I've been on the road for nearly a week -- trade show for my company. I've just updated the wiki. Unfortunately, due to my dead PC and schedule, it's going to be at least two more weeks before I can make major progress.

@Wayne: I guess it has the new version of the sensor chip and the USB interface does not do pixel packing

I'm not sure.


@Wayne: How fast can you run the shutter and not drop below 24FPS in 8 bit and in 10 bit modes, and can the software be setup to force a set synced frame rate (24fps)?

You're clearly not reading the posts -- if you catch up on the discussion, you'll see where we're at.


@Wayne: Real Range and light sensitvity estimates?

I won't even comment on this until we have the 10bit log transfer happening. That will make a big difference.


@Wayne: What is the perforamnce difference when running global shutter to compared to rolling shutter.

Again, we've been over this, for now it's rolling shutter only.

@Wyane: Does sumix know of any front end pro video capture/control software.

The hardware is so specialized that it's much easier for them to write it themselves. That's why I'm getting involved now, to brainstorm with them as that software evolves.

- ben

Does anybody know about someplace to look at image test samples from the Altasens chip??

My main goal is to make a 2K or 1920x1080 camera, and up to now I'm heading for the SI-3170 which uses a chip called YM-3170 from a company called Y-Media situated in California which doesn't have a Web page.

Anyway it seems to me that this is a rather old chip.It was presented September 2000, so I don't know.
More data welcome.....

JVC AltaSens camera
 

The JVC KH-F870U camera definitely looks interesting if it really is in the $2000 price range.

According to the information I could find, here are the drawbacks for our project:

• 1080i60 or 720p60 only (no 24p for filmmaking)
• HD-SDI output only (4:2:2)
• Probably incompatible with standard lenses

They could easily have put in 24p to make it suitable for low-budget filmmaking and they deliberately left it out. Jerks.

@Rob: "They could easily have put in 24p to make it suitable for low-budget filmmaking and they deliberately left it out. Jerks."

24p is one of the last remaining things they won't put in inexpensive cameras. I think the fear is that people will flock to the cheap stuff rather than the high-margin expensive stuff.

Which, of course, is true. What they don't realize is that if they don't provide 24p, somebody else will. Like Steve or Sumix.

- ben

Yup, I know it. It's like IBM in the early days of the PC -- they didn't want to add advanced features because they didn't want to cannibalize their higher-end market. But then other companies cannibalized it anyway.

What sort of data transfer rate would be required to transmit 4:2:2 footage to harddisk?

10 bit data, 4:2:2 for 1920x1080 takes @24 fps takes 119 MB/s.

Two WD Raptor disks in RAID 0 are enough to sustain that transfer rate.Although I think you would need a motherboard with a data bus higher than 133 MB.


Does anybody know about someplace to look at image test samples from the Altasens chip??

My main goal is to make a 2K or 1920x1080 camera, and up to now I'm heading for the SI-3170 which uses a chip called YM-3170 from a company called Y-Media situated in California which doesn't have a Web page.

Anyway it seems to me that this is a rather old chip.It was presented September 2000, so I don't know.
More data welcome.....

What about 12bit 720p 60fps (ie the JVC camera)? How does one calculate this? I've never needed to do it before.

@Juan: 10 bit data, 4:2:2 for 1920x1080 takes @24 fps takes 119 MB/s.

True, but remember that we're transferring and storing 10 bit monochrome data from the sensor (ie, before it's de-Bayered into RGB data).

So to calculate 4:4:4 for 10 bit at 1920x1080 at 24fps, we do this (this works for any Bayer calculation):

1920 * 1080 * 10bit / 8bits-per-byte == 2592000 bytes per frame
2592000 * 24 = 62208000 bytes per second.
62208000 / 1024 = 60750 kilobytes per second.
60750 / 1024 = 59.326 megabytes per second

So under 60 MB/sec for uncompressed 4:4:4 10-bit 1080p @ 24fps. If we add lossless compression, we can probably cut that almost in half.


