Industrial CMOS camera as an entry-level HD cinema camera (1080p)

[Troy Warr]

After stumbling upon a post in another forum, I've gained new interest in an idea that I've been kicking around for a while. I'm interested in the viability of using an industrial CMOS camera as an HD cinema device.

 

Does anyone have experience with a camera anywhere along the lines of the Silicon Video 9T001C from EPIX? My budget doesn't have a hard ceiling (roughly $3-5K), but a $1000 1080p setup (actually more like 2K) that includes camera, capture card and frame grabber software is very intriguing. The sensor specs can be found here.

 

The type of stuff that I shoot (MOS shorts, music video, experimental) often requires the small camera form factor, so that's a big plus. I don't mind being tethered to a PC, and I love progressive scan and the capacity for variable frame rates. However, I'm unsure of several things about this kind of setup:

 

- What kind of image quality should I expect? Obviously this won't compete with any high-end 1080p cameras, but using good glass and low compression, could it possibly get me anywhere near the current crop of HDV prosumer cameras ($3-5K price range)?

- If I mount some super-fast lenses (e.g. f/1.0 ballpark) and shoot at large apertures (maybe using ND filters to reduce the light), would the 1/2" sensor give me any shallowness to the DOF? Or, could I find a way to buy/build a 35mm lens adapter to interface with the C mount (same concept as the P+S Technik or the Red Rock Micro)?

- Given a fast, modern PC, would I face any bottlenecks or extra expenditures involved with live footage ingest? Would this be prohibitive when considering this setup from a budget standpoint?

 

An obvious alternative to something like this would be a low-end HDV camcorder like the Canon HV10, but the potential advantages to this setup (variable frame rates, progressive scan, lower footage compression, interchangeable lenses, live footage ingest) are very desirable to me.

 

Thanks in advance for any feedback!

[Ilmari Reitmaa]

I would expect some difficulties with capture device codecs, which are unlikely your regular video edit codecs. Conversion software for proprietary/unusual stuff could be difficult/expensive to come by. Also industrial cameras come in greater variety as industrial applications are much more varied than cinematography, where image quality at the visual range of the spectrum is at premium; the criteria of quality are different for industrial cameras and you may end paying for wrong kind of quality (low signal to noise ratio but poor dynamics, etc.) or features you really don't need. Furthermore, opportunities to resell your investment are fewer than with a camcorder; you're probably stuck with the camera and the capture device and whatnot, even if you won't need them later.

 

Still, I wouldn't so much doubt the feasibility of the idea, but I would expect all kinds of funky workflow-related issues, be they digital (proprietary codecs, proprietary capture devices, driver problems, insufficient RAID...), or practical, such as having to haul a monitor around more than usual (no viewfinder), having to customize bridge plate/mattebox, etc.

[Troy Warr]

Thanks, Ilmari, for your feedback!

 

These are all excellent points to consider, and I definitely have my work cut out for me if I decide to take this route.

 

EPIX provides a demo version of the XCAP software that's used to acquire images from the camera, and though I'm just starting to learn its features, it appears that the formats of choice for image acquisition are either TIFF or BMP. I'm assuming that I would be able to capture a sequence of still images that I could then place on a timeline in Adobe Premiere or After Effects, edit and then export in the desired HDTV format.

 

Also, for some reason I'm under the impression (though I could be dead wrong - I need a little more research here, too) that the CMOS chip used in this particular camera is the equivalent of a 3-megapixel still camera chip, likely from a digital point-and-shoot or the equivalent; my guess is that it's just controlled through the included PCI board in such a way as to allow for rapid capture of still images, to essentially form the equivalent of progressive-scan video, just captured as a sequence of stills rather than with a traditional video codec. If that's the case, I'm hoping that speaks favorably to the image quality - and at least that would get me in the ballpark of a decent HD video camera, albeit one sourced from the imaging chip of a point & shoot digital still camera.

