Sorry guys I did not intend to send this thread into a different direction.
DAN
For your info the NLE was M100 iFinish which came with the real time Sony MPEG encoder SD the menu option allowed the use of 15 to 50 Mb I-frame 4:2:2 in CBR. It also allowed the selection of MPEG-1, MPEG 2 4.2.0 or MPEG 4.2.2. This system was only used to create archival copies of finished projects and not used as an editing medium. I used the VBR option for storage allowing control of file size vs quality. I now use the NanoFlash to store my finished projects.

Dear Lance,

Thanks for the extra information.

Hi Rafael: Yes I understood this was so, all I was wondering if there is room in the Sony XDCAM HD specification for the utilization of a VBR encoding scheme. Does anyone know or have a link to any possible web page which would display the full Sony XDCAM HD 4:2:2 spec so we could check ?

Hi Mark,

To cheat the NLEs shouldn't be a problem. In fact the NANO files are off-standard (data rate, I-frames,..) but they work in most of the applications. At least FC can manage ot of stuff that in theory is not supported.
rafael

I agree that "cheating" NLE's should not be a problem. After all, all we need is the highest possible quality - and NOT a strict compliance with any particular "industry standard"...

Why?

Well - simply because it's our delivery formats that ultimately counts; this must be absolutely compliant with some industry standard (be it HDTV broadcast, BluRay disk, ....).

So, does it make sense to bother Convergent Design an push them to upgrade their encoding schemes from the "constant bitrate" to the "constant quality"?

Well, I think no, it does not (strange does it sound from somebody who initiated this thread, doesn't it).

I'll explain: given that all cameras will inevitably feed the nanoFlash with some noise (XDCAM EX in particular, as this is their Ahilles heel when compared to other great specs) , the only solution seems to be getting rid of it in post.

I'm editing in Vegas Pro, and I've just purchased the Neat Video Pro plug-in. I have run it on both the SxS clips, and the simultaneously recorded, 220 Mbps I-Fo nanoFlash clips...

And I can tell you the results from the nanoFlash clips (not just I-Fo 220, but also L-GoP 100) are much, muuuuch better than those from the native XDCAM EX. The obvious reason being that with much less compression artifacts, the ingenious Neat Video filter can really show all its power on what it's supposed to deal with: the noise.

Add to it that Neat Video can also sharpen your video at the same rendering run, and you can shoot with ALL detail enhancement OFF - thus further relieving the nanoFlash codec from dealing with detail enhancement, and allowing it to encode even more efficiently. The motion, as well as static high frequency details, can be better handled by the encoder - resulting with even less compression artifacts.

Do you hate noise, that looks augmented along with detail in the nanoFlash files? I do. Do you want your video sharper? Well, I do too.

The answer for the time being is:

- record in L-GoP, at 100 Mbps max, or
- record in I-Fo, at 220 Mbps min, and

- render your material with the Neat Video (or any other, equally capable, tool)!

Well, of course Convergent Design may come with a better solution any time now, but frankly, I'd rather they spent their man-hours on finally implementing CF card hot-swapping, SDI embedded + analog audio mixing, and alike...

Just my opinion, YMMV.

Piotr

I've been using this software and it does very good noise reduction. I use it on everything out of the EX3.

Guys,

Thank you very much for your advice. Of course I hope you realize I'm talking subtleties here - just like with the (dark area) noise, the shimmering fine detail ("translucent" grain) can only be seen on a fairly large display, like my 50" plasma.

Kris,

Could you please send me a PM? I tried emailing you, but for some reason I can't. It so happens I do live in the SE of Poland, so I'd be more than happy to meet and share experience.

Hello Piotr, I would love to know what is the outcome of the NF trade/test with Kris? You never come back to clear up the suspicion of faulty NF or camera ensue?
Cheers

Hi Luben,

My meeting with Kris is planned as soon as he comes to Poland (he's in UK now, and supposed to give me a phone call when he arrives here). Of course I will update this thread accordingly - but frankly, I'm 99.99% positive both my EX1 and my nanoFlash are working up to specs.

