HC encoder 0.21 released

[hank315]

A new release is out: HC021

- added SSSE3 support
- added some new functions: *LUMGAIN, *SILENT and *SCDFILE
- higher encoding speed for best profile
- as usual some bug fixes


*NOTES
Normal profile using MMX/SSE was broken, is fixed now.
Don't expect too much from the SSSE3 support, speed gain is only marginal.

[canuckerfan]

great! thanks:)

[odyn1982]

Thank you! HC is my favourite/best mpeg2 encoder :D

[~bT~]

good stuff! cheers hank. :)

[Blue_MiSfit]

Thanks Hank! The best just got better!

~MiSfit

[Guest]

Can I join the chorus, please? Thank you, Hank!

[Rippraff]

Me too: Hup Holland, hup. Bedankt, Hank! :)

Cu Rippraff

[Digga]

Can I join the chorus, please?only if I'm also allowed to ;-)
thank you for a great and freely available MPEG-2 encoder!

[sh0dan]

Thanks for your incredible useful tool!

[buzzqw]

:thanks: :thanks: :thanks:

thanks! Hank!!!

BHH

[sockeye]

You're awsome Hank! :thanks:

[techreactor]

Thanks, excellent!!!

[Fishman0919]

At the risk of sounding a little repetitive... ;)


Thank You for your wonderful tool and all your work!

[tom942]

Just say the same thing..., thanks!!! :)

[45tripp]

I'm using HC more and more :thanks:

[Trahald]

excellent. will begin using immediately. thanks hank

[Malcolm]

@hank315
First: Thanks a lot for your hard work!

i didn't had the time to play around with the 0.21 version but i'm curious what 'lumgain' is good for.
This command adapts the quantization matrices on dark scenes so the bitrate will be raised.
Does this prevent blocking in dark areas?
Can you please explain?

greetings,
Malcolm

[halsboss]

Thankyou hank315. Have been using HC for years for home video and it's EXCELLENT. You are a champion person.

[AlanHK]

I installed v.21 and found a problem.

HC crashed and exited a few seconds after starting an encode of a particular file.

I tried the same script with v.20 and it worked, but on the "info" line it reported "Avisynth error in frame 59" (and earlier frames) which is apparently what choked v.21.

Looking at the file a new scene begins on frame 60, possibly a join, but otherwise I can't see any problems; VirtualDub, MPC Classic have no problems with the raw avi or the avs file.

I did a frame scan of the avi with VirtualDub and it found nothing wrong.

After Trimming the first 59 frames in the AVS, v.21 ran.

There must be some minor glitch with the AVI, but HC is being a little oversensitive I think.

[hank315]

@Malcolm
This is how it works:
If the average luminance is small the matrix values (Intra and Inter) are lowered.

Example, in a normal scene the Intra matrix could be:
08 16 19 22 26 27 29 34
16 16 22 24 27 29 34 37
19 22 26 27 29 34 34 38
22 22 26 27 29 34 37 40
22 26 27 29 32 35 40 48
26 27 29 32 35 40 48 58
26 27 29 34 38 46 56 69
27 29 35 38 46 56 69 83

In a dark scene it may be something like this:
08 11 13 15 17 18 19 23
11 11 15 16 18 19 23 25
13 15 17 18 19 23 23 25
15 15 17 18 19 23 25 27
15 17 18 19 21 23 27 32
17 18 19 21 23 27 32 39
17 18 19 23 25 31 38 46
18 19 23 25 31 38 46 56

The luminance check/matrix change can be at every GOP, the matrix change will depend on the average luminance per GOP and the sensitivity of the *LUMGAIN command.
It might help to prevent blocking in low lit scenes and fades.

@AlanHK
The "Avisynth error in frame xxx" means Avisynth couldn't serve the requested frame, the returned data is unpredictable.
If you can post the file with the first 100-200 frames I can have a closer look to see what goes wrong.

[techreactor]

@Malcolm
This is how it works:
If the average luminance is small the matrix values (Intra and Inter) are lowered.

