How much is MOBILE HDMI or HD-SDI PLUS analog SD/HD aquisition worth to you?

[Alex Maranda]

Originally Posted by Alex Maranda

@Dave

For those people wanting an uncompressed HD workflow this solution is indeed of no interest; that market is catered by the Wafian HR-1, at $17,500 a pop.

I want to correct myself, the Wafian is also Prospect HD with 10 bit Cineform DI. Just to muse a bit on the question posed by Dave Perry, who exactly wants to store 1.485 Gbit/s? Noone is doing this, not Silicon Imaging with a 2k sensor (they use Cineform RAW), and not RED (they have REDCODE). Everybody compresses to various degrees.

[Alex Maranda]

Originally Posted by Greg Hartzell

Why only 80gigs of storage? How many batteries will you have to carry with you for a days of shooting?

I'm sorry, I just don't get these indie film solutions. Why not just shoot on a better format?

120 and 160 GB HDD are available as well in 2.5'' form factor. It can be a build option. The point of 80 GB was to provide an intuitive reference for two hours of shooting, also being the sweet spot $/GB in 2.5'' form factor. Build option for sure.

This product concept targets a niche market where MOBILITY, low budgets and high resolution (both pixels and color) are important.

If you shoot with CineAlta or with SI or RED, then this is not your thing. If you shoot with the SR1, the V1U, the Canon G1/H1 (I listed them from the low to high end), then this could a product of value for you. Unless HDV@25Mbit is already good enough. What was the better format you were refering to?

[Alex Maranda]

Originally Posted by Jad Meouchy

I think heat dissipation and power consumption will be a very critical issue for that Core 2 (and the HDD). Intels are not known for their efficiency. You should consider an AMD mobile processor and a more embedded OS, probably stored on a locked flash drive. Also, why limit to a 2.5" drive? It's a very minor incremental cost to get a 350gb or even 500gb 3.5" drive and store 4-5 times as much data at a much higher transfer rate.

Wifi ! That is what devices like the firestore are missing. Even if it's not realtime transfer, do it to dodge the risks and engineering hurdles of hot connectors.

I hate to say that these kinds of solutions, while ideal for people like me, do not sell well. Good luck though.

Jad, I take your post in good faith; it is apparent you diagonally read the thread.
I will address your concerns one by one:
- it isn't a Core 2 Duo desktop it is the mobile Merom version (T7400 if you must), with a thermal envelope under 25W. AMD Turion has a higher thermal envelope than this; besides, I am using a SBC (which stand for Single Board Computer) designed around the Centrino mobile platform. I don't design this from scratch. I understand you are a fan of AMD laptops.
- a more embedded OS; as in what? vxWorks? Nucleus? Monta Vista Linux? and more exactly WHY, when the Blackmagic DRIVERS and the compression software (namely Online mjpeg or Prospect HD) runs only on Windows XP (Embedded or not). Just to stress the point: Off the shelf software you can buy now.
- The custom software image IS locked and stored on a CF card, which is field swappable. This CF contains a Windows XP Embedded build with only the required drivers and compression software, and power management. Re-read the thread.
- the 2.5'' HDD comes from the current form factor of the box, which has a 2.5'' shock-mount HDD bay. Also the higher storage is not really useful unless what you want is 12 hours of continuous HD shooting. Power consumption is also a factor - this box is meant to run off a battery belt.
- Wifi: transmitting to what exactly? and why? (I'll ignore the bit-rates for this discussion; I actually won't - 800.11g is 54Mbit/sec peak with 25 Mbit/sec effective). The solution proposed has a builtin 100 Mb Ethernet and optional Gbit ethernet (which I will provide anyway since it's like $30 a card). This Gbit Ethernet will allow accesss to the internal HDD (when offloaded to other storage) at its peak burst rate or 40 MB/sec (which - pay attention to bits and bytes), is an order of magnitude more than Wifi. Since we're discussing bitrates, both the mjpeg and the Cineform codecs are under 15 MB/sec, well under the average sustained rate of a 2.5'' HDD.



I would also like to make a general comment to you and other posters who replied to this thread. This wasn't meant as an applications engineering wiki; If you guys care to comment on the real hurdles I would face engineering-wise I will happily take the discussion offline with you; so far you haven't touched any of it, probably because none of those who replied are practicing electronics engineers.

Also, (this is actually more frustrating :), none of you touched the question that I did ask! What is this worth to you?
What is an attractive $XXXX to you IF YOU ARE IN THE MARKET?
And how many of you are out there?