4:2:2 is not a good idea for transfer and storage of Bayer-filter data. It's a false economy, because 4:2:2 refers to luma and chroma. Those terms only work if you're dealing with YUV/YCbCr or RGB. So you wind up with twice as much data for 4:2:2 than 4:4:4 uncompressed. Here's the calculation for 4:2:2 10bit 1080p @ 24fps:

1920 * 1080 * 10 / 8bits == 2592000 bytes for the luma channel
1920 * 1080 * 10 / 8bits == 2592000 bytes for the two chroma channels
(each chroma channel is half-size, so together, they equal one full-size frame)
5184000 bytes per frame, 124416000 bytes per second
124416000 / 1024 / 1024 = 118.65 MB/sec.


4:2:2 takes twice the data to store and transfer, with no benefits. 4:4:4 is the way to go.


By the way, I talked to Sumix about a FPS lock, to which Farhad responded "We are implementing exactly this."

And Sumix seems to be very open to implementing a Cineon exporter in their 10-bit software update. That would rock.

- ben

Does Sumix have support for RAW bayer data, or it only outputs RGB???

Will they give us some kind of open source applet or an SDK???

Have you thought about using a 16x9 adapter to use all of the sensor´s area to get a better image???



Does anybody know about someplace to look at image test samples from the Altasens chip??

My main goal is to make a 2K or 1920x1080 camera, and up to now I'm heading for the SI-3170 which uses a chip called YM-3170 from a company called Y-Media situated in California which doesn't have a Web page.

Anyway it seems to me that this is a rather old chip.It was presented September 2000, so I don't know.
More info welcome.....

@Juan: Does Sumix have support for RAW bayer data, or it only outputs RGB???

You can output RAW B&W bayer data or de-Bayered RGB data via nearest-neighbor, bilinear or laplacian de-mosaicing algorithms. I only output RAW.


@Juan: Have you thought about using a 16x9 adapter to use all of the sensor´s area to get a better image???

Absolutely. I have a 16:9 adapter and a 2.33:1 adapter I'll be testing with in the coming weeks and months.

- ben

GREAT!!! thank you :).

I believe that getting the image thru anamorphic, be-bayer it, and resize to 1280x720 would give amazing quality!! (in fact almost 4:4:4!! You will be getting 512 instead of 360 lines for Red and Blue)

Will they give us some kind of open source applet or an SDK???

Definitely! My thinking is that if you do anamorphic 1280x1024, you should be able to expand it to 1920x1080 and have it look okay too!

Sure!!!

Anyway in that situation I would prefer using a 1920 bayer.

My idea is using that 2048x1536 chip with anamorphic to get a final 2048x1107 image that will rock HDCAM!!!!

Or going to 1920x1080 for TV and get better results!!
(Remember that in fact HDCAM is 1440x1080)

@OBIN

Could you give us an exposure test of your camera. I´d prefer an outdoors image on a sunny day with some deep shadows somewhere.
That way we would be able to determine the camera characteristics...
Also it would be nice if you could tell us at what aperture your problems with streaks start...and if they disappear at lower apertures or remain the same...

@Juan: Remember that in fact HDCAM is 1440x1080

And that's just luminance -- the chroma is 480x1080! That's essentially 3:1:1. And it's highly, highly compressed -- MiniDV has 5:1 compression, but at 135megabits/sec (only 16.8Megabytes/sec), HDCAM is 8.4:1 compression.

Less color resolution than DV, and more compression? I think I'll stick with our 4:4:4 10bit uncompressed, at 1/100th the cost! :)

- ben

I am not at work today but I am going down for a few min to pickup the camera and bring it home for a few TESTS


I may get the SI-3300 camera for testing soon ..less streaking I hope ;)

OBIN,

Is it the 3300 or the 3170???
Is it the 3300 2048x1536?
Does it support 30 fps at full resolution??

Do you have a photometer?
If so, could you measure its sensitivity too?

What is your usual gamma correction??

I guess somebody at SI should make these test for us, but.......

Anyone who wants more test images with HIGHCONTRAST stuff please provide your email...I can't upload ftp from home (don't know login)

I have been fooling around and it turns out that yes they more you GAIN the camera the worse the "smear" is...I now question if I had gain REALLY high on the shoot with the backlights....it was so much pressure I could have overlooked that I think...it's weird ...testing at home here I have to push it HARD to get that amount os smear again...

