My EVE

First, the programs. Of all the low end editing suites aimed at consumers or prosumers there are only two that aren't complete and utter dogsh*t. Those are Adobe Premiere Essentials and Sony Movie Studio. Of the two, I prefer the sony product because they offer something called the sony movie studio suite that includes some really great extras like some basic composting features and a decent sound editing program. At 60 bucks from Amazon, you really can't beat it. Here's the link- http://www.amazon.com/gp/aw/d/B008MIMIY8/ref=mp_s_a_1?qid=1363195037&sr=8-1π=SL75

Also, you should rethink your method of acquiring FRAPS. It's an awesome program made by an individual dude who you should support. At $20 for life, and it's incredibly cheap.


As far as the codecs etc. go, I don't really know what you are looking for, so you're going to have to give me a bit more detail about what you're looking for. That being said, I think what you're asking is what codecs and programs get used at the different steps along the way. Here is how I do it.

Step 1. Import FRAPS footage to a program called virtualdub. (as an aside, I downloaded a "lossless" codec called lagarith. It's great.) I then go to the video options and change the codec to lagarith. I then export this video and let it transcode. I do this step because most editing suites don't seem to like the native codec for FRAPS.

Step 2. Bring the footage into your editing program. Do all of your edits. Only at this point do I use what I called a finishing codec. These are the codecs that you may be used to seeing like .x264, which is the codec that I like to use. The reason I use lossless codecs up to this step is because the fewer times that you compress the footage, the better it will look. A finishing codec is a lossy codec. It will give you great compression for viewing, but is really sh*t to use for editing. You are going to need to use a container as well, and and these are going to be the things that you are used to seeing like mp4, avi, h264, etc. I personally use mp4 for my videos because YouTube likes it and won't take x264.


Let me know if you have any more questions.

About FRAPS, that's what it sounded like you were doing, I just wanted to throw my two cents in really more for other people reading this.

As far as the sharpness goes, are you keeping the resolution the same the whole way through? Is your final product the same resolution as what you are encoding in? Does the video look like that on your computer or only on youtube. If it's just on youtube, youtube doesn't really like any resolutions other than 360p, 480p, 720p, and 1080p. When you say "really high bitrate" what exactly do you mean? I use a 4mbit bitrate, which usually works fine for youtube videos.

It also looks like you aren't compensating for the difference between rec709 color space of the final footage and the computer RGB that the footage is recorded in. That is what is making your footage appear darker than it is. (This sh*t gets complicated, but essentially it's because in rec709 reference black is 16 instead of 0 and white is 235 instead of 255).

In movie studio, you compensate for it like this:
Step 1: http://i.imgur.com/VUG1Idy.jpg
Step 2: http://i.imgur.com/dh4EPsv.jpg
Step 3: http://i.imgur.com/OJID0E3.jpg
Step 4: http://i.imgur.com/SWVQk4I.jpg

Make sure that you do the above for all the tracks that you have, and you'll be fine.

Yeah, 20 mbps bitrate is silly huge and really won't help you all that much considering that you are uploading to YouTube. I use a 4 mbps bitrate and I am perfectly content with the results it produces. The quality is completely fine and the file size is much more acceptable.

One thing that you might want to check is the pixel aspect ratio. It should be square or 1.00000, depending on how it is displayed. It will be in your render settings. Something else that you might want to check is that your framerate matches the source footage.

You should also check your FRAPS footage to make sure that isn't blurry as well.

At some point, you are going to have to accept the fact that like all encoding, there will be a loss in quality. It's just a function of the process, and if you think about it, it literally IS the process.

As far as the lagarith thing goes, that is a relic from when FRAPS used the avi 1.0 container and couldn't hold more than 4 GB of data, so you had to stitch the footage together in virtualdub. I will have to check out whether or not I still need to do that step. Thanks for the heads up.

You are welcome for the feedback. It its always great to see more people making videos. If you want any more specific advice or help in the future, feel free to contact me in game if I don't respond here.

FYI, I found a simple way to record my screen using Free & Open Source software, as well as seemingly minimal resource overhead and much smaller initial 'uncompressed' files.

As for editing, I haven't really tried much, Avidemux is ok for very basic cuts, crops, overlays & framerate changes. Blender, well I still haven't found the time to learn the UI in all these years that it's been like that. I don't recall which other FOSS-on-Windows apps I looked at, except that really the 'pro' apps seems to dominate w.r.t. features, for a cost & having to be really quite recent versions to cope with the latest codecs & containers (the accidentally officially free Adobe Premiere Pro 2.0 couldn't seem to import common HD). Possibly the more powerful FOSS options are still Linux/Unix(Mac) native atm.

