Log in

View Full Version : No one talks about divx 6.7?


Pages : 1 [2] 3

Elias
12th October 2007, 21:21
DivX. They should just buy a pure MPEG-4 license fee and play MPEG-4 codecs (as in AVC/ASP/AAC in mp4), and that's it.
You have some pretty strongly held opinions, so I'm not hear to persuade you specifically, but I would like to respond generally as to why hardware companies value DivX and put our logo on their products.

OK, this is already longer than I had intended. But let me know if you have any other questions and I'll do my best to answer them. Or I'll grab Gej and make him answer them. :pMy qualm with DivX, is simply that you're trying to make your codec more and more, proprietary, rather than MPEG-4 specs compliant. That is my big issue with you guys. It was all fine at first. You even supported mp4 output. It was when you decided to make your codec avi specific, where I began losing respect for you guys. Then came packed bitstream, and other non-MPEG-4, specs breaking nonsense with your codec.

Now, your actual DivX codec, is not bad per se. It's actually a good codec, at least for ASP. It has its qualities, so to say. But like I said before, your refusal to go fully MPEG-4, your disregard of AAC and mp4 container, well, that's not okay with me.

Now, I know you're doing this with a certain motive. You're trying to create a brand. You're trying to make people choose your codec; your specific flavour of MPEG-4. It's all about making money, of course, and the way DivX Networks has decided to go, is possibly, more money making in the long haul, compared with how for instance, Nero went with their codec.

But Nero, respect standards, and they have a lot more respect for the MPEG-4 standard than DivX Networks has shown so far.

I remember back in 2002, when you supported mp4 output, and wrote on your site about the benefits of the mp4 container, and how it prevents the codec from being locked to a certain company, and that it is a free container, with interoperability as its strongest key point.

Well, suddenly, you decided to ditch mp4 and now you're trying to do just that, lock your codec into your company.

Sorry, but you have no credibility in my book any longer, and I just hope your codec loses more and more popularity. If you're not going to follow standards, I am not going to use your codec, regardless of if it's better than XviD or not. XviD gets the job done, and it supports direct mp4 output. When I use an MPEG-4 codec, I want to use all of its standard, which includes its native MPEG-4 container, its MPEG-4 audio, etcetera.

Now you perhaps understand why I have strong opinions about this.

Oh and by the way, to my knowledge, Sony doesn't support DivX in PS3 and PSP because of your license fee. Last time they mentioned it, they were satisfied with supporting only standards.

kosmonaut
12th October 2007, 23:01
I respect your position on standards, but just to give you a bit of historical background to our decisions.


When I use an MPEG-4 codec, I want to use all of its standard, which includes its native MPEG-4 container, its MPEG-4 audio, etcetera.


All of those elements of the MPEG-4 standard are good in theory, but proved very difficult to implement in hardware. Remember, this was back in 2002-2003 when we were putting DivX Certification together. Our position is that had we insisted on precisely the elements you want, .mp4, AAC, etc., widespread hardware adoption of MPEG-4 ASP would not have happend, at all. Getting this stuff into actual chips that could be replicated at a sane price (something extremely important in real world business decision making) was only possible with certain compromises. And don't forget, the vast majority of the content people wanted to watch at that time did not, and does not, have those things either. It's mostly in .avi containers and with MP3 or AC3 sound.

The DVD player market is brutal, with miniscule margins. If what we came up with in DivX Certification increased the end price of the device appreciably, it would never have been adopted widely. Other companies have tried to get their technology adopted, and have even spent millions of dollars to do so, like Microsoft with WMV. But because they could not do it inexpensively, it has never really been implemented into low cost devices. DivX Certification hits a sweet spot, allowing the chip makers, ODM's, and OEM's to use it while keeping their prices competitive.

Oh and by the way, to my knowledge, Sony doesn't support DivX in PS3 and PSP because of your license fee. Last time they mentioned it, they were satisfied with supporting only standards.

Sony has its own history with standards, both pro and con. Also, Sony does indeed sell DivX Certified DVD players. The PS3 and PSP however have entirely different strategic visions behind them. Yet don't be surprised if they support DivX one day.

DeathTheSheep
13th October 2007, 00:55
I think that's very well argued.

I also think that if somebody really wants an mp4, they can easily (often in one step) mux their a/v into mp4. That's why those tools exist and are often quite easy to use.

