I've seen a lot of debate regarding the best compression method for videos.
I have movies taken by a digital camera in Uncompressed AVI format (some HD some not).
I want to reduce the size but not loose quality (as much as possible).
What would be the preferred method and tools?
From what I've learnt in the past, opposite to audio, video has a lot of redundancy, so lossless will have hight compression rate in relation to lossless audio compression, will that be the way to go? or should I go with MPEG2/4, DIVX etc.?
TMPGEnc, Handbrake, something else?
I appreciate your help and recommendations
+ Reply to Thread
Results 1 to 12 of 12
Choose container mp4 or mkv and encode your video into H.264 using Constant Quality encoding. Higher quality - bigger files, so it is up to you to find Constant Quality value that works for you. It could be marked as CRF in other GUI encoders using x264 encoder.
Start with factor 18. Or choosing 17, for example, whatever feels good to you, lowering the number means better quality but bigger filesize. HD source gets away with one or two step higher number as oppose to SD resolution (where everything gets blown up on screen, including bad artifacts). So you do couple of short tests to find your constant rate factor number, then just use it all the time.
Handbrake or Vidcoder would be fine, not sure if TMPGEnc has constant quality encoding.
EDIT: sorry for not reading the question properly, sure, you might try to go lossless, in VirtualDub (if it is going to load that video) what compression ratio you are going to get, then you have to sort it out , comparing what you got, what room you are going to give it. If bitrate is too much, you can try very low Constant Quality number (handbrake), not sure what would give about 50Mbit/s (example), depends on video, perhaps value less than 15, also I'd put I-frames closer together (30 frames) , low ref frame number 2-3, preset default, do not forget to make sure that buffers are set depending what bitrate you want bitrates to be at its highest, so you do not have buffers set to 40000 expecting 50Mbit bitrates , better a bit higher ...
Last edited by _Al_; 21st Nov 2015 at 16:41.
Typically, for archival purposes, you should leave the video untouched to maintain the full quality. You never know in which format you may want/need the video in future.Google is your Friend
Thank you both.
@_Al_ that is the exact type of answer I was looking for, thank you.
@Krispy Kritter I fully understand what you mean, but sometimes you just want to save some space and compromise on the quality, having 250MB file compressed to 150MB doesn't seem to bad (720P), especially if you have 100s of them
Storage space is relatively cheap. And the current market for TV's is 4k. Who knows what it will be 5 or 10 years from now. Take all factors into consideration, after you delete the originals there is no going back.Google is your Friend
You are right but you are always limited by the original resolution, moreover there is no end to that (in a few years we'll have 15K TVs ), and one must learn to say enough is enough....
True. But all things being equal, your original source files are very likely going to scale better to a larger resolution. The amount of data present is just as important as the resolution, and compressing to save space removes that data.
This forum is littered with topics of people trying to make their old media look better on their new big screen TV. ALL of those people wish they had a higher quality source. I'm just trying to save you (or your kids or grand kids) some grief in the future. It's your time, money, and memories, do as you please.Google is your Friend
Most of the cams/devices that DO record AVI usually use either MJPEG (like one of the cams I have) or DV, or possibly Cineform/DNxHD if you're really lucky (rich).
Since what you are starting with is ALREADY compressed (often ~5-8:1), compressing them even further just makes them that much more "fragile". That may be ok for dime-a-dozen TV clips, but it would be a shame to do that to precious family/vacation (or similar personal) footage.
BTW, we will probably not have 15k TVs for decades, if ever (as there is no visual difference beyond 4k for standard home TV sizes). Not worth the effort/expense.
It's really a change of thinking that is necessary. Good practice in this hobby is to keep the source. Things change all the time, and you always have the highest quality option to work with this way.
My suggestion is that you should typically have at least two collections. (And of course, their respective backups.)
A) Your archival/acquisition/source. This may not be optimized for playback or compression, but it doesn't matter. And it's always typically the first file you obtained, whether by camera, capture, download, etc, whatever it is - this should always be your master.
B) Your playback/deliverable version. This is what you use for playback, and it could be in multiple formats whether you choose DvD, blu-ray, media player, mobile, etc. It doesn't matter what you do here as long as it fits your playback needs and you have kept a copy of the source.
The mindset here is dissociating the two into separate departments. They are not the same. Professionals do this all the time, and so do advanced hobbyists. If you only look at it as "extra space" you are not thinking this correctly, because they really are two sets, not one doubled. For phycological reasons you can keep the source tucked away into a back corner of your closet so you won't feel like you're using all that HDD space. But keep it safe for the future.
As well, you will save yourself from a ton of potential regret of damaging quality - that you will never get back, as mentioned - from irreplaceable family moments (if they're that precious) just because you wanted to save a few TBs of storage today, which is cheap and compact now, and will be ever more so in the future.I hate VHS. I always did.
Corucopia - yes, I screwed that one up, in many cases users can use term uncompressed for their camera original footage, I realize that too late, that is why I corrected myself in #8 input, so my #2 answer makes little sense, because encoding with lower constant rate factor, would give him same volumes, if not bigger and his footage would be worse (after one encoding generation) ...
op should figure it out himself actually, as soon as he'd started to encode his footage using those lower CRF numbers and getting volumes that are bigger than original , even with compressed audio, that might be PCM in originals