RipBot264 v1.18.3 - Simple and easy to use GUI -> IPOD . PSP . CONSOLES . BLURAY

[TDS]

Like I said, I was going to do some tests of my own, and here are the results !!

Nothing is perfect, it seems

Removed screen shot, as it has been copied in Atak's reply, below.

That's just a bit too random IMO.

AFAIC, if you don't have any encoding & playback problems, who gives a toss about "test's".

[Pauly Dunne]

Hey guy's,

I have 2 new re-hashed builds for you to try.

If you want a simpler experience, with a few Custom SMDegrain scripts ready to go, then the "Lite" build will be for you.

https://forum.doom9.org/showthread.p...64#post1983764

If you what to boggle your mind with almost endless possibilities, then the full build is for you.

It's only limited by your imagination, experience & requirements.

There are some very exciting new filters & scripts to explore.

https://forum.doom9.org/showthread.p...64#post1983764

Not sure what else to say, the ppl that already use these builds, (and I know there's 100's of you), will know what to do.

Newbies just need to unpack the .7z, do not rename the extracted folder, and put it where you want (preferably in the root of C:\ drive), read the instructions & install everything, and it should be up and running in no time.

Set your .ini files or use you existing files.

Good luck.

[MCFish]

Originally Posted by Pauly Dunne

Hey guy's,

I have 2 new re-hashed builds for you to try.

Thx mr. dunne

[Atak_Snajpera]

Originally Posted by TDS

Like I said, I was going to do some tests of my own, and here are the results !!

Nothing is perfect, it seems



That's just a bit too random IMO.

AFAIC, if you don't have any encoding & playback problems, who gives a toss about "test's".

With FFMS2 YOU WILL DEFINETELY have problems at split point in DE mode!
Trust me! I've been there and I was extremely frustrated with FFMS2 until I switched to LSmash. All my problems were gone when I dumped FFM2S! Older users of DE mode can testify how buggy was FFMS2. Corrupted frame when chunk starts, duplicated frames at chunks start and so on...

Missing frame means that frame for some reason was not extracted for testing. Maybe ffmpeg.exe crashed... Maybe you ran out of memory because UHD consumes a lot especially with 32 thread CPU! On my 16 thread CPU memory consumption reaches 18 GiB!
Focus on CORRUPTED FRAME ERROR and FRAME INSERTED ERROR

[TDS]

Originally Posted by Atak_Snajpera

With FFMS2 YOU WILL DEFINETELY have problems at split point in DE mode!
Trust me! I've been there and I was extremely frustrated with FFMS2 until I switched to LSmash. All my problems were gone when I dumped FFM2S! Older users of DE mode can testify how buggy was FFMS2. Corrupted frame when chunk starts, duplicated frames at chunks start and so on...

Missing frame means that frame for some reason was not extracted for testing. Maybe ffmpeg.exe crashed... Maybe you ran out of memory because UHD consumes a lot especially with 32 thread CPU! On my 16 thread CPU memory consumption reaches 18 GiB!
Focus on CORRUPTED FRAME ERROR and FRAME INSERTED ERROR

Did some more SeekTester runs today, keeping a very close eye on the memory usage.

With a 1080p clip, FFMS2 ran between 7 & 9Gb, whilst LSmash ran between 8 & 10Gb.

However, a totally different story with 2160p.

A 10bit HDR clip (Dirty Dancing), using LSmash, the usage was sky high, all but maxing out my 32Gb of DDR5 memory.

Didn't have any FFMPEG errors, and the memory usage was jumping aroung wildly, and when the test completed,
there were 998 errors out of the 1000 frames tested.

Then using FFMS2, there was even higher memory usage, and it actually crashed the PC, I had to turn it off manually.
There were multiple FFMPEG error popup warnings, before the crash.

Needless to say, I won't be running that test again.

So I starting to think that with these new Gen multi core CPU's, that 32Gb of RAM just isn't enough.

I will try a test on the Ryzen 5900X 12 core, and see what it does.

This may just be a reason why the 16 core Ryzen's are "falling off a cliff" when encoding 2160p.....

