On 05/22/2015 01:15 PM, Chris Osborn wrote: Do you have anything like Freecycle in your area? Usually, if you say you're looking for an old-style TV, people will jump at the chance to give away the old sets. --Chuck

On Fri, 22 May 2015, Chris Osborn wrote: A little over half a century ago, UHF was an extra price option! My brother and I pooled our money and bought a 19" Philco "portable". Our father chipped in the additional to get UHF. I remember watching the Cuban missile crisis on it, with my father muttering, "The SOB has gotten us into war!"

On Fri, 22 May 2015, Mark J. Blair wrote: If'n you want composite, instead of RF on Coco, it is truly trivial to connect, once you open the case (guess what the input to the RF modulator is)

The main problems stem from the fact that these computers output anything but broadcast-standard video. In some cases it was because they were built to a price and it was 'what can we get away with'. In others it was more a case of getting extra features (like colour) almost for free. Incidentally, one of the better things (for me) about Amstrad machines is that service manuals existed for them. Said manuals are essentially a schematic and a parts list, but that is all that is normally needed. Certainly for the older machines (all the CPCs and PCWs and the earlier PCs) they are not just boardswapper guides. What are you doing for the PSU (the computer ran off the SMPSU in the monitor IIRC)? One problem with SCART (and I don't think it's the cause of your problems) is that it is several interfaces on one connector. In particular there is composite video, RGB video (using the composite pin for sync) and later S-video (using the composite pin for Y and IIRC the 'red' pin for C). Not all devices implement all parts of the interface. In particular UK TVs almost always have the RGB inputs, VCRs did not. The CPC output is RGB video, and AFAIK the CPC-SCART cable is a simple cable with perhaps level-shifting resistors inside. So it will use the RGB pins on the SCART connector. -tony

Assuming that I don't find an off-the-shelf converter that Just Works for our poorly-behaving vintage computer video outputs, what I have in mind is this: A converter that is specifically designed to emulate the response of an 80's TV or 80's composite monitor when driven by a vintage computer output, and translate that as well as practical to modern displays (particularly, 1080p via HDMI). I have some ideas about how to accomplish this, but I will need to do more work to see if I can create a solution that is not absurdly expensive. On the input side, I envision having two RCA jacks and an F connector, accepting composite video, Y/C separated video (for C64, etc.), or TV RF input. It should be able to accept NTSC or PAL, so the US and UK folks can play. Is there a need for SECAM, or any other video standards? What about other electrical interface options? I'm only hands-on familiar with US machines. I understand that Atari computers were especially popular in Poland, so I'd like to support those... anybody here know what format/channel a Polish Atari 8-bit would output? In the middle will be some FPGA to perform any necessary magic. I've been looking at a prohibitively expensive ($115) one that has enough dual-port RAM blocks to support a frame buffer. I need to see if I can push the frame buffer out into external RAM in order to move to a cheaper FPGA. It would be ideal if the video parameters could be figure out automagically, but I have a feeling there will be a need for user-adjusted parameters, and possibly even loading up different FPGA programming to handle some odd-ball signal. Output would be HDMI, at 1080p. Are other interfaces and/or resolutions desirable? I'd like to keep it as feature-simple as practical. Handling the VHF/UHF tuner economically may be another sticky point. Maxim makes a tuner chip that's available at Digi-Key, but I refuse to design Maxim parts into anything on account of off-topic reasons. Mouser has stock of a very inexpensive ST tuner chip that looks very promising, but the full datasheet isn't openly available. I need to contact ST to see if I can talk them out of it. Their site mentions an NDA for the eval board, so it might be tough, particularly since my intention would be for my design to be open to allow off-label uses. Assuming I don't lose interest before completing this (a high-risk caveat, naturally) and that I can find a way to make the price bearable, what do y'all think about this silliness? I'm particularly interested in learning about non-US TV formats so I can design in maximum utility. -- Mark J. Blair, NF6X <nf6x at nf6x.net> http://www.nf6x.net/

