NanoFlash, Ninja or the AJA Ki Pro MINI?

[xmephestox]

Thanks for the explanation and clarification, but not sure I understand your concept of oversampling. As an engineer, I understand how you can oversample an analog signal, for example use a 12-bit A/D converter to produce a very accurate 10-bit digital result. But this same concept does not apply to a digital signal like HD-SDI.

By definition, HD-SDI always has a 10-bit word length. However, the actual video signal is not required to have 10 effective bits. In the case of many (most) cameras, the lower two-bits are simply set to zeroes. So if you capture 8-bit video using a 10-bit CODEC, all you will do is compress the lower two zeros, there is no gain on the quality of the video. You can't oversample data that does not exist in the first place.

Mike Schell

I'm not saying you are gaining information that isn't there, in the form of oversampling i'm describing here comes from a workflow where you overshoot the existing source quality and information so when you start pushing it around through the vigors of post, it has more room to get pushed around.

8 bit color gamut gives you a 0-255 amount color information, 10 bit gives you about 4x more. Converting your 8 bit source to 10 bit, will stretch out the 0-255, it's the same amount of color, but when you start let's say, adding blue to balance your very warm scene. Working natively with a 8 bit codec a slight nudge and you may move let's say for the sake of saying it, 20-30 increments in the color. Stretching it out will let's say let you move in increments of 5 rather than 20. (the math is not exact i'm just saying it as a brief example) Now the you may not have as many shades of blue, but you have finer control over what you do have.

It's the same when it comes to your shadows and luminance. You gain more control, not information, which is why in a lot of VFX software, they work in 32 bit float with their filters, plugins, etc. even when they are only work with a 8, 10, 12 bit source. They need the extra head room to make finer calculations, they are clear and proven results.

You can say well you can convert in post, and that's true, but that's transcode time you could have got in one step, on set, without redundant hard drive usage, and adds a generational loss. Now you can sit there and just look at the footage comparison, and start at avchd native, nanoflash xdcam intra frame, and pro res all day and not see much of a difference if any, nor am I saying that you will. But as an editor, and colorist I always want to be working with the highest quality footage I can, which every weapon in my arsenal as allowed, and it's my firm belief that productions if solid high standards should want that as well.

Even if you bring 8 bit footage to a finishing house to get it color corrected, they'll always convert it to 10 bit before working on it. The advantages are there, even if it's just a technical stand point.

 

[Steve Kahn]

Laptop for capture???

Laptop for capture???

My laptop has an e-sata port.

Would it be theoretically possible to capture just using a laptop and encoding software via laptop? HD-SDI to e-sata

 

[Amr Rahmy]

I'm not saying you are gaining information that isn't there, in the form of oversampling i'm describing here comes from a workflow where you overshoot the existing source quality and information so when you start pushing it around through the vigors of post, it has more room to get pushed around.

8 bit color gamut gives you a 0-255 amount color information, 10 bit gives you about 4x more. Converting your 8 bit source to 10 bit, will stretch out the 0-255, it's the same amount of color, but when you start let's say, adding blue to balance your very warm scene. Working natively with a 8 bit codec a slight nudge and you may move let's say for the sake of saying it, 20-30 increments in the color. Stretching it out will let's say let you move in increments of 5 rather than 20. (the math is not exact i'm just saying it as a brief example) Now the you may not have as many shades of blue, but you have finer control over what you do have.

It's the same when it comes to your shadows and luminance. You gain more control, not information, which is why in a lot of VFX software, they work in 32 bit float with their filters, plugins, etc. even when they are only work with a 8, 10, 12 bit source. They need the extra head room to make finer calculations, they are clear and proven results.

You can say well you can convert in post, and that's true, but that's transcode time you could have got in one step, on set, without redundant hard drive usage, and adds a generational loss. Now you can sit there and just look at the footage comparison, and start at avchd native, nanoflash xdcam intra frame, and pro res all day and not see much of a difference if any, nor am I saying that you will. But as an editor, and colorist I always want to be working with the highest quality footage I can, which every weapon in my arsenal as allowed, and it's my firm belief that productions if solid high standards should want that as well.

Even if you bring 8 bit footage to a finishing house to get it color corrected, they'll always convert it to 10 bit before working on it. The advantages are there, even if it's just a technical stand point.

thank you for that.

and that 8bit to 10bit comparison thing. the one with the zeros at the end of an 8 bit sequence(not sure if it was on this thread), to be polite was questionable.

 

[GRENCH]

I belive the AJA is going to be my choice!

 

[Jim Brennan]

For those of us already in the P2 world, is the HPG-20 an option?

 

[trez]

Wait - did Jan confirm 8-bit? I hadn't heard that...
...Disappointing if it ends up being 8-bit, I'd really rather it be 10-bit.

Panasonic stated it clearly in their presentation:

http://mirrorlessforum.com/microfourthirds-rumors/2755-download-ag-af100-presentation-file.html

Really disappointing, but... I'm gonna buy one anyway...

