Norman Black
Forum Replies Created
-
[Danny Hays] “I am a GoPro user as well. I shoot in 60p by default as none of my cameras can shoot 30p. You are right in that we may want to slow down the clip. Have you tried the Flux feature in GoPro Studio?”
Yes, I have looked at Flux. Before Flux existed I purchased Prodad Respeedr and I have used that so I have not actually used Flux other than simple tests. I will say that Respeedr is WAY faster than Flux.
-
Any tutorial should be fine, as the titler has not changed over time. The location of the collections button has changed but the features are now as they were.
I’ll bet nearly all tutorials are “old” because they were done when the titler was a new thing.
-
[Martin Stanesby] ” I have also found that whenever panning at 25fps, or using dolly track, it is not as smooth. I know that effectively scaling down from 50fps to 25fps is losing frames anyway, but it just works better for me. “
Curious. If you are shooting 50p and an appropriate shutter of say 1/100 and then disabling resampling and rendering down to 25p it should look worse than 25p with an appropriate shutter of 1/50th. The slower shutter of 25p would have motion blur to smooth things out.
In the GoPro world, shutter speeds are 1/1000 and faster outdoors so there is no motion blur in video. For normal video that shutter is not ideal as we don’t get motion blur, but when wanting to use something like Twixtor/Respeedr those fast shutters are ideal.
-
[John Rofrano] “Those of us who shoot at the same frame rate that we deliver don’t have this problem so it never bothered me.”
Since I am one of the GoPro guys, I often shot in 1080p60 intending to final render at 30p. The 60p was only just in case I wanted slow mo. You never know what might happen on an MTB ride.
This is a situation where “Smart” resampling needs to be smarter, IMO. When going from 60p(59.94) to 30p(29.97) the resampling does frame blending for the result when it should probably just drop every other frame. I think if the smart resampling were smarter, there would be a lot less complaints about it.
-
[Greg Holman] “So I’ve been using Sony Movie Studio Platinum 13 to make videos for my YouTube channel. The videos look great after rendering, however once they are uploaded to YouTube a ‘blurry flicker’ appears in the background image layer.”
Youtube re-encodes everything you upload to their own spec. So if your encode is fine but theirs is not then creating a “better” encode on your end is not going to help Youtube any since their are not preserving your quality. Generally they cannot since their bitrate is quite low.
[Greg Holman] “I then render it using the Sony AVC template without changing the settings (I have tinkered with them to try and solve the problem but to no avail).
“There are a lot of Sony AVC templates. Which one? Internet 1080? Per chance, did you render to an interlaced format. That might mess with things on the Youtube side of things. They recommend progressive uploads.
-
[John Rofrano] “You want to keep it on for it’s original purpose which was to go from 29.970 to 23.976 or from 23.976 to 29.970 back in the say when those where the only two frame rates that matter in the NTSC world. For those of us who are not shooting 60p, Disable Resample works just fine. I’ve never turned it off.”
I like the OP thought it (resample) was a worthless feature, but when I looked at the Sony Red car GPU benchmark project I noticed that there were two clips in slomo with smart resample, and they look nice and smooth. The Project is 1080p30 and one clip was 1080p30 and the other 1080i60. Both clips were .632 playback speed. So it can and does work, but of course not for everything.
SCS can improve the Smart in Smart resampling, and they should probably give a preference for global disable and default it as current.
-
[Aaron Star] “@Norm – h.263(Dvix)- you should look at Dvix today, they are well past h.263, and have been one of the 1st h.265 systems for some time now. The .MKV container still limits peoples wide acceptance of the format.”
Yea. DivX is a company name and not a codec name but the MPEG-4 part2, aka H.263, codec was the first thing for them and they were primary of that at the time, I just lumped the company name and original codec together. Almost like making a copy and Xerox.
DivX currently owns Mainconcept. A name many Vegas users all know and love, err hate, since nearly every video/audio encoder/decoder in Vegas is sourced from Mainconcept. The AVC decoder in Vegas is quirky/problematic to say the least.
x264 wipes the floor with every other AVC encoder out there, in quality and speed, and over time x265 will likely get to a similar place. Multicoreware/x265 has licensed rights to the x264 code/algorithms/concepts. Even for commercial purposes. I doubt SCS will ever license either one.
-
H.265 is an extremely new codec and not older than H.264. Did you mean to type H.263 (which is Xvid/DivX), or mpeg-2 per chance?
Anyway Vegas cannot ever import Xvid/DivX in an MP4 container. Maybe in an AVI container with proper Video for Windows AVI codecs installed.
40 fps? Is that a typo?
[Kyle Le] “After I right click on create proxy file, what else am I supposed to do? “
Wait for the proxy encoding to finish. Vegas proxies are XDCAM mpeg-2. If you cannot edit that smoothly then it is not likely anything will edit smoothly on your machine. MPEG-2 is very fast/easy to decode.
Editing 4K is likely to involve proxies for edit/cutting the video at least. Hey it is done like that in big budget films. In cutting you want smooth play to get the feel for the flow and you don’t need 4K to cut/edit.
-
[Blake Gibson] “I stumbled across this thread in my travels, it looks interesting, this guys basically found an app that lets you use more than 2gb memory on 32bit exe/dll’s… apparently it only works to get around the crashing during render issue in Vegas. I wonder if it would work for the fileiosurrogate.exe:”
Yes, you can try setting the large address aware flag on the executable (fileiosurrogate), for 3GB user space, but you have no idea what the result might be.
Many applications wrongly assume addresses above 0x80000000 (2GB) are system addresses. Even with Sony knows their fileiosurrogate application does not make any such assumptions, they have no idea what assumptions the quicktime code makes. It goes even further. Quicktime has no idea what assumptions the third party codecs, like DNxHD, might make.
It is most likely the case that everyones code does not make any silly assumptions about address range. You can try it, but the millisecond something does not work I would switch it out.
-
[Blake Gibson] “I’m guess they only mean specifically TO Quicktime FROM non quicktime files? Like AVCHD to DNXHD or something?”
That said, wouldn’t the 32bit memory limit apply because its using quicktime?
I think they mean rendering to Quicktime. Which means one output file, and the problem being only on large/long projects. The sources are not likely relevant.
On the front end of Vegas you have the file decoders. In the middle you have the video engine. On the back end you have the file encoder (render as). The file encoder has no clue what is going on in the video engine and no clue about the decoders.