 

You're definitely correct about the problems with resale - I can't imagine having anything but a hard time trying to recover any money on the investment should I decide to sell this down the road. Fortunately, though, it doesn't appear that I would need to upgrade my current computer to support the bandwidth required for image acquisition, and any image enhancement methods that I try (purchasing a high-quality C-mount lens or two, finding a way to adapt SLR-based 35mm lenses for increased DOF) *hopefully* wouldn't mean much more than an investment in some popular optics, e.g. Nikon F-mount lenses. I would think that the $1000 investment in this system would be the only major part that I couldn't expect to mostly get back with resale.

 

That's a good point about the pitfalls of working in the field, too. I would have to haul a PC and monitor around tethered to the camera - but I'm prepared to do that. Another alternative that I have considered would be something like the Sony HDR-FX7 attached to the Blackmagic Intensity HDMI card, which would also require a tethered PC. I really don't like the implications of the HDV format (high compression/low bitrate, long-GOP interframe codec, 4:1:1 sampling, no 1080p) so the Blackmagic card would give me a higher quality solution, but at a greatly increased cost (probably $5-8K after camera, 35mm adapter, card, PC upgrades) vs. this solution at about $1-2K, if it would in fact work as I hope it can.

 

Provided that the camera captures TIFF or BMP sequences, can handle 1080p/24 (at least at 2:35 ratio), and has image quality at least comparable to a 3 megapixel digital point & shoot camera - all of which appear to be possible but I still need to verify with EPIX - can you see any other potential pitfalls, better alternatives, or drawbacks to this system that I might not be considering? Keep in mind that I don't need sync sound.

 

Thanks again for your feedback!

[Luke Prendergast]

The examples given are all odd frame rates eg., 20/27/34fps, so unless the software allows arbitrary setting of the pixel clock and/or resolution you won't be getting 'normal' video rates out of it.

[Troy Warr]

Thanks, Luke! That's a good point. I *believe* that you are able to set most features of the camera, including resolution and pixel clock. I've tinkered more with the demo software, but there's not a whole lot to gain from it without having a camera attached and its drivers installed. But, there are settings for horizontal and vertical resolution, among others - so I hope that's a good sign. It appears that there is a maximum datarate that can be handled by the CMOS chip and/or controller card, as well as a maximum practical number of frames per second - but it seems that if you don't exceed those limits, you're able to set almost everything manually.

 

I am encouraged by the fact that nobody has pointed out any major pitfalls yet - not that that's any proof of concept, but I think that at least means that it's worth investigating further. Though I'm not holding my breath for the best image quality, it would be awesome if I'm able to squeeze out decent 1080/24p video from a ~$1000 solution!

 

Thanks again for your help!

[Phil Rhodes]

Hi,

 

It's an interesting option. A year or so ago I did some image-processing work for a company doing industrial process imaging, and the little machine vision cameras are actually quite decent. Probably not stunningly good dynamic range, but lots of pixels and it's worth pursuing just out of interest.

 

Phil

[Hal Smith]

I've emailed EPIX to see if they're interested in demoing their camera in a Cinematography environment.

 

Copy of email:

 

There has been a thread on a professional Cinematography Forum discussing use of a 9T001C and support software as a low budget High Definition camera. Would it be possible to obtain a demo unit for testing? Results would be posted on Cinematography.com. I would envision the camera being bicycled through several Cinematographers for evaluation, possibly an ASC (American Society of Cinematographers, an invitation only professional society) member or two.

 

See: http://www.cinematography.com/forum2004/in...mp;#entry143849

 

Such an arrangement could be a win-win for both EPIX and the Cinematography community. As well as a budget HD solution your gear possibly could become the digital crash camera of choice for HD origination. At your prices a large budget production could afford to kill a couple just to get a unique shot, all would depend on just how good the pictures can be in the highly controlled lighting and design environment of motion picture production

[Troy Warr]

Hi,

 

It's an interesting option. A year or so ago I did some image-processing work for a company doing industrial process imaging, and the little machine vision cameras are actually quite decent. Probably not stunningly good dynamic range, but lots of pixels and it's worth pursuing just out of interest.