Looks like DivX is sticking with what has been proven successful and still fairly open-standard. In fact, the best/most popular/easiest to use/most powerful tools out there still support avi, and they do so better than mp4. I know of one free tool that supports real mp4 editing (avidemux, I think it was called), but it is notoriously unstable and less powerful than the alternative VirtualDub.

Most people don't give a hoot about so-called "full standard" compliance as long as its usable anywhere and everywhere they need with virtually no restrictions (and actually quite a few benefits) while still retaining the option, if for some reason they care, to manually go "full standard" with a simple mux or two.

you have no credibility in my book any longer, and I just hope your codec loses more and more popularity. If you're not going to follow standards, I am not going to use your codec, regardless of if it's better than [others].
Nonsense. This is attack mode. There is no neutrality here. :rolleyes:

prOnorama
13th October 2007, 04:23
Now, why do companies like Sony, Samsung, LG, Philips (essentially every major DVD manufacturer) work with us? Why have they put out into the market over 100 million DivX certified products so far? These guys are tough, smart business people. If they thought it wasn't absolutely necessary to their bottomline, they would not do it. Period.

Sure but I bet DivX has smart, tough business people as well ;)

Basically Xvid is the only competitor in the MPEG-4 ASP market. And it's a big competitor. When negotiating deals with major DVD manufacturers it could be possible DivX would claim "exclusivity" i.e. they want to be the sole recognized brand for MPEG-4 ASP. Having no Xvid logo's on DVD players suits DivX well. Mind you this is all speculation on my side but it seems logical to me. If I were a DivX marketing manager I would probably suggest such a route. ;)

On the other hand: DVD manufacturers aren't stupid, it will probably hurt their sales if they don't at least "non-officially" support Xvid. (which means they are sticking to the standards for MPEG-4 ASP which is a good thing IMO).

In the manual of my Philips DVP 5980 Xvid is not mentioned at all, not even MPEG-4 ASP (only DivX/DivX Ultra whatever that is), but it plays every Xvid file I throw at it (except when the resolution is not supported). :)


My qualm with DivX, is simply that you're trying to make your codec more and more, proprietary, rather than MPEG-4 specs compliant.

That seems logical to me. The most sucessfull businesses use some form of "vendor lock-in", see Microsoft, Apple (iPod) etc.

DivX going "more proprietary" seems a logical thing for them to do in terms of business strategy (monopolize a market and make it difficult for the competition). Can't blame 'em for trying.

Meanwhile I stick to Xvid because it suits my needs fine and even though DivX 6.7 might be better (I doubt the difference will be huge since the MPEG-4 ASP potential basically has been "maxed out" already I think, though others might disagree) I'm just not willing to pay for it.

(I don't use a PDA, iPOd etc. maybe DivX has more of an advantage on that hardware in terms of hardware compatibility)

Mtz
13th October 2007, 07:16
All of those elements of the MPEG-4 standard are good in theory, but proved very difficult to implement in hardware. Remember, this was back in 2002-2003 when we were putting DivX Certification together. Our position is that had we insisted on precisely the elements you want, .mp4, AAC, etc., widespread hardware adoption of MPEG-4 ASP would not have happend, at all. Getting this stuff into actual chips that could be replicated at a sane price (something extremely important in real world business decision making) was only possible with certain compromises.
Yeah, and now, near the DivX logo we can see the Nero Digital logo. For the users, Nero Digital = MP4.

enjoy,
Mtz

ReinerSchweinlin
13th October 2007, 08:47
Did somebody else some comparisons with recent XVID and LAVC?

Elias
13th October 2007, 10:29
I respect your position on standards, but just to give you a bit of historical background to our decisions.That's very appreciated.All of those elements of the MPEG-4 standard are good in theory, but proved very difficult to implement in hardware. Remember, this was back in 2002-2003 when we were putting DivX Certification together. Our position is that had we insisted on precisely the elements you want, .mp4, AAC, etc., widespread hardware adoption of MPEG-4 ASP would not have happend, at all. Getting this stuff into actual chips that could be replicated at a sane price (something extremely important in real world business decision making) was only possible with certain compromises. And don't forget, the vast majority of the content people wanted to watch at that time did not, and does not, have those things either. It's mostly in .avi containers and with MP3 or AC3 sound.I can understand that it was more difficult back in 2002-2003 to put together MPEG-4 ASP with AAC and mp4. Certainly, the MPEG-4 standards had not come as far is it has today. I still think you chose the easy way out, and I see no reason why you don't go fully MPEG-4 standard today. Today, MPEG-4 (and I'm not talking about the mp3/ac3+avi nonsense, mind you), has come a long way compared with back then. More and more hardware devices like cell phones, DVD-players, video game consoles, and operating systems, have a lot wider support for MPEG-4 ASP/AVC/AAC in mp4.