Speaking about Ninja vs AJA - don't really understand why people consider touchscreen to be an advantage. I'd much prefer to have solid buttons. No accidental stops etc... Ninja still seems appealing, though (considering its price). If it had HDMI passthrough I'd probably get one to save some $. Without this feature I won't be able to connect my DP1 monitor

 

[trez]

Thanks for the explanation and clarification, but not sure I understand your concept of oversampling. As an engineer, I understand how you can oversample an analog signal, for example use a 12-bit A/D converter to produce a very accurate 10-bit digital result. But this same concept does not apply to a digital signal like HD-SDI.
Mike Schell

You both seem to confuse things. Oversampling means using higher sample rate - in case of video it relates to resolution. Bitdepth of single pixel has nothing to do with oversampling. Converting 8-bit to 10-bit should be called zero-padding. Of course it doesn't produce new information.

 

[Barry_Green]

For those of us already in the P2 world, is the HPG-20 an option?

Yes, and it's the option Panasonic is promoting for those who want a higher-end recording format. You would get two of the major benefits of AVC-Intra, being the higher bitrate/better quality compression, and the 4:2:2 color sampling. But you wouldn't be able to take advantage of the 10-bit bit depth that AVC-Intra is capable of.

 

[GRENCH]

Yes, and it's the option Panasonic is promoting for those who want a higher-end recording format. You would get two of the major benefits of AVC-Intra, being the higher bitrate/better quality compression, and the 4:2:2 color sampling. But you wouldn't be able to take advantage of the 10-bit bit depth that AVC-Intra is capable of.

Is the Panasonic unit as portable and lite weight as the AJA, Nano or Ninja for that matter? It looks a bit bulkier and is a bit more expensive at $4,000.00.

 

[GRENCH]

I may be switching from the AJA mini to a Nano Flash. Nano just looks alot more portable and customizable.

 

[James Bridges]

"I may be switching from the AJA mini to a Nano Flash. Nano just looks alot more portable and customizable."

I think you will really like it. I have had mine for over a year now and the footage looks great. Contact me if you have any questions.
Jim

 

[Yak]

Has anyone compared the Codec between these recorders? I Know the AJA mini records in ProRess while the Nano is based on Sony's Mpeg2 Codec. While size does matter if one will be using a external recorder on the AF-100 the the Codec of the recorder should also be considered.

 

[Bandido]

Panasonic AG-HPG20 msrp is $5,560, the average street price is $4,520. CD Nanoflash msrp is $2,995, AJA Ki pro mini msrp is $1,995 and the Atomos Ninja msrp is $995. As you can see, they all compete at different price levels. The closest competitors, based on price and features are the Nanoflash and the Ki Pro Mini.

I can see Convergent Design lowering the msrp on the nanoflash to compete better against the Ki Pro mini. Panasonic will have to come up with a new P2 portable recorder to complement better the AF100; more compact and affordable than the current version. There is a huge market out there for this kind of devices and a matching AVC-Intra P2 portable recorder would be a good option until they release a version of the AG-AF100 with AVC-Intra and P2 media on-board, especially if they make the HD-SDI output 10 bit to match the AVC-Intra codec.

 

[Duke M.]

The Nano can record in QuickTime, MXF or MPG file formats. The others can't making the Nano more versatile IMHO.

The Nano is 4.2 x 3.7 x 1.4". The Ki Mini pro is 1.8 x 5.4 x 3.5" so they are both small enough to mount on the camera and avoid a tether.

 

[Matt Gottshalk]

Just having manual pots on the Ki Pro mini, XLR audio inputs and hard physical buttons are the reasons I will use it.

 

[James Bridges]

The Nano can record in QuickTime, MXF or MPG file formats. The others can't making the Nano more versatile IMHO.

The Nano is 4.2 x 3.7 x 1.4". The Ki Mini pro is 1.8 x 5.4 x 3.5" so they are both small enough to mount on the camera and avoid a tether.

I couldn't agree more, the video quality, recording flexibility and the fantastic customer support is reason enough (good luck calling AJA at 2am with a question, thanks Dan Keaton!!)
Lack of xlr inputs shouldn't sway you on the purchase of one over the other.
Also, the Nano uses very little power, a major plus for me. I agree price could come into play, a grand more can be a bit steep for some.
But, I feel at this point it is the best flexible option.
So far my clients have been very happy with this.
I use this on my HDX900 and so far no issues.

 

[Matt Gottshalk]

XLR inputs ARE a big thing, especially when you want to use the recorder on a source that has analog outputs. Not every device has SDI embedded audio.

I have never had anything but great CS from AJA.

 

[GRENCH]

XLR inputs ARE a big thing, especially when you want to use the recorder on a source that has analog outputs. Not every device has SDI embedded audio.

I have never had anything but great CS from AJA.

Great point mcgeedigital. I was taking for granted that the Nano had XLR inputs. I am an audio guy 1st and for most so that is a shocker for me. I'm just having a problem thinking up a proper on camera mounting solution for the AJA due to it's size and build. I need to have this thing and my AF-100 as 1 unit.

 

[David Saraceno]

The Nano can record in QuickTime, MXF or MPG file formats.

Which QuickTime codecs will the Nano record to? QT, again, is a container for codecs. It would be great to identify which codecs.

thanks

 

[Postmaster]

XLR inputs ARE a big thing, especially when you want to use the recorder on a source that has analog outputs.

What´s the reason to have XLR inputs on the recorder, when you have them on the camera you use with the recorder?
Do I miss something?

Frank