 

Phil

 

Hi Phil,

 

Thank you for your input! That's encouraging to hear. I have very little experience in the field, but part of what drove me to investigate this option was my experience with a very small composite video camera that I bought a few years ago. I used it in conjunction with an entry-level Canon miniDV camcorder (as a deck) to record some footage from the POV of my car's front bumper. I taped it underneath, about 4" from the road surface, and ran the cables into my car, plugging the DC input to my car's cigarette lighter (using an adapter) and the RCA output to the camcorder's analog-in port.

 

As you mentioned the video wasn't stunning, but it was surprisingly good for a $50 solution, and honestly, the limiting factor in video quality seemed to be the Canon camcorder itself; i.e. the video that I captured looked indistinguishable from the camcorder's native footage (albeit monochrome). Granted, this was a budget model (a ZR70, as I recall), but if I this concept is scalable and I can squeeze entry-level HDV quality (e.g. on par with a Canon HV10) out of this camera from EPIX - especially if it's capable of 1080/24p - I'll be a very happy man!

 

I've emailed EPIX to see if they're interested in demoing their camera in a Cinematography environment.

 

Hal - thank you so much for taking this step! I'm really excited to hear what the response from EPIX will be. You're certainly right that this could be an ideal option for a crash cam should it prove to be practical and compatible with (or at least adaptable to) current HD formats.

 

Since the camera is tethered to a remote PC via Ethernet cable, the possibilities would be great: no onboard media recovery to worry about, a very compact form factor, and the option to operate the camera from a great distance relative to the controller PC.

 

I remembered an article in American Cinematographer that detailed the use of Sony XC999 lipstick cameras in the movie XXX. Several were strapped to a car that was driven off a bridge, and the trunk was filled with crash-proofed miniDV recording decks. I've seen the XC999 camera heads alone retail for around $1000 each (likely more for the PAL versions used in XXX), and miniDV decks are certainly not cheap. Combine that with the fact that you're risking the shot by crashing your recording media along with the recording devices, and on an expensive/complex stunt, that seems awfully risky.

 

Using that same stunt as an example, I could envision using several of these 9T001C cameras in the car, running Ethernet cables in a bundle to a remote location (maybe even still on the bridge?) where a PC can capture their output (the EPIX site somewhere mentions using up to 8 PCI cards and attached cameras per PC) until they hit the ground. I believe that Ethernet cable can be run for fairly long distances before a repeater/amplifier is needed, and it's compact enough to be hidden or rotoscoped from external shots, and cheap enough to be disposable. If you rig up a simple "breakaway" section of cable, you could even eliminate any damage to the recording PC should the cable become snagged or ripped away from the PC along with the camera head at the moment of impact.

 

Anyway, please let us know what you hear back from EPIX - I'd be more than thrilled if some of the pros from this forum were able to test a demo unit and provide feedback!

 

Thanks again!

[Thomas James]

If you are looking for a good entry level progressive scan high definition camcorder why not get the JVC GR-HD1 for about 1350 bucks ? While the resolution is only 720p it features a much better short 6 gop HDV compression than the 1080i long 15 gop. The compression of the JVC is like halfway in between long gop HDV and DVC pro HD. The resolution of 720p is the equivalent of the progressive like frame modes such as Canons 1080F. If you want 1080p you will have to get the Sony HVR-V1u because the FX-7 can only record 1080i interlace.

Edited December 27, 2006 by Thomas James

[Troy Warr]

If you are looking for a good entry level progressive scan high definition camcorder why not get the JVC GR-HD1 for about 1350 bucks ? While the resolution is only 720p it features a much better short 6 gop HDV compression than the 1080i long 15 gop. The compression of the JVC is like halfway in between long gop HDV and DVC pro HD. The resolution of 720p is the equivalent of the progressive like frame modes such as Canons 1080F. If you want 1080p you will have to get the Sony HVR-V1u because the FX-7 can only record 1080i interlace.