OK, more testing, but this time on a 5900X 12 core with 32Gb DDR4.

Testing a 4K Game of Thrones episode.

FFMS2 is a memory hog, it used close to all the 32Gb at certain points, and resulted in 498 errors.

However, LSmash used up to a high of approx 22Gb, and had ZERO errors.

I need to test that same file on the 16 core PC's, just for a bit of consistency.

Later on in the day, I did some tests on the 5950X, and I upped the RAM from 32Gb, to 64Gb...at first I had some crashes, but shuffling the RAM sticks around, fixed that.

So, the same episode of GoT.

FFMS2, is still a memory hog, it peaked @ 52.5Gb usage, but only had 111 errors, so maybe more RAM helps FFMS2.

LSmash, memory usage peaked @ 35.3Gb, and had ZERO errors.

No FFMPEG errors

So, I guess the next thing is to test some encodes of 2160p, while I have 64Gb available.

https://rpg-handbook.readthedocs.io/...#frame-serving

[TDS]

Another day, another interesting observation on memory use.

Obviously SeekTester is also a MAJOR "stress test", as it maxes out both the CPU & RAM, and like I mentioned in the previous post, SeekTester using FFMS2 is a usage hog.

So today, I am just doing a "normal" encode over 4 PC's, on a 2160p HDR movie, decoded with FFMS2.

Memory use, using only 1 DE server per PC:-

5900X - approx 13Gb (32Gb DDR4 installed)

5950X - approx 13.2Gb (64Gb DDR4 installed)

7950X - approx 15.3Gb (32Gb DDR5 installed)

13900KF - approx 14.8Gb (32Gb DDR4 installed)

And I have to mention that this is first time I have done significant encoding since the RB update that darkened the text colour in the Encoder screens.

Very nice, thanks, Atak.

[JThomasG]

hi,

ripbot264 is not quitting properly, when i exit it hangs and i need to kill it, does this happen to anyone else?

[TDS]

So regardless of the errors that SeekTester produces when using FFMS2, (and sometimes with L-Smash), the encodes I did yesterday did not have any encoding issues, and I'm using DE, and the whole movie played without issue.

I still think it would be appreciated if this could be added back as a decoding option....it's already a part of RB, it just needs the "switch" in the code to be enabled, and made available.

Leave it up to the user to decide, please consider.

[TDS]

Originally Posted by JThomasG

hi,

ripbot264 is not quitting properly, when i exit it hangs and i need to kill it, does this happen to anyone else?

So is this a regular occurrence ???

I sadly have this happen every so often, not too sure how, or why, but the best way to deal with it, is to "kill" it, and re-start.

I would be shocked, if nobody else had this issue, from time to time.

[JThomasG]

Originally Posted by TDS

So is this a regular occurrence ???

I sadly have this happen every so often, not too sure how, or why, but the best way to deal with it, is to "kill" it, and re-start.

I would be shocked, if nobody else had this issue, from time to time.

Yes, happens more often then not for me.
i even noticed a ripbot process in suspended mode that i could not kill i process manager (access denied) .

[Atak_Snajpera]

Originally Posted by TDS

So regardless of the errors that SeekTester produces when using FFMS2, (and sometimes with L-Smash), the encodes I did yesterday did not have any encoding issues, and I'm using DE, and the whole movie played without issue.

I still think it would be appreciated if this could be added back as a decoding option....it's already a part of RB, it just needs the "switch" in the code to be enabled, and made available.

Leave it up to the user to decide, please consider.

Encode vc-1 which is very common on blu-rays and then see what weird things happen at split points. I've told you before. EncodingClient always modifies script and ffms is replaced with lsmash! Ffms is unrealiable. Stop asking for this inferior and broken filter. I WILL never allow ffms to be used in De mode again.

[TDS]

Originally Posted by Atak_Snajpera

Encode vc-1 which is very common on blu-rays and then see what weird things happen at split points. I've told you before. EncodingClient always modifies script and ffms is replaced with lsmash! Ffms is unrealiable. Stop asking for this inferior and broken filter. I WILL never allow ffms to be used in De mode again.