I'm baaaackkkk.
I've also read this thread and caught up - bouncing up and down in my chair wanting to answer things but I figured that I should read to the end first. So:

Juan: The SI-3170 has been correctly stated as the Y-Media chip - very fast (100Mpix/sec) 3.2 mpix. It uses two 12 bit A/D converters but has a fair amount of FPN and low sensitivity. I put it on par with the IBIS-5. I wouldn't suggest it for studio work with the Altasens coming out. But hey, I wouldn't suggest the IBIS-5 either and we have an external A/D for lower noise and higher data rates. If someone wants to try it on camera link at 12 bit, let me know. The SI-3300 is a Micron camera - 3.2Mpix at 12fps but 1920x1080@24fps.

More on IBIS-5: Yes you can use it in global shutter mode for continuous capture. The integration time is sequential to the readout time so every time you need to change exposure you must recalculate the pixel clock to get the frame rate you want. And, the readout time must be fast enough to handle that. This is probably why Sumix doesn't do it on USB.

Jason: I think Rob explained the 10/12 bits on the SI-1300. It is a 10 bit A/D with the data shifted up by two. This is because most industrial apps only know 8 and 12 bit data.

Obin and all on smearing: Obin, try running at only 24fps instead of 48. Micron suggests lower clock rates for less smearing. This is the rate the IBIS-5 is being tested at anyway. I've tried to be open about this as soon as we heard about the problem. I'm also letting anyone who has an SI-1300 upgrade to either the SI-3300 or SI-1920HD if they need to (the Altasens is more money).

Here is a link for an Altasens monochrome 8 bit image to get everyone thinking:
http://host196.ipowerweb.com/~silico...20in%20HD3.jpg

steve please provide a smear test if you can

take the camera and shoot a light that is taking about 5x5 pixels of less of the screen and see if you get dark streaks from it with the 3300

I am fooling with a footage clip on the machine in Combustion now..I will try and send a WMV via email if you want it...

Ok...I think I am starting to understand the smear thing...I have footage that is 16bit tiff files in Combustion..everything looks good..when I BOOST the gamma in the SHADOWS the "smear" comes out..if I just boost the MIDS and HIGHS the smear never shows...so how hard would you guys think you need to push the darks for this camera in a Cine type shoot for a film? also if you DON"T have a HOTSPOT in the frame and you push the darks hard they are fine....

things are looking better from my POV now....maybe I can work with this after all....

The new lens is a HUGE improvement over the cheap CCTV lenses I had...new one is a $350 8-51mm C-mount zoom

guys I need some FTP space if you want to see some clips/video.... I can't send files via email, they will not fit..anyone have FTP??

The whole point is to push the shadows. If you're not pushing the shadows (of course there's a point to where you can push), then you're missing a couple bits in the low end-in other words you taking your 10-bit image and making it 8 or even 6-bits, cause those darks a hiding a huge amount of bit depth. You're looking at a linear image. If you check back a couple pages ago to where I placed some linear images in my post as samples, linear images are very dark, and you have to raise those darks to get the proper image. You have to apply an overall gamma of around 1.7-2.2. If the darks from that gamma gain go nuts, and you have to crush them, then you're loosing a lot of dynamic range. I've worked with a lot of linear images, and they should be clean when you do the gamma correction to match them to the gamma of the monitor/display device. There could be some non-linearity in the bottom, and I'm not sure how that skews the results or limits the dynamic range. Maybe if Les Dit could check out some of the image, he could find out what the true dynamic range of the image is.

ok 1.7-2.0 gamma is ok for the darks in what I shot today

someone needs to allow this message board atleast a way to upload stills for viewing on the boards!

Oh, Okay,

Then that's good.

BTW, I have ftp space (not much, but enough for stills).

Email me at jrod@mindspring.com and I'll post it.

You can upload images to the wiki, but they are currently limited to 1 MB. This is probably way too small -- tomorrow I'll try to remember how to change the setting.

Ok Jason..when I push the gamma in the DARKS at about 3.8 it looks really bad

shadows are ok at 1.6 gamma. No higher then that...what do you think?

