Does anybody know if this new Sony will output HD footage through SDI or HDMI? And what kind of quality that output would be? 4-2-0 or 4-2-2?

Can't speak to the NanoFlash capabilities, but a quick look at the camera specs shows it's a 60i interlaced camera - no progressive frame rates so color space isn't really that much of an issue.

As I understand it the 60i or 50i is only a 'wrapper' and can easily and without loss be turned into 25 or 30 p. Is that correct?
And, my main question, could this camera 'feed' my NanoFlash?

Dear Roelf,

Real Interlaced footage originates at the sensor.

Two "Fields" are produced at separate times. One with the odd scan lines, one with the even scan lines. Thus, the sensor is exposed twice.

The two fields make up one Frame. In 1080i60 (True), there are two fields every 60th of a second and one frame every 1/30 of a second.

(1080i60 is fairly rare, 1080i59.94 is much more common, and is the typical format even when one discusses 1080i60 unless one says something like 1080i60 True).

Read Progressve footage originates at the sensor.

There is only one "Frame", where all of the scan lines are scaned progressively at one time.

CCD sensor have a Global Shutter, thus they actually expose the frames or fields all at once.

CMOS sensors generally are still exposed to the light while they are scanning.

One can deinterlace footage in post, but it can not overcome the fact that with a true interlaced sensor the even scan lines were obtained from a different point in time from the odd scan lines.

It seems to be like this: this shoots progressive segmented frames, not interlaced, two identical streams of progressive material in the container. So no deinterlacing needed, no blending two streams, they’re both identical.
Could it be a good camera to combine with the NanoFlash?

Dear Roelf,

If it is Progressive Segmented Frames, then the nanoFlash should be able to record it.

PSF is used when the sensor and electronics create progessive frames, but the video is to go out HD-SDI (or maybe HDMI).

PSF takes the odd scan lines and puts them into a field, and the even scan lines in a field, both are in one frame.

The nanoFlash can take PSF and convert it back to true progressive before it is recorded. This is what our "Record PSF>Prog(ressive)" menu option does.

We would be very happy if this camera puts out a high quality image via HDMI.

ProVideo Coalition.com: FRESHDV by Matthew Jeppsen & Kendal Miller (http://provideocoalition.com/index.php/freshdv/story/sony_nex-vg10_announced/)

In this review you can see that the HDMI signal from the Nex-VG10 is uncompresed, so will work perfect with nanoflash.

www.gecoproducciones.cl

Dear Friends,

The absolute best assurance that the nanoFlash will work with this camera is for us, or anyone else with a nanoFlash and access to the camera to test it.

Dear Friends,

Our friend, Alister Chapman has a mini-review of the NEX-VG10 on his website XDCAM-USER.com (http://www.xdcam-user.com) (Go to Alister's Blog).

Alister reports that the camera definitely shoots progressive and the output is progresive in a 50i wrapper. I assume that it also supports a 60i wrapper.

So, far me, the question is if it is outputting Progressive Segemented Frames, which appear to be interlaced to a monitor or TV.

Dear Dan,
Do you have more information about the NEX-VG10 an the posible conection with NanoFlash?

Dear Gerardo,

I have not found a review indicating the quality of the HDMI output. You may know of such a review.

If it is clean, and if it is of high quality, then the nanoFlash would be ideal for use with the NEX-VG10 camera.

Dear Dan,
But you know about some serius test with nanoflash? some video to download on the web to test the resulting conbination?

Dear Friends,

Late last night I received an email from a person that has both the Sony NEX VG10 and the nanoFlash.

The VG10 puts out a Green "Standby" Dot on the HDMI output when the camera is not recording.

When the HDMI output is in use, the LCD and the Viewfinder go blank.

HDMI never carries timecode as this is not included in the HDMI spec.

One must press record on the camera to get rid of the Green Dot and then manually start the nanoFlash.

There is controversy as to whether the HDMI output is before or after compression in the camera.

It appears, from examining the footage it is after processing in the camera as artifacts in the recorded images (in-camera recording) are also in the nanoFlash recordings.

Please consider all of this information to be preliminary. We have the results of one reliable tester, but we need others to confirm these results.

Here is the original email, reproduced here with permission:

I just came across the thread at DVINFO.NET concerning the use of a Nanoflash with a Sony NEX VG10. I thought I'd drop you a line to let you know that I purchased a VG10 a few weeks ago and have been testing recordings with the VG10 HDMI output port into the nanoflash.

There are clams that the HDMI out of this camera is pre-compression 4:2:2. In my testing, I have observed aliasing artifacts in the native AVCHD recordings, as well as in the MXF mpeg2 files, as recorded by the Nanoflash. This would indicate to me that the HDMI output stream has already been processed by the VG10 processor.