The luminance check/matrix change can be at every GOP, the matrix change will depend on the average luminance per GOP and the sensitivity of the *LUMGAIN command.
It might help to prevent blocking in low lit scenes and fades.

@AlanHK
The "Avisynth error in frame xxx" means Avisynth couldn't serve the requested frame, the returned data is unpredictable.
If you can post the file with the first 100-200 frames I can have a closer look to see what goes wrong.

Can this not be made automatic, wherein the luminance is checked before the encode and then based on the values, a suitable *lumgain parameter is invoked automatically.

[Malcolm]

@Hank315
Thank you for your explanation. it's clearer to me now. so by lowering the matrix values (divisors) you produce less quantized frequency values in low luminance gops. but this means also that you need more bits to encode (huffmann, lzw) these gops, right? (since the resulting quantized values are bigger). So in turn you have to steal them elsewhere!? Does this happen automatically by the vbr algorithm over the whole length of the video? Or does the huffmann/lzw encoding handle the bigger quantized values good enough, so that you can (sloppy said) use additional n bits for low luminace gops but just have to use <n bits less for other parts?
If this would be true, then this would mean you can user lower matrix values for the whole movie. so there must be an error in reasoning with me...

from the practical side of view: i've managed to do some first tests yesterday (had to build up my new Core2Duo PC first). My findings: in nearly all cases (except some single frames) the encodings with lumgain=4 were better that the comparative encodings with lumgain=0! i tested low, medium and high bitrates, as well as clean videos and noisy videos.
Conclusion for me: Always leave lumgain at 4!

P.S. HCEnc rocks! Thank you for your hard work!!

side note @Hank315
I just wanted to point out that i have some issues with MT() and HCEnc. See this thread (http://forum.doom9.org/showthread.php?t=125550) for more if you would like to comment on it.

Also: i think HCEnc counts the processing time 'wrong' for smp processing: If avisynth processes some filters in parallel using mt("filter to run parallel"), HCEnc sums the time for each thread as if they would run in serial, but in reality they run in parallel. So HCEnc really counts the time all cores were processing, but not the time the user has to wait until the encode is finished.

[Boulder]

Wouldn't lumgain=4 be a sort of adaptive quantization equivalent (compared to what XviD uses) since it steals bits from the dark scenes.. I'll try to find a good sample clip to test it on, something that has both dark and bright, detailed scenes to see how it performs.

[techreactor]

Wouldn't lumgain=4 be a sort of adaptive quantization equivalent (compared to what XviD uses) since it steals bits from the dark scenes.. I'll try to find a good sample clip to test it on, something that has both dark and bright, detailed scenes to see how it performs.

It will be interesting to know the results, please keep us posted on your tests. :thanks:

[gizzin]

I've noticed after multiple encodes your encoder still has problems with fade in or out scenes. They are still blocky even at higher bitrates 3000 or more. I've noticed it only seems the with flat areas that are in the fade in and out scenes. Also maybe even flat areas generally but not as noticeable. I fiddled with the vbr but it doesn't seem to make much of a difference. Just thought I'd point that out to you.

[Asmodeus]

Hi.

Thx for the next version, Hank.
But I have problem with runing it.

With the same settings as for 0.20
*INFILE d:\mpeg\some.avs
*OUTFILE d:\5\some.m2v
*FRAMES 0 60760
*PROFILE best
*ASPECT 4:3
*GOP 18 2
*CQ_MAXBITRATE 3.000
*CQ_PFACTOR 1.50
*CQ_BFACTOR 3.00
*CLOSEDGOPS
*PULLDOWN
*LASTIFRAME
*PREVIEW
*MATRIX hvsbest

I get
ERROR, input not complete, required input: input file, output file, bitrate, max bitrate.
and program of course exits.

[hank315]

*MAXBITRATE is missing...

[Boulder]

Out of interest, what causes the "adjusting average bitrate..." message in the info window?

