Home > Uncategorized > Convergent Design NanoFlash Review Part 2 – Picture Quality

Convergent Design NanoFlash Review Part 2 – Picture Quality

So here is part 2 of my NanoFlash review as promised. This time I am going to talk about the picture quality and what recording at a higher bit rate offers. I have been using my NanoFlash with both a PDW-700 XDCAM HD camcorder (recording 50Mb/s 4:2:2) and an XDCAM EX1 recording at 35Mb/s 4:2:0. I have recorded from both at 100Mb/s long GoP and from the EX1 at 50Mb/s. I did a brief test with the I frame codec, but have not really gone into any great depths with that because I think the Long GoP (essentially XDCAM but at higher bit rate) is for this type of device the ideal type of codec offering stunning picture quality with reasonable sized files.

So, what do I think. Well the biggest surprise was the fact that I could see the difference between the 4:2:2 50Mb/s from the PDW-700 and the 100Mb/s from the NanoFlash on a 42″ 1920×1080 monitor, before I did anything to the material. The NanoFlash pictures were just a teeny tiny bit “cleaner” with less jitter on pans across areas of high detail. Now I will say that these were shots I chose as I knew they would stress the codec, lots of very fine detail and a medium-slow pan. It’s not a big difference, but there is certainly less mosquito noise in the 100Mb/s pictures. The difference between the EX 35 Mb/s footage and the 100 Mb/s footage is greater still, which was no surprise really, but again it’s not a massive difference, it still amazes me just how good an EX1 or EX3 looks, even at 35Mb/s.

So on to some further tests. Now I will state here that I would be perfectly happy to see my XDCAM EX, 35Mb/s footage being broadcast, and indeed I have supplied many hours of stock footage from my EX’s for broadcast and everyone has been perfectly happy with the quality. BUT the BBC in particular are keen to get all footage shot at, at least 50Mb/s which is one of the main reasons for my NanoFlash purchase. The argument being not so much that there is a big difference at the first generation, but that 50Mb/s and above holds up better in post production and after transmission. So for my tests I decided to take a pair of clips, 35Mb/s from the EX1 and 100Mb/s from the Nanoflash and then do a heavy grade (gamma correction plus color correction) and then render that out to Apple ProResHQ. The frame grab below shows the whole scene, which was a slow pan across some Arizona scrub land with lots of very fine detail including bushes, cactus and gravel with a blue sky that gradually darkens slightly from left to right.

graded XDCAM EX frame grab

graded XDCAM EX frame grab

When playing the graded and rendered clips back I could see quite a bit more mosquito noise in the 35Mb/s EX clip than the NanoFlash clip, although again I would say that given the rather extreme grade the EX footage still look pretty good. looking more closely at the sky I could see a very noticable difference between the two clips. the 100 Mb/s NanoFlash footage was much cleaner and exhibited less blocking as can be see in these two pixel for pixel grabs from the top right of the full size video frames.

Frame grab from XDCAM EX 35Mb/s clip

Frame grab from XDCAM EX 35Mb/s clip

Frame Grab from NanoFlash 100Mb/s clip

Frame Grab from NanoFlash 100Mb/s clip

So clearly there are some clear advantages to using 100Mb/s for projects that are going to require grading or other post production work. Next I took a look at some of the more detailed parts of the image. The extra color bandwidth of the nanoFlash’s 4:2:2 over the EX1′s 4:2:0 was apparent with areas of fine color detail looking less muddy and overall the NanoFlash images appeared sharper. I believe this is down to the reduction in mosquito noise at 100Mb/s as opposed to any actual change in the resolution of the recorded images.

Frame grab from 35Mb/s graded EX footage.

Frame grab from 35Mb/s graded EX footage.

Frame grab from 100Mb/s graded NanoFlash footage.

Frame grab from 100Mb/s graded NanoFlash footage.

Conclusions:

Well while there is a small difference in the first generation footage from the EX and the NanoFlash, it is not huge and unless you view the material on a large screen it will not be very apparent. For simpler projects, news and many many applications the raw EX footage will be just fine and I will often shoot that way. However the footage from the NanoFlash is cleaner, has a little better color resolution  and is more robust. So for projects that will be graded or go through challenging post production stages as well as chroma key work the NanoFlash will make a big difference. Considering the low cost of both the NanoFlash and it’s readily available compact flash media it really does make a lot of sense to have one in your production kit. I will be using mine with both my EX cameras and my PDW-700. Given that the PDW-700 at 50Mb/s is already regarded as the superior of HDCAM, then the 700 with a NanoFlash is edging towards the realms of HDCAM SR. I now own the tools to produce HD images at a quality level that I never dreamt I would be able to afford myself. Even the EX1 and NanoFlash looks stunning. Well done Convergent Design, I’m sure this product is a winner.

Categories: Uncategorized Tags: , , , ,
  1. Julio Veas
    August 28th, 2009 at 12:34 | #1

    Alister

    Did you think on the Ki-Pro before buying the nanoflash and if you did what made you decide betwen one or the other.

    J. Veas

  2. August 28th, 2009 at 13:27 | #2

    The Ki-Pro certainly has to be considered if you are looking for improved picture quality, but it’s just too big for the kind of work I do. I had also considered the Flash XDR but again this is also too big. What I like about the NanoFlash is that it can sit on the back of the camera, effectively becoming a part of the camera which makes it suitable for run and gun, fast setup scenarios and everyday use. In addition my tests with the ProRes codec have indicated that to get picture comparable to 100Mb/s long GoP you would need to use the full HQ version at 220Mb/s so for solid state media your storage cost are increased, and I would not want to use a hard drive for my line of work.
    Ultimately the Ki-pro may offer a slight quality advantage, although I’m not sure of how much (if any) difference there actually is, as to record 10 bit over 8 bit does require more bandwidth and the NanoFlash’s Long GoP is more efficient than ProRes.
    Probably for studio and controlled shoots where you could traditionally use a VTR the Ki-Pro may be the better choice, but for location and field work I think the NanoFlash is the more useable product.

  3. John Richard
    August 28th, 2009 at 14:23 | #3

    Another consideration is that the Nano and XDR recorders are more versatile than the Ki-pro as they can output both .mov or .mxf files for use with just about any NLE and either Mac or PC.

    Thanks Alister for the analysis.

  4. August 28th, 2009 at 14:31 | #4

    To be fair to the Ki-Pro, ProRes .mov files can be opened on a PC with quicktime installed. Most Direct Show compatible PC apps (vegas, premiere etc) should be able to edit them.

  5. September 1st, 2009 at 15:53 | #5

    Do you, or have you ever bumped the 35Mb/s EX files up to 4:2:2 before working on them?
    Not to say this will be as good as 100Mb/s, but for us “poor folk” without a NanoFlash I think this might help if doing heavy post processing…
    I did this with some old SD video I was re-working on the other day and the final output was much improved.

  6. September 1st, 2009 at 16:08 | #6

    I don’t bump up to 4:2:2 before editing, but I if I am outputting to a 4:2:2 format then in FCP I set the timeline to render to ProResHQ (or XDCAM HD422). There is no point in changing the colour space prior to working on the clips as that just adds another compression step which will degrade the final image.

  1. No trackbacks yet.
Tweeter button Facebook button Technorati button Myspace button Linkedin button Digg button Stumbleupon button