It's just that - unlike most "real pros" here who assess their material using super expensive production monitors with HD-SDI input - I always base my opinions on the results I'm getting with my audience in mind, and this means a really big display. Not necessarily with professional inputs (just HDMI), but of 50" and up in size - while most persons trying to convince me there is no noise problem are using something like a 20" incher... What they don't realize is that it's not the type and quality of the input, but the very size of a pixel which decides whether or or not some artifacts will be noticeable or not.

I have almost come to terms with it, and am right now working a new workflow involving obligatory use of the Neat Video filter in Vegas. The additional benefit is that it can de-noise and sharpen at a single go, thus making it the obvious choice to shoot with in-camera sharpening (detail) off. This further enhances results by making the nanoFlash encoder less burden (no sharpening artifacts), while the noise can be removed in post and the picture sharpened to taste.

It's a pity though that all those ideas myself and other people presented in this thread (like fine tuning the Long GoP structure, or even introducing VBR) have not even been commented or considered seriously by CD...

Hi Piotr & Luben:
I can now confirm that VBR encoding *is* part of the *Official Sony XDCAM HD 4:2:2 SPEC ! I have a copy of the spec straight from Sony entitled,

"Codec Technology for XDCAM Tapeless Products and Systems
By Hugo Gaggioni, Chief Technology Officer, Sony Broadcast & Production Systems Division"

On Page 5 it States:

"3. Adoption of VBR
For HQ and LP mode, a Variable Bit Rate compression method was adopted in order to take advantage of the random access nature of tapeless media. It is different from the 2 pass* VBR compression technique used in DVD authoring. This codec was specifically developed for MPEG to perform high-speed efficient picture coding.
The XDCAM product line is based on the MXF file format and MPEG-2 Long GOP for exchanging and transmitting Video, Audio and Metadata. In the VBR (Variable Bit Rate) encoding method adopted for XDCAM HD products, the maximum bit rate that can be processed in real-time is fixed. This results in the high level of picture quality exhibited by XDCAM HD products over different picture material while minimizing the file size and maximizing transfer speed."

Apparently, VBR encoding was introduced by Sony as soon as the XDCAM *HD* version of cameras came out.

Dear Piotr,

We have to maintain compatibilty with a wide range of Non-Linear Editors.

Using a non-standard number of frames in a Group of Pictures would require substantial amount of testing to ensure compatibility.

We use a "Closed GOP". Switching to an "Open GOP" has ramifications when editing. With Closed GOP, the I-Frame that starts each group does not depend on or need information from the preceeding GOP.

Dear Mark,

There are three flavors of XDCam:

XDCam HD
XDCam EX
XDCam

We support both XDCam EX (35 Mbps, 4:2:0 35 Mbps VBR) and XDCam (50 Mbps, 4:2:2, CBR).

XDCam HD is different.

XDCAM - Wikipedia, the free encyclopedia (http://en.wikipedia.org/wiki/XDCAM)

Any chances of supporting XDCAM HD format - at least with the higher bit rates, Dan?

Dear Piotr,

I have not been able to determine if XDCam HD is 1440 x 1080. The PDW-350 is a 1440 x 1080 camera and it is a XDCam HD camera.

We want to continue to record in 1920 x 1080, thus we are using XDCam format and not XDCam HD format.

Dear Dan,

I'm a bit confused with what you're saying. Even in the Wikipedia article you linked to, it reads:
"In August 2009, Convergent Design began shipping the nanoFlash Portable Recorder, which uses the Sony XDCAM HD422 codec at bit rates ranging from 18 Mbit/ second to 35 Mbit / second (in the 4:2:0 colorspace), and bit rates ranging from 50 Mbit/ second to 280 Mbit/ second (in the 4:2:2 colorspace).

So, why are you saying nanoFlash is only capable of XDCAM EX and XDCAM, but not XDCAM HD?

As to your other point: of course I'd be for the full raster - just VBR option for the high bitrates. From what Mark quoted, the Sony chip you're using is capable of it...

Thanks,

Piotr

Hi Dan:

Uhh, I understood the Nano Flash & Flash XDR are using the Sony XDCAM HD 4:2:2 codec, therefore, VBR is also possible. (??) So what codec are you using ? XDCAM or XDCAM HD ? XDCAM is an SD only codec according to the Sony White Paper on the spec.

Dear Dan,

Could you please explain your confusing post? I'm sure it's not just myself or Mark who are waiting...