In my tests using lumgain=4, I noticed that HC actually uses matrices with lower coefficients than the one I use as the base (MPEG standard). I haven't done any serious comparisons yet. It would be interesting to hear if anyone has done so (with screenshots as well).

[hank315]

Out of interest, what causes the "adjusting average bitrate..." message in the info window?
After the first pass HC generates a bitrate curve which dictates how many bits have to be generated in the second pass for every GOP.
During the second pass, quantizers are estimated based on the bitrate curve, also an adaptive algorithm is used to correct for short-term deviations.
If this fails because it can't hold the bitrate (mostly because of too small quantizers or max. bitrate/VBV issues) the average bitrate will be lowered causing the message.
From that point on, the average bitrate will again be equal to the given average bitrate, the encoder will not try to insert the bits that couldn't be allocated in the previous part.

In my tests using lumgain=4, I noticed that HC actually uses matrices with lower coefficients than the one I use as the base (MPEG standard).
For those who are interested, here's the source which adapts the matrices using the LUMGAIN command:

subroutine luminance_gain(luminance)

use HCdata

implicit none

integer*4 luminance

! local var
real*4 b
real*4 lum1(0:64) &
/0.85,0.85,0.85,0.85,0.85,0.85,0.85,0.85,0.85,0.85,0.85,0.85,0.85,0.85,0.85,0.85,0.85, & ! 0 - 16
0.85,0.85,0.85,0.85,0.85,0.85,0.85,0.86,0.86,0.86,0.87,0.87,0.88,0.88,0.89,0.89, & ! 17 - 32
0.90,0.90,0.91,0.91,0.92,0.92,0.93,0.93,0.94,0.94,0.95,0.96,0.97,0.98,0.98,0.99, & ! 33 - 48
0.99,0.99,0.99,1.00,1.00,1.00,1.00,1.00,1.00,1.00,1.00,1.00,1.00,1.00,1.00,1.00/ ! 49 - 64
real*4 lum2(0:64) &
/0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75, & ! 0 - 16
0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.75,0.76,0.77,0.78,0.78,0.79,0.79, & ! 17 - 32
0.80,0.81,0.82,0.83,0.84,0.85,0.86,0.87,0.88,0.89,0.90,0.91,0.92,0.93,0.93,0.94, & ! 33 - 48
0.94,0.95,0.95,0.96,0.96,0.97,0.97,0.98,0.98,0.98,0.99,0.99,0.99,1.00,1.00,1.00/ ! 49 - 64
real*4 lum3(0:64) &
/0.65,0.65,0.65,0.65,0.65,0.65,0.65,0.65,0.65,0.65,0.65,0.65,0.65,0.65,0.65,0.65,0.65, & ! 0 - 16
0.65,0.65,0.65,0.65,0.65,0.65,0.65,0.65,0.66,0.67,0.68,0.69,0.70,0.71,0.72,0.73, & ! 17 - 32
0.74,0.75,0.76,0.77,0.78,0.79,0.80,0.81,0.82,0.83,0.84,0.85,0.86,0.87,0.88,0.89, & ! 33 - 48
0.90,0.91,0.92,0.93,0.94,0.95,0.96,0.96,0.97,0.97,0.98,0.98,0.99,0.99,1.00,1.00/ ! 49 - 64
real*4 lum4(0:64) &
/0.50,0.50,0.50,0.50,0.50,0.50,0.50,0.50,0.50,0.50,0.50,0.50,0.50,0.50,0.50,0.50,0.50, & ! 0 - 16
0.50,0.50,0.50,0.50,0.50,0.50,0.50,0.50,0.51,0.52,0.53,0.54,0.55,0.56,0.57,0.58, & ! 17 - 32
0.59,0.60,0.62,0.64,0.66,0.68,0.70,0.72,0.74,0.76,0.78,0.80,0.81,0.82,0.83,0.84, & ! 33 - 48
0.85,0.86,0.87,0.88,0.89,0.90,0.91,0.92,0.93,0.94,0.95,0.96,0.97,0.98,0.99,1.00/ ! 49 - 64

if (luminance > 64) return ! adapt matrix only for luminance <= 64

! select luminance change
select case(lumgain) ! select case based on *LUMGAIN setting
case(1); b=lum1(luminance) ! mild change
case(2); b=lum2(luminance) ! moderate change
case(3); b=lum3(luminance) ! heavy change
case(4); b=lum4(luminance) ! hit dark scenes real hard
end select

