Joined: 6/25/2007
Posts: 732
Location: Vancouver, British Columbia, Canada
It looks better, has wide compatibility (LAV Filters, ffdshow, mplayer2, and VLC), and has equal or smaller file size. 10-bit 4:4:4 should definitely be the primary downloadable encode, with possible 8-bit 4:2:0 as secondary. The quality improvements are huge in low-res video like NES/SNES/GBC/Sega Genesis/Mega Drive. I support options, but I don't support the idea that old technology should be considered "primary" while new, widely-supported, and greatly improved technology is only secondary. We switched to H264 early in its existence and nobody had an issue with that. It was a huge improvement over xvid or divx.
Joined: 6/25/2007
Posts: 732
Location: Vancouver, British Columbia, Canada
"AR correction", as mentioned by Ilari, is a container flag which specifies an aspect ratio for playback as one would see on a 4:3 TV. This is based on the assumption that most people play NES and SNES games on 4:3 TVs which would stretch the 256×240 or 256×224 image (both of which are not 4:3) to a 4:3 width:height ratio. This can be specified in the header on video muxing. Some games' graphics are designed for 4:3.
In my personal opinion, it's not really necessary since decent video players allow the viewer to override the aspect ratio to 4:3 if they want to see the distorted image. Also, I've found that most games' graphics are not designed for 4:3, including Super Mario Bros., so I don't think it's a good idea for Super Mario Bros. Also, it's far less common to be able to have the player ignore the "corrected" aspect ratio flag than it is to be able to override the aspect ratio with 4:3.
"PAR" as mentioned by Nahoc above means "pixel aspect ratio". This is the ratio of a single pixel's width to its height; width:height. Without any aspect ratio flag in the container header, all video players assume a 1:1 PAR (square pixels), meaning the NES image would be 256×240, assuming it wasn't cropped.
"DAR" is "display aspect ratio", meaning the ratio of the entire image's width to its height. A 256×224 image with square pixels stretched to 4:3 DAR would end up as a 298×224 (incorrectly rounded) image. These (298 and 224) are the dimensions you would find in an "AR corrected" mkv's "video display width" and "video display height" header fields, as seen in mkvmergegui's header editor. Video players use this flag to override the display aspect ratio on playback without the actual video stream data being affected.
Joined: 6/25/2007
Posts: 732
Location: Vancouver, British Columbia, Canada
goofydylan8 and amaurea, I thoroughly enjoyed your posts and completely agree with everything you've both said. To be honest, I skimmed this thread except for those few posts, so I'm sorry if I didn't mention others who agree. Anyway, count me in for accepting all games as "publishable". We have a rating system. Let's start using it effectively.
Joined: 6/25/2007
Posts: 732
Location: Vancouver, British Columbia, Canada
You're definitely right. YUV 4:4:4 10-bit is far better for lossy compression. I agree with your conclusion here.
I would still like to see a comparison using -qp 18 (or some other qp value), as I said in the channel, just for an numerical way to compare.
Joined: 6/25/2007
Posts: 732
Location: Vancouver, British Columbia, Canada
I still think a PC tower case is most representative of the system. Of course, most PCs were horizontal a long time ago, but the ATX mid tower or full tower case is almost standard among PCs nowadays, making it an easily-identifiable icon.
Joined: 6/25/2007
Posts: 732
Location: Vancouver, British Columbia, Canada
Have some more faith in your CPU. Intel and AMD CPUs are very robust. They are designed to be able to run at 100% capacity for months, if not years, on end. Your CPU won't burn itself out from a few encodes. Also, I'm not sure about AMD, but Intel CPUs throttle themselves before they get to a temperature which would damage them.
Joined: 6/25/2007
Posts: 732
Location: Vancouver, British Columbia, Canada
Use normal numbers. Roman numerals don't add enough flavor to warrant their usage. Whenever I install a game with Roman numerals and am prompted for directory name, I just use the real number instead ("heroes of might and magic 3", "civilization 2", "master of orion 2", etc.). Not only does it cause my file manager to properly sort the directories by name, but it's also easier to read at extreme speeds (such as when scanning through a list). Anyway, I like "1", "2", "3", "4", "5", "6" etc. "I", "II", "III", "IV", "V", "VI" can be relatively confusing when you're quickly scanning, especially with the possibility of confusing "IV" and "VI".
No. This post will not hijack this thread.
Touch-me, your TAS is amazing and I voted yes, as you could probably guess since I already posted. I am very happy you decided to submit another previously-unpublished game TAS to TASvideos (after Secret Of Mana).
Joined: 6/25/2007
Posts: 732
Location: Vancouver, British Columbia, Canada
This is one of my favorite games. I have been dreaming of a TAS for this game for a long time. I just watched the entire TAS and enjoyed everything contained. This TAS is amazing.
I noticed you credited Enix for publishing the game, but not ChunSoft for actually making the game. :o
Anyway, Touch-me, you're a genius.
Joined: 6/25/2007
Posts: 732
Location: Vancouver, British Columbia, Canada
Congratulations to all the fantastic winners of this year's TASVideos awards. You all inspire awe with your dedication and quality of production. Thank you for making these exciting pieces of work. :)
Joined: 6/25/2007
Posts: 732
Location: Vancouver, British Columbia, Canada
If you count the total number of votes by adding 32 to 1, you will find that it is 33. Either vote count divided by 33 is the resulting percentage you're discussing. In the case of the 32 "Yes" votes, the result of 32 divided by 33 is 0.96 with the last two digits repeating, or 96.96% with the last two digits repeating. If this is the real value for the percentage of voters which voted "Yes", which it is, that means that the forum software printing the vote percentages is simply truncating the decimal and all following digits. This results in inaccuracy in totals, but is sufficient in describing with some accuracy the overall result.