Thanks,

Piotr

Dear Friends,

As I understand it, there are three flavors of Sony XDCam:

XDCam HD, 1440 x 1080 4:2:0
XDCam, 1920 x 1080 4:2:2
XDCam EX, 1920 x 1080 4:2:0

Here is an excerpt from an article on Wikipedia:

XDCAM is a tapeless professional video system introduced by Sony in 2003.

The first two generations, XDCAM and XDCAM HD, use the Professional Disc as recording media. This disc is similar to Blu-ray Disc and holds either 23 GB of data (PFD23, single-sided) or 50 GB (PFD50, double-sided).

The third generation, XDCAM EX, uses solid-state SxS cards instead.

XDCAM - Wikipedia, the free encyclopedia (http://en.wikipedia.org/wiki/XDCAM)

--------------------------------------------------------------------------------------------------

The nanoFlash uses XDCam, 1920 x 1080, 4:2:2 at all bit-rates 50 Mbps and higher.

And XDCam EX, 1920 x 1080, 4:2:0 at bit-rates of 18 Mbps and 35 Mbps.

And we have an option to record in 1440 x 1080, 4:2:0 at 35 Mbps.

--------------------------------------------------------------------------------------------------

1. The Sony PDW-F330, PDW-F350, PDW-F355 are "XDCam HD".

Here is some text from an offical Sony site discussing the Sony PDW-F330:

HD 1080 Recording at Selectable Frame & Bit Rates
XDCAM HD products can record video signals in 1080/59.94i, 50i, 29.97P, 25P and native 23.98P - using a "MPEG HD" codec with industry standard MPEG-2 MP@HL compression.

The user can also select the following bit rates; 35, 25 or 18Mb/s depending on picture quality and recording length requirements. Choosing the highest bit rate of 35Mb/s results in a recording time of approximately 60 minutes, while choosing 18Mb/s provides a recording time of approximately 120 minutes - the longest recording time offered by any current HD camcorder.

MPEG HD images are recorded at 1440x1080 pixels with 4:2:0 sampling.

Sony : PDW-F330K (PDWF330K) : Features : United Kingdom (http://www.sony.co.uk/biz/product/xdcamcamcorders/pdw-f330k/features)


--------------------------------------------------------------------------------------------------

2. The Sony PDW-700 and PDW-F800 are "XDCam".

The Sony PDW-700 is titled:

"PDW700 Sony Professional XDCAM®
HD Camcorder"

The Sony PDW-F800 is titled:

"PDWF800 Sony Professional XDCAM®
HD422 Camcorder".

While the title for the Sony PDW-700 camera includes "HD" on the next line, it is in the same class as the PDW-F800 camera which uses "HD422" on the next line. I interpret these to indicate that these two cameras are "XDCam" cameras as opposed to the "XDCam HD" cameras that are 1440 x 1080.

The following is from an offical Sony website for the PDW-700:

MPEG HD422 (MPEG-2 422P@HL) (CBR: 50 Mb/s)
MPEG HD (MPEG-2 MP@HL):
HQ mode (VBR, maximum bit rate: 35 Mb/s)
SP mode (CBR, 25 Mb/s)
LP mode (VBR, maximum bit rate: 18 Mb/s) (Playback only)
MPEG IMX (MPEG-2 422P@ML) (50/40/30 Mb/s)
DVCAM (CBR,25 Mb/s)

Sony Product Detail Page - PDW700 (http://pro.sony.com/bbsc/ssr/product-PDW700/)


The follow is from an official Sony website for the PDW-F800

MPEG HD422 (MPEG-2 422P@HL) (CBR: 50 Mb/s)
MPEG HD (MPEG-2 MP@HL):
HQ mode (VBR, maximum bit rate: 35 Mb/s)
SP mode (CBR, 25 Mb/s)
LP mode (VBR, maximum bit rate: 18 Mb/s) (Playback only)
MPEG IMX (MPEG-2 422P@ML) (50/40/30 Mb/s)
DVCAM (CBR,25 Mb/s)

Sony Product Detail Page - PDWF800 (http://pro.sony.com/bbsc/ssr/product-PDWF800/?XID=O:BPSEG_PDW-f800:Google_OptNet)

