Experiences with HC Encoder

[jdobbs]

I was wondering if any of you have experimented with hank315's new HC encoder and DVD-RB (support was added in the latest Pro version)? While it is slower than CCE and QuEnc, I did some testing (Alien vs Predator) -- it looks very promising, and the sizing is extremely accurate. If you have some free time it would be nice to get some reports on your experiences.

Thanks.

[DD51]

Originally posted by jdobbs
I was wondering if any of you have experimented with hank315's new HC encoder and DVD-RB (support was added in the latest Pro version)? While it is slower than CCE and QuEnc, I did some testing (Alien vs Predator) -- it looks very promising, and the sizing is extremely accurate. If you have some free time it would be nice to get some reports on your experiences.

Thanks.

This is embarrassing but here goes...

I'd love to try but I don't see how to set it up in the setup portion of DVD-RB. There is no HC setup.
I'm using DVD-RB0.78 Pro

[NobbyNobbs]

There is no setup, you have to put the HC encoder in the same folder as RB, and name it HC.exe :)

[DD51]

Originally posted by NobbyNobbs
There is no setup, you have to put the HC encoder in the same folder as RB, and name it HC.exe :)

Thanks for responding...how embarassing...

But now I get the following error message"
Cannot create file "C:\Program Files\DVD-RB\HC.exe". The process cannot access the file because it is being used by another process.

Man this is frustrating.

[jdobbs]

Reboot your computer. Then download the version that supports the "*BATCH" parameter (it prevents having to hit ENTER for each encode). It is located here. (http://hank315.dyndns.org/HC_0.01.zip) You'll have to rename it to HC.EXE.

[DD51]

Originally posted by jdobbs
Reboot your computer. Then download the version that supports the "*BATCH" parameter (it prevents having to hit ENTER for each encode). It is located here. (http://hank315.dyndns.org/HC_0.01.zip) You'll have to rename it to HC.EXE.

Thank you that works great...It now works without any problems.
BTW I had rebooted my comp and also checked in the processes. It would not start!

Anyways...I'm going to backup "24" Season 3 Disc 5. I have already done disc 1-4 with CCE SP with 2 pass.
I'm going to run it tonight and get back with my personal findings.
With CCE SP it went like this:
Phase I: Prepare took 3 minutes
Phase II: took AVG. 105 minutes
Phase III: took AVG. 15 minutes

Will let you know with HC.exe my findings when all is done.
I will specifically be looking for quality differences between the 2 on a frame per frame basis.

Thanks again for the help...

Regards,

[jdobbs]

Thanks! I'm interested to see what you think.

[TheSeeker]

It'll be hard to topple a giant like CCE though. As far as mpeg2 encoders go its pretty much the best engine alive isn't it?

I think I may also give HCEnc a try. See how it stacks up.

[jdobbs]

Oh yeah. CCE is (in my humble opinion) the best on the planet. But HC is free, and never hurts to check out new things.

[dragongodz]

As far as mpeg2 encoders go its pretty much the best engine alive isn't it?
yes possibly but its a fallacy that no other encoder,especially a free one, can approach its quality. for example if you have to do a frame by frame analysis to notice any differences of encodes done by 2 different encoders then its pretty safe to say they are of similar quality in that instance. :D

[TheSeeker]

Ok. I am playing around with HC encoder now. And I setup DVD RB 0.78 in hc mode, set the quality to best and left the quant mat. setting alone. (Default). Finally, I downloaded the version of hc referenced by JDobss in this thread, and renamed it to HC.EXE and put it in the same directory as dvdrb:

http://forum.doom9.org/showthread.php?s=&threadid=91282

My source was a totally unpreprocessed version of Saw. I did a prepare phase and that went fine. But when I hit encode, immediately I get a Runtime Error 5: Invalid Procedure Call or Argument. Not sure why Im getting the error...

Using avisynth 2.55
dgdecode 1.10

EDIT: Ok after some more messing around I found I can manually run the encode by copying the contents of the HC.ini from the D2VAVS directory into the hc.ini that is in the same directory as dvdrb and hc.exe so it seems its something about the way dvdrb is calling on hc.exe that my setup doesnt like.

[hobyho]

I have to say HC is a very good encoder, quality is excellent... especially with low bitrate encodes.

I've done Man On Fire with HC (like a 50% compression due to DTS track), where both CCE and Procoder showed a degree of blocking at two particular places in the movie, whereas HC produces none.