So really, I don't see the problem with supporting it today. Sure, it breaks your DivX certification with your older versions. That's not a problem, just release a modified MP4Box solution so that people can convert their old encoded DivX 4/5/6 content into pure MPEG-4, and an AAC encoder.The DVD player market is brutal, with miniscule margins. If what we came up with in DivX Certification increased the end price of the device appreciably, it would never have been adopted widely. Other companies have tried to get their technology adopted, and have even spent millions of dollars to do so, like Microsoft with WMV. But because they could not do it inexpensively, it has never really been implemented into low cost devices. DivX Certification hits a sweet spot, allowing the chip makers, ODM's, and OEM's to use it while keeping their prices competitive.I have so far never seen a 3G cell phone capable of playing either WMV or DivX. Yet, all of them, play mp4 Simple Profile and AAC in mp4.Sony has its own history with standards, both pro and con.That is true, I'm quite aware of how they sometimes have been doing some screwing with the mp4 container, for instance. But that is corrected now in the PSP.Also, Sony does indeed sell DivX Certified DVD players.Well then, I guess nobody's perfect :)The PS3 and PSP however have entirely different strategic visions behind them. Yet don't be surprised if they support DivX one day.I hope not. Truth is, both the PS3 and the PSP support DivX/XviD quite fine, if you put the encoded content into the mp4 container; DivX Networks could easily get PS3 decoding by just stop being this stubborn about your DivX certification. But that would mean Sony won't pay you for it and your DivX license fee, because they're already paying for MPEG-4. Of course, you've wasted lots of money on this DivX certification, and it seems you don't want that money to go to waste by simply doing the right thing: supporting standards, which means you'll have to trash your DivX certification, or modify it so that it'll become incompatible with your current DivX certification.

You want Sony to support your brand on the PS3, rather than just be another player in the MPEG-4 market. I think it's quite arrogant of DivX Networks to expect Sony to break standards for you. Because let's face it, what DivX is, basically, is a good MPEG-4 ASP content, in a limited, old, and outdated container where it really doesn't belong because it breaks the standard.

Look at how Nero did it, they followed the standard all the way (if you ignore their stupid choice of VOB subtitles in mp4 rather than MPEG-4 Timed Text), and now their codec implementation of MPEG-4 works perfect in the PS3.

Let's face it, the PS3 is a good hardware decoder for your business. Why not just support mp4 and AAC, remove that stupid packed bitstream, and get a new user base on the PS3? Why all the hassle you're going with now?

I don't understand the way you're handling your business right now, but I certainly wouldn't call it smart.

This is what Sony said last time about it:Q.
One thing about video file playback. It seems that quality is not constant, and picutre quality falls from time to time.

K1
We don't have an answer to how much quality we can guarantee from playback of video files created by PCs. Although there is standards such as "mpeg-4", there are many cases where the files contain data that are way off the standard.

K2
We have codecs in place that play back data following proper standards, and functions to improve the quality of these picutres. We need more work on handling data that is outside the standard.

K1
For now, out main goal is to allow playback of as many standard as possible.

Q.
From that point of view, how about playback of WMV and Divx?

K1
The problem with those formats are that they are not industry standards like "mpeg-4", but format that is locked in by various corporations.
But I understand the demand (laugh).

Q.
How about overcoming the licencing fees for such formats by selling update for WMV playback at the playstation store for 500yen?

K1
We don't want to resort to such measures (laugh). And its not a matter of not implementing because it is by another company.Source: http://www.neowin.net/forum/index.php?showtopic=567411

Seriously guys, get along with the program, start supporting MPEG-4 AAC, AVC and mp4, and you will gain my respect again, not to mention, your content will play on the PS3 and elsewhere by default. Your codec is almost 100% MPEG-4 standards compatible anyway; stop crippling it with packed bitstreams, avi, and other stupid decisions.Nonsense. This is attack mode. There is no neutrality here. :rolleyes:The reason why I want your codec to die out, is because it's breaking standards and it's not specs compliant. I'm really quite neutral to be honest, if you do the right thing, I will give you credit for it, if you screw up, I will criticize you for it. It's really as simple as that.