--------------------------------------

3. The Sony PMW-EX1, PMW-EX1R, PMW-EX3, PMW-320 and PMW-350 use "XDCam EX".

The following is from an offical Sony site, and describes the Sony EX1R:

Video:
MPEG-2 Long GOP HD HQ mode: VBR, maximum bit rate: 35 Mb/s,
MPEG-2 MP@HL HD SP mode: CBR, 25 Mb/s,
MPEG-2 MP@H14 SD mode: DVCAM
Audio: Linear PCM (2ch, 16-bit, 48-kHz)

Sony Product Detail Page - PMWEX1R (http://pro.sony.com/bbsc/ssr/product-PMWEX1R/?XID=O:BPSEG_PMW%20EX1:Google_PVPA)

Dear Friends,

As I understand it, there are three flavors of Sony XDCam:

XDCam HD, 1440 x 1080 4:2:0
XDCam, 1920 x 1080 4:2:2
XDCam EX, 1920 x 1080 4:2:0.....Dear Dan: I have to tell you that no High School, Community College, or especially ANY University will accept written papers or essays with citations from Wikipedia, because Wikipedia is often presenting erroneous data as factual and accurate.

According to Sony's White Paper data: XDCAM HD is 1440 x 1080 AND 1920 x 1080
**XDCAM = Standard NTSC Definition @ 720 x 480, or 486 and PAL Spec ONLY.
The Nano Flash therefore *must possess a Sony hardware encoder of the XDCAM HD 4:2:2 persuasion. The XDCAM HD spec can ALSO do Standard NTSC Definition. EDIT: XDCAM is 4:2:0 and XDCAM HD is 4:2:2
-------------------------------------------------------------------------------------------------- [/QUOTE]

Hi Dan,

I am out of my league here, but as I understand the principle, the nanoflash pulls the signal from the HD SDI stage and so circumvents the compression by the camera, thus giving the option of compression to the nanoflash. I realize this is a bit of an oversimplification because not all HD SDI signals are the same.

The key to the puzzle would presumably be to compare the HD SDI signals from the XDCAM EX and the XDCAM HD cameras (not the final codecs). If the HD SDI signals are the same, then the request seems to be to have the nanoflash compress the signal from 1920x1080 to 1440x1080.

At that point I am lost because I am not sure why you would do that....???

Alan

Hi Dan:
Another reason why I am thinking the Sony hardware encodrer in the Nano/XDR is of the XDCAM HD 4:2:2 verity, is because the XDCAM regular encoder *cannot ALSO do XDCAM EX. Only the more modern XDCAM HD 4:2:2 encoder can do the extra XDCAM EX 4:2:0 codec as well.

Dear Mark,

For each quote, I referrenced the source, Many of these were from official Sony websites.

I am well aware that Wikipedia has errors.

Sony's labeling is confusing.

In any case, I listed what the nanoFlash can do.

Whether you want to call it Sony XDCam, XDCam HD, or EXCam EX does not really matter.

I have listed the modes that we can do, which includes the formats used by the Sony PDW-330, 350, and 355, and the modes used by Sony XDCam EX (EX1, EX1R, EX3, PMW-320 and PDW-350), and the modes used by the Sony PDW-700 and PDW-F800.

Thus we handle the 1920 x 1080 and 1440 x 1080 as well has 720p 1280 x 720.

Would you like to include a link to the Sony White Paper you reference?

Dear Alan,

We added the 1440 x 1080 mode, to match the output of the PDW-330, PDW-350, and PDW-355 cameras at the request of a major Hollywood studio.

Dear Mark,
Would you like to include a link to the Sony White Paper you reference?Dear Dan: Yes Dan I would. You can download the white paper from http://pro.sony.com/bbsccms/assets/files/micro/xdcam/solutions/XDCAM_WhitePaper_F.pdf

The document confirms what I had been saying many times - that for 50i, 25p the GOP length should be 12, not 15.

Dear Mark,

Sony's labeling is confusing.