I'm on the list, but it's so high-volume that I rarely read it. I'll look at the RGB2VGA board to see if I might learn anything from it. His mention of a line buffer is already my "Oh, duh!" moment about how to use cheaper external SDRAM instead of on-FPGA dual-port memories for the frame buffer. The dual-port memories are very convenient, but having enough to form a frame buffer pushes the design up into over-$100 FPGAs. I'm in Riverside, CA, but I already have enough real monitors. :) It's similar, but probably not right for this application. It outputs VGA, which is already obsolescent, at a maximum resolution of 1360x768. The goal of this project is to drive modern 16:9 monitors and TVs at 1080p, over HDMI. I believe that board on eBay is intended for replacing monitors in coin-op video games, which generally have very different video hardware compared to vintage 8-bit home computers. Has anybody tried this board with home computers that are known to be troublesome with modern displays? I'm at least interested in seeing how they got the price down to $40, and whether anything in their solution might be usable for this application without being able to read Chinese datasheets. :) On May 23, 2015, at 08:59, Jochen Kunz <jkunz at unixag-kl.fh-kl.de> wrote: I initially considered using SDR techniques for the TV demodulation. Using an off-the-shelf TV tuner IC would probably be much cheaper than building an appropriate RF front end from scratch, though if it outputs an IF instead of baseband then SDR techniques might still be the cheapest way to extract the video baseband, since the design would have a couple of ADCs in it in any case to handle Y/C from a C64. Don't want the ADCs to be too expensive, though... + Latency I think that's a non-starter due to latency. I can't imagine the overall latency being lower than several video frames, and that would be a killer for games. If it could work with sufficiently low latency, then using something like a Raspberry Pi or Beaglebone Black instead of a PC could lower the price and make it more stand-aloney. -- Mark J. Blair, NF6X <nf6x at nf6x.net> http://www.nf6x.net/

On May 23, 2015, at 9:33 AM, Mark J. Blair <nf6x at nf6x.net> wrote: The GBS-8200/8220 doesn?t support composite input, only RGB. I?ve used the board on quite a few of my computers that output RGB and it works fine. I?ve even got a couple of blog posts: ZX Spectrum 128: http://www.insentricity.com/a.cl/233 Commodore 128 CGA: http://www.insentricity.com/a.cl/219 BBC Micro: http://www.insentricity.com/a.cl/211 I still need to hook it up to the CoCo 3 RGB, but I can?t imagine any reason it wouldn?t work. It also works fine with my Apple IIgs. -- Follow me on twitter: @FozzTexx Check out my blog: http://insentricity.com

On May 23, 2015, at 12:05 PM, Steven Hirsch <snhirsch at gmail.com> wrote: I believe mine is a knock-off clone, and not a ?genuine? GBS-8200. -- Follow me on twitter: @FozzTexx Check out my blog: http://insentricity.com

On May 23, 2015, at 12:35 PM, Chris Elmquist <chrise at pobox.com> wrote: The GBS-8200/8220 supports analog signals, not digital. It will directly support the 15khz misnamed ?CGA? of arcade monitors (which is the same RGB signal the IIgs puts out), but that?s an analog signal, not digital. For use with the Commodore 128 I had to build a small circuit which converts the RGBI signal to analog RGB. In the case of the BBC Micro which also puts out 5V TTL digital RGB, I didn?t bother with lowering the levels since the GBS-8200 didn?t care and I didn?t need to deal with a 4th intensity signal. -- Follow me on twitter: @FozzTexx Check out my blog: http://insentricity.com

There are probably a fair number of "TV cards" in both ISA and PCI wandering about, since they're not terribly useful with the advent of digital TV (and the web). Has anyone hooked up an ordinary NTSC modulator with one of those and an 8 bit PC that relies on the peculiarities of NTSC chroma encoding? If so, how's the reception quality? --Chuck

On Sat, 23 May 2015, Chris Osborn wrote: Getting an acceptable combination of crisp 80-column text and proper color aliasing from a converter is decidedly non-trivial. I own one of just about every commercially available (and hobby) converters and precisely none of them provides a universal solution. Some give great displays from an Amiga and suck for anything else. Of my two (pricey) CVP CM-345S converters, only one provides useable display from an Apple IIGS. My GBBS-8220 can occasionally be coaxed into giving a solid display from a Color Computer 3. The list goes on... Luis's converter comes the closest to providing usable display from an Apple II, but (for me at least) only with certain levels FPGA code. Still, he is putting a huge amount of effort into the project and I have high hopes this may be a robust one-size-fits-all solution. If transparent, tweak-free emulation of a classic CRT display were easily doable, it would have been done by now. --