! adapt matrix
intra_matrix=NINT(b*intra_matrix) ! set to nearest integer for whole intra matrix
inter_matrix=NINT(b*inter_matrix) ! set to nearest integer for whole inter matrix

! clip to 'normal' values
intra_matrix=MAX(intra_matrix,8); intra_matrix=MIN(intra_matrix,255)
inter_matrix=MAX(inter_matrix,8); inter_matrix=MIN(inter_matrix,255)
intra_matrix(1)=8

end

[DrP]

Not strictly a HCENC post but related... when I run the output of HCENC 0.21 through project-x it identifies it as MPEG AUDIO and fails to process the video properly. PVAStrumento 2.1.0.17 just hangs.

[Boulder]

From that point on, the average bitrate will again be equal to the given average bitrate, the encoder will not try to insert the bits that couldn't be allocated in the previous part.Would it be reasonable to allocate it elsewhere? I just encoded First Blood part I and it tends to undersize as at certain scenes HC needs to lower the average bitrate probably due to the VBV buffer or max bitrate issues.

I'm using DVD-RB so the encoding is done per cell instead of the whole video in one go.

For those who are interested, here's the source which adapts the matrices using the LUMGAIN command:Thanks for the code snippet:) If I understand it correctly, the values in the table are multipliers? I actually thought that lumgain 1-3 would lower the coefficients for the low-luma frames and lumgain 4 would raise them greatly - at least that's what "hit dark scenes real hard" means to me ;)

An interesting thing is also the fact that based on a quick peek I'd estimate modified matrices being used 95% of the time. I need to doublecheck that to be more precise.

And one more thing: I've got some kids cartoon material which triggers a scene change very often, many GOPs end up 2-4 frames long. If you're interested, I can snip a sample to analyze. If you happen to know "Maisy Mouse", that's the one.

[Asmodeus]

*MAXBITRATE is missing...

Ok. It works now. But AFAIR at first I just used settings via GUI without any major changes, and it didn't worked, and now without any changes it's woking :? Is it possible that MAXBITRATE was grayed in GUI for some reason?

[DrP]

And one more thing: I've got some kids cartoon material which triggers a scene change very often, many GOPs end up 2-4 frames long. If you're interested, I can snip a sample to analyze. If you happen to know "Maisy Mouse", that's the one.

I've seen the same thing happen with Family Guy during the start of the show. Characters move their heads to and fro slightly and HCENC uses an I frame at each movement.

[Cunhambebe]

AlanHK:

I installed v.21 and found a problem.
HC crashed

Same here - The HC encoder previous version starts but it crashes as I try to encode from avs scripts (that work very well with VDMod and Xvid). The newest version doesn't even start. The application crashes on my OS (Windows XP SP1). I'd like to know what's wrong with it.

[Boulder]

Could you post the script that crashes, please. Also let us know what Avisynth build you use.