I guess having the option to do anything with gamma is great when you compare this with videotape stuff...

@Rob: 1 MB. This is probably way too small

Not necessarily, depending on the purpose of the upload -- 720p frames only take up about 800k with JPEG at 100% quality

- ben

FTP Space -- Have a little to share...
 

Hi Obin,

I can give a little FTP space and bandwidth, send me email and I give you a login and password.

marjamar@speedwing.net

Okay, here's Obin's pic:

http://home.mindspring.com/~jrod/gamma_DARKS_3.8.jpg

BTW Obin, at 10bits, and a gamma of 3.8, there is no way you're going to make those blacks look good.

In Combustion you should go to the "Master" setting (not shadows, mids, highs), and then apply a gamma of 1.7.

Actually even better, you should go to the LUT in the footage tab (under output), and then give that a gamma of 1.7 so that there's no "filters" being applied. That's the nice thing about Combustion, you can apply LUT's to images, something that AE doesn't let you do. Just watch out, because the default LUT clips the shadows and highlights, you don't want that, so stretch the curve over to the end of the highlights and shadows. But try the LUT feature. And again, you need an overall gamma of 1.7, not separate gammas for each of the different sections of the image, because those gammas are ON TOP of the gamma of the master. So you're applying a gamma of 3.8 to the base gamma of 1, not applying an overall 1.7 which would look much different.

ohhhh....hmm I did not know that...

cool I will dink around with the LUT...


thanks..good stuff!

<<<-- Originally posted by Ben Syverson :

@Wayne: How fast can you run the shutter and not drop below 24FPS in 8 bit and in 10 bit modes, and can the software be setup to force a set synced frame rate (24fps)?

You're clearly not reading the posts -- if you catch up on the discussion, you'll see where we're at.

@Wayne: What is the perforamnce difference when running global shutter to compared to rolling shutter.

Again, we've been over this, for now it's rolling shutter only.
- ben -->>>

Sorry, I did but I think I mean a few more advanced things then were covered.

I know you have talked about it not doing 24fps exactly, but I'm curiiouse if the config and options have a more advanced realtime mode that forces the sytem to let it sync to an exact FPS. I don't remember you mentioning how fast the shutter could run. In Obin's camera faster shutter means more bandwith (which maxes out), but if yours had a frame buffer you should be able to do much faster shutters.

When I said performance difference between Global and rolling, I was meaning in terms of any effects on picture quality and max shutter speed (global leaks a bit, and who knows what else).

Thanks

Wayne.

So jason you don't think that smearing is all that bad ?

I will send you a raw tiff file of that 3.8 DRAKS gamma pic for you to take a look at

Wayne, like I mentioned before, currently there is no way to "lock" the frame rate, but as I mentioned, Sumix is currently planning on incorporating such a feature. If you want to run at 24fps exactly, you'll have to run the camera at 40mhz, and take only 24 of the 40 or so frames per second. At 24mhz, you're always going to be brushing up against 24fps. Personally, I'd rather run at 24mhz/23.5fps and take my increased exposure even if it means a few more editing headaches. But folks who want absolute sound sync on 50 seconds of footage :) can set that up at 40mhz.

I haven't tested global vs rolling because currently the firmware/software is rolling-only for motion. Global would clearly be preferable, and I've brought that up with Sumix, but until that's a possibility, I won't waste time testing it.

- ben

clip that has been CC and gamma set at 1.7

http://obin.weet.us/cc-1.wmv
http://obin.weet.us/raw image.tiff
http://home.mindspring.com/~jrod/gamma_DARKS_3.8.jpg

2 above are the same shots..if you can open the raw black and white take a look at that...the dark is the one I pushed the crap out of the dark GAMMA

Obin,

I think you need to shoot an 18% grey card.

Because right now we have NO idea how much dynamic range we're getting out of this camera.

Your images look good (sans the bad bayer demosaicer), but they might be too bright, you might be clipping too much of the shadow dynamic range.

Get a light meter, and shoot a grey card. I'm not sure what the ISO of these cameras is, you might be able to find that out from Steve. But once you find that out, go shoot a grey card, and see where that shows up in your file. Then you can see how much gamma correction it takes to make that 18% grey 50% grey inside Combustion. A white card is 2 and 2/3 stops over 18% grey. So then you can see how many stops over white that you're getting before it clips.

