Next NanoFlash Firmware Update Request
 

Hello Mike & Dan,
Thank you for the firmware update I just love the SD playback it's great if you are taking orders for the next NanoFlash firmware update can have, Hot Swap or 75 Mbps record rate, Six Audio channels and hold the fries.

Dear Lance,

We may be able to meet your requests in our December release.

No promises yet.

Yes, thanks for the next upgrade step. Used it last night with perfect results.

Hot Swap would be my no 1 pick, followed by a few more audio channels followed by more Mbps selections followed by on screen data (but not recorded to the CF cards of course)

All in good time.

Cheers

I’m going to be testing the latest firmware today, now that I have read the manual. I’d like to see these features in the nanoFLASH if possible;

1. Hot Swap
2. Long GoP Time Lapse
3. Long GoP Over and Under Cranking 720/25p

Here's my top requests :) :

1. Over and undercranking in 720p 24
2. Flip function - so that it would be easier to use the nanoflash for 3d recordings
3. 75 Mbps record rate
4. You guys keep being you.

Ed -

The problem with flip (for 3D utilizing a beamsplitter) is that it takes TIME. If one camera is flipped, the other isn't, the time delay in flipping the one side would put the two signals out of sync.

So unless the nanoFLASH has a built-in video (and audio) delay for the UNFLIPPED side...

Hi Daniel-
Our plan is to implement real-time image flip. In other words we flip the image as it comes into our video processor chip (FPGA) before being processed by the MPEG2 Encoder. So there won't be any additional delay.

We don't have a roll-out date for the image flip, but it's on the roadmap.

Best-

Without being flippant.....
 

Dear Santa -Could the C.D. roadmap have Flip and Flop - so 'left to right' and 'top to bottom' for those with 35mm lens adapters trying to gain a little more 'stop'.

Yours - one of the elves.

Dear Godfrey,

Yes, we will try to meet your needs.

This is not currently planned for our December release, but, as Mike says, it is in our plans.

Mike -

Excuse my ignorance.

Let's say you have two cameras feeding two nanoFLASH. Genlocked.

Signal comes into each recorder, heading for the CF.

BUT, one recorder goes straight to the CF, while the other is doing something ELSE first, which SHOULD take at least one video line, based on my experience.

If you do ANY "first in last out" to the video signal, it should take time, and the other recorder doesn't have such a delay (normally) when not inverting.

Or, are you saying there's truly (and amazingly) no delay? I fully understand doing flips normally, in non-stereo situations. Some cameras have that. But there's always a price to pay (time) which is unacceptable in professional 3D.

Dear Daniel,

During our normal processing of the video signals, unless one has E to E selected, then there is a built in delay from HD-SDI in to HD-SDI out.

If one selects E to E, then this delay is bypassed, and the resulting delay from input to output is around 1 microsecond.

So, if you want us to peform an image flip on one nanoFlash and no image flip on a second nanoFlash, we will use the same built in delay for both.

In other words, it is our intention to not introduce any more delay than we already have built into our normal processing.

As long as the delay is exactly the same for both, obviously no problem.

I look forward to seeing this feature.

Hi Daniel-
You bring up a good point. As Dan mentioned, we have an E to E mode which produces a zero delay from HD-SDI In to HD-SDI Out during record mode. So, no problem viewing the video during record mode, as long as you don't mind fliping one of the monitors upside down!

Ideally, we would add the same exact processing delay to both the flipped and non-flipped nano and send two non-flipped SDI streams to their respective monitors. I can't guarantee this is possible in the current architecture.

With regard to actually recording to the CF cards, it will not effect the final video if one nanoFlash CF card write is delayed by one or more frames relative to the other, so long as we write the same exact frames to both cards. On the other hand, it is critical that both frames are played back in exact synchronization. But, playback does not involve an image flip, so there is no difference in the processing delay.

This can be a bit confusing and certainly something we will test.

Best-

Mike -

My point is, I doubt your magic box can delay as we are discussing. That's a lot of delay and data, even if only one line.

If the two streams aren't in exact sync (a bad condition for 3D), 3D monitoring equipment, such as the Transvideo 3D, monitor will not work.

Both signals must be in exact sync.

So unless you truly have an internal (adjustable) delay to take into account the longer signal path for the flipped signal, it isn't likely to do the job.

However, this is strictly 3D. For all other shooting, the flip should be no problem.

Image Flip Processing Delay
 

Hi Daniel-
We have 128 MBytes of high-speed DDR2 memory, which is sufficient to store over 25 uncompressed HD frames. So, we should be able to easily adjust for any processing delays.

Best-