Christine - you're very welcome. The offer for help is open to whoever wants it. I would prefer to keep it on the forum because that can help whoever needs it, but I am more than happy to respond to in game mails.

Daneel - As strange as it initially sounds, a smaller file size for capture footage is actually COUNTER-productive. Raw footage is f'ing HUGE by design. It's nice to have all of that information to use to edit with. The merits of having all of that information to edit an amateur internet spaceships video is a different matter and one of personal opinion more than anything else. I prefer to use the use the larger footage since the "correct" way is to keep as much information intact as far into the process as possible. If your computer chugs when editing the larger files, you can use something called proxy files. That's a separate topic in and of itself, and I've gone far enough on a tangent already, but if you or anyone else is interested there are some great videos on YouTube explaining the process.

As far as the free and open source stuff goes, I really don't have any experience or interest with any open source NDEs. There are some freaking spectacular open source ancillary programs though. MEGUI is an AWESOME encoder, for instance. Isn't blender a 3d modeling program, though?

l0rd Carlos - I've never tried out VLC for screen capture, I'll have to give it a shot.

I can't get that VLC to work, tried editing the location (I'm no computer wizard, not with things like this). The program itself starts fine but when I try and use that bat nothing happens.

- edit -

I uploaded a new test here with what seems to be maxed out (but not silly) settings, could do with some critiquing on quality. On that note, I've been looking at many eve videos so far to see if someone got the quality exactly right but haven't found anything just yet.

Also, I had a look at that Adobe program but that's way outside my comfort zone.

Daneel, I definitely understand wanting to use an open source video codec and a free program. Honestly if you're getting good results, that's really all that matters. I do, however want to clear a few things up. The codec that it looks like you are using, mp4v, is part of the MPEG-4 standard. Which, despite being a great standard, has the unfortunate property of being F*CKING IMPOSSIBLE TO IDENTIFY. For reasons that are beyond me, both the container and the codec are called MPEG-4, and they pretty freely interchange the file extensions as container and codec too. In addition, there are like 20 "parts" that each have their own properties and they never tell you which one is being used. That being said, it looks like mp4v is most commonly used in the MPEG-4 Part 2 profile, which is really outdated. You should be using something like part10 (aka .h264) or part14. You also need to be really careful when using the word 'raw' when discussing photo and video because RAW is a file type, which this definitely is not.

Another huge concept is that in almost all (barring a few specific and debatable) cases MPEG-4 ONLY uses lossy codecs for video. One could argue that most footage shot in real life is shot in a lossy format, but I would counter that those cameras give you full control over your shot including things like exposure, white balance, color filters etc. that your capture method doesn't provide for. It should be noted that this kind of stuff really won't apply for most of the people reading this, and honestly, I've gone off on another tangent.

I want to be clear that I'm not arguing against your capture method, I just wanted to note the differences for people reading this topic. Like I said before, if it looks good, that's really all that matters.

I'm also going to give your method a shot when I get home from work.

You are of course right, by 'raw' I was meaning the initial output of any capture app. And yeah iirc VLC with those options is doing MPEG-2, but do you know the difference between it in high bitrate usage, and the FRAPS output quality? Also can you name any common lossless codecs?
I tried recording directly to part10/h264/VC-1 as per the FHC thread, but it's a little too taxing to do in real-time, for a loss in quality and also unneeded reduction in file size.

I'm just using notepad to create/edit the .bat iirc, as it's just a simple windows-line-endings text file. Win 7 64bit here. VLC on an SSD, writing to a diff HDD to my OS/Eve installs.
Feel free to try different FPS and bitrate values, or even play with the transcoding options in general if you're familiar with them/doesn't take you long to compare results, with your experience.

P.S. For anyone wondering, the difference between MPEG stardard parts is explained out there. part14 being a container, part10 being the better video compressed encoding than the older part2, etc.

Daneel, I tried your method and I'm going to have to pass on it. For one, I play in fullscreen mode and it is just recording my desktop and not the game. The second, and much bigger issue though is that my computer chugs pretty hard when trying to use your method. I've got a much lower end system than the monster that you are running, and trying to encode and record at the same time is not working for me.

H264 is indeed open source and it is a FANTASTIC codec. As a general rule, it's got a great compression rate with excellent quality. It also absolutely supports 4:2:2 and 4:4:4. However, h264 has these things called profiles. The profiles are what tells the program how to encode the file. It is the profiles that I'm having trouble finding free versions of, specifically the Hi422 and Hi444.