I can't wait for further improvements to it, hank is doing a terrific job, keep up the good work.

[jdobbs]

My source was a totally unpreprocessed version of Saw. I did a prepare phase and that went fine. But when I hit encode, immediately I get a Runtime Error 5: Invalid Procedure Call or Argument. Not sure why Im getting the error... That ain't good. :( I'll see what I can find.

[DD51]

Originally posted by jdobbs
Oh yeah. CCE is (in my humble opinion) the best on the planet. But HC is free, and never hurts to check out new things.

OK! Just ran HC with the following:
Quality/Speed Selection: BEST
Quant Matrix: HC Matrix (default)
Movie ONLY: "24" Season 3 disc 5.
Please note I do ONLY movie backup! Everything else removed!
Here's the log from DVD-RB 0.78.

Transcoding...
-----------------
[13:32:21] Phase I, PREPARATION started.
- VTS_01: 955,457 sectors.
-- Scanning and writing .D2V file
-- Processed 64,476 frames.
-- Building .AVS and .ECL files
- VTS_02: 922,676 sectors.
-- Scanning and writing .D2V file
-- Processed 63,220 frames.
-- Building .AVS and .ECL files
- VTS_03: 990,586 sectors.
-- Scanning and writing .D2V file
-- Processed 63,925 frames.
-- Building .AVS and .ECL files
- VTS_04: 991,941 sectors.
-- Scanning and writing .D2V file
-- Processed 65,302 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 54.2%
- Overall Bitrate : 2,929Kbs
- Space for Video : 3,831,280KB
- HIGH/LOW/AVERAGE Cell Bitrates: 3,984/393/2,929 Kbs
[13:35:42] Phase I, PREPARATION completed in 3 minutes.
[13:35:42] Phase II ENCODING started
- Creating M2V for VTS_01 segment 0
- Creating M2V for VTS_01 segment 1
- Creating M2V for VTS_01 segment 2
- Creating M2V for VTS_01 segment 3
- Creating M2V for VTS_01 segment 4
- Creating M2V for VTS_01 segment 5
- Creating M2V for VTS_01 segment 6
- Creating M2V for VTS_01 segment 7
- Creating M2V for VTS_01 segment 8
- Creating M2V for VTS_01 segment 9
- Creating M2V for VTS_01 segment 10
- Creating M2V for VTS_01 segment 11
- Creating M2V for VTS_01 segment 12
- Creating M2V for VTS_01 segment 13
- Creating M2V for VTS_01 segment 14
- Creating M2V for VTS_01 segment 15
- Creating M2V for VTS_01 segment 16
- Creating M2V for VTS_01 segment 17
- Creating M2V for VTS_02 segment 0
- Creating M2V for VTS_02 segment 1
- Creating M2V for VTS_02 segment 2
- Creating M2V for VTS_02 segment 3
- Creating M2V for VTS_02 segment 4
- Creating M2V for VTS_02 segment 5
- Creating M2V for VTS_02 segment 6
- Creating M2V for VTS_02 segment 7
- Creating M2V for VTS_02 segment 8
- Creating M2V for VTS_02 segment 9
- Creating M2V for VTS_02 segment 10
- Creating M2V for VTS_02 segment 11
- Creating M2V for VTS_02 segment 12
- Creating M2V for VTS_02 segment 13
- Creating M2V for VTS_02 segment 14
- Creating M2V for VTS_02 segment 15