[Cunhambebe]

This is the avs script:
LoadPlugin("C:\unzipped\dgmpgdec148\DGDecode.dll")
LoadPlugin("C:\avs_filters\ColorMatrix\colormatrix.dll")
MPEG2Source("C:\brazil_intro.d2v").AudioDub(WavSource("C:\brazil_intro.wav"))
ColorMatrix(d2v="brazil_intro.d2v")
Sharpen(0.2)
Lanczos4Resize(720,480) # I've decided to left this...
FadeOut(5)
a=last

LoadPlugin("C:\unzipped\dgmpgdec148\DGDecode.dll")
LoadPlugin("c:\avs_filters\TomsMoComp\TomsMoComp_Avisynth\Debug\TomsMoComp.dll")
LoadPlugin("C:\avs_filters\ColorMatrix\colormatrix.dll")

MPEG2Source("C:\brazil_selvagem.d2v").AudioDub(WavSource("C:\brazil_selvagem.wav"))
ColorMatrix(d2v="brazil_selvagem.d2v")

Trim(424,690) + \
Trim(29544,29594) + \
Trim(29775,30862) + \
Trim(695,2231) + \
Trim(2475,2682) + \
Trim(3037,3138) + \
Trim(3303,4796) + \
Trim(4962,5595) + \
Trim(5638,5886) + \
Trim(6035,6132) + \
Trim(6530,6605) + \
Trim(6225,6258) + \
Trim(6333,6472) + \
Trim(6609,6964) + \
Trim(7010,7858) + \
Trim(7935,7975) + \
Trim(8209,8637) + \
Trim(8659,8763) + \
Trim(9168,9250) + \
Trim(10406,10619) + \
Trim(10911,11146) + \
Trim(11204,11983) + \
Trim(12233,12526) + \
Trim(12746,13276) + \
Trim(13422,15204) + \
Trim(15452,15662) + \
Trim(15782,16163) + \
Trim(16227,16958) + \
Trim(16982,17073) + \
Trim(17203,17274) + \
Trim(17425,18033) + \
Trim(18396,19412) + \
Trim(19426,19632) + \
Trim(19636,19796) + \
Trim(19845,20386) + \
Trim(20390,21004) + \
Trim(21064,21269) + \
Trim(21370,21436) + \
Trim(21748,21846) + \
Trim(26380,26493) + \
Trim(22062,22283) + \
Trim(22287,22456) + \
Trim(22511,22611) + \
Trim(22622,23005) + \
Trim(23150,23183) + \
Trim(23347,24745) + \
Trim(24938,25539) + \
Trim(25646,25880) + \
Trim(26498,26985) + \
Trim(26990,27789) + \
Trim(27808,28162) + \
Trim(43875,44673) + \
Trim(55187,55237) + \
Trim(52011,52188) + \
Trim(57633,57881) + \
Trim(55133,55163) + \
Trim(59030,59310) + \
Trim(55805,57478) + \
Trim(57912,59020) + \
Trim(59431,59518) + \
Trim(28167,28473) + \
Trim(59523,59591) + \
Trim(28478,28538) + \
Trim(29549,29590) + \
Trim(28682,29468) + \
Trim(30866,42946) + \
Trim(43016,43272) + \
Trim(44355,44490) + \
Trim(52281,52926) + \
Trim(53000,53350) + \
Trim(53676,53995) + \
Trim(54261,54665) + \
Trim(54714,54892) + \
Trim(59595,59950) + \
Trim(60006,60442) + \
Trim(60546,63488) + \
Trim(63508,63902) + \
Trim(63916,65526) + \
Trim(65572,67100) + \
Trim(79036,80072) + \
Trim(67108,69033) + \
Trim(69172,69237) + \
Trim(69244,70510) + \
Trim(70516,74042) + \
Trim(74098,76315) + \
Trim(76325,78964) + \
Trim(80099,80155) + \
Trim(44898,52003) + \
Trim(43330,43462) + \
Trim(54930,54981) + \
Trim(90,337) + \
Trim(43467,43660)

FadeIn(35)
TomsMoComp(1,5,1) # video is TFF
Sharpen(0.2)
Crop(8,4,-8,-4)
Lanczos4Resize(720,480)
FadeOut(145)
b=last

