Ki Pro vs Nanoflash

[Ed David]

Has someone ever done a test recording from a camera to a nanoflash and then also to a KI Pro to see what the difference was in terms of color space and noise, etc? I would love someone to end any argument going around between the two acquisition formats? This will be especially useful now that the KI Pro Mini will come on the scene, which I see as a direct alternative to the Nanoflash.

[Dan Keaton]

Dear Ed,

We welcome the competition between the nanoFlash and the Ki Pro and the Ki Pro Mini.

The Ki Pro Mini, appears to be an excellent product with very aggressive pricing.

We congratulate AJA on producing a very attractive product.

We also feel that the nanoFlash is a very attractive and versatile product with many applications.

One advantage that the Ki Pro and Ki Pro Mini have is the fact that ProRes is 10-Bit codec.

Thus, if one has a low-noise 10-bit or higher camera, this theoretically can be an advantage.

But, there are two things to take into consideration, the noise level of the camera and the codec itself.

In December 2009, Sony had a special presentation in New York City for "heavy hitters".

The 220 Mbps AJA ProRes files were compared directly, on $25,000 Sony monitors, to Sony's 50 Mbps XDCam Optical Disk Format, which is the codec that we use.



Since I was not there, and since I am obviously biased, I feel that it would be inappropriate for me to disclose hearsay.

I do feel that head to head tests will show just how good the nanoFlash is and how competitive both products are.

I also want to keep this discussion and comparison accurate, polite, and respectful to both AJA and Convergent Design.

[Daniel Symmes]

This is the Convergent Design nanoFlash forum. I would prefer to read about AJA products in the appropriate AJA forum.

[Steve Phillipps]

I get the feeling that there's no real need for a head to head test here - I'm sure the outcome would be that there's very to separate them on image quality, and that both are more than good enough for any broadcast application - after all even the 50 mb/s 8 bit 422 XDCam codec is accepted at the highest broadcast levels and by the EBU. Both these products have specs well above that.
The differences will likely be operational and compatibility ones - Nano looks a little more compact, Ki Pro maybe a bit more rugged etc. And then I suppose - but saying that for broadcast kit both are amazing value for money.
Steve

[Ed David]

I think this is the right forum to talk about KI Pro vs Nanoflash since I have two nanoflashes and no KI Pro. Anyway I also think a test is relevant.

If you are shooting a feature film, which device to use for the highest quality look for the grading? Or for T.V. No offense to the BBC and EBC and all those standards - but I am not an engineer, I'm a cinematographer and I want my cameras to look the best. Just because some engineer says 8-bit 50mbps is the "silver" standard doesn't mean I'm going to shoot a television show at that bitrate - I will do what looks best. And I can't wait to use either the nanoflash or the KI Pro Mini with the new Panasonic AF100 - and thousands of others can't either. If there was a test that shows which device gives the best quality - people will work around the form factor.

At the end of the day quality wins. Bulkiness doesn't. Furthermore I found out the KI Pro Mini weighs 1 lb and uses around the same amount of energy as the Nanoflash. So the decision between the two becomes harder and harder.

I hate seeing everyone jumping on the bandwagon of KI Pro so quickly especially when they don't remember the delays of the big daddy KI Pro.

We are the pioneers of recording to a higher-quality acquisition format and we should be on top of different devices to record to when our producer asks us too and should know how to defend our choice of using the xdr or the nanoflash. There is no better forum than this.

[Steve Phillipps]

If you're going to do what looks best to you then the only way is for you to do the test. If you're not going to listen to the BBC and EBU then why would you listen to someone on an internet forum?

Steve

[Chris Hurd]

Originally Posted by Ed David

I think this is the right forum to talk about KI Pro vs Nanoflash since I have two nanoflashes and no KI Pro.

Sorry Ed, but that's not how this site works.

We organize topics based on what those topics are, not on what you own. It doesn't matter that you have two Nanos and no Ki Pro. On this site, a Ki Pro topic belongs in our AJA forum, not our Convergent Design forum. If the topic concerns both, then post it one level up, in http://www.dvinfo.net/forum/tapeless...ing-solutions/

One of the biggest battles I face daily in managing this site is getting folks to post in the correct forum, for the benefit of those who come in later looking for something and expecting to find it in its proper place. Please, please, please use the *entire* site. It doesn't matter what gear you have at home. Thanks in advance,

[Ed David]

oh okay. sorry about that. will do.

[Steve Kalle]

Even though the Aja is $800 cheaper, there is the recording capacity that needs to be factored into the cost. Pro Res 422 is 220Mb/s (I think) and is roughly equivalent to XDCAM 50/422 in image quality. Factor in the added cost of lets say 4 32GB with the Nano and for equivalent recording time in the Aja, you'd need 16 32GB or 8 64GB cards which adds a significant cost.

At $190 each for SanDisk 32GB 60MB/s cards, that is an additional $2280. So subtract the $800 difference of the Nano & Aja, and now the Aja costs $1480 EXTRA! I know someone will bring it up, so lets equate the difference with the Nano recording at 100Mb/s L-GOP. The Nano still wins on cost at roughly $700 less.