- Creating M2V for VTS_02 segment 16
- Creating M2V for VTS_03 segment 0
- Creating M2V for VTS_03 segment 1
- Creating M2V for VTS_03 segment 2
- Creating M2V for VTS_03 segment 3
- Creating M2V for VTS_03 segment 4
- Creating M2V for VTS_03 segment 5
- Creating M2V for VTS_03 segment 6
- Creating M2V for VTS_03 segment 7
- Creating M2V for VTS_03 segment 8
- Creating M2V for VTS_03 segment 9
- Creating M2V for VTS_03 segment 10
- Creating M2V for VTS_03 segment 11
- Creating M2V for VTS_03 segment 12
- Creating M2V for VTS_03 segment 13
- Creating M2V for VTS_03 segment 14
- Creating M2V for VTS_03 segment 15
- Creating M2V for VTS_03 segment 16
- Creating M2V for VTS_03 segment 17
- Creating M2V for VTS_04 segment 0
- Creating M2V for VTS_04 segment 1
- Creating M2V for VTS_04 segment 2
- Creating M2V for VTS_04 segment 3
- Creating M2V for VTS_04 segment 4
- Creating M2V for VTS_04 segment 5
- Creating M2V for VTS_04 segment 6
- Creating M2V for VTS_04 segment 7
- Creating M2V for VTS_04 segment 8
- Creating M2V for VTS_04 segment 9
- Creating M2V for VTS_04 segment 10
- Creating M2V for VTS_04 segment 11
- Creating M2V for VTS_04 segment 12
- Creating M2V for VTS_04 segment 13
- Creating M2V for VTS_04 segment 14
- Creating M2V for VTS_04 segment 15
- Creating M2V for VTS_04 segment 16
- Creating M2V for VTS_04 segment 17
- Creating M2V for VTS_04 segment 18
- Creating M2V for VTS_04 segment 19
- Creating M2V for VTS_04 segment 20
[18:56:54] Phase II ENCODING completed in 321 minutes.
[18:56:54] Phase III, REBUILD started.
- Copying IFO, BUP, and unaltered files...
- Processing VTS_01
- Rebuilding segment 0 VOBID: 1 CELLID: 1
- Rebuilding segment 1 VOBID: 1 CELLID: 2
- Rebuilding segment 2 VOBID: 1 CELLID: 3
- Rebuilding segment 3 VOBID: 1 CELLID: 4
- Rebuilding segment 4 VOBID: 1 CELLID: 4
- Rebuilding segment 5 VOBID: 1 CELLID: 5
- Rebuilding segment 6 VOBID: 1 CELLID: 6
- Rebuilding segment 7 VOBID: 1 CELLID: 7
- Rebuilding segment 8 VOBID: 1 CELLID: 8
- Rebuilding segment 9 VOBID: 1 CELLID: 9
- Rebuilding segment 10 VOBID: 1 CELLID: 10
- Rebuilding segment 11 VOBID: 1 CELLID: 11
- Rebuilding segment 12 VOBID: 1 CELLID: 12
- Rebuilding segment 13 VOBID: 1 CELLID: 13
- Rebuilding segment 14 VOBID: 1 CELLID: 14
- Rebuilding segment 15 VOBID: 1 CELLID: 15
- Rebuilding segment 16 VOBID: 1 CELLID: 16
- Updating NAVPACKS for VOBID_01
- Rebuilding segment 17 VOBID: 2 CELLID: 1
- Updating NAVPACKS for VOBID_02
- WARNING: VOB Count Old: 1 New: 2
- Updated VTS_C_ADT.
- Updated VTS_VOBU_ADMAP.
- Updated IFO: VTS_01_0.IFO
- Processing VTS_02
- Rebuilding segment 0 VOBID: 1 CELLID: 1
- Rebuilding segment 1 VOBID: 1 CELLID: 2
- Rebuilding segment 2 VOBID: 1 CELLID: 3
- Rebuilding segment 3 VOBID: 1 CELLID: 4
- Rebuilding segment 4 VOBID: 1 CELLID: 5
- Rebuilding segment 5 VOBID: 1 CELLID: 6
