@roadzy said:
I was just given an old Dell T3500. It has 4gb of ram and a 2.53 GHz Dual-Core Intel Xeon. I believe the CPU is this: W3505 – 4M Cache, 2.53 GHz, 4.80 GT/s Intel® QPI. I found a list of CPUs i could upgrade down the road to but I’m wondering if it is the bottle neck.
Under 2000 passmark points. Many modern Corei3 cpu’s have more power than this.
You will need at least 4000 overall passmark points
and over 1350 passmarks point for a single cpu core (this is crucial! Don’t look only after the overall performance rating.)
Althoug I recommend you to use a more powerful cpu of course to have power for more than 1 transcoded playback at once.
When you rip your BRs, you can extract the ‘forced’ subtitles and convert them into SRT format.
Then you could continue with your current mp4 files.
Movie Title (year) - 1080p.mp4
Movie Title (year) - 1080p.eng.forced.srt
@roadzy said:
I was just given an old Dell T3500. It has 4gb of ram and a 2.53 GHz Dual-Core Intel Xeon. I believe the CPU is this: W3505 – 4M Cache, 2.53 GHz, 4.80 GT/s Intel® QPI. I found a list of CPUs i could upgrade down the road to but I’m wondering if it is the bottle neck.
Under 2000 passmark points. Many modern Corei3 cpu’s have more power than this.
You will need at least 4000 overall passmark points
and over 1350 passmarks point for a single cpu core (this is crucial! Don’t look only after the overall performance rating.)
Althoug I recommend you to use a more powerful cpu of course to have power for more than 1 transcoded playback at once.
When you rip your BRs, you can extract the ‘forced’ subtitles and convert them into SRT format.
Then you could continue with your current mp4 files.
Movie Title (year) - 1080p.mp4
Movie Title (year) - 1080p.eng.forced.srt
I tried he script and the mp4 files is still 25gb! Was the point of the script just put it in a different format or to shrink it in size?
Also I was curious what people do with 5.1 or 7.1 audio ? Do you have to have a specific type of format to have that?
Check the Bitrate - I bet is like over 10m bps
Try converting to say 3,000 bps - check the size and the picture
And yes it will take time !!
Ok I’ll try that! Is there a sweet spot that you like? Thanks! Also how do you get forced subs to work? For instance in Argo, they talk in another language in the one part. If you are watching the dvd/bluray, it shows up as subtitles. I don’t want subtitles on the whole movie though. I’m checking subtitles and force subtitles when ripping in makemkv. Even as a mkv I can’t get them to show up. Am I doing something wrong?
The easiest way to get forced subtitles is to use Sub-Zero subtitles and enable it for just forced subtitles. Rip the movie using handbrake to get a small-ish file, and then Sub-Zero will grab only the forced subs. Or, you can download the forced subtitles and use mkvtoolnix to mux them in after you rip with Handbrake. Takes no time at all to do that.
If you use JuiceWSA’s guide you’ll find great settings for Handbrake.
@roadzy said:
Also how do you get forced subs to work? For instance in Argo, they talk in another language in the one part. If you are watching the dvd/bluray, it shows up as subtitles. I don’t want subtitles on the whole movie though. I’m checking subtitles and force subtitles when ripping in makemkv. Even as a mkv I can’t get them to show up. Am I doing something wrong?
Also, if you don’t want the file to transcode then you will have to convert the forced subtitle to plain text SubRip, *.srt.
As pictured above. That episode has a single line of non-english dialogue, so I converted the IDX/SUB file. No need to transcode the whole file for 1 line of text…
How accurate is Subtitle Edit with OCR of vobsubs? How much correction are you having to do? Shame SRT cannot have onscreen location parameters as VOBSUBS can have placement to avoid covering an important part of a scene etc.
After some training it is pretty accurate.
It has some weaknesses with italic text and very short lines. (e.g. “Hey.” always ends up garbled.)
You can save some editing work if you load an existing srt file from somewhere else and only use the OCR’d version to sync up the other file.
(That’s a very cool feature, IMO.)
Just use plexkodiconnect
2, Batch convert all VOBSUBS and PGS to SRT (OCR isn’t perfect and will lose subtitle dynamic x, y placement)
3, Move to emby which doesn’t have VOBSUB subtitle issue.
If you don’t mind, I’ve a slightly related question; is there still any issues at all with sidecar SRT and android TV; @anon18523487 mentions it a few times . I know a while back direct stream of sidecar SRT was disabled (BURN IN only) and wanted to know if android tv clients (shield tv, sony tv) can now direct stream these?
Also, if i moved ahead with this while waiting for plex to fix this: VOBSUBS stop showing; then plex app on shield tv will crash I’d prefer not to embed the SRTs but sidecar them (for ‘extras’ I would embed, due to a bug plex has in this regard) and wanted to know the syntax for ‘default’ subtitles?
I know forced is: Movie Title (year).eng.forced.srt
Is ‘Default’: Movie Title (year).eng.default.srt
I don’t know specifics for Android TV, because I’m not using it myself.
But you always want to make sure that your SRT files are:
are encoded as UTF-8
valid according to the spec (which means no empty lines, no overlapping show times and no invalidly encoded html or other stuff inserted from opensubtitles.org )
Most of these issues can be rectified with SubtitleEdit.
There is no syntax for ‘default’ external subtitles.
Plex simply ignores the ‘default’ tag when its set in MKV files and uses the first subtitle stream that fits the requirements.
I closed down the OCR window (the one where you can see the OCR text alongside the bitmap text); how do I get that back so I can continue corrections? I dont want to have to start over on the file that I have already part corrected, but i do need the bitmap subtitles sidebyside.
You can start a second instance of Subtitle Edit
open the bitmapped subtitle again and go into the OCR window
You can scroll down to the right line number/time stamp. There is no need to perform OCR again.
I found that you can copy in the SRT file into the subtitle text window of the OCR window, and do it that way too; obviously the time codes all match, so it’s perfect.
)