Can I please get $$$$ replies from here on ;) This is market research, luck shouldn't be a factor when considering expensive gear. If there isn't a margin to make a business out of it, I will just use the prototype for my own video productions. It certainly is cost effective and future proof to me.

If you plan to use or are using one of the following cameras:
Sony HDR-SR1
Sony HDR-FX7
Sony HVR-V1U
Canon XH-G1
Canon XH-H1
...and future HDMI or HD-SDI cameras

this post MAY be for you, if you're interested in 1920x1080 4:2:2 10 bit YUV acquisition.

[Alex Maranda]

add to the camera list
JVC GY-HD100u
which has analog component output, supported by the DeckLink HD cards.

[Ken Hodson]

Originally Posted by Alex Maranda

Can I please get $$$$ replies from here on ;) This is market research, luck shouldn't be a factor when considering expensive gear. If there isn't a margin to make a business out of it, I will just use the prototype for my own video productions. It certainly is cost effective and future proof to me..

My answer to that would have to be cost wise, the same amount as you pay for it. Like yourself many of us here in these forums have been discussing and salivating and waiting for price/performance of uncompressed or better yet visually lossless capture systems to become doable. Many of these same people are extreemly up to date on hardware/software. Now that there are extreemly capable cams on the market and the intoduction of mobile dual cores has made everything extreemly doable now with high compression low bandwidth codecs. There have been many threads on this with many different means to an end explored. I think it fair to say that nothing that is not off the the shelf, is needed here.
This is not to say that some eager individual(you?) can't make a strong go of it (alienwear makes gaming PC's with completely off the shelf items, and the non techie loves it.) My advice to you is to build a working system, spec it out, include a worth while markup (bare minimum), and advertise it at the price YOU arrive at. I personally believe there is a market as long as the price is reasonable. If it sells you can always increase the price. If it doesn't sell, drop it like a rock.

[Alex Maranda]

Originally Posted by Ken Hodson

My answer to that would have to be cost wise, the same amount as you pay for it.
<snip>
My advice to you is to build a working system, spec it out, include a worth while markup (bare minimum), and advertise it at the price YOU arrive at. I personally believe there is a market as long as the price is reasonable. If it sells you can always increase the price. If it doesn't sell, drop it like a rock.

Hi Ken, Vancouver is my previous home :) I understand you guys have had quite the weather recently...

As per your reply: The same amount I pay for it? At cost? I don't quite get your brand of humor. All the knowledgeable people can go the DIY route then. Or not??
Ken, look, this is not a home improvement project you go to Home Depot and get "components" from. Alienware (I like your comparison!) makes PCs other ENGINEERS could build, if they can handle the logistics.
Of course, other skilled engineers who are not put off by sourcing components from distributors all over the world could have a go at this, just like I can.


This is clearly a niche market, one I don't expect any volume manufacturer to enter - the consumer doesn't know better or care. At the other end of the scale there is the Wafian.

This comes in between, low(er) cost, wearable!, highly mobile. Speaking of mobility: it can be made self-contained!
I can remove the PSU and place a battery inside, with an external charger. No need for a battery belt. I would love to get feedback on usability factors like this one.

Show of hands please! I will not start building these in series without a 50 unit waiting list. I'll just show off pictures of my prototype, as inspiration for other DIY builders :)

[David W. Jones]

Originally Posted by Alex Maranda

Show of hands please! I will not start building these in series without a 50 unit waiting list. I'll just show off pictures of my prototype, as inspiration for other DIY builders :)

With all due respect Alex, it would be just as foolish for us to get on a waiting list for a product that has yet to be designed, tested, and priced, by someone we know nothing about,

as it would be for you to set a pricing structure by asking a forum "what is it worth to you" without knowing your cost and time involvement, or setting up a support structure.

[Greg Hartzell]

as it would be for you to set a pricing structure by asking a forum "what is it worth to you" without knowing your cost and time involvement, or setting up a support structure.

"Setting up a support structure," now that's a brilliant question. Do you plan on fixing 100 some odd machines for their useful life span or are you going to send people to the wayside when their harddrive crashes (hypothetically)?

I think heat may be an issue as well. All the best of luck in the world to you Alex. I think this is a cool idea, but this is one area where smaller would seem to not to be better.

[Ken Hodson]

Originally Posted by Alex Maranda

The same amount I pay for it? At cost? I don't quite get your brand of humor. All the knowledgeable people can go the DIY route then. Or not??