So how come it shows it to be using FFMS2, in 1.avs ??

https://forum.doom9.org/showthread.p...77#post1983177

[avinewbie]

[QUOTE=Pauly Dunne;1983226]Hey guy's,

I have 2 new re-hashed builds for you to try.

Hello P Dunne

I would like to try out your special builds but unfortunately my AV BitDefender keeps blocking the download and flags the file ImageSeq.dll in avisynt/libraries/x86/plugins/ImageSeq.dll as infected.
It does clean it and leave the file in the folder( presumeably sans infected bits).
Its probably a false +ve and the file has been uploaded to BitDef but would you mind checking?
The same warning occurs for all your recent uploads. I havent yet tested if every thing works.
Thanks

[Pauly Dunne]

Originally Posted by avinewbie

Hey guy's,

I have 2 new re-hashed builds for you to try.

Hello P Dunne

I would like to try out your special builds but unfortunately my AV BitDefender keeps blocking the download and flags the file ImageSeq.dll in avisynt/libraries/x86/plugins/ImageSeq.dll as infected.
It does clean it and leave the file in the folder( presumeably sans infected bits).
Its probably a false +ve and the file has been uploaded to BitDef but would you mind checking?
The same warning occurs for all your recent uploads. I havent yet tested if every thing works.
Thanks

Hi avinewbie,

Thankyou for wanting to try my builds

I would doubt that there is anything wrong with that file, or the downloads in general.

I would suggest that it's AV BitDefender being way to severe with certain files.

I only use Microsoft Defender (the default Windows AV app), so maybe just as a test, at your end, disable or remove AV BitDefender, and see what happens then.

Having said that, I think you'll find that RipBot doesn't need the files in the folder that that file is located, but if that's what triggering your AV, then that's a problem.

PM me, if you're still having issues.

Cheers

[Pauly Dunne]

In light of "avinewbie" 's post, I will remove the offending files & folder's, as it appears that RipBot does not required them, anyway.

And getting valuable feedback, can just make things better

Also, and yet again, there has been some significant updates to the working's of my build's, mainly Dogway...he just can't keep fiddling & updating his scripts, and he has basically re-written SMDegrain in the passed couple of days.

So to keep up with his changes, I've needed to reflect those changes in ALL the SMDegrain based scripts (and there's a LOT of them, especially in the full build).

So I will be posting updated builds, shortly, and unfortunately, the changes may not be compatible to the current builds.

I knew I should have waited a bit longer before posting, but you just never know what the others are doing, and when they will update their app's.

So when it comes time, I will leave the current builds available, and leave it up to you, whether you want to try the updated builds.

Sorry guy's, I know this is a PITA.

Regards

[avinewbie]

[QUOTE=Pauly Dunne;1983491]In light of "avinewbie" 's post, I will remove the offending files & folder's, as it appears that RipBot does not required them, anyway.


Hi PD:
I can confirm this new upload was successfully downloaded--no Bitdefender warnings!
Thanks you so much for your efforts, Much appreciated.

[Pauly Dunne]

Originally Posted by avinewbie

Hi PD:
I can confirm this new upload was successfully downloaded--no Bitdefender warnings!
Thanks you so much for your efforts, Much appreciated.

Thanks avinewbie, that is good to know

I had a good look around the files & folders of Avisynth, and found some older stuff in there that didn't need to be there.

Now I can update the "full" build, and post that shortly.

[Pauly Dunne]

It's all been updated

https://forum.doom9.org/showthread.p...64#post1983764

[TDS]

I have run into a bit of a problem when importing a DNxHR HQX clip..

Getting this error message :-



File info :-



Or is it that RipBot264 doesn't support these types ??

Even after running it thru MKVToolNix, it has the same error

EDIT:- more info...I tested this .mov with Staxrip, and as it has several "Source" options, I tried FFMS2 & L-Smash, and it imported the file with no issue.

But when I changed it to DirectShowSource, it displayed the exact same error as RipBot does.

So RipBot must somehow "connect" DirectShow with both FFMS2 & L-Smash, when decoding.

Regards

[Atak_Snajpera]

Send me that file.