source frame rate should generally be preserved rather than changed as a side effect of compression. If the video has interlaced footage, YouTube’s upload guidance says to deinterlace it before uploading. A setting that changes frame rate or deinterlaces can affect the sample, so note those choices and keep them consistent across comparison encodes.
Your choice is a trade-off rather than a hunt for a magic value. A smaller file can be easier to store and transfer, while a less compressed encode can retain detail better for YouTube’s later processing. If you are deciding between two nearby values, compare them on the same difficult scene and choose the one that looks acceptable at a file size and encoding time you can manage.
Test a short representative section first
Do not encode a long recording just to discover that your chosen RF softens text or turns a gradient into visible bands. Make a short sample from the full file first. Choose a part that contains typical material and at least one demanding moment, such as motion, fine detail, grain, or a scene transition.
HandBrake’s Preview feature can help you inspect a frame and check framing or filters, but a still image is not a substitute for watching an encoded moving sample. For a meaningful comparison, encode a short section that includes the motion you want to judge. If you use a middle section, check that it represents the rest of the programme; a quiet title card alone is not enough for a music video or news loop.
A disciplined test can be simple:
- Choose a representative section and record its start point and duration so you can repeat the same test.
- Set the intended output resolution, frame rate, x264 encoder, and preset, then encode at RF 22 for 1080p.
- Watch the result at normal viewing size, then inspect any questionable areas more closely. Check motion, fine texture, text, gradients, and audio sync.
- If you see a problem, encode the identical section at a nearby RF value. Change only RF so the comparison remains useful.
- Compare appearance, file size, and encoding time, then decide whether the trade-off makes sense for the full file.
The size of the short sample is not a reliable promise of the complete file’s size. A longer programme may contain scenes with very different complexity, and constant quality lets the bitrate vary with the content. Use a sample to compare options, not to claim a guaranteed output size.
If the sample fails, identify the failure before moving the slider. A soft image may be present in the source; clipped highlights or wrong colours may be a separate processing issue; stutter may relate to frame-rate handling; and unreadable text may need a larger source or redesigned graphics. Lowering RF cannot fix every defect in the original or every workflow choice.
Once the sample looks acceptable, encode the full file with the same settings and watch the beginning, a few demanding sections, and the end before scheduling it. For a continuous channel, the longer-running question is also whether the playback arrangement stays reliable overnight; the article on monitoring a continuous church stream remotely addresses that separate operational concern. File quality and stream monitoring solve different problems.
YouTube upload guidance is not an RF setting
YouTube’s official recommended upload encoding settings specify format and bitrate guidance, not a HandBrake constant-quality value. For SDR 1080p, YouTube lists 8 Mbps for standard frame rates of 24, 25, or 30 fps, and 12 Mbps for high frame rates of 48, 50, or 60 fps. Those are YouTube recommendations for upload encoding; they are not equivalent to setting RF 22.
For SDR 2160p, YouTube lists 35–45 Mbps at standard frame rates and 53–68 Mbps at high frame rates. Those figures help explain why a fixed-bitrate workflow and a constant-quality workflow are different ways of making an encode. Do not treat a bitrate recommendation as a target RF, or assume that converting one directly into the other gives the same result.
YouTube’s guidance also names other specifications, including MP4, H.264 High Profile, progressive scan, 4:2:0 chroma subsampling, CABAC, two consecutive B frames, and a closed GOP. It recommends AAC-LC, Opus, or Eclipsa audio and BT.709 for SDR. One RF control cannot set all of those properties, and not every listed property is a quality slider. Check YouTube’s current page before preparing an upload because official recommendations can change.
This distinction matters when you use HandBrake’s constant-quality mode. RF asks the encoder for a quality level and lets the bitrate vary; YouTube’s table gives suggested bitrates for upload encodes in defined resolution and frame-rate classes. If you need to follow a bitrate-based workflow, use a suitable bitrate mode and verify the rest of the export settings rather than trying to force constant quality to match one number.
HandBrake offers Creator presets intended for uploads to hosting platforms, including YouTube. They can be a starting route if you would rather not set RF manually, but preset names and exact settings may vary by HandBrake version. Check what your installed preset selects, then inspect an output sample instead of assuming the preset is right for every source.
If you are specifically comparing a pre-recorded file with live-streaming settings, keep the workflows separate. A YouTube Live bitrate chart addresses live streaming bitrate by resolution and frame rate; it is useful context for a live encoder, but it does not turn those live recommendations into a HandBrake RF setting for a pre-recorded upload. Likewise, the video may later be used in a continuous live channel, but the quality decision for its file is still made before the channel plays it.
Prepare the file for a continuous channel
Once your sample passes, keep a note of the settings that produced it: output dimensions, frame rate, encoder, preset, RF, audio choice, and any filters or deinterlacing. This makes it easier to reproduce the result for the next programme and to spot an accidental difference. A saved preset can help, but review its selections if you change HandBrake versions or source material.
Before committing the full encode, confirm that the file plays from beginning to end and that its audio is in sync. If the video is meant to loop, watch the ending transition back to the opening. A harsh cut, a silent gap, or a brief black frame may be a playlist or edit issue rather than an RF issue. Resolve that in the source or playback plan instead of repeatedly increasing quality settings.
For a 24/7 channel, decide how the encoded file will be made available to the broadcast and what happens if the source or connection stops. Encoding quality does not by itself keep a live stream running. If the repeated task of leaving a computer on is the part that tends to fail, StreamNeo can take the uploaded-file and continuous-broadcast burden off that computer, while leaving you responsible for preparing a suitable file and checking your YouTube channel settings.
If you prefer to keep the process local, plan for storage and transfer as well as encoding. A larger output takes more space and may take longer to upload on a limited connection; a smaller output may ask you to accept more compression. Neither trade-off should be guessed from the RF number alone. Use the sample, then judge the full export and upload in your actual workflow.
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
Is RF 22 the best HandBrake setting for YouTube?
RF 22 is a sensible first test for typical 1080p x264 material, within HandBrake’s documented 1080p range of RF 20–24. It is not a YouTube-prescribed setting or a universal best value. Judge it on a representative sample from your own video.
Does a lower RF always make the file much bigger?
A lower RF generally asks the encoder for higher quality and often produces a larger output, but the actual size depends on the footage and encoding choices. RF does not provide a fixed size prediction. Compare short samples using the same source and settings.
Should I use the same RF for 720p and 4K?
No single RF should be assumed to give the same trade-off at different resolutions. HandBrake documents separate ranges for 720p and 2160p, and the content still matters. Begin within the range for your intended output, then inspect a sample.
Does YouTube specify a CRF or RF number?
YouTube’s upload guidance specifies format and bitrate recommendations by resolution and frame-rate class, not a HandBrake RF value. Use the official page for current upload details, and treat HandBrake’s RF as a practical quality control rather than a YouTube rule.