Now, XviD also has some issues, and it's mostly, because XviD is aping DivX. The XviD coders are just trying to follow your lead, but they've been doing some things right, like for instance, the possibility to directly encode to mp4.

I don't understand why you're limiting yourselves to VirtualDub. Avery Lee isn't going to implement mp4 output because he doesn't care, unless of course, you pay him to do it. His encoder is quite good, but it's based on Video for Windows, which is old and outdated. You're basically forcing yourselves to be stuck in old software if you're doing this simply because of the few pros you get out of WfV.

In my honest opinion, DivX Networks is not only locking its users into their codecs, but you're making yourselves dependent, on the software that's available, rather than creating your own, independent, MPEG-4 standards compliant codec.

Well, I think your decisions aren't tactical.That seems logical to me. The most sucessfull businesses use some form of "vendor lock-in", see Microsoft, Apple (iPod) etc.

DivX going "more proprietary" seems a logical thing for them to do in terms of business strategy (monopolize a market and make it difficult for the competition). Can't blame 'em for trying.And I don't like Microsoft either for doing that. All these corporations should support standards, but they don't, because they care more about making money, and screwing us consumers over by screwing with the standards.

Ranguvar
14th October 2007, 03:02
XviD is not better. IMO that is simply not true. That was the case when Divx was 5.x but certainly not now. XviD has not improved in quality in over two years while the DivX people have been improving on a quarterly basis.

In the only double blind MOS score test that I know of says that DivX is better.

http://www.compression.ru/video/codec_comparison/subjective_codecs_comparison_en.htmlThat's because they used the defaults... Xvid's deblocking is default off. DivX's is default on. They said so themselves.

Dark Shikari
14th October 2007, 03:41
That's because they used the defaults... Xvid's deblocking is default off. DivX's is default on. They said so themselves.But DivX is surely drastically better; we can't let facts get in the way of our bias can we? ;)

buzzqw
14th October 2007, 08:03
and a command line interface could be quite handy

BHH

SeeMoreDigital
14th October 2007, 09:06
and a command line interface could be quite handy

BHHAgreed....

That way, if people want to encode DivX with B-VOPs and place the video stream within the .MP4 container, they can do it correctly (as per the MPEG-4 spec).


Cheers

Kurtnoise
14th October 2007, 09:14
and a command line interface could be quite handy
exists already...

buzzqw
14th October 2007, 09:35
exists already...

tweaking registry ? using dll ?

i mean an interface like xvid_encraw.exe or x264.exe

BHH

Kurtnoise
14th October 2007, 10:05
In the Dr DivX package, you can find a command line tool.

and yes it uses the DivX libraries just like xvid_encraw and x264 do but no avisynth support as input.

Kurtnoise
14th October 2007, 10:07
That way, if people want to encode DivX with B-VOPs and place the video stream within the .MP4 container, they can do it correctly (as per the MPEG-4 spec).

we can do that flawlessly with MP4Box...

buzzqw
14th October 2007, 10:10
but no avisynth support as input

that will kill almost any simple gui

... but thanks for the tip.. i will search for it

BHH

Vasyek01
14th October 2007, 10:58
I do no, but after read all the posts still don't know why Divx 6.7 so bad? :helpful:

Dark Shikari
14th October 2007, 11:12
I do no, but after read all the posts still don't know why Divx 6.7 so bad? :helpful:
Most importantly, more than anything else: It costs money, but is no better than Xvid in any real way. In a sense it is what Mainconcept is to x264--they're the same "style" of encoder in terms of both the standard they use and the manner in which they compress the source, so the only real difference is that one costs money and the other doesn't.

weaver4
14th October 2007, 13:19
That's because they used the defaults... Xvid's deblocking is default off. DivX's is default on. They said so themselves.