Yes we have had some big winds on the coast this year. Also more snow then we have had for a long time(and it stayed for more then a day! Yech)

But back to the topic. My point about cost was that many people will be in the D.I.Y category, and once you consider a reasonable markup for yourself it wouldn't be worth it for many of us. Just like my previous point of Alienwear gaming PC's. Nice PC's but I would waste extra money on their product when I could build it far cheaper myself. But obviously many people do buy them.
The factor many of us are saying is build it. Get a web page. Show picks. Spec it out. And give us a final price. You may very well be surprised at your level of response and have a ton of orders lined up. It is no surprise that people want a product like this. But do they want your product at your price? Show it to us with a price and we will give you feed back.

[Alex Maranda]

David, Greg, Ken,

these are excellent questions; I intend to provide warranty for the whole system, which cannot be longer than the warranty I get for the individual parts; outside that I will offer paid extended warranty. Any firmware upgrades will be free under warranty.
The warranty of a HDD is tipically two or three years; it depends on the manufacturer. The cost of shipping under warranty repair will be split 50% with the customer. At a minimum there will be a two year system warranty included in the purchase price, for "normal" failure modes (that is - no visibile physical abuse).

About heat dissipation: yes, as I already alluded to this, it is a challenge. Because of the very small size of the box, active cooling is not an option. However, the enclosure is made of an alluminium alloy, with good thermal conductivity. I plan to use the whole enclosure as a heatsink - it will be quite warm to touch, by design.

Do you have an oppinion on the usability (mobility) factor? As I mentioned, I could remove the PSU and replace it with a heavy duty battery INSIDE the enclosure. It will make the box quite a bit heavier, but just one item to carry. However leaving out the PSU will make the design exclusively mobile; that is, when the battery runs out (wont last the hour I'm afraid) you have to recharge. With the PSU plugged to AC mains one could shoot until the HDD is full, but for mobility would have to carry a separate battery pack/belt.

I will take your advice and build the prototype; my attempt at market research was more about volume projections and feature set than what price the market will bear. The price for the first units will carry retail markup from my parts sources. If I get any decent volume going, I will be able to pass on the discounts.

Thanks for the feedback so far.

[Jad Meouchy]

Originally Posted by Alex Maranda

Jad, I take your post in good faith; it is apparent you diagonally read the thread.

Guilty.

Originally Posted by Alex Maranda

- it isn't a Core 2 Duo desktop it is the mobile Merom version (T7400 if you must), with a thermal envelope under 25W. AMD Turion has a higher thermal envelope than this; besides, I am using a SBC (which stand for Single Board Computer) designed around the Centrino mobile platform. I don't design this from scratch. I understand you are a fan of AMD laptops.

25W is very high, in my opinion, for a CPU in a device of this footprint. I was suggesting something more embedded so that you could offload processing from the main CPU and thereby reduce the overall power requirements. With a daughter encoder board, this should not be a processor-heavy task. A 400mhz celeron would be plenty of power for managing I/O.

Originally Posted by Alex Maranda

- a more embedded OS; as in what? vxWorks? Nucleus? Monta Vista Linux? and more exactly WHY, when the Blackmagic DRIVERS and the compression software (namely Online mjpeg or Prospect HD) runs only on Windows XP (Embedded or not). Just to stress the point: Off the shelf software you can buy now.

How about Windows CE? And I had no idea the drivers only worked on XP; I would have expected them to work on 2000, Vista, and Unix (Mac). Windows Embedded is no more 'off-the-shelf' than QNX Neutrino..

Originally Posted by Alex Maranda

- The custom software image IS locked and stored on a CF card, which is field swappable. This CF contains a Windows XP Embedded build with only the required drivers and compression software, and power management. Re-read the thread.

I was poorly implying that locking the boot image away from the user would allow the storage medium to be field-swappable instead.

Originally Posted by Alex Maranda

- the 2.5'' HDD comes from the current form factor of the box, which has a 2.5'' shock-mount HDD bay. Also the higher storage is not really useful unless what you want is 12 hours of continuous HD shooting.

What about shooting three consecutive days of a trip? 12 hours of continuous shooting sounds like a lot until you talk about more than one shoot.

Originally Posted by Alex Maranda

- Wifi: transmitting to what exactly? and why? (I'll ignore the bit-rates for this discussion; I actually won't - 800.11g is 54Mbit/sec peak with 25 Mbit/sec effective). The solution proposed has a builtin 100 Mb Ethernet and optional Gbit ethernet (which I will provide anyway since it's like $30 a card). This Gbit Ethernet will allow accesss to the internal HDD (when offloaded to other storage) at its peak burst rate or 40 MB/sec (which - pay attention to bits and bytes), is an order of magnitude more than Wifi. Since we're discussing bitrates, both the mjpeg and the Cineform codecs are under 15 MB/sec, well under the average sustained rate of a 2.5'' HDD.