 

Hi Thomas,

 

Thanks for the feedback and suggestions!

 

I had looked at the specs of the JVC GR-HD1 a while back as an option, but several limitations led me to consider other options instead:

 

- 720/30p is less resolution and more fps than I'd prefer. Though 1080i is, by some accounts, only modestly superior to the effective resolution of 720p, I'd still even prefer one of the pseudo-24p pro-sumer cameras that records to tape in 1080i with reverse 3:2 pulldown. Ideally, I'd like to find a solution that can offer 1080/24p within my relatively low budget, which is a tall order on any camcorder, but that's what led me to the industrial cam route.

- The shorter GOP compression is a nice feature, but the GR-HD1 is still limited to a peak 25Mb/s bitrate because of the use of miniDV tape. With an industrial cam and a remote PC, I'd be able to surpass that greatly, and could even record uncompressed HD footage if my PC can handle it. At the very least, I can use a lower-compression codec than DV or MPEG-2.

- Lack of interchangeable lenses - though this is not an essential feature for me, I am drawn to the flexibility of using a C-mount industrial cam. This would allow for a much greater assortment of optics (instead of just telephoto or wide angle attachments), and the greater sensor size of this industrial cam in particular would mean reduced DOF.

 

Aside from that, since I do have $5k or so to play with, I'd sooner invest in a more modern Canon or Sony HDV camcorder should this idea fail. Perhaps a more accurate title for my post should have been "Industrial CMOS camera as an alternative HD cinema camera" since I'm ultimately looking to circumvent the limitations of the HDV pro-sumer arena to get the highest quality, highest-resolution image possible within my budget, without regard to the ease of use that a major-manufacturer camcorder lends. It's not so much that I'm looking for an "entry-level" HD cam, than I have a limited budget compared to the usual funding necessary for a 1080/24p camera but still want to achieve picture quality roughly comparable to the lower-end 1080/24p devices out there.

[Jason Rodriguez]

Not wanting to nip this discussion, as it is a very viable idea, but not with the Micron CMOS sensors.

 

I'm not sure if you're familiar with the notion of a rolling shutter, but that is what the Micron sensor uses, and it's read-out rate is too slow to get rid of the "skew" that a rolling shutter with too slow a read-out rate will deliver.

 

You can read more about it here:

 

http://digitalcontentproducer.com/hdhdv/de...h_hdv_10092006/

 

But basically the Micron is no good for cinematography . . . it produces some excellent images, and has great color reproduction, but is not designed with the high-speed readouts necessary for cinematography work, and any panning, etc. will show a really bad "skew" to the footage like you see in Steve Mullen's examples. For instance, the Altasens chips we're using in the SI-2K read out almost three times as fast as the Micron sensor, which prevents any issues with rolling shutter skew.

 

Also I've used XCAP . . . ugh . . . their frame-grabbers are very nice and quite economical, but the software is really bad. I would suggest the EPIX frame-grabbers, but get StreamPix or Video Savant as recording software instead of XCAP.

[Troy Warr]

Not wanting to nip this discussion, as it is a very viable idea, but not with the Micron CMOS sensors.

 

Hi Jason -

 

Thanks so much for your input! That's exactly what I needed to hear. I figured that it was possible in theory, but as I have so very little experience with industrial cameras, I can't really visualize the drawbacks without hearing from someone who has witnessed them firsthand.

 

I'd heard the term "rolling shutter" before, but hadn't investigated it until now - gotcha. I can see how that would be a terminal problem for my intended use.

 

Do you happen to have any recommendations for a particular camera and/or integrated sensor that would fit my budget? (~$5000) I guess what I'm really trying to do here is set up a "Baby SI-2K Mini" if you will - ideally 1080p resolution, remote capture/framegrabber to PC, but with obvious compromises in terms of image quality, flexibility, workflow, and power as an imaging tool when compared to the SI-2K itself. Does Silicon Imaging make anything that may fit the bill? Is that asking too much for my budget?