Don't know how much impact deblocking would have, not much I would guess.

weaver4
14th October 2007, 13:39
I do no, but after read all the posts still don't know why Divx 6.7 so bad? :helpful:

Well it is not "so bad", I would not say that XviD is better either, both are very, very close in quality. I have stuck with DivX over the years because:

they have several "paid" developers always working on the codec.
they have a good certification program.
they have sales staff that was dedicated to working with manufactures.
they use to have a nice community forum which had a lot of DivX representation.
and they use to have a strong commitment to tools; like the Doctor.
I believe companies have more duration than open source projects.

delacroixp
14th October 2007, 15:25
I do no, but after read all the posts still don't know why Divx 6.7 so bad? :helpful:
Well it is not "so bad", I would not say that XviD is better either, both are very, very close in quality. I have stuck with DivX over the years because:

they have several "paid" developers always working on the codec.
they have a good certification program.
they have sales staff that was dedicated to working with manufactures.
they use to have a nice community forum which had a lot of DivX representation.
and they use to have a strong commitment to tools; like the Doctor.
I believe companies have more duration than open source projects.

Perhaps the question is not so much that DivX 6.7 is so bad or that Xvid may or may not be better or worse than DivX but ... what's best... that eternally subjective quest for greatness...
DivX has served the global community well... most consumers (who have money) still use DivX... geeks, who possibly live on an island, a world apart, push the latest codecs... which will eventually be the money spinner of tomorrow...

DivX H263 still has some life left... some people are still using DivX 5.2 (http://forum.doom9.org/showthread.php?t=129720)...


:):devil::D
Pascal

codeguru
15th October 2007, 10:00
1.)
they don't send the 15 day eval code if the request is made from european IP adresses
2.)
the full install kills the divx3 audio support in all apps except the new divx player

related on 1.)
as an owner of a Q6600 quadcore CPU I would like to test if the image filters of 6.7 are really multithreaded but I can't because this f*in code does not come

but filtering in virtualdub with the dnr2 and temporal cleaner gives much better results than the builtin-filters of 6.2 pro so the community edition with a quality of 8 is good enough, and I only have to enable multithreading in the vdub filters, because actually the filters are singlethreaded and all of them run on a single cpu core.

At least I can now encode my nearly 100 VHS tapes at 8 hours each at one pass, saving lot of time, that's something that works in 6.2 only with a very poor image quality due to less compute power left for filtering prior to encoding.

The multithreading is working quite well and stable, so some additional filters can be applied while capturing.

For offline encoding there's more a competition between h.264 and WME9, both are quadcore-capable and have no restriction about image quality. For the best encoding quality for divx I would take 6.2, but that is wiped by the 6.7 install.

red5goahead
15th October 2007, 10:29
imho Divx is the best decoder for mpg4 material. I use it into KMPlayer (not with 6.7 version). I use with full deblocking+sharpening. simply perfect.

falcon2000eg
16th October 2007, 12:13
Divx is not bad personally im using their codec to play my movies. But encoding is another issue Xvid is faster at the same quality "single core chip".Maybe i can get more from Divx with more tweaking and i think it will be faster on multicore chips.But what is the point if I do not care about speed and want high quality I will go for h.264.So now it is a matter of quality/speed factor between Dixv and Xvid because both nearly reached the maximum quality I can get from an ASP codec.

jsquare
16th October 2007, 23:24
I do all of my encodes with XviD using GKnot/Stax, but recently converted a HD source with the DivX 720HD profile and it came out beautiful at around 1/10 the size of the original.
I'm starting to like the features and recently developments with DivX, and I don't see any enthusiasm with the XviD team on improving the codec any further, much of the effort is being pour into x264.

Vasyek01
17th October 2007, 06:33
Thx!
Maybe some one did compare HD quality, not SD? I compressed a movie in HD and used the encoding type two pass, average bitrate with 5000 kbps on the Divx encoder. I find that it works faster than Xvid encoder. Maybe I did something wrong?

delacroixp
18th October 2007, 08:18
Maybe I did something wrong?
Maybe not...


:):devil::D
Pascal

weaver4
18th October 2007, 15:09
I have found later versions of DivX (6.X and above) to be faster than XviD by 10 to 20%.

Vasyek01
19th October 2007, 03:02
I have found later versions of DivX (6.X and above) to be faster than XviD by 10 to 20%.
Me too. I like quality of Xvid - super! But it's the really slow encoder.
Divx works about 30% faster than Xvid for me.

Dark Shikari
19th October 2007, 03:14
Me too. I like quality of Xvid - super! But it's the really slow encoder.
Divx works about 30% faster than Xvid for me.Perhaps you need to use faster settings?