I would also note that a 1:1 scaling and comparison of the native vs. Nanoflash files shows, quite clearly to my eyes, that the resolution of the Nanoflash recorded files exceeds the native AVCHD file until there is significant frame motion, such as a camera pan. With full frame motion, the Nanoflash files appear to suffer a degredation in image resolution as compared to the native AVCHD files.

I should note that I recorded at 220Mbps, I-frame only mode. Both files were read into Avid Media Composer v 5.03. The AVCHD files were transcoded to DNxHD, the Nanoflash files were linked via AMA. and the two images were compared after applying a scale filter.

I'm not sure if this will help you any, however, I thought I'd give you the info as I've seen. Somewhat subjective results, I admit, but, this is what I've observed.

Dear Dan,
Can you tell your friend to make this test: "go in to a dark room with a big lamp, start recording in both, camera and nanoflash, then turn on and off every 3 seconds the big lamp, must to be a lamp that can be turn it off and on fastly, that way you don't give time to the codec to process all the information about colors and shadows; then compare the two recorded footage, specialy in the moments that light turn on and off; in the camera footage tou must find some compresion blocks; if you have the same in the nanoflash footage, then the HDMI is after-compression, if is not is pre-compression." I must to tell you that the "aliasing artifacts" is a problem on the down-scale process from the big senssor to HD; this senssor is almost full 35, so to accomodate to HD 1920x1080 must to be processed in a down-scale way, and aliasing artifacts is not a codec compression problem. Please ask your friend to do the test and let us know. Thanks

Dear Gerardo,

I forwarded your request this morning.

I'm curious about this too.

I believe the VG10 to be seriously compromised as a video camera because i don't believe it has a OLPF designed for video. I suspect that the aliasing spoken above a few posts ago is not compression but aliasing moire. I've seen this on some full size sample videos.

The key to this is the fact that sony says you can take stills with it! 14MP stills. Well if you can take stills then it doesn't have a filter so it's going to be recording as much aliasing as the little cameras.

(i've been thinking about getting a mini nex for a walk around camera with rangefinder lenses but i've been curious about the VG10 too)

cheers
paul

the problem is not the full size sensor, you can see RedOne or others cameras that take the image from the sensor and record it like it is in full size or 4k and don't pressent moire or alliasing, but in the down-size process the image sufer this problem; there is in the market some filters to eliminate or reduce alliasing, and work very well. But the thing I must concern about it is if the HDMI output is really pre-compression or not.

Gerardo,

The RED sensor is completely different to mass produced dSLR sensors like NEX, massive price difference too.

I don't believe there's a photo orientated sensor that can read out fast enough to scale full resolution frames down to HD size. At the moment they're all skipping lines/columns in order to get the data rate needed for 24/25/60 fps.

Panasonic seem to be closest with their m43 but i don't think even that is scaling, perhaps the GH2 is?

Am i wrong?

As for the HDMI i would love to know. Can someone just plug their NEX into a TV and wave it around and look for compression artefacts!? I'll grab one of those little NEXs soon enough and try myself.

cheers
paul

I don't believe there's a photo orientated sensor that can read out fast enough to scale full resolution frames down to HD size. At the moment they're all skipping lines/columns in order to get the data rate needed for 24/25/60 fps.

Thomson Viper camera sensor is 3x 9.2 Megapixel (That's 27 megapixels of data), scaled to 1080p in realtime.

If a DSLR took 27 megapixels of data and scaled it to 1080p at video frame rates, in real time, it would probably be as big, expensive and power hungry as a Viper.

DSLR do an amazing job for what they are though.

Thomson Viper camera sensor is 3x 9.2 Megapixel (That's 27 megapixels of data), scaled to 1080p in realtime.

If a DSLR took 27 megapixels of data and scaled it to 1080p at video frame rates, in real time, it would probably be as big, expensive and power hungry as a Viper.

DSLR do an amazing job for what they are though.

I did say excluding dedicated video sensors :)

Although that's 3 separate sensors, whereas we're talking single 14 megapixel compared to a mere 9 of the viper and a dSLR has to debayer full res *before* it can scale. Plus i wonder whether the viper sensor combines the data in hardware before the readout. You never get to see the full res from them.

cheers
paul

Dear Dan,
in the camera footage tou must find some compresion blocks; if you have the same in the nanoflash footage, then the HDMI is after-compression, if is not is pre-compression." Thanks
I don't think the HDMI would out put "post-compression".
That means would mean that the processor would be compressing (to AVCHD on recording) and de-compressing (AVCHD to HDMI) at the same time time. Too much, IMO, for that processor.
Would be easier to route that signal directly to the HDMI.
rafael

Dear Rafael,

I agree.

It is logical to assume that Codec Modules work in either encode or decode mode, not both encode and decode at the same time.

I know for a fact that our Sony Codec Module can do encode or decode, not both at the same time.