Without these tests, and no pun intended, we're shooting in the dark :-)

But your images do look good :-)

BTW, you know what I'm lov'n?

I'm lov'n those highlights on the skin!

That's very hard to do on Sony HD cameras without it looking nasty, cause they tend to clip and get color-shifts in the highlights. I've done it, but it never looks that good.

Chalk one up to Uncompressed!

ok... Joe Dunton Cameras have a grey card you think?

I live 2 blocks away from them

Thanks Jason...as you can tell that image has been WHACKED on by me in Combustion...not bad eh?

post the images Jason if you would for everyone

Obin,
That streaks in the black have always the same spatial placement.
If the answer is yes it would be posible to make a plugin to remove them if necessary..

Also, if you can and will, please remember the exposure test I've told you....

Yah Man,

You're in Willmington! Next to NY this is the film hot-spot on the east coast (not sure why it's not Norfolk, Virginia Beach, less hurricanes ;-)

BTW, here's Obin's images he sent me:

http://home.mindspring.com/~jrod/original.jpg
http://home.mindspring.com/~jrod/re-touch.jpg

Ben thanks for this advice, I see what you mean by the 40Mhz, clock limit, I had forgotten that, and I posted before I saw your subsequent post about Sumix looking at a lock. But really Sumix should have had somebody here to find out in times past, like SI's Steve has. I have been unable to get around to contacting them yet (as Steve I suggested) to talk with them.

Rob L is still on holiday, maybe he can help us with some space (like the 35mm adaptors have) when he gets back, good to see you all back.

The JVC altsens, is a 3chip camera, not bayer. Bayer is worse than 4:2:2, it is more like 4:2:0, red and blue is only sampled every 4 pixel block. If it is really under $2K, that would help offset the extra price of a HD-SDI card, but I don't really see JVC selling a HD camera for a price so close to the card. The thing is not even a true HD spec camera (I'm just being controversial here) if it only supports 30p and 60i. For lense, make an SLR adaptor.

I have talked to the local JVC pro distributor and that camera looks like an update to an existing model, and not likely to be under $2K. He informs me that the new HD1/10 like 3ccd camera maybe released at CVI?? show in Amsterdam, in Steptember.

Juan, if you want Altsens images, ask Altsens I'm sure they would have links on there site. I think you are going to be gaurantted to get better results from an Altsens, anyway.

Steve, arty B&W 1080 image, keep it away from Woody Allen ;), actually thats probably the best way to promote your camera, get him to use it (when the software is done).

Aaron thanks for the links but http://www.ccd.com/kaf-1602e.pdf is not found. I actually turned down a secondhand scsi HD rack thing for $20 recently, I wonder if it was fibre channel. If we could only get the speed and capacity in one drive at a cheap anough price (including standard interface like SATA 300, high speed doesn't need fibrechannel anymore) we can go three chip no worries (very nice for case design and drive swapping). I want to research this too.

Rob S what make of PC do you have, ussually they just exchange main boards to repair.

<<<-- Originally posted by Obin Olson : Anyone who wants more test images with HIGHCONTRAST stuff please provide your email...I can't upload ftp from home (don't know login)

I have been fooling around and it turns out that yes they more you GAIN the camera the worse the "smear" is...I now question if I had gain REALLY high on the shoot with the backlights....it was so much pressure I could have overlooked that I think...it's weird ...testing at home here I have to push it HARD to get that amount os smear again... -->>>

I posted questions ages ago, asking about how the gain effects latitude as well, as the more gain the more current issues. Your earliest outside shoots (with the dark face) did not seem to have any smear/streaking problems when lifted in post. Maybe we need to do standardised tests on each camera to test lattitude and problems at: 0 gain, the highest gain without smearing/streaking/colour banding, highest gain, lowest problem free negative gain, and lowest negative gain, for each channel. Testing for quality, and also hotspots and shadows.

Obin, would a 35SLR prime relayed (condesed) into 1/2 inch format condensed, help with dark shots?


Thanks

Wayne.