Storage media and formats.

Forum for discussion of narrow-bandwidth mechanical television

Moderators: Dave Moll, Andrew Davie, Steve Anderson

Storage media and formats.

Postby Steve Anderson » Fri Jun 22, 2007 2:54 pm

In the thread "Missing sync pulse and the club standard --- Why?" discussion has gone off at a tangent into the realms of storage media for NBTV that satifies the requirements.

So that thread can continue in it's intended course I've started this one which I hope will deal with the problems that conventional storage media poses for NBTV and a means to circumvent them.

For the time being it's best to stick with Klaas's and Vic's colour encoding format to be used within .wav files rather than an audio CD standard. There is in theory no reason why these couldn't be DC coupled which would eliminate the problems inflicted by AC coupling and a frequency response that doesn't go all the way down to DC.

Using a sample rate of 48kHz rather than 44.1kHz means we get 120 samples per line rather than the awkward 110.25. Files of the same duration will be 9% bigger, but in todays world that's no real issue. An hour of two 16-bit data streams sampled at 48kHz comes out at 691.2 Megabytes which would just fit on a CD and more recent flash drives.

The use of a .wav file format means that it can be handled just like any other file, on (data) CD, flashdrives, it can be e-mailed, posted to this forum or placed on a web site for download. It might be a little impolite to post a file of that size in this forum without checking with Andrew first!

The use of a PC sound-card or a conventional audio CD player is easy and convienient. Plus with the PC you can edit and do other things to the data you might want to. However, they are not DC coupled, at least I haven't come across one that is.

So here's the challenge...

An interface that takes signals from the source and generates a DC-coupled .wav file. This can be in several formats, all compatible

1) Mono 8-bit at 48kHz, video only as the greyscales I have posted.
2) Stereo 8-bit mono video and sound at 48kHz. (8-bit sound can be quite good!)
3) Stereo 16-bit, video encoded as per Klass & Vic's standard and 16-bit mono sound.

There are other combinations, but that's enough for now.

On playback it should simply do the reverse of above and output up to four DC coupled signals, R, G, B and sound. (OK, the sound doesn't need to be DC coupled and it's probably better that it isn't, but that's easy).

Another advantage is that those that don't want to use the interface can still play the files back on a PC with a soundcard, but the colour information would be as good as lost resulting in a compatible monchrome signal.

So gents, let's get on with it.

As a footnote, I noticed some years ago that some audio editing sofware has a 'software capacitor' within it thereby completely undoing all we are setting out to do here. Caution is advised.

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Postby Steve Anderson » Fri Jun 22, 2007 5:39 pm

I've got several spare operational hard disc drives kicking around....I wonder?

Steve A.


This wasn't really a serious consideration, beyond the headache of interfacing with the disc there's the fact it's not easy to pass around and give to someone else to use.

The CD/DVD seems the ideal media, it's cheap, can be stuck in the post and exchanged easily. The use of the .avi file file format means it can be handled like any other.

This is really a hardware issue, the file format can easily carry DC-coupled signals, it doesn't care what the data actually represents.

If only we could get a data stream in and out of a PC simply then the rest of the hardware is easy. But thus far I keep hitting barriers.

What I'm trying to do is keep this "Convention Compatible". I don't want to force the use of parallel ports as that would usually mean having to drag along a desktop PC, monitor etc.

The only I/O this laptop has is analogue audio I/O, dial-up modem I/O, PCMIA, USB and the network port. Now that sounds quite a lot, but they're all somewhat of a headache to work with.

I'm trying to keep it as simple as possible, such that the average NBTVer could tackle it with a reasonable degree of confidence. Which led me to mention a stand-alone device and the dabbling with the CD drive I did a couple of weeks ago.

So at the moment I've come to a bit of a halt. And it's frustrating!!!

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Data I/O.

Postby Steve Anderson » Fri Jun 22, 2007 6:10 pm

Doing some further research on the Elexol site I came across a little device which might be the answer to our (or at least my) prayers. The USBIO24R Digital I/O Module. (PDF attached).

This has three 8-bit I/O ports, sadly not four, but that can be got around in hardware.

I'm not sure what the cost is but I think it's worth persuing. The rest of the hardware, although not trivial, would be quite possible for most NBTVers I think.

Where I come unstuck is the software for it...any takers?

Steve A.

Having studied the circuit diagram a bit more closely I'm a bit concerned that Port C only shows 6 lines connected, not 8. I'll bung them an e-mail.
Attachments
USBIO24Rds1.pdf
(226.83 KiB) Downloaded 789 times
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Data I/O.

Postby Andrew Davie » Fri Jun 22, 2007 7:29 pm

Steve Anderson wrote:Having studied the circuit diagram a bit more closely I'm a bit concerned that Port C only shows 6 lines connected, not 8. I'll bung them an e-mail.


I've had a read of the spec, and surely that's an error in the circuit diagram. The documentation clearly states 24 independant I/O lines. Also note, the command table talks about writing "1 byte port data" for all three of the ports.