Yes. I agree it is.
In any case, I listed what the nanoFlash can do. Yes you did, but I found your post confusing.

you want to call it Sony XDCam, XDCam HD, or EXCam EX does not really matter....Yes I think it does, because depending on which one it is, it can neither do HD or XDCAM EX 1440x 1080 codec. (Regular XDCAM can't). XDCAM HD 4:2:2 hardware encoders *implement one pass realtime VBR encoding.

have listed the modes that we can do, which includes the formats used by the Sony PDW-330, 350, and 355, and the modes used by Sony XDCam EX (EX1, EX1R, EX3, PMW-320 and PDW-350), and the modes used by the Sony PDW-700 and PDW-F800.....You listed the Sony PDW-F800. OK, but that's a well known XDCAM HD 4:2:2 camera, and this is part of what I find confusing.

Thus we handle the 1920 x 1080 and 1440 x 1080 as well has 720p 1280 x 720....OK, and you also can do regular SD and HD .MPG, which is also a cool feature.

The document confirms what I had been saying many times - that for 50i, 25p the GOP length should be 12, not 15.Hi Piotr: I wonder if this is yet another reason why Avid Media Composer has difficulty recognizing the Nano/XDR recorded CF card media in AMA mode ? Avid does not *see* the file structure of XDCAM HD files as the same as XDCAM HD files recorded on discs out of the XDCAM HD cameras.

I have no idea whether this is the reason, Mark, but wanted to thank you for finding out this white paper.

At least now we can start talking facts here, not suppositions, assumption, and wishes...

I hope CD will catch up. Dan is absolutely right when saying that no matter what the names are for standards and codecs, nanoFlash is capable of full raster, 422, at high bit rates (other formats aside). But if this is so (and we all know for the fact it is), this means the nanoFlash codec is in fact the XDCAM HD one - and thus can be fine-tuned and optimized (more optimal GoP lengths, VBR in addition to CBR, etc.).

I really think it's time for some official declaration from CD now.

The nanoFlash uses XDCam, 1920 x 1080, 4:2:2 at all bit-rates 50 Mbps and higher.

Dear Dan,

I read again through your summary of formats/codecs, and I think where you went wrong in your above statement, is when you correctly quoted the PDW-700/F800 as using "XDCAM HD", but interpreted it as XDCAM instead (I have underlined the key fragments of your own quote):


The Sony PDW-700 and PDW-F800 are "XDCam".

The Sony PDW-700 is titled:

"PDW700 Sony Professional XDCAM®
HD Camcorder"

The Sony PDW-F800 is titled:

"PDWF800 Sony Professional XDCAM®
HD422 Camcorder".


AFAIK, XDCAM was the first, Standard Definition, incarnation of the XDCAM "Optical" technology (using Professional disks). The cameras using it have been the PDW-535p or 510p.

And obviously, as the nanoFlash is capable of the 422 HD at bitrates of up to 280, it clearly is of the "XDCAM HD" flavor (and not XDCAM, as you stated earlier).

Also, It's been mentioned many times before that CD recorders use the same encoding chips as the 700 and 800 cameras do.

So all in all, I think all that the article linked to by Mark says is applicable to the nanoFlash - including the VBR option viability.

Dear Dan,

I read again through your summary of formats/codecs, and I think where you went wrong in your above statement, is when you correctly quoted the PDW-700/F800 as using "XDCAM HD", but interpreted it as XDCAM instead (I have underlined the key fragments of your own quote):...I also think it *must XDCAM HD hardware encoder, because the regular XDCAM hardware encoder **cannot do the XDCAM EX codec @ 1440x1080,"" which is also an HD format of course.




AFAIK, XDCAM was the first, Standard Definition, incarnation of the XDCAM "Optical" technology (using Professional disks). The cameras using it have been the PDW-535p or 510p.....This statement is correct.


And obviously, as the nanoFlash is capable of the 422 HD at bitrates of up to 280, it clearly is of the "XDCAM HD" flavor (and not XDCAM, as you stated earlier).....Your deductive reasoning is quite correct.

Also, It's been mentioned many times before that CD recorders use the same encoding chips as the 700 and 800 cameras do....Yes. This is what threw me for a loop as well. These are well known XDCAM HD 4:2:2 hardware encoding cameras, so why does Dan list them and seem to be see, this proves it ??