- Rebuilding segment 6 VOBID: 1 CELLID: 7
- Rebuilding segment 7 VOBID: 1 CELLID: 8
- Rebuilding segment 8 VOBID: 1 CELLID: 9
- Rebuilding segment 9 VOBID: 1 CELLID: 10
- Rebuilding segment 10 VOBID: 1 CELLID: 11
- Rebuilding segment 11 VOBID: 1 CELLID: 12
- Rebuilding segment 12 VOBID: 1 CELLID: 13
- Rebuilding segment 13 VOBID: 1 CELLID: 14
- Rebuilding segment 14 VOBID: 1 CELLID: 15
- Rebuilding segment 15 VOBID: 1 CELLID: 16
- Updating NAVPACKS for VOBID_01
- Rebuilding segment 16 VOBID: 2 CELLID: 1
- Updating NAVPACKS for VOBID_02
- WARNING: VOB Count Old: 1 New: 2
- Updated VTS_C_ADT.
- Updated VTS_VOBU_ADMAP.
- Updated IFO: VTS_02_0.IFO
- Processing VTS_03
- Rebuilding segment 0 VOBID: 1 CELLID: 1
- Rebuilding segment 1 VOBID: 1 CELLID: 2
- Rebuilding segment 2 VOBID: 1 CELLID: 3
- Rebuilding segment 3 VOBID: 1 CELLID: 4
- Rebuilding segment 4 VOBID: 1 CELLID: 5
- Rebuilding segment 5 VOBID: 1 CELLID: 6
- Rebuilding segment 6 VOBID: 1 CELLID: 7
- Rebuilding segment 7 VOBID: 1 CELLID: 8
- Rebuilding segment 8 VOBID: 1 CELLID: 9
- Rebuilding segment 9 VOBID: 1 CELLID: 10
- Rebuilding segment 10 VOBID: 1 CELLID: 11
- Rebuilding segment 11 VOBID: 1 CELLID: 12
- Rebuilding segment 12 VOBID: 1 CELLID: 13
- Rebuilding segment 13 VOBID: 1 CELLID: 14
- Rebuilding segment 14 VOBID: 1 CELLID: 15
- Rebuilding segment 15 VOBID: 1 CELLID: 16
- Rebuilding segment 16 VOBID: 1 CELLID: 17
- Updating NAVPACKS for VOBID_01
- Rebuilding segment 17 VOBID: 2 CELLID: 1
- Updating NAVPACKS for VOBID_02
- WARNING: VOB Count Old: 1 New: 2
- Updated VTS_C_ADT.
- Updated VTS_VOBU_ADMAP.
- Updated IFO: VTS_03_0.IFO
- Processing VTS_04
- Rebuilding segment 0 VOBID: 1 CELLID: 1
- Rebuilding segment 1 VOBID: 1 CELLID: 2
- Rebuilding segment 2 VOBID: 1 CELLID: 2
- Rebuilding segment 3 VOBID: 1 CELLID: 2
- Rebuilding segment 4 VOBID: 1 CELLID: 2
- Rebuilding segment 5 VOBID: 1 CELLID: 3
- Rebuilding segment 6 VOBID: 1 CELLID: 4
- Rebuilding segment 7 VOBID: 1 CELLID: 5
- Rebuilding segment 8 VOBID: 1 CELLID: 6
- Rebuilding segment 9 VOBID: 1 CELLID: 7
- Rebuilding segment 10 VOBID: 1 CELLID: 8
- Rebuilding segment 11 VOBID: 1 CELLID: 9
- Rebuilding segment 12 VOBID: 1 CELLID: 10
- Rebuilding segment 13 VOBID: 1 CELLID: 11
- Rebuilding segment 14 VOBID: 1 CELLID: 12
- Rebuilding segment 15 VOBID: 1 CELLID: 13
- Rebuilding segment 16 VOBID: 1 CELLID: 14
- Rebuilding segment 17 VOBID: 1 CELLID: 15
- Rebuilding segment 18 VOBID: 1 CELLID: 16
- Rebuilding segment 19 VOBID: 1 CELLID: 17
- Updating NAVPACKS for VOBID_01
- Rebuilding segment 20 VOBID: 2 CELLID: 1
- Updating NAVPACKS for VOBID_02
- WARNING: VOB Count Old: 1 New: 2
- Updated VTS_C_ADT.
- Updated VTS_VOBU_ADMAP.
- Updated IFO: VTS_04_0.IFO
Correcting VTS Sectors...
[19:13:29] Phase III, REBUILD completed in 17 minutes.