Comparing two encoders' speed is useless if you ignore the fact that they have different encoding options.

weaver4
19th October 2007, 14:55
Yeah, I should of said that my results are with AutoGK and AVI.Net, default configuration, Constant Quality, Q=4 for DivX and Q=3 for XviD, 640 width. I believe this is a fair comparison.

Vasyek01
19th October 2007, 22:23
I compared the same HD clip (NOT SD) with the same Quality, same bit rate, same size out... It's no question, the Xvid encoder can be setup for way better quality than the Divx encoder. Especially for Standard Definition (720x480 dpi) format, but I'm talking about a High Definition format like 1280x720 dpi.
So, I found out for myself, if I have some really good HD quality movie, I use Xvid, for ordinary stuff Divx is OK.

Dark Shikari
19th October 2007, 22:58
Yeah, I should of said that my results are with AutoGK and AVI.Net, default configuration, Constant Quality, Q=4 for DivX and Q=3 for XviD, 640 width. I believe this is a fair comparison.Quantizer isn't the only setting--motion search settings and so forth?

Many apps automatically set Xvid up for the max quality settings while ignoring speed; you can easily gain a lot of speed without losing much quality with Xvid.

Why totally different quantizers with Xvid and DivX btw? :confused:

weaver4
20th October 2007, 02:57
Quantizer isn't the only setting--motion search settings and so forth?

Many apps automatically set Xvid up for the max quality settings while ignoring speed; you can easily gain a lot of speed without losing much quality with Xvid.

Why totally different quantizers with Xvid and DivX btw? :confused:

I just used the defaults, I figured the authors of avi.net and autogk know what is the best setup.

Divx Q of 4 is about the same Bits-Per-Pixel as Xvid Q of 3; so they are about the same filesize and same quality.

Dark Shikari
20th October 2007, 04:24
I just used the defaults, I figured the authors of avi.net and autogk know what is the best setup."Best" is a relative term, which is why it is banned on this forum.

For example, I could use my Q50 settings that I came up with for x264 and be about 10% worse quality than MeGUI's HQ-Slow settings, but be about 5 times faster. You probably would have minimal notice of the 10% difference, too--its diminishing return--lots of speed for little quality gain. The same is true with Xvid, and stating "Xvid is higher quality but is slower" is meaningless because you could use faster settings and sacrifice a tiny bit of that quality.

Ice =A=
20th October 2007, 11:14
Well said!

DeathTheSheep
21st October 2007, 03:07
Xvid is almost unusable for me due to jumping/flickering b-frames issue in still anime scenes, which seems to be present in all settings cases except 0 bframes (obviously) and high bitrate.

DivX is fine...

Dark Shikari
21st October 2007, 05:29
Xvid is almost unusable for me due to jumping/flickering b-frames issue in still anime scenes, which seems to be present in all settings cases except 0 bframes (obviously) and high bitrate.

DivX is fine...I've never seen or heard of anything like this. Example?

CruNcher
21st October 2007, 12:36
Xvid is almost unusable for me due to jumping/flickering b-frames issue in still anime scenes, which seems to be present in all settings cases except 0 bframes (obviously) and high bitrate.

DivX is fine...


did you tried my EDP build yet ? i especialy tweaked for that case p and b-frame flickering in low bitrate situations was a major point in my optimizations and detail preservation but i did it for real source footage im not sure if it helps with artificial source, but anyway at least it should visible reduce p and b-frame flicking :)

weaver4
21st October 2007, 21:21
Well said!

In the Doom9 codec shootout they stated that XviD was slightly faster than DivX 6.1. DivX claims they are 30% faster now than 6.1.

Inventive Software
22nd October 2007, 01:44
6.1 also didn't have multithreaded encoding IIRC, so it was a fair comparison. But correct me if I'm wrong. I hate being wrong. ;)

dannyv
22nd October 2007, 21:19
You have some pretty strongly held opinions, so I'm not hear to persuade you specifically, but I would like to respond generally as to why hardware companies value DivX and put our logo on their products.

It's not just the technology, although that is important of course. Far more than that, DivX Certification is a pretty intense suite of testing. Many products have failed it, especially on the first couple of go arounds. And it goes quite deep. We work with the chip guys first, then the ODM makers, and then finally with the OEM's, all with goal of ensuring a certain level of quality and interoperability that is very, very reliable.

Now, why do companies like Sony, Samsung, LG, Philips (essentially every major DVD manufacturer) work with us? Why have they put out into the market over 100 million DivX certified products so far? These guys are tough, smart business people. If they thought it wasn't absolutely necessary to their bottomline, they would not do it. Period.