So all in all, I think all that the article linked to by Mark says is applicable to the nanoFlash - including the VBR option viability....I thank you Piotr for getting to the heart of the matter here. It doesn't matter about listing a whole bunch of XDCAM cameras, or going into what the Nano Flash does- ***The heart of the matter is does the Nano Flash (and by extrapolation, the XDR) possess an XDCAM HD 4:2:2 Full Raster hardware encoder, and thus the Sony specification calls for the implementation of realtime hybrid V B R encoding. Will CD enable this feature ?***

Hi Piotr,

If I understand correctly, you anticipate that reducing the number of pictures in the GOP will reduce the shimmering you observe because the reference image will be renewed more often. Have you observed this shimmering in the final 1920x1080 product? Is it mostly on shiny objects? I am not sure what to look for.

What problem will the variable bit rate solve?

Many thanks,
Alan

Dear Alan,

If you read the entire thread, you will find answers to both your questions. The VBR encoding superiority over CBR can also be studied in numerous Internet sources. The 12 frames long GOP has long been the industry standard for 50i/25p/24p.

One thing I'd like to add: While I understand CD is currently very busy with maintaining the current firmware quality and reliability (and hopefully, also with introducing the long promised features like card hot-swapping or SDI-embedded + analog audio mixing), I still hope that once they're done with this (and the "important project" they've chosen never to discuss on this official forum of theirs), they will look into optimizing the code so that the maximum quality is squeezed out of the Sony encoding chips.

Piotr

Hi Piotr,
During my testing of the NanoFlash I also observed the same shimmering when recording Mpeg Long Gop 1920x1080. The same scene was also recorded in I-Frame and the shimmering was no longer an issue, now I record everything in 100Mbps I-Frame as standard.

Lance, please give it a closer look (and on a really large display, too - your clients may be using huge displays):

- 100 Mbps is way too low for I-Frame Only. In this mode, I recommend 220 and up.

I do not know if this is of any help but here is a screen grab of the pre set from CS5.

Dear Friends,

I just returned from a 10-day business trip which included a 3D-Workshop.

It is hard to determine the official "Family Name" for Sony's PDW-700 and PDW-F800 cameras.

For the PDW-F800, they put "XDCam" on one line and on the next line they put "HD422" and list the codec as MPEG-2 4:2:2P@HL.

For the PDW-700, the Put "XDCam" on one line and on the next line they put "HD".

Both the PDW-700 and PDW-F800 use the same codec module, and we use the same codec module.

The white paper presented by Mark is from 2007, and it appears that was written before the introduction of the PDW-700 and PDW-F800.

Ron Little's post shows that Adobe CS5 makes a distinction between XDCam HD, XDCam EX, and XDCam HD422.


I know that for 720p we use a Group of Pictures (GOP) of 12.

As I understand it, for 1080, we use a GOP of 15.

If this is wrong and causing any problems, we will look into it.


We also use a "Closed Gop" and I feel that we should. Otherwise one GOP will depend of the images from the previous GOP.

We use VBR where Sony uses VBR. Otherwise we use CBR, the same as Sony, as far as I know.

Adobe - Premiere Pro CS5: Third-party hardware compatibility (http://www.adobe.com/products/premiere/search_result.html?manuSearchVar=1&modelSearchVar=0&supportSearchVar=0&brand=&dataFormat=&version=CS5&device=cameras&format=NTSC&buttonSubmit=)

Dear Friends,

For 720p we use a GOP of 12.

For 1080p29.97, 1080p30, 1080i59.94, and 1080i60,we use a GOP of 15.

For other 1080i or 1080p frame rates we use a GOP of 12.

Thus 1080p23.976, 1080p24, 1080p25 and 1080i50 are a GOP of 12.

This information was obtained from our Chief Engineer.

Dear Friends,

We also use a "Closed Gop" and I feel that we should. Otherwise one GOP will depend of the images from the previous GOP.

That means that any GOP will use his own I frame and the one before as reference.
Two "I" frames instead of one.
Thats GREAT.
Professional MPEG-2 software compressors by default uses Open GOPs.
rafael

We use VBR where Sony uses VBR. Otherwise we use CBR, the same as Sony, as far as I know.