 

If nothing like that exists from SI, any idea if AltaSens manufactures a lower grade of imaging chips and/or cameras with these capabilities that are in my budget?

 

Though I'm no marketing pro, I tend to feel like there would really be a market for this kind of setup. With storage prices plummeting and bandwidth rising, and with a "hole" in the market between highly compressed HD like the HDV or H.264/AVC codecs and higher-quality formats like DVCPRO HD and Cineform, I think that independent filmmakers would really eat up a solution that would allow for at least an intermediate quality level, albeit with a non-traditional hardware configuration and workflow. I just don't see the logic in investing in HDV hardware and high-quality optics, when the compression is what's really going to bite you, and there's no (easy) way around that. But, the next useful step up from there is a *big* step - you're in the realm of a $15-20K setup like the SI-2K or (potentially) RED.

 

Thanks again Jason -

 

Troy

[Charlie Dijak]

Hello Cinematography Forum:

EPIX, Inc., has added an ?FAQ? page to its description of the SILICON VIDEO 9T001C camera. The URL is: http://www.epixinc.com/products/sv9t001.htm. Most of the questions will look familiar since they were pulled from your discussions here.

 

EPIX, Inc. would like to thank Hal Smith for contacting us and thus giving us a chance to respond to your questions.

 

Best Regards,

Charlie Dijak

EPIX Sales Dept

E-mail: cad@epixinc.com

Tel: 847-465-1818

[Jason Rodriguez]

Hi Charlie,

 

BTW, I want to qualify my statement about XCAP being really "bad" . . . it's quite powerful and has a very full-featured assembly of engineering and analysis tools (we use it internally at SI quite a bit), it's just not very straight forward when it comes to recording as a cinematographer in the field (which is not what it's intended for).

 

Sorry, since this is a Cinematography forum, I was making a general statement to the use of XCAP as one would use it in this profession, not it's overall capabilities, and *definitely not* it's capabilities as a sensor profiling, engineering, and data analysis tool.

 

And of course you guys make great frame-grabbers :)

[Jason Rodriguez]

Hi Troy,

 

I would combined the EPIX frame-grabbers with some sort of CCD camera that had a full-frame shutter, rather than the rolling shutter of the Micron . . . not that rolling shutter is no good in and of itself, but the Micron's read-out rate is too slow for cinematography use. As I mentioned before, the Altasens has a rolling shutter, but the read-out rate is 3x faster to prevent any visible skew.

 

There are a number of cameras on the market based on the 4-megapixel Kodak KAI-4021, I would try some of those. Imperx for example, or Pulnix. They can do 24fps at up to 2K resolutions, and will give you pretty clean results out of the box. Combined those with the EPIX frame-grabbers and some recording/VTR software, and you'll have a pretty powerful and high-quality recording combination based on industrial camera technology.

[Phil Rhodes]

Hi,

 

You have to understand that this is basically what Silicon Imaging have already packaged up and sold, but of course they've done a lot of work with LUTs and image manipulation.

 

I do wonder if you could combine something like this with a Pro35-style groundglass adaptor, and offset some of the resolution loss by evaluating what would happen if you treated the groundglass as part of the low-pass filtration. I don't know if the LPF is glued irremoveably on the front of the chip at manufacture, but it'd be an interesting thing to try. Most of the time you're effectively low-pass filtering the image twice with these groundglass devices.

 

Phil

[Cesar Rubio]

Hi everybody:

 

First day and post. I hope I won't offend anybody or break a forum rule with this.

I have some Images to show on my forum that were taken with AVT Cameras, just in case somebody wants to see them at:

 

http://www.davidrubio3d.com/view_forum.php?id=3

 

Thanks and nice forum.

 

Cesar Rubio

[Phil Rhodes]

Hi,

 

That's interesting.

 

I notice some very severe colour moire pattering, especially in the resolution targets but also in the small white text on the colour swatch. What software are you using to capture and de-bayer the images? Have you provided for proper low-pass filtering on the cameras?