Many wireless technologies are reaching 100MB, even on an 802.11b foundation. Having a high speed wired transport is great, but what I would like to see is a low-speed (i.e. real-time) wireless transport, so I could duplicate to a high-density data storage solution at the same time, or even use a wireless, completely digital, field monitor. This feature was more of a personal request than something with mass appeal.

Originally Posted by Alex Maranda

I would also like to make a general comment to you and other posters who replied to this thread. This wasn't meant as an applications engineering wiki; If you guys care to comment on the real hurdles I would face engineering-wise I will happily take the discussion offline with you; so far you haven't touched any of it, probably because none of those who replied are practicing electronics engineers.

Well if you're going to try and sell me a tech device that you've built and assembled, I'm going to need to know what's in it. Your post sounded like a feeler thread. You put enough specifications in your post to make it sound like a tech doc, but then you criticize the tech people who read it! I'm just trying to help a fellow EE, sheesh.


I would be prepared to pay about 2500-3000 for it (not including cost of storage medium). I don't know if that's reasonable or not, but there's one number.

It's a good idea for a solid project; I didn't mean to say otherwise.

[Alex Maranda]

Originally Posted by Jad Meouchy

Well if you're going to try and sell me a tech device that you've built and assembled, I'm going to need to know what's in it. Your post sounded like a feeler thread. You put enough specifications in your post to make it sound like a tech doc, but then you criticize the tech people who read it! I'm just trying to help a fellow EE, sheesh.


I would be prepared to pay about 2500-3000 for it (not including cost of storage medium). I don't know if that's reasonable or not, but there's one number.

It's a good idea for a solid project; I didn't mean to say otherwise.

Jad, I may have been a wee bit abrasive...I guess I need to pick up some customer support skills. Thing is you didn't come off as an EE (I stand corrected), rather more like fishing for buzz features. I don't have OEM R&D resources or budget, hence the locked down spec right down to form factor, OS, and connectivity.
Look, if Wifi is really a cool feature to have then it can become a build option (a Wifi PCI card is $90), but at the expense of Gigabit Ethernet! I thought Gb Ethernet is a no-brainer, but if you can make do with the onboard 100 Mbit plus a Wifi card then I will be happy to build your rig.

$3000 you say....not with one of the $995 DeckLink HD cards I'm afraid - remember I don't get any volume discount since there's no volume. It may be a feasible price point for the Intensity HDMI card though.

You do have a point with the 25 W TDP for Merom; fortunately the CPU socket is also pin compatible with Yonah cores, which go down to 9-11 W in battery mode. OTOH Merom can be undervolted by some margin....this is one for the lab tests to tell.

[Greg Hartzell]

I still don't get al of these quables over form factor here. The way I see it, if you are already carrying around lights, props, support, grip equip, etc.

Here is my idea: For under a $5000 price point, seems like you could build and market a dedicated capture machine into a 2u or 3u rack case. In this rack you could have a hot swappable raid (optional maybe?), you could output either timecode or a reference video signal routable to an external recorder (maybe in the same rack?), have plenty of cooling and power options. While you wouldn't have the form factor, you have a very scalable solution that would work for a variety of situations, and have the option of capturing uncompressed if there is the need. This option might appeal to people who regularly use larger cams like the xdcams and the F900. Just an idea though.

[Ken Hodson]

Personally I would be happy with a ATX microATX form factor. Larger components would give the option of full sized drives and the capability of using less cpu intensive codecs like sheervideo or mjpeg2000 at the sacrifice of drive space(which is cheap and plenty full at the 3.5" size(two disk RAID). Ultra small form factor would be nice, but I don't know if the cost/performance would be worth it for many. I have become used to shooting HDV with a laptop and HDV rack. I need that large LCD monitor and clip control. But then again I don't do run and gun, so finding wired power isn't a problem and if it is I'm not adverse to packing along an inverter and some deep cell batteries. If I have to go fleet of foot then I have the HDV tape to carry me through without worry.
As soon as someone figures out how to get HDMI into a laptop, with rapid adoption of HDMI by the new cams now, I think ultra portable uncompressed will be in the reach of everyone. Now if we can just get HDVrack to work uncompressed. Of course I am behind anyone who is willing to push the envelope in developing anything lossless and portable.

[Phil Stone]

I would buy one. Even a low quality HDV hard drive recorder costs €2000. If this thing is going to be about €4-6000 (card dependent) then its obviously going to be a hugely popular product. If it had some kind of video out you could plug in something to monitor the video.

Phil