Done.


First impression:
1. Wow! Can't beleive how long it took! Encoding 321 minutes.
On my computer vs. the other discs that I backed up with CCE, it took 3 times longer to do! I have a pretty fast computer and I would be affraid to see how much time it would take on a say 2.0GHZ comp! With that being said it better be worth it. IMO thats "WAY" too long!

2. Wow! "LESS" MACRO BLOCKS! (This to me is very impressive!) This I'm truly surprised! There are less macroblocks vs. CCE backups! There's alot of fast motion scenes in "24" and I see less macro blocks with HC encoding! Trully impressive to me.
When I looked at previous backups with CCE, You could see more macro blocks in the fast motion scenes.

3. WOW! Quality looks pretty good! I'm very impressed with the picture quality! I mean looking at it on a frame per frame basis against the original source was pretty astonishing to me. The quality to me is very good and very close to the original source! (Man please don't flame me! This is JUST my opinion)

4. Wow! (Purple bars!!!) I can't beleive it's doing purple lines/bars at the bottom right of the picture! (This puzzles me and is unacceptable for me!) I haven't got the slightest clue why this happened or is happening! NEVER seen this before. Is this a bug?

I back up all my movies with CCE SP 2.67 with the following settings:
VBR_Bias =20
Quality Prec =16
VBR_Passes =2

Please take this all with a grain of salt...this is JUST my humble opnion. Thats all!

I will eventually burn this copy and look at the final result on my 7-foot wide screen (NEC HT1000 projector) and will give my opnion again. To me this is a better test! I will however do some test's in fast action scenes on a frame per frame analysis to see more detailed info. I'll let you know what MY EYES see.

Right now I have other things to attend to but will get back to you all if anyone cares.


Sincerest regards,
A Rookie!


:)

[jdobbs]

Thanks for the look! I noticed there is a new version out now that (I believe) fixes the purple bars.

[dragongodz]

I noticed there is a new version out now that (I believe) fixes the purple bars.
yes but it should be noted this is the gui version that has the fix. hank315 does however say
Still working on a new batch version which will have all options of this latest GUI version.
so you will have to wait a while to use a fixed version with DVD-RB.

[TheSeeker]

Aside from the fact that I have to kick off each segment manually right now Im rather impressed with HC as an Encoder. The output video looks darn nice, I love the easy configurability (just paste options into .ini) I also like that you can specify custom quanti mats. (again in the .ini). Its all so sleek and simple and the quality looks darn nice. I havent noticed the purple bars on the batch version but maybe I missed them somehow? though i dont see how thats possible. Once I get this to work with DVD RB, I may have a nice alternative to CCE, and Procoder. Im anxious to see how it does with heavily compressed anime. As that is a nightmare for any encoder.

[manolito]

The thing with with the colored bars on one side of the encode was discovered by ebotron (see the HC thread under "Other MPEG encoders"). It seems that it is not present in PAL and also depends on the frame resolution. I never had this problem, but I live in PAL country of course.

If jdobbs implementation for using HC does not work for you, there still is my little HC_CLI utility which also works with earlier versions of DVD-RB. Have a look here:
http://forum.doom9.org/showthread.php?s=&threadid=89440&perpage=20&pagenumber=1

Like jdobbs' implementation HC_CLI also does not work with the GUI versions of HC. You need the batch enabled "non-GUI" version.


Cheers
manolito

[TheSeeker]

@Manolito

I will give your utility a try. What is the newest version of RB that works with it?

[manolito]

@TheSeeker
HC_CLI works with any application that supports QuEnc. So even ancient versions of RB should work.

All that HC_CLI does is intercept the QuEnc command line arguments, pick only the ones that are essential, convert the format as required, write a new HC.INI file and call HC.EXE. Nothing fancy...

Just keep in mind that HC_CLI was written before Hank315 published the GUI versions of HC, so these versions (which do not support batch encodes or templates) will not work.

Cheers
manolito

[TheSeeker]

Sweet. It looks like that should do the trick. I will give 'er a try when I get home from work. I will be using the newest RB (0.78) can I expect this to work fine with hc_cli as long as i make sure and use it in QuEnc mode?

[DD51]

Originally posted by DD51
I will eventually burn this copy and look at the final result on my 7-foot wide screen (NEC HT1000 projector) and will give my opnion again. To me this is a better test! I will however do some test's in fast action scenes on a frame per frame analysis to see more detailed info. I'll let you know what MY EYES see.:)

Well I did burn it and have been playing with the end result on my big screen and "must say" for a free encoder the picture quality is very very good (Awesome). I have no complaints aside the purple lines that it did at the bottom right! I had alot of difficulty seeing any macroblocks anywhere! Even looking at some fire scenes there were NO blocks! Again this is just me but the picture quality is Awesome.

My one BIG complaint is how slow the encoding process is! For me this won't do it for me. The speed of encoding is unacceptable to me.
If futur versions increase the speed, then this is a true winner for me!
I'll be waiting for futur updates and will keep a close eye on this little program. This is the best freeware encoder I have ever scene soo far!
Congrats to the author.

Sincerest regards,

[TheSeeker]

@DD51

What kind of processor do you have? Because I think that you can get a very nice speed boost if your Proc. support SSE2 or SSE3 instructions. But if you have a little bit older AMD then your proc supports neither and you could see some slow speeds.. But on my AMD64 the speeds werent all that bad. I mean they werent as good as CCE but I wouldnt say they were super slow either. I think Procoder is slower than HC.

