Suggestions to lower file size without affecting quality

[lovetorecord]

Hi, I often take video recording of my 4K screen for 1-1.5 hours/session. The average file size for an hour is about 1.5 GB. Any good suggestions to reduce the file size without affecting the text and graphics image qualities? Currently under Settings>Video, I have:

Base (Canvas) Resolution 4096x2160
Output (Scaled) Resolution 4096x2160
Downscale Filter [Resolutions match, no downscaling required]
Common FPS Value 60

What is the difference between Base (Canvas) and Output (Scaled) resolutions?

Thanks

 

[Lawrence_SoCal]

My 1.5 hour recordings of a 1080p30 base canvas is 11GB.. so your video is already highly compressed to only be 1.5GB for 2160p60
With that said
- base canvas equals your starting point, which presuming 4K monitor and 4K content, you have correctly set
- output resolution allows you to re-scale that larger or smaller (with extra computation required to accomplish). For example, if you were to livestream at 1080p, your output could be scaled in Output resolution or later in stream settings... Having base canvas and output be the same means no scaling, which is better for recording

Now, as to file size:
1. you ignore pinned post in this forum regarding posting your OBS log when asking for assistance. You didn't do that, so my feedback based solely on your original post, vs others more knowledgeable than I possibly making suggestions based on specific encoding settings you are using
2. In general, for a given video codec, smaller file size = lower image quality/resolution/framte rate, etc... there is no free lunch.. extra data = more image fidelity. period. Which means cutting your frame rate, if 60fps not needed, would result in smaller files
However... it depends
3. There are optimizations one can make to the encoding, based on amount of motion in the video, color gamut desired, etc, etc. This can get fairly technical real quick. And I'd advise thinking about how much time you want to spend learning to make marginal gains. There are people who make careers out of such optimizations.
Oh, and this is a very high-level commentary... this is NOT my area of expertise.. just commenting so you have something to think about before someone can provide a better reply
4. You can change to newer codecs to improve compression. H.265 uses less bandwidth (smaller file size) for a given image quality, compared ot H.264, in most cases (there are always exceptions). The expectation (hope) is that AV1 will provide a similar improvement. AV1 improvements tend to be more noticeable at lower bitrates (per EposVox comparison YouTube videos/analysis). The newer codecs take a lot more compute power to pull off (of consumer gear, only the latest Intel ARC has AV1 encoder built-in. The upcoming AMD GPU is reported to have AV1 encoding as well. Time will tell if nVidia RTX 40xx has AV1 encoding built-into the GPU, just not turned on/announced yet (ie enable in a firmware update), or whether we'll have to wait for a later hardware release/iteration/revision/model or ??)
With that said, H.264 video encoding is reported to have improved over the various generations of NVENC... so a newer generation nVidia GPU with NVENC could improve quality for a given bitrate, or smaller file size for given image quality. I have not compared myself, but my recollection is that the difference is noticeable, but not necessarily large

I hope this helps

 

[Lawrence_SoCal]

Another thought to add to the above:
When choosing encoding codec, beware that H.265 is a bit of a licensing mess, hence it not being adopted for streaming. Most video editors can handle it fine, but you will hear of 'gotchas' more often with a H.265 encoded video (especially if getting fancy with audio tracks) vs H.264. AV1, being relatively new, is likely to have more bleeding edge teething pains, especially with editors. So when considering file encoding format to use, think about how you will use that file. The new, more compressed codecs will take more computation to edit... Ie smaller file size, slower to edit and save in a typical video editor ... a trade-off .. I recommend considering your entire workflow, and final video usage, when choosing initial recording approach