LoadPlugin("C:\unzipped\dgmpgdec148\DGDecode.dll")
LoadPlugin("C:\avs_filters\ColorMatrix\colormatrix.dll")
MPEG2Source("C:\brazil_credit_roll.d2v").AudioDub(WavSource("C:\brazil_credit_roll.wav"))
ColorMatrix(d2v="brazil_credit_roll.d2v")
Sharpen(0.2)
Lanczos4Resize(720,480)
FadeOut(10)
c=last


a++b++c


And here's the Ini file:
*INFILE c:\savage_brazil5a_hdencoder.avs
*OUTFILE C:\brazil_mpg.m2v
*LOGFILE C:\HC.log
*BITRATE 4000
*MAXBITRATE 8000
*FRAMES 0 72183
*PROFILE best
*ASPECT 4:3
*GOP 18 2
*DC_PREC 10
*SCANMETHOD alt
*MATRIX mpeg
*CPU mmx


AviSynth Version: 2.56 Build Oct 28, 2005

[pandy]

Thank You, Hank! Great work!
I have only one question is there any chance to set level in HC?
I have video 1440x1080 and HC by default set MP@H1440L - but i prefer use a MP@HL instead ancient and rare MP@H1440L?

[Boulder]

@Cunhambebe: what happens if you load only one segment at a time (i.e. only use the portion that produces your clip a, b or c)? Have you tried upgrading to Avisynth v2.5.7? Have you tried leaving the CPU detection to automatic in HC? (Also, you are using alternate scan but encode as progressive?)

[Cunhambebe]

Thanks Boulder for taking time to respond;
what happens if you load only one segment at a time (i.e. only use the portion that produces your clip a, b or c)?
-I'll try it and post back soon with the results.

Have you tried upgrading to Avisynth v2.5.7?
-No, since the original homepage is off, anyway I know they have another homepage for it. I'll try it. But one thing is certain, HC last release can't even start here. - Still don't know why.

Have you tried leaving the CPU detection to automatic in HC?
-No, I'll ahcen this option from MMX, that was broken in 2.0 to auto :) Let's see if it works.

(Also, you are using alternate scan but encode as progressive?)
-I guess this is a mistake, isn't it? Since the source is deinterlaced (progressive) I'll change it to zigzag :confused: Anyway, do you think this could crash the application?
Thanks for your help,
Mark

Edit: I've tested everything:
-loading one segment at a time;
-leaving CPU detection as auto;
-zigzag;
Well, absoltely nothing, nothing makes it work.

There's only one option left: installing the new AviSynth version 5.7 - but I'll think about it. Is it stable?
On the other hand using QuEnc has shown excellent results, but some undesirable blocking (dark and white blocks) when you have fast motion on the screen (ex: waterfalls, birds beating wings, etc).... Maybe, changing to another matrix or tirn off "Use Trellis Quant", but that's another story for another topic.
Thanks,
Mark

[hank315]

Have you tried to load the avs file in the GUI?

Without the source I can't do that much.
Is it possible for you to upload the first segment (or part of it) somewhere?

[Cunhambebe]

Thanks so much for taking time to respond.
Have you tried to load the avs file in the GUI?
-In fact, that's what I'm doing, loading the avs file in the GUI.
Without the source I can't do that much.
Is it possible for you to upload the first segment (or part of it) somewhere?
-You mean the source file? I'll upload it and post back as soon as I finish it - thanks;
Mark

[Cunhambebe]

Here's the source file (about 10 MB)
Source (http://www.litoralsulvirtual.com/brazil_intro.zip)
Edit: since hank315 has already downloaded the source file, it was just deleted from my server. Thanks

[FreshMeat]

Hi!
I just installed DVD-RB Free and HC Encoder 0.19 (or what is default) and I upgraded it to 0.21. I just replace old HCbatch.exe with new 0.21 file. OK, I started to backup my DVD, but when it started to encode DVD, I get this error:
http://i4.tinypic.com/4r06moo.png

It's finnish, but I think you understand it. It's the most common windows error, program has executed an illegal function etc.

It works OK with old version, but I want newest. Any suggests?

[hank315]

@Cunhambebe

Thanks for the clip...