[onesoul]

Hi,

At the early stages of DVD-RB I posted here about the "Robin Williams Live on Broadway" Pal Interlaced DVD backup, then I used CCE 2.67.00.27 and the outcome was awful (average bitrate of ~4300 kbs), after many different settings (bias, IQP, matrixes, etc, I think the only I thing I didn't tried was encoding it as progressive but I doubt it would be any useful), I had just given up trying backing up interlaced content in CCE. Even transcoding programs like Dvdshrink or recode are better with interlaced content. (Just have to add that encoded progressive content with CCE looks great)

I read procoder was very good with interlaced content but I didn't try it yet (going to sometime). I didn't try quenc yet too but I am eager to try it also (0.54 still produces better quality but less accurate on refraining high bitrate, right?)

Now I tried dvd-rb 0.78 with HC to backup the same DVD and from my early impressions I couldn't spot differences between the HC encoded and original. I am so amazingly surprised with the hank315's HC encoder, it may well be my all purpose encoder. I will still do some frame to frame comparison just for the sake of it. The encoding took 323 minutes but I don't mind waiting for such a good output (It's done over the night anyway).

The average bitrate was again ~4300kbs, chose HC best quality and matrix was HC default. All other settings were chose by the mighty jdobbs's DVD-RB, (interlaced, gop, etc).

I put a couple of questions though:
- The close all gops setting is not chosen, when does it need to be applied?
- GOP settings were 12 2, pal interlaced content could benefit with gop 15? (as default interlaced NTSC setting by DVD-RB), what about autogop?
- It's good to have different matrix options but the problem is I don't know when to use them, is one of them better with interlaced content? If so, tell me please :)

Cheers

[edeus]

I dont understand, wouldnt macroblocks be not visible in CCE anyway?

Using HC, is it very developed or in a testing phase? It sounds so promising but CCE gives such good quality already, i want a reason to go to HC.

If DVD-RB recommends it maybe I'll go for it then :)

Btw could it have been that the encoders way back were crap compared to todays ones, and so when we recompress we are noticing far superior results. When we try newer dvd's, it is less impressive?

[jdobbs]

I'm certainly not suggesting anyone switch from CCE to HC. I just think it is always good to get a range of opinions on a new encoder. The reason DVD-RB has multiple encoders listed is so people can choose...

[DD51]

Originally posted by TheSeeker
@DD51

What kind of processor do you have? Because I think that you can get a very nice speed boost if your Proc. support SSE2 or SSE3 instructions. But if you have a little bit older AMD then your proc supports neither and you could see some slow speeds.. But on my AMD64 the speeds werent all that bad. I mean they werent as good as CCE but I wouldnt say they were super slow either. I think Procoder is slower than HC.

@TheSeeker,
Here are my Computer specs.

P4C800-E Deluxe - MB Version, 2.00
Bios Version 1021 beta6
P4 3.0C GHz
OC'd to 3.554GHZ 948 FSB
Kingston memory kit: PC3200 @ 474MHZ Dual CH enabled
2 Sata drives 120GB each
Video Card: MSI 6800GT 256MB
Operating System : Win XP Pro SP2
Instructions: MMX, SSE, SSE2 (I don't have SSE3)

Yes I love to overclock! It makes a big difference.

All I can tell you is that I redid the Disc 5 with CCE SP 2.67 and
it took 104 minutes to encode vs. 321 minutes with HC. BUT! The quality of HC is better IMHO.



Update: After redoing My disc 5 with CCE...IMHO HC looks better! HC's encode produced "LESS" macroblocks compared to CCE encode.

[onesoul]

After hearing some opinions about cce and interlaced content I decided to make more tests, well the output isn't so bad as I have said before, I am not sure what went wrong on my earlier tests. I read on dvd-rb file that the handling of ecl file with 2.67.00.27 was fixed at 0.60. On my earlier tests I used an older version, I have no idea if could be it, but I am suspecting that the bad output was due to some filter.

I apologize for providing bad information before.

Now I have done more testing with no filtering at all! But actually I think still cce wasn't the best. I am going to upload some m2v files somewhere later.

@jdobbs
Is the latest Quenc 0.59 b2 suported? Could you check if settings are getting selected right? I noticed a problem when you select qlb matrix. Also on mpeg2 settings of quenc it isnt checked top field first but that I suspect it is the way it should be?
And I noticed the dc precision is set at 8, on other encoders dvd-rb automatically sets to 9, why?