Dear Dan,

My understanding is that XDCAM HD as used by Sony, has only 4 bitrates available:

- 18 Mbps (LP mode, VBR used to ensure maximum recording times)
- 25 Mbps (SP mode, CBR used to ensure compatibility with HDV)
- 35 Mbps (HQ mode, VBR used to ensure maximum quality)
- 50 Mbps (4:2:2 mode, CBR used to ensure compatibility with the legacy HD422 format)

Since nanoFlash can use higher bitrates as well, and their very purpose is quality - all Long-GoP modes above 50 (like 80, 100, 140 and 180) should be VBR, if this is only possible from the chip programming viewpoint. As no such modes exist in Sony XDCAM HD, compatibility is irrelevant. The average bitrate would give the user an estimate of the recording time in a given mode, while the maximum bitrate actually used by the chip could in all cases go as high as the CF card can handle.

I guess this is what someone called a "Constant Quality" mode earlier in this thread.

Dear Friends,

I get conflicting information when I search on-line for Open versus ClosedGOP's.

In some cases, they indicate that the Closed GOP does not referrence previous frames.

In other cases, they indicate that the "B" frame at the end of an Open GOP can reference the next frame, thus allows this one "B" frame to be slightly smaller. Tektronix's website lists this description.

What is a closed GOP? > Frequently Asked Questions : Tektronix (http://www2.tek.com/cmswpt/faqdetails.lotr?ct=FAQ&cs=faq&ci=6372&lc=EN)

I feel that this is most likely correct and this is different than what I previously posted.

Since, if one is using our MOV or MXF files for editing, I feel that Closed GOP is better as it does not need the next GOP to decode the previous GOP.

But, for creating a DVD, and not editing, using Open GOP's may be ok.


Here is another link to a discussion of the differences:

GOP closed/open ???? What does it mean ? [Archive] - Doom9's Forum (http://forum.doom9.org/archive/index.php/t-105129.html)


Here is my position:

In some limited cases, such as when creating a DVD, without multiple angles, and without the need to fast forward, the Open GOP, offers some slight advantages.

However, when editing the files, I feel that Closed GOPs are definitely better.

I am reluctant to change from Closed GOPs to Open GOPs for a slight advantage, which may cause problems in editing our files.

If we decide to try Open GOP's it would require substantial testing on on part, with every editor and condition to prove that it did not cause problems, and this would distract us from delivering other features that have been requested.

This is a moot point when using our I-Frame Only files as this applies only to our Long-GOP files.

Dear Piotr,

For 50 Mbps and higher 4:2:2, we follow the Sony codec standards, which is CBR for compatibility reasons.

We have proven that we can increase the bit-rate, and the professional NLE's, other than Avid, handle our Long-GOP files at these increased bit-rates, and Avid handles our 50 Mbps Long-GOP files and all of our I-Frame Only files.

We have not (yet) proven that any NLE can handle the Sony 4:2:2 codec if we modify it to be VBR.

The Sony XDCam EX codec is VBR, thus we use VBR for that codec.

At some point in the future, we could experiment, but this would require quite a lot of testing to ensure that all NLE's will work with VBR.

At this time, we need to concentrate on the other features that have been requested and we have promised.

Dear Dan,

I realize introducing VBR now would take a lot of testing; I just hope CD will keep such a possibility in mind when you are done with the more urgent things....

Thanks,

Piotr

Dear Dan:
The information I have at hand from Sony indicates realtime, one pass VBR to be part of the XDCAM HD standard. I still don't understand what type of encoder we have in our recorders from CF, but you wrote that it doesn't matter, so I'm assuming it could be done (??). Perhaps some experimental beta test software could be put out (Which I would be happy to test in my spare time to see what happens in Avid NLE with Long GOP and whatever ?

The information I have at hand from Sony indicates realtime, one pass VBR to be part of the XDCAM HD standard. I still don't understand what type of encoder we have in our recorders from CF, but you wrote that it doesn't matter, so I'm assuming it could be done (??).

This is exactly what I meant when stating:

Since nanoFlash can use higher bitrates as well, and their very purpose is quality - all Long-GoP modes above 50 (like 80, 100, 140 and 180) should be VBR, if this is only possible from the chip programming viewpoint. As no such modes exist in Sony XDCAM HD, compatibility is irrelevant.

And it certainly can be done, the 35 Mbps VBR being the proof....