 

Phil

[Cesar Rubio]

Phil:

 

You are not the one from the Sopranos right?...I am just kidding (lol):D

 

The charts images were taken in the AVT installations on Massachusetts USA.

Tomorrow I will have Scott Smith's answers for some of your questions.

 

The ugly Images of me, were taken with a Markin F-033c which has a IR (Infrared) cut filter by default. The non "c" (color) version, does not have this filter and it gives better resolution for monochrome (black & white) images.

 

I used the AVT SmartView 1.5.1 to grab single images.That software comes free (among other software applications) with every AVT Camera.

 

For recording to disk (continuos video) I use the Stream Pix 4.3 software.

 

Thanks for your interest, we'll have more info later so stay "tunned".

 

Cesar Rubio.

[Cesar Rubio]

NOTE: This is a edited post. In the previous one I had problems editing it.

 

Phil:

 

You are not the one from the Sopranos right?...I am just kidding (lol):D

 

The charts images were taken in the AVT installations on Massachusetts USA. Tomorrow I will have Scott Smith's answers for some of your questions.

 

The ugly Images of me, were taken with a Markin F-033c which has a IR (Infrared) cut filter by default (you can remove it, and put it back if you want) The non "c" (color) version, does not have this filter and it gives better resolution for monochrome (black & white) images.

 

I used the AVT SmartView 1.5.1 to grab single images.That software comes free (among other software applications) with every AVT Camera.

 

For recording to disk (continuos video) I use the Stream Pix 4.3 software.

 

Thanks for your interest, we'll have more info later so stay "tunned".

 

Cesar Rubio.

[Troy Warr]

EPIX, Inc., has added an ?FAQ? page to its description of the SILICON VIDEO 9T001C camera. The URL is: http://www.epixinc.com/products/sv9t001.htm. Most of the questions will look familiar since they were pulled from your discussions here.

Charlie, thank you! I read through the FAQ and that answered most of my questions about the 9T001C. It seems like a wonderful camera - though unfortunately, it sounds like the rolling shutter would indeed inhibit its use from a cinematography standpoint.

 

I'm curious to know, Does EPIX have any plans to offer an HD resolution (1280h x 720v, or 1920h x 1080v) camera with a global/snapshot shutter in the future? I noticed that the SV642 has a global shutter but is a standard definition camera. I'm still thinking about using that in specialty applications due to its high frame rate, but I would definitely be interested in an HD camera (capable of 24 frames/second) if one should be available in the future.

 

Thanks again,

 

Troy

[Troy Warr]

Hi Troy,

 

I would combined the EPIX frame-grabbers with some sort of CCD camera that had a full-frame shutter, rather than the rolling shutter of the Micron . . . not that rolling shutter is no good in and of itself, but the Micron's read-out rate is too slow for cinematography use. As I mentioned before, the Altasens has a rolling shutter, but the read-out rate is 3x faster to prevent any visible skew.

 

There are a number of cameras on the market based on the 4-megapixel Kodak KAI-4021, I would try some of those. Imperx for example, or Pulnix. They can do 24fps at up to 2K resolutions, and will give you pretty clean results out of the box. Combined those with the EPIX frame-grabbers and some recording/VTR software, and you'll have a pretty powerful and high-quality recording combination based on industrial camera technology.

Jason,

 

Thanks so much for your help! I contacted Imperx and Pulnix as you mentioned, and also came across Prosilica, who offers at least 2 models that meet my resolution goal, one of which features the Kodak KAI-4021 sensor. I'm waiting to hear back - hoping that these solutions are within my budget.

[Troy Warr]

You have to understand that this is basically what Silicon Imaging have already packaged up and sold, but of course they've done a lot of work with LUTs and image manipulation.

 

I do wonder if you could combine something like this with a Pro35-style groundglass adaptor, and offset some of the resolution loss by evaluating what would happen if you treated the groundglass as part of the low-pass filtration. I don't know if the LPF is glued irremoveably on the front of the chip at manufacture, but it'd be an interesting thing to try. Most of the time you're effectively low-pass filtering the image twice with these groundglass devices.