However, does the Aja support hot-swap? If it does, then that is an important advantage for some including me.

Also, the 10 bit of Pro Res eats up wasted space with most cameras under $20k because they have more noise.

[Billy Steinberg]

Pro Res 422 is 220Mb/s (I think) and is roughly equivalent to XDCAM 50/422 in image quality.

I'd love to know more about this test Dan mentioned, and I'll try to ask my Sony friends in the broadcast division when they get back from IBC. I have some ProRes tests I want to make myself, but I am a huge fan of the 100Mb Long GOP CD/Sony codec. I have recorded a lot of stuff from 10-bit cameras with very low noise (actually 12 or 14 bit cameras, but 10 bit HDSDI), and the 100Mb CD codec produces essentially the same picture as the raw uncompressed stuff.

If you'd like to learn more about Apple's ProRes codec, they have a pretty nice white paper on it. There are four bit rates available, all VBR. 220, 147, 102, 45.

does the Aja support hot-swap? If it does, then that is an important advantage for some including me.

And me. Sorry to say, the answer is No. And I have a question in to the aja rep about whether the KiProMini can seamlessly transition from one card to the other while recording. If you have KiProMini specific quesions, you should take advantage of the aja forum here. You can get a pretty good feel for what the company and their products are all about, and how they respond to their customers. I'll warn you that nobody lets you into the thoughts of the company like CD does...

And that said, I wouldn't mind a really good, in-depth comparison between the (Convergent Design controlled) Sony hardware MPEG2 codec and the Apple ProRes codec...

Billy

[Rafael Amador]

Originally Posted by Steve Kalle

Pro Res 422 is 220Mb/s (I think) and is roughly equivalent to XDCAM 50/422 in image quality.

Also, the 10 bit of Pro Res eats up wasted space with most cameras under $20k because they have more noise.

Hi Steve,
I don't know where you get those ideas from.
You are comparing the lowest XDCAM 422 option (8b Long GOP at 50 Mbps) with what is becoming the industry quality standard (10b Intraframe at 220 Mbps).
You can compare Prores with the NANOs Intraframe or with the AVC-Intra.
But in the end, the codec is not the only think to consider, but the processor. The capability of compressing a 10b Uncompress stream in RT at full quality and without tricks (the blurring that we are seeing in XDCAM).
If the process is properly done, you can NEVER compare a picture with 10b Y' with a picture with 8b Y'. The luminance makes the picture. The color is not that important for our eyes.
Another thing to consider is that Prores has become the production & intermediate codec for all the Apple platform: Whatever the acquisition format, people end up in Prores.
Also any PC can open Prores just having QT installed And don't forget neither, that Prores offers 4 levels of quality for 422.
On your comment about noise/space, I really don't see the relation.
rafael

[Steve Phillipps]

I don't really know the answer to the codec comparison but I'd guess that the regular ProRes 422 would be the equivalent of the XDCam 50 mb/s. From what I've heard from the editing side of things ProRes HQ is ultra-high quality, and near the quality of uncompressed so should easily compare to the MPEG 100 mb/s Long GOP codec or Panasonic's AVC Intra 100. I don't have evidence of this though - but suffice to say I'm sure that the codecs within the Nano and the Ki Pro are both way above anything most users could ever need.
Steve

[Mike Marriage]

You may be interested in this test I did recently, if anyone wants to bring along a nanoflash, we can compare that too:

XDCAM EX vs Prores

The advantage of the Ki and Prores is that you are shooting straight to a reliable post format. That is great if you are going through a multi-stage post environment as Prores is designed for multiple generation use, MPEG2 is not. Of course you can convert XDCAM to Prores but that takes a lot of time and takes up more space as you still want to keep your original files. I often need to produce 24fps and 25fps versions of commercials and trailers. If I shoot XDCAM, I have to convert everything to Prores in order for Cinema Tools to change the frame rate; which takes ages!

Both will produce excellent results but with a low noise camera I'd guess that the Ki would have a slight advantage. However the Nano offers a substantially less data hungry workflow for very little loss in quality - the choice is yours!

[Steve Phillipps]

My thoughts exactly Mike.

I get the feeling the Nano may have the advantage if you want variable frame rates? The over/under crank function will give you slow motion etc. from any 720/60 stream, where as on the Ki Pro I think your options are limited.

Steve

[Dan Keaton]

Dear Mike,

Thank you for posting your test results.

I would like to make one thing clear.

While the title of this thread is Ki Pro versus nanoFlash, your test compares Sony XDCam EX footage versus Apple ProRes footage.

To be clear, no nanoFlash footage appears to have been used in your testing,

Your test compared 35 Mbps 4:2:0 footage to Apple ProRes 4:2:2 footage.

A different test, one which uses nanoFlash 4:2:2 footage, using bit-rates from 50 Mbps to 280 Mbps may produce different results.