Do VBR files exist? May I have a special firmware that would allow my Nano to make them? I'll sign up as a tester for Sony Vegas - my regular platform.

Gints,

The nanoFlash is already capable of VBR - in the XDCAM EX mode (35 Mbps).

Gints,

The nanoFlash is already capable of VBR - in the XDCAM EX mode (35 Mbps).Dear Piotr: I suppose we have already at least one mode in which VBR is encoding away, so I should go test this. Personally, I've never used the EX Mode on my XDR. I guess it's time to go try it. :-) Isn't there also an 18 Mbps mode ?

Dear Friends,

This is from Piotr's post (I added the item numbers):

1 - 18 Mbps (LP mode, VBR used to ensure maximum recording times)
2 - 25 Mbps (SP mode, CBR used to ensure compatibility with HDV)
3 - 35 Mbps (HQ mode, VBR used to ensure maximum quality)
4 - 50 Mbps (4:2:2 mode, CBR used to ensure compatibility with the legacy HD422 format)

Items 1 & 2 are XDCam HD. We do Item 1, we do not do Item 2, as far as I know, as we do not support 25 Mbps.
(Note, this is not XDCam, or XDCAM HD422.)

Item 3 is XDCam EX and we do this item, and it is VBR.

Item 4 is XDCAM, or XDCAM HD422, but it is not XDCam HD. We use this codec for 50 Mbps and higher, 4:2:2, CBR.

Our I-Frame Only also uses the codec in Item 4, but we put out only I-Frames.
Sony does not have an I-Frame Only option in their current cameras, as far as I know.

I feel confident that we have the technical ability to turn on VBR in our 50 Mbps and higher 4:2:2 modes.

But, there is a price to pay.

We either have to prove that this VBR mode is acceptable to all NLE's and this becomes our standard, or have to support both CBR and VBR.

If we have to have both, then this adds a level of complexity and an opportunity for someone to create a file that will not work with their editor.

And, not everyone knows which NLE will be used when the files are recorded.

If we need to have both, then it will take us substantially longer to test each new release as we would be almost doubling the testing that we do before each release.

This quality control testing takes one employee, running many nanoFlashes simultaneously, about two weeks or longer to test each and every variation that we have now.

Each file type, each bit-rate, each variation is recorded, then individually tested in each of the professional NLE's.

Multiply 2 (MOV and MXF) by the different frame rates we support, (23.976, 24 25, 2997, 30, 50, 59.94, and 60), by the number of codec type / bit-rate variations (I-Frame Only, 100, 140, 180, 220, 280, plus Long-GOP 18,35,50,80,100,140,180), and you get the number of files that we have to create for testing (approximately).

Then add the testing for all of the bit-rates, etc. for MPG mode.

Then take each appropriate file and test it in the various NLE's.
(For Avid we need to only test some of the Long-GOP files).

Now, when we find a problem, we fix the problem in our code, then start testing every option over again.

Thus, at this time, we want to concentrate on the other features that have been requested.

Please remember that the nanoFlash is an extremely versatile device. No camera, or device that I know of offers as many recording options / flavors / bit-rates, frame rates, etc.

I hope this explains fairly why we need to concentrate on the other features that we have promised.

Dear Dan,
As I said before somewhere, for me VBR makes no sense when Data Rate is not an issue.
VBR makes sense when you have to put 3 hours MPEG-2 in a DVD.rafael

I have been losely following the thread since its beginning. Nothing wrong with trying to learn and improve.

I do wonder about this proposal to move the VBR long GOP recording. Has anybody measured how much better, I mean noticeably better for all of the hassle it would take for CD to add it?

Or is a spec improvement that would improve the 200% freeze frame blowups?

Because unless there is a visual improvement there is no point in following this path imho.

Dear Rafael and Tim,

I agree with both of your comments.

At high bit-rates I do not expect much of a difference.
(But, unless we test it, we do not actually know, and I think that would be Piotr's position, which is valid.)

At SD DVD bit rates, it could make a difference, in my opinon.
(But so far, we have not been including our MPG recording in these discussions)
And I do not know if our MPG is VBR or CBR at this moment. I am checking on this.

Dear Friends,

Currently, our MPG files are CBR.

In our upcoming release, our next firmware release, they will be VBR.