[jdobbs]

Yes QuEnc 0.59B2 is supported. I'll take a look at the qlb problem. I wouldn't worry about the top field setting. Since the file is being reencoded that really doesn't matter.

[thomasphoenix]

Hi Jdobbs,
I tested the Hc encoder , the version u recommended,on Spiderman2 at 65% compression,at quality setting good,I also encoded with cce 2.70.02 ,with OPV.
I didnt get any purple corners everything was ok,And to answer many doubts about the speed ,you have to disable hyperthreading for HC to encode with full cpu load otherwise it uses only one thread of the multithread ie 50%.so expect to gain almost 40% speed with hyperthread disabled.The Quality is extremely good,It compares very well with CCE,Though I will not say its better,in some small things, maybe CCE seems to be a teeny weeny bit sharper,(or maybe im imagining it ,looking too hard).

[archaeo]

it would be nice to see some side by side image captures...

[hank315]

Just some remarks...

It's good to hear people seem to like the quality of HC, some don't like the speed.
Goal was to backup a DVD in good quality, even a 3 hour movie, in about 2 * real time using my own hardware using the best settings.
I'm using a P4 3.2 GHz, Prescott core which can do SSE3 and it can do a 3 hour movie in 6 hours.
In future releases I will try to speed it up while keeping the same quality, just don't know yet if that's possible.
The speed gain using SSE2 instead of MMX is just about 5%, using SSE3 may speed it up by approx. 20% over SSE2, SSE3 is able to load unaligned memory very fast.

Development isn't over yet, still have a lot of ideas of improving it, especially at low bitrates.

About the flashing color error, it will show if (hor.res * vert.res)/1024 isn't a whole number.

ATM I'm finishing a new batch version which will take arguments but can also use the ini file.
The encoding engine will be exactly the same as the latest GUI version.

Also did some testing:
1) LOTR - Fellowship of the Ring - PAL 720*576
encoding time 5.5 hours, final DVD size 4.28 GB

2) Matrix - Reloaded - PAL 720*576
encoding time 5 hours, final DVD size 4.29 GB

All done with *PROFILE BEST
And the final quality, just what the profile says :D :D

[TheSeeker]

Just wondering.... Any plan to support 3DNow! optimizations for AMD users?

[jdobbs]

About the flashing color error, it will show if (hor.res * vert.res)/1024 isn't a whole number. That's really a problem for NTSC -- since it's 720 x 480. Can it be fixed?

[hank315]

That's really a problem for NTSC -- since it's 720 x 480. Can it be fixed?It's already fixed in the new 0.12 GUI version.
It is still in the old CLI version but will ofcourse also be fixed in the new batch version, was just a stupid programming error :mad:

[jdobbs]

Well, just read a few of the pages on this forum and you'll see I know that feeling... :eek:

[quantum]

I haven't posted in ages, but I feel compelled to for this.

Keep up the great work Hank. Getting results that are competitive with CCE is truly an achievement. Going even slightly farther is more than an eyebrow raiser. I'll be looking forward to the continuing development.

I'll throw my 2 cents in and say do what you need to do with regards to time. I'll choose a slower encode for better results every time. If someone wants it faster, then CCE or DVDShrink is available.

[dragongodz]

I'll choose a slower encode for better results every time.
i agree. infact HC's second pass may be slightly slower in the future ,depending on things i wont discuss here, for the sake of possibly consistantly better quality, will only happen if hank315 finds the time to implament it and if it proves better of course. maybe as an option even. there is so much going on behind the scenes to try and improve this encoder that people do not know. :D

[jdobbs]

Originally posted by quantum
I haven't posted in ages, but I feel compelled to for this. Hey! quantum! Dude. Glad to see you're still alive and kickin'

[thomasphoenix]

Dear Hank,
Great Job,Can u add hyperthread support ,since I can test CCe and Hc without changing bios settings each time,like im doing now.

[DD51]

Originally posted by hank315
It's already fixed in the new 0.12 GUI version.
It is still in the old CLI version but will ofcourse also be fixed in the new batch version, was just a stupid programming error :mad:

Hank315,
I can't wait to try the fixed batch version. May I ask when will it be available? Congrats on your new project. I'm impressed with the quality!

About the speed issue.
Please this is JMO!