The truth is that they came to us initially. When users began to request the ability to playback MPEG-4 files (at the time almost exclusively DivX, now many other flavors, of course) on inexpensive devices, some innovators began to enable that, and to put it on their boxes. But users had no way of knowing what exactly that meant. MPEG-4 is a vast suite of standards. What DivX Certification allowed was for a precise, real world "standard" to exist that told everybody, from simple viewers to content creators the precise compression choices they should make in order to be inter-operable in the widest manner possible.

That is not a small thing and it is something that the CE companies valued DivX for enormously, and they continue to do so. Combine that with the rigorous testing we do in our Certification lab and hopefully you can recognize why the hardware companies look to us, and our logo, to demonstrate to consumers that their products are innovative, inter-operable, and high quality.

OK, this is already longer than I had intended. But let me know if you have any other questions and I'll do my best to answer them. Or I'll grab Gej and make him answer them. :p

Very well said. I am an avid DIVX user. I encode a lot of 1920x1080 HD content to 1280x720 and the results are fantastic. The gripe I do have is the divx converter is very limited compaired to DR DIVX. I use DR DIVX because its the only easy program that allows me to copy stream the AC3 5.1 sound and it allows me to custom resize the resolution. Many people have said that DR DIVX does not work with divx 6.7 but I have no problems with it.

What my biggest gripe is it seems divx has turn there attention mostly to stage6 and have stopped assisting in the development DR DIVX. And as far as the divx converter its usless for what I need to do.

DR DIVX has its problems such as if you encode in theater mode then switch to High definition mode it locks up and the 6.7 codec needs to be reinstalled.

My question to you kosmonaut is dr divx dead? Dookl1 (developer) and most users over at the divxlabs forum have asked for assistance many times and the same question "is dr dead" as well has been asked and there requests and questions mostly go unanswered.

CruNcher
23rd October 2007, 10:43
also people shouldn't forget their are other very good ASP encoders worth mentioning :P lavc ASP, Nero Recode (Ateme ASP), 3ivX and as a special case encoder hdx4 come to my mind, all of them improved since the last shootouts.

weaver4
23rd October 2007, 14:26
I agree with dannyv. The Dr. is a very good tool, it is 90% of the way there before activity stopped on it. The question has been asked many times over, without answer; "Is the Doctor Dead?"

kosmonaut can you answer?

Inventive Software
25th October 2007, 14:58
Couple of bugs to report. Custom PARs with values above 255 aren't supported? I have a PAR that's 368:267 (see, this is why I hate cropping material most days :p) for one source, trying to get it 16:9. No I'm not kidding.

Secondly, why aren't those custom PARs applied correctly? If I want a PAR of 64:45, I should be able to get a PAR of 64:45 and not something screwy and incorrect like 20:11 or 16:11. Is it going to be sorted for the next release?

delacroixp
25th October 2007, 15:20
"Is the Doctor Dead?"

The Doctor is Immortal... maybe he's on sabatical getting a second PhD at MIT (http://web.mit.edu/)... go figure...


:):devil::D
Pascal

Ranguvar
27th October 2007, 00:17
Don't know how much impact deblocking would have, not much I would guess.It makes a huge difference.

Ranguvar
27th October 2007, 00:20
@DeathTheSheep: I have literally thousands of Xvid anime clips, 2 B-VOPs, no flickering or stuttering or anything. How slow is your PC?

Brother John
27th October 2007, 00:48
Custom PARs with values above 255 aren't supported?
That’s a bitstream limitation. MPEG-4 Visual’s custom PAR fields are unsigned 8-bit integers, i.e. only values between 0 and 255 are possible.
Btw: For H.264 it’s unsigned 16-bit integers: 0 to 65535.

I have a PAR that's 368:267 (see, this is why I hate cropping material most days )
But cropping has zero effect on PAR...

Inventive Software
27th October 2007, 15:08
Yes it does, and allow me to attempt to explain.

After cropping, the original source material is no longer the 720x576 (1.25:1) that it was encoded to, it's now 698x572 (1.22:1 rounded). To get that source 16:9 after cropping, you need a new PAR since 64:45 wouldn't be correct.

I can't remember how I got 368:267 (that PAR doesn't apply to the above calculations BTW), or for what source, but it does matter, honest! ;)