Jan

Jonno said:

I'm having fun trying to get Adobe Encore to import the video stream without crashing. I'm not sure the designers of this software even considered feature film file sizes when they put it together...

On x264bluray.com they say that Adobe Encore CS5 has issues with BD compatible streams if --open-gop is enabled, which it is in this case. I don't know if this applies to all versions of the software or if they fixed it, but as we're talking about Adobe here, the issue is probably still present.

I still want to do a few more test encodes for the upcoming AVCHD release. I'll then send the settings to Harmy in a few days and also post some comparison screens between Source, BD/MKV and AVCHD.

bilditup1 said:

 

Jan said:

 The par2 files however are different, so you'd need to redownload them at myspleen

 

Jan - not sure if I understood or not, but the par2 files are not part of my torrent. I didn't try downloading them till earlier today, and when I did, most of the downloads seem to have failed mid-way.

Sorry, I don't have an account at myspleen, so I didn't know that. What I meant though was that the par2 files on usenet differ from the ones Harmy created. The rar files are the same but renamed.

The Griff said:

Thank you, Harmy, and happy birthday.

Thanks, also, to everyone else involved.

Now, does anyone know if I should download this from usenet, would I be able to return to myspleen and help seed? That is, will the two versions match?

 

For the usenet upload the rar files have simply been renamed from Deed2.0 to Fixed_Deed2.0. The par2 files however are different, so you'd need to redownload them at myspleen.

Besides, 28% (50 files) have been uploaded to usenet so far, with approximately 1.5 days to go. So I guess downloading it from my spleen would be faster anyway.

So, for the upcoming AVCHD and BD, it'd probably be a good idea to not only create par2 files for the rar archive, but also an sfv file (contains CRC32 checksums) allowing us to verify the extracted content. What do you guys think? Would it help if Harmy created one for SW2.0 now?

Ok, I just extracted my rar archive which is definitely ok, and then verified the torrent at TPB. I'm really sorry, but I'm afraid it's indeed faulty. The first 3.83GB (24.3%) of the main movie seem to be ok, but from then on every single byte differs.

As a side note concerning the readme: Simply change the file type to .nfo if the tracker does not allow text files.

EDIT: When I advised Harmy to create par2 recovery files, my primary intend actually was that uploaders could verify and repair the files BEFORE creating mirrors or torrents.

Ok, here's how to verify the integrity of the rar archive and repair it if necessary.

First of all, download every part of the DeedV2.0 rar archive and at least DeEd.MKV.par2.part01.rar. Extract Deed2.par2 from said rar file to the same folder.

Get Multipar from http://hp.vector.co.jp/authors/VA021385/ and install it. Once it's finished right click on Deed2.par2, select "Open with..." and choose Multipar. It needs a few minutes for the process the finish, but in the end the program window should look like this:

If instead it says that some blocks are lost, you need to download further parts of the DeEd.MKV.par2 rar archive. You do not need to get all par2 files, it's sufficient to just download the same number of slices as those damaged. When looking at the file names in the above screenshot, f.e. +003 means that said par2 file contains three additional slices.

As a last step extract the par2 files, repeat the above steps and once checking is complete, simply click on "Repair" and wait for it to finish.

That's weird. Maybe the uploaders can check if their files are ok using a tool called Multipar.

I think it might fit if you remux it leaving out some of the audio streams.

Happy Birthday Harmy :-)

I'll start to upload a fixed post to usenet in a few minutes. The rar files are simply renamed, the par2 set was recreated though. I hope we'll get a good posting now. Sorry for the confusion.

Sorry that's my fault. I messed up something during the upload tests. There's only one posting in progress.

EDIT: Somehow the indexers combine the parts from a.b.test and the other groups. I didn't want that...

EDIT2: I'll post a fixed upload now starting from scratch. Sorry for all the confusion.

Is there such a thing as favorite prequel movie?

Darth Tater said:

Harmy said:

[...] I'll set an alarm to get up in two hours, to hopefully start the download of h_h's audio, so that it's done in the morning and I can start muxing right away. [...]

Harmy,

So excited to hear of the pending release. Did the above plan work out, or did you end up having to use the old h_h audio? I combed back through, and wasn't able to find an answer between here and there. Either way, thanks a ton.

-DT

Yes, hairy_hen's new mix is included.

All right, and good luck with your thesis :-)

Harmy, don't you want to wait a few more hours until hairy hen's recreated 70mm mix is available?

Actually Harmy, your screenshots are a great example of how you tweaked the colors, but they don't allow you to determine the loss of detail. They'd need to have the same colors, brightness, contrast etc.

What the hell? I've never seen an encoding flaw which looks like this. Which decoder do you use?

That's the plan.

The MKV should be about 13GB in size without audio steams.

I usually take the eac3to route for chapters, plus I use Chapter Grabber to get chapter titles.

It looks good so far, I hope the speed doesn't decrease too much later on. At such low percentages it's normal however since the first few thousand frames are the credits and the oping crawl which are easier to encode and require a smaller bitrate. That's why this part of the movie is faster to encode.

Ok. So let it finish it and then you'll see how fast the second pass is. If it's considerably slower than previously you can still abord it. Then you have the statistics file and can run the second pass seperately.

Have you checked the RAM usage by any chance?

That's strangely slow. How much did the first pass progress?

What kind of error did you get?

I think the reason for the low CPU usage in the first pass is that we use single-thread Avisynth. A multi-thread version exists, but I'm not sure by how much the encode will speed up. Plus we'd have to modify the script.

By the way, as long as you kept the files .stats and mbtree.stats from your last encode, you wouldn't have to redo the first pass again, just the second. 

Nice to see the encode is running :-) I sometimes get greatly differing encoding time for the same video as well and I don't know why either. It slowing down gradually is pretty normal too. The second pass of my test encode ran at 1FPS average while the first was almost exactly as fast as on your end.

If you want to work with your computer while it encodes just use the task manager and change the priority of x264.exe to either "Lower than normal" or "Low". I advise against doing CPU heavy stuff, but for the rest it should be ok.

Well, the settings for the Bluray are done. me=umh is for the faster settings and me=esa for the slower settings, while subme is set to 10 is for both.

I'll work on the AVCHD settings later.