Looks like a nice device, but where I have a problem with this is that it is Windows-only!
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Re: Data I/O.

Postby Steve Anderson » Fri Jun 22, 2007 7:42 pm

Andrew Davie wrote:Looks like a nice device, but where I have a problem with this is that it is Windows-only!


Yes, it's somewhat depressing. But on the positive side it's a step in the right direction. Perhaps there are other companies that make similar products with drivers for other OSs. And perhaps a forth or more port. Two could be used to get six ports I guess.

What OS were you looking for it to work with?

I've found what seems to be an updated version of the same device, the circuit diagram has been amended to show all three ports as 8-bit. Plus using what is called 'Header 14' it might be possible to get four or more ports out of the device with a little hardware fudging.

What concerns me now is are we potentially pushing USB1.0 to it's max? Four 8-bit data streams at 48kHz = 1.536 Megabits/sec, and that's without any control data contained within. Or is USB1.0 specified in Megabytes/sec? I'm never sure without a timing diagram if MB or Mb means bits or bytes.
Attachments
USBIO24.pdf
(477.5 KiB) Downloaded 747 times
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Data I/O.

Postby Andrew Davie » Fri Jun 22, 2007 10:57 pm

Steve Anderson wrote:What OS were you looking for it to work with?


Well, I'm a Windows person -- but NBTV isn't :) I would like to see any NBTV-related developments OS-independant.

Steve Anderson wrote:I'm never sure without a timing diagram if MB or Mb means bits or bytes.


Not that you can particularly rely on people to use the correct abbreviation -- but in any case, MB is megabytes, and Mb is megabits.
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Re: Data I/O.

Postby Steve Anderson » Sat Jun 23, 2007 3:08 pm

Andrew Davie wrote:Not that you can particularly rely on people to use the correct abbreviation -- but in any case, MB is megabytes, and Mb is megabits.


Well I opened up "USB in a nutshell" and thankfully within a few pages it defines the speeds...

Low Speed - 1.5Mbits/s
Full Speed - 12Mbits/s
High Speed - 480Mbits/s (USB2.0 only)

They wrote the words just as above, confusion gone. So to use the USB to multi-port device we would need to use the Full Speed mode (USB1.1).

I'm trying to stick to USB1.0 or USB1.1 as not everyone has USB2.0 (me!).

Plodding on through the spec I came across the clock speed used for each mode, for Full Speed it's logically 12.000Mb/s (their abbreiviation) with a tolerance of +/-0.25% (2,500ppm) which is a bit naff really. In real life one would hope that it's a bit better than that.

There is naturally some overhead within the 12Mbits/s stream but on first glance plenty of bandwidth for our snail-like 1.536Mbits/s. Then there is the driver for the Virtual Comm Port, I wonder if that slows things down?

I guess there is only one way to find out is to buy one and give it a blast.

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

USB Speeds.

Postby Steve Anderson » Sat Jun 23, 2007 6:41 pm

Before I rushed out and bought one of those multi-port devices I wanted to know if the USB ports here actually supported the Full Speed mode of 12MBits/s. It appears they do. The machines both are running Win XP Pro and only use the drivers contained within.

One is my laptop, a very entry-level machine by todays standards, the other is my wifes desktop which I built out of a pile of bits. They are both about three years old.

I used two flash drives as the USB external devices, one of 128MByte, also three years old, the other 2GByte, only six months old. I created a 100MByte file and timed how long it took to read and write this file to the flash drives on each machine.

They all came out as just shy of 8MBits/s, just under 1MByte/s, with the exception of the older flash drive that in writing to was about half that speed. So I think I'm OK as far as the hardware running in Full Speed mode is concerned.

Time to go shopping...

Steve A.

Hmm, maybe not. Having downloaded the datasheet for the FTDI interface chip (FT245BM) it appears that using VCP drivers slows it down to 300kBytes/s which is becoming marginal. Using D2XX drivers it will go at 1MByte/s.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Data I/O.

Postby Steve Anderson » Tue Jun 26, 2007 2:09 pm

Andrew Davie wrote:The documentation clearly states 24 independant I/O lines. Also note, the command table talks about writing "1 byte port data" for all three of the ports.


I had a very quick reply from Elexol and as suspected it was an omission on the circuit diagram. They will upload a corrected version.

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Postby AncientBrit » Thu Jun 28, 2007 5:44 pm

FTDI site is not the easiest to negotiate but I found this device which is a USB flash pen interface.

http://www.vinculum.com/prd_vdrive1.html

Drive1 is only available to developers, Drive2 is available as a boxed unit at around 14 uk pounds.

The bumf says its controllable using 4 wires with a "simple VDAP Command set". with either RS232 or SPI interfacing.

The implication is that it could be used with a microcontroller rather than a PC/Windows.

Might be worth a punt,


Graham
AncientBrit
Green padded cells are quite homely.
 
Posts: 858
Joined: Mon Mar 26, 2007 10:15 pm
Location: Billericay, UK


Return to Mechanical NBTV

Who is online

Users browsing this forum: Google [Bot] and 21 guests