Thanks for the feedback, Phil - I have considered that, and realized that I'd be foolish to try to approach the quality of the SI-2K Mini in a substantially cheaper package. They've priced their cameras at a level that reflects the equipment involved and all of the man-hours of research required to build them. As I mentioned I'm mainly looking into the viability of an "entry-level" version of the SI-2K - a camera with understandably more limited features, image quality and applications - but for those same reasons, cheaper. I have more time than money to invest at this point, so I'm more than happy to invest days/weeks/months into building and configuring such a system, should it prove to be potentially "worth it" from a results standpoint. If I find that I might as well go with a pro-sumer HDV cam for my applications, I'll most likely just drop the idea and pick one of those up.

 

The SI-2K Mini is exactly what I'm looking for, and if I had about $15K to spend I'd certainly pick one up. But, I'm hoping that there's a way to emulate at least some of the spirit of the SI-2K in a cheaper package.

 

I'll have to look into your LPF idea, BTW - I'm not sure that I understand the technical requirements of doing that just yet, but since I do plan to use a 35mm ground glass adapter should this work, I'm going to be interested in getting the best results from that. I expect I'll have to spend some time communicating with RedRock, Cinevate, etc. to see what their thoughts are on the compatibility of such a system, but I'll be sure to bring that point up.

[Troy Warr]

Hi Cesar,

 

Thanks for your input! It's great to meet you - and it sounds like you're already well into implementing the type of solution that I'm wondering about.

 

I'm very interested to learn more about the AVT cameras and your experiences with them, most of all the PIKE F-210C. Like Phil, I'm a little concerned that the image quality does not immediately appear to be comparable to what I'm used to seeing from HD camcorder frame grabs. It may just be the fact that I'm used to "cinematic" images of softly lit faces, nature, scenery, etc.; but, the moire patterns, unusual "shadowy" feedback on the edge of the "TARGET... MULTI RANGE" chart (near the top right of the 90-degree rotated, full-size image), and the (for lack of a better term) "blocky" anti-aliasing are sort of troublesome.

 

I noticed in a post in your forum that you mentioned proper setup for industrial cameras as a way to achieve more film-like/cinematic results - are you able to elaborate on that a bit? I also noticed that the PIKE F-210C has a 1" sensor - that's very large; are you able to capture shallower depth of field to at least rival 16mm film? Do you need special C-mount lenses for that camera, or is 1" a standard size (e.g. like lenses for 2/3" or 1/2" applications)?

 

Lastly, in your experience, do companies like AVT readily offer price lists? I've contacted a few companies about their industrial/machine-vision cameras, and it seems to be an industry custom to be tight-lipped about price. I don't have a ton of money to spend (maybe ~$3K for the camera itself, ~$5K for a complete setup, though I just bought a PC that should be adequate), so I'm not sure if I'm wasting my time in pursuing such a solution over the readily available HDV pro-sumer camcorders out there.

 

Thanks again for your input, and hope to hear from you soon!

[Cesar Rubio]

Troy:

 

Nice name...I like the movie very much!

 

I don't want to steer away anybody from this excellent forum, but all the questions & concerns that you have, it would be better to post them on the images topic , so Scott Smith could answer them for you (and everybody interested) He has limited time and he does not usually participate on forums.

 

For the price list, I will soon post it on my forum to everybody to see. One thing that I can tell you is that the Pike F-145 c list price is $3490 and I honestly prefer 720p over 1080p if you are not going to show your work on a big screen Digital Cinema.

 

For the sendor size comparison see this please:

 

http://www.dalsa.com/dc/origin/dc_sensor.asp

 

The small HD (in red) sensor is a 2/3" the same as the F-145c and the 1" on the Pike F-210 is a little bit bigger than that. the 16mm film size is precisely that 16mm in width. A 1 inch CCD is only 12.8mm. So you get a better DOF for 3-D work with Digital Cameras.

 

Thanks.

Cesar Rubio.

Edited January 12, 2007 by Cesar Rubio