I tested it with the next script:

LoadPlugin("F:\HC_projects\HC021_run\DGDecode\DGDecode13.dll") # fileversion13 DGDecode
LoadPlugin("C:\Program Files\AviSynth 2.5\plugins\Colormatrix.dll") # load colormatrix plugin
MPEG2Source("E:\movies\testfiles\brazil_intro\brazil_intro.d2v") # wavsource dumped
ColorMatrix(d2v="E:\movies\testfiles\brazil_intro\brazil_intro.d2v")
Sharpen(0.2)
Lanczos4Resize(720,480)
FadeOut(5)
HC ini file:

*INFILE e:\movies\testfiles\brazil_intro\brazil.avs
*OUTFILE F:\HC_projects\HC021_run\run\brazil.m2v
*LOGFILE F:\HC_projects\HC021_run\run\brazil.log
*BITRATE 4000
*MAXBITRATE 8000
*FRAMES 0 942
*PROFILE best
*ASPECT 4:3
*GOP 18 2
*DC_PREC 10
*MATRIX mpeg
*CPU mmx
Runs OK, don't know why it doesn't work for you...

[Cunhambebe]

hank315
Thanks for the clip...
-You're welcome.
-I still don't understand why you used DGDecode13.dll instead of :
LoadPlugin("C:\unzipped\dgmpgdec148\DGDecode.dll"). Anyway, I've loaded the DGDecode13.dll exactly as you've done but unfortunately HC crashes on my OS (I still don't run SP2 - maybe this might cause this problem, I don't know).
Please note that this is happenning with HC020 because the newest version 021 doesn't even start here.
Thanks for your help. What a pity, because this must be a terrific encoder since its input is YV12 (no color space conversions). I'll keep trying. Maybe I'm doing something wrong.
Thanks again,
Mark

[hank315]

I still don't understand why you used DGDecode13.dll instead of :
LoadPlugin("C:\unzipped\dgmpgdec148\DGDecode.dll").
I just wanted to be sure the right DGDecode.dll was loaded, your d2v file mentiones fileversion13.
After some searching it seems dgmpgdec148 is also fileversion13, so that's OK.
Using SP1 shouldn't make any difference either...

What happens if you load the d2v file?
Or the most simple avs script like:

LoadPlugin("C:\unzipped\dgmpgdec148\DGDecode.dll")
mpeg2source("C:\brazil_intro.d2v").

[Cunhambebe]

Thanks for responding, hank315. What happens? The same thing. A Windows warning screen pops-up to report the problem to Microsoft. That means, the apllication crashes. What a mystery!
QuEnc was under the same folder that HC; I've created a new one just for HC. Same thing again. Same result.
I've tried QuEnc too, it's fast, much faster than TMPGENc, there's absolutely no video banding (there are some in TMPGEnc) because threre's no color space convertion BUT.... some unacceptable white blocks (or artifacts) are produced when you've got very fast motion on the screen such as birds beating their wings. It's hard, isn't it?
Well, I'll try to install an older versionof HC, maybe it works here and start another topic on QuEnc.
Thank you very much for your attention and valuable help. I hope one day this can be fixed because I know your encoder is a great piece of software!
Cheers,
Mark

[Cunhambebe]

Hi there,
I have installed HC version 019 and it works as a charm. Great encoder, no video banding at all (only if present on the original source). Great! HC really rivals TMPGEnc and MainConcept in terms of speed and quality (no color space conversions = no video banding and no loss of quality).
Now, I've got to find out why the 2 new versions (020; 021) crash on my OS (Windows XP SP1).
Thanks and congrats to hank315.
Cheers,
Mark

[Boulder]

Would you mind posting some information regarding your CPU, CPU-Z screenshots might be interesting to see.

(also, upgrading to WinXP SP2 is not a bad idea in general ;))

[FredThompson]

I'm seeing HC 0.21 and QuickPar not "playing nice" together. HC will run just fine but QuickPar, even when set to idle priority, draaaaags.

Are other people seeing this? Is there a fight between the two?