Dan
Your bottom line on Nex VG10 out via HDMI to nanoflash? Compared to say HV20/30 up to EX1 or 3.
Terry

Dear Terrance,

I would love to help, but I have not tested the Nex VG10, sorry.

Dan
Your bottom line on Nex VG10 out via HDMI to nanoflash? Compared to say HV20/30 up to EX1 or 3.
Terry
Hi Terrance,
I've shot a lot of material with HV40 with NF at 50Mbps and clean EX1, a whole documentary that way and the end results of HV40 with NF at 50 are much much better then EX1 at 35 Mbps. The image, the quality and the colours. Reason is HV40 via HDMI provides 8bit 4:4:4 colour space uncompressed (if I remember corectly), compared to EX1 4:2:0. That is a huge difference increadably noticible. I use HV40 with DOF adapter or naked with zoom half way in for off focus (DOF ilusion) back ground. HV40 does shoot 24F (NF does the pulldown) and I have 23.976 on the NF. I believe HV40 was the best and last of Canon, with so many fitures in that smal price. HV40 with NF it's an increadable combination for a second camera and places you can't place your large primery recording unit (car trunk for instance) and it is much much better choise then 5/7D for video.
Cheers

Luben
Thanks for the reply. Good information. I have been using my HV20 and NF for small inserts on some TV shows that where shot with HDcam and agree how great the combo is. I now have the NEW VG10E (25p) and intend to use it for a feature project with the NF. I jumped the gun and bought the camera and now have to see if it was a good choice. Sony held back a few things that Canon managed to give freely with HV series. The NEX VG10 turns off EVF and LCD wen connected to HDMI being the obvious one, the 24p being the other. I wanted the bigger sensor and removable lens. Even when I get one of the new EVF from redrock it is less expensive than the AF100. I am looking for input now on how best to make my new system work. Maybe best if I start a new thread and see if there are others out there trying to do the same.

"The NEX VG10 turns off EVF and LCD wen connected to HDMI being the obvious one, the 24p being the other. I wanted the bigger sensor and removable lens. Even when I get one of the new EVF from redrock it is less expensive than the AF100".
The only similarity between the SONY and the PANA is about having "bigger sensor and removable lens".
For the rest, the SONY is a domestic camcorder while the PANA is a professional machine.
rafael

I shot some ftg on Saturday. Not perfect test situations. Just seeing if my rig worked.
NexVG10E (25p)
Canon FD lens adapter with Tokina 28mm/2.8
NanoFlash (140mbit, I Frame at 25p)
Lower-end Marshal monitor.
Shot at 5.6 with no ND, camera was set to manual with 0db gain and 50 shutter.
It is a brief shot from CU to WS through two chain-link fence with people walking by. I was looking for artifacts and morie. I exported two versions from original one uprez to Prorez HQ size: 750 mb and one XDCAM HD422 at 50mbits size: 220 mb. I can also do an output to order. I can post or am willing to "yousendit" it to a few interested people.
To my eye on a 120 " projection screen I was impressed with the colour and lack of artifacts. If anything, I think the monitor was the poor player. I hope to fix that with one of the new EVF coming out.
Terry

pleasse can you send the file via yousendit to videos@gecoproducciones.cl or put it in vimeo, so I can test it, because I want to buy the nex-vg10, but I need some power reasson to do it and work with my nanoflash. Thanks

sorry, I didn't mention, I work in PC no Mac, so when I tryed to see the file that you send me nothing happened; can you send me the MPEG xdcam file please; thank you so much, videos@gecoproducciones.cl

I am sending it this afternoon. It's in the original codec that the NF captured.
Terry

Hey Terrence, i am considering a nanoflash purchase to complement my VG10. Could you send me a yousendit link to me as well? I would love to take a look at the footage you shot. If so, my email is:

jburkhimer1@cox.net

Thanks!!

-burk

Jason
I will send it along. This ftg is limited. and at 25p. I used it to check out aliasing and morie in the field not the lab.
Terry

For anyone else here is the "yousendit" link it's big over 500 mb. I don't know if this works, just pasting the link. It will be up for 14 days.


Link to file:
https://download.yousendit.com/cEd2NnFEVEh6RTljR0E9PQ

I runing many test with my new NEX-VG10 and Nanoflash, they comunicated perfect. The best performance betwen is to put the Nanoflash in PSF progress check: yes; and the nanoflash recognized the signal as 29,97p not 59i; then undercrank to 24p, then you have a beatifull 24p from your Nex-VG10 in the nanoflash quality.
www.gecofilms.cl

Dear Gerardo,

Thank you very much for posting your results.

I am assuming that the HDMI output is clean and you like the qualiity of the images produced by the Sony NEX-VG10 camera + nanoFlash.