What I don't like about slow encoders is that they beat the shxx out of the CPU's. These encoders do quit a number on processors. They push CPU's to the MAX! I don't think running a CPU at 100% for 5 hours + straight on several occassions can be good. That's my worry about slow encoders!:confused:
Just my 2 cents worth.:confused:

[onesoul]

Nice to hear such goods things about HC which is already great!

I said before I would put m2v files from my tests, I din't yet, I was hoping someone could show some webspace where I could send around 70mb, my isp provider doesn't offer me enough space.

But I have put a screenshot comparison online here (http://pwp.netcabo.pt/onesoul/enctest.html), it will take some time to load though (around 4mb).

I gave some opinions there. Feel free to do the same (I am a just a noob speaking) :)

@jdobbs
From the tests I have made, quenc seemed to me to produce better results (at least with this setup) with dc precision of 9 (over the default 8).

[onesoul]

Originally posted by DD51
What I don't like about slow encoders is that they beat the shxx out of the CPU's. These encoders do quit a number on processors. They push CPU's to the MAX! I don't think running a CPU at 100% for 5 hours + straight on several occassions can be good. That's my worry about slow encoders!:confused: Lol, are u kiddin? CPU's are made to live at max 24/7, of course it needs to have good heatsink, and a good power supply along a ups helps too.

Just to you know, my computer is always 24/7 on at 100% load running united devices and prime95 (from mersenne.org).

[jdobbs]

Originally posted by onesoul
@jdobbs
From the tests I have made, quenc seemed to me to produce better results (at least with this setup) with dc precision of 9 (over the default 8). Hmmm... currently I'm not forcing dc precision to any value -- just leaving it at the default. Maybe I should add it as a choice on the QuEnc setup.

[DK]

having done some conversion before with hc_enc i used the latest rb version now with two titles so far and the result is:

awesome ;)

keep up the great work, hanc!



one more time the gop length question came to my mind, when looking at the ini files

hc states 15, rb seems to set 12 again

will a change to the rb.ini after preparation work?

[jdobbs]

You can change the rebuilder.ecl file after prep. An HC.INI is generated from that during encode. But as always, I recommend 12 for PAL, 12 for NTSC/FILM, and 15 only for NTSC interlaced. I believe that's also the way the studios do it most of the time.

[dragongodz]

What I don't like about slow encoders is that they beat the shxx out of the CPU's. These encoders do quit a number on processors. They push CPU's to the MAX! I don't think running a CPU at 100% for 5 hours + straight on several occassions can be good. That's my worry about slow encoders!
sorry but i find that amussing after you have said
P4 3.0C GHz
OC'd to 3.554GHZ 948 FSB
Yes I love to overclock! It makes a big difference.
and you dont think you are stressing your cpu there ? :D

also did you turn off HT to compare the speed as thomasphoenix suggested ?

[onesoul]

Originally posted by onesoul
I said before I would put m2v files from my tests, I din't yet, I was hoping someone could show some webspace where I could send around 70mb, my isp provider doesn't offer me enough space.

But I have put a screenshot comparison online here (http://pwp.netcabo.pt/onesoul/enctest.html), it will take some time to load though (around 4mb).
I have managed to put on my webspace the original m2v file:

http://pwp.netcabo.pt/onesoul/enctest/original.zip (~22mb)

I can try put the encoded files online too, if it's needed.

[manolito]

Originally posted by jdobbs
You can change the rebuilder.ecl file after prep. An HC.INI is generated from that during encode. But as always, I recommend 12 for PAL, 12 for NTSC/FILM, and 15 only for NTSC interlaced. I believe that's also the way the studios do it most of the time.
Well, different users may have different needs. For very low bitrates a longer GOP will improve compressibility.

Also HC has an "AutoGOP" mode which is supposed to make some intelligent assumptions about the optimal GOP length and adjust it accordingly. But if RB always issues a GOP command, this AutoGOP mode will be overridden. IMO Rebuilder should support this Auto mode, too.

Cheers
manolito

[jdobbs]

Noted. Unfortunately most users of DVD-RB don't know (or care) what a GOP. I'm trying to avoid adding more-and-more-and-more options that confuse people. It's getting hard... my fear is that the "One Click" folks for whom it is primarily written are being overruled by the experts. I don't want to make DVD-RB as complicated as some of the other CCE methods...

Example: Have you ever used one of the default FFMPEG encoders? Most people haven't. It's mainly because the 9 million options are intimidating.