This means that, the NEX-VG10 with the nanoFlash, one can record in

1080p24 (23.976p),
1080p30 (29.97p) and
1080i60 (1080i59.94)

If the camera also supports 1080i50, then the camera + nanoFlash will be able to record

1080i50 and
1080p25 and
other frame rates.

Since you were using Over and Under-Cranking to achive 24p, you may not have audio.

If you record 1080p30 (1080p29.97) without Over and Under-Cranking, then you will have audio.

In our Over and Under-Cranking modes, we support audio at certain crank rates, but not all.

In our PSF to Progressive modes, we support audio in every case.

Gerardo and Dan
This great news. Sounds like I jumped the gun buying the 50i/25p Vg10. However, do you think I could get 23.98 or 24p buy undercranking on the nanoflash? BTW I love the image you get with VG10 - nanoflash and good old prime lens with adapter. Closest I have seen to super16 or 3 perf-35 at an indie budget. And alot lighter than an Arri 2C.

Dear Terrance,

Yes, I think you can get 1080p24 from your 1080i50 version of the NEX-VG10.

I do not think you can get 1080p23.976 by using our undercranking feature, however, just 1080p24.

We would have to test this to be certain.

The other question is whether we would have audio or not.

Dan
Thanks. I will give it go. I had planned to shoot at 25p and then in post slow down to 24p and do the 4% audio adjust to get the sync and pitch. It work with my last feature and the Europeans have been doing it for years. However getting 24p would be the ticket. As the audio when under or over-cranking: does the audio just dis-engage?
Terry

Dear Terrance,

We only support audio in a few crank options.

I need to discuss this with our Chief Engineer.

Each and every cranking combination (Base Frame Rate + Crank Rate) requires special code in the nanoFlash to support audio.

We may, in the future, be able to support more of the popular combinations with audio.
This is not a promise at this time.

To answer your question, "Does the Audio Just Disengage":
I think this depends on the combination.
I will have to get some more information from our lab before I can answer your question accurately.

Dear Dan and Terrance

hey, vg10 + nanoflash are a great companion togeter: test nanoflash + nex-vg10 on Vimeo I just confirm that the signal is precompretion in a chroma key test; and you can pull out 24p from your vg10 in the nanoflash, you have to turn on the PSF recording mode in the nanoflash and then undercracking to 24p 24, and the audio is sincronic too.
www.gecofilms.cl

Gerardo
Did you get any sense of how much it is pre-compressing, like HDCAM? Your end result, they are Progressive Segmented Frames then?
Terry

Dear Terrance,

While I am not in Chile performing the tests with Gerardo, I am certain that the output of the NEX-VG10 is Progressive Segmented Frames

I read Gerardo email wrong, so that's great - PSF 24P, precompressed and good chance of audio sync. For the price and knowing how Sony is a bit tight with features, the camera with the feature packed nano is a great deal!

Gerardo, what are you thinking of using for a field monitor? Bought a lower end 7" marshal but not the greatest. Are you going to look at the Redrock or Zacuto EVF when they get out?
Terry

Dear Terrance,

As I understand your posts, by Precompression, you mean that the signal out of the NEX-VG10 is uncompressed.

And yes, that would be true.

I expect the signal out of the Sony NEX-VG10 to be 4:2:2 uncompressed.

It could actually be 4:4:4 uncompressed also.

The Canon HV30, HV40 and possibly others actually output 4:4:4 over HDMI and the nanoFlash accepts and processes the signal using only the 4:2:2 material for recording.

Yes, uncompressed is what I meant. With regards to the HV20. I have been using one for a few years and with the nano in the last while. The Nex VG10 beats it out sensor-wise big time. Just in detail alone. I am going to start playing with the audio on nanoflash. Only need mono or 2 track at best. Any tips for compact powered submixers to the 1/8 input on nanoflash?

Dear Terrance,

The nanoFlash can accept:

One Balance Mic/Line Input

Two Unbalanced Mic/Line Inputs

These analog inputs are recorded at 24-bits / 48K.

The nanoFlash does not provide Phantom Power.

Thus, certain mics can be connected to the nanoFlash directly, others may need a cable, or others may need an adapter that provides Phantom Power.

The Juicedlink and Beachtek products could solve certain scenarios.

But many situations may not need an adapter.

So, what mics or audio sources do you have, so I can be of more assistance?

Dear Terrance,


The Canon HV30, HV40 and possibly others actually output 4:4:4 over HDMI and the nanoFlash accepts and processes the signal using only the 4:2:2 material for recording.

Dear Dan,
8b 4:4:4 YUV means 2.5 Gbs out of the HDMI, so very close to a Dual-link SDI.
Can you confirm that?
Best,
rafael

Dear Rafael,

I can confirm that Canon informed us that the output of the HV30/HV40 is 4:4:4 and this is via HDMI.

I have not personally tested this.