Even back in the day, I believe it involved swapping the cable between a monochrome monitor and a color one. I expect that this universal converter idea will require a way for the user to tell it whether they want color or monochrome video at the moment (and might as well let them choose what color phosphor to emulate). That way, an Apple II user who wishes to use modern displays could switch between 80 column mono mode for text or color mode for graphics, without swapping cables or displays. Aha! I've heard Nth-hand accounts of trouble getting vintage computers to play with video converters, but you sound like one of the folks with firsthand experience. Agreed! I envision that the converter should have some sort of smarts to help it figure out what sort of input it sees, but even in the ideal case it'll probably need some capability for user tweaking or mode switching for corner cases. The SAW filters I used are small enough that my 46-year-old eyes have trouble seeing them without a magnifying glass. :) I do almost all of my soldering under a stereo microscope any more. But I can solder down to 0201 discretes that way, although I don't like going smaller than 0402. I don't have a lot of experience with hot air soldering yet, but it has become quite available to hobbyists, with hot air systems now available for around $100. The "Maker" movement has done a lot of good in the area of making surface mount soldering more approachable by Joe Random Hobbyist. Now, one of the engineering problems I'd need to solve would be how FEW PCB layers I could get away with using that FPGA. 16+ layers is not uncommon for full die escape, but layer counts like that are far too expensive at the volumes this thing might sell in. Cell phones have high layer counts, but only because they sell in huge volumes. I favor Xilinx FPGAs since they're what I use in my day job, and those are all BGAs. I would not offer this device as a kit in any case; I've helped Ian out doing rework for US FreHD customers, and even good old through-hole soldering can be hard for folks without a lot of experience. I'd rather design for automated assembly than spend a lot of time on tech support for kit builders who had trouble. With a BGA on it, the board's going through a reflow oven anyway. No, I think it's FM. I recall listening to TV audio on my US FM tactical mil radios whose frequency coverage extended over the bottom 2-3 VHF TV channels, back before they turned off the analog broadcasts. I agree that stereo support isn't needed, as stereo TV post-dates the computers in question if I'm not mistaken. But if I do the final demod digitally in an FPGA, then adding AM support wouldn't be a big deal if needed. That sounds like a deep rabbit hole to fall down! It might result in a case of "if you want me to add support for the computers from your country, send me one so I can develop with it". But this is also a point in favor of using an FPGA: Fairly major architectural changes just look like firmware upgrades to the end user, who can remain blissfully ignorant of my development pain. :) If I use one of the Zynq chips (which I'm currently favoring), then upgrades can probably be as simple as swapping an SD card for the end user. Don't want to force people to buy a programming cable and download a 15 gig development platform! That evil thought is definitely worth exploring! The thing I'd worry about is the strong local AM broadcast station getting in through the computer's lousy shielding and ruining the plan. I have a local AM station that confuses my MFJ antenna analyzer if I don't use a broadcast band blocking filter, or wait until they turn down the output power at 8PM. Dealing with them is part of my day job, so maybe I'm a good person to tackle this one? :) I think that the Apple ][ application needs to handle both mono 80 column and color graphics as a user-selectable and easily-toggled option, since those machines were very commonly used with both mono and color monitors. Switching configuration files should be an available option, but I think that color vs. mono selection is important enough to get a mode switch in the standard configuration. I think that configuration file changing would be better for oddball cases that are so unlike standard NTSC or PAL that supporting them in the same design would make the architecture messy. -- Mark J. Blair, NF6X <nf6x at nf6x.net> http://www.nf6x.net/

On 23 May 2015 at 16:35, Chris Osborn <fozztexx at fozztexx.com> wrote: Is it me or was there a trick missed in the ADC section? If you have 8 comparisons to spend I wonder if a modified binary chop would have been more efficient? A standard binary chop would take three comparisons, but I would have thought if you repeated each stage twice (and a third time for up to two of them if they differed), that would give you a more robust output...

On Jun 1, 2015, at 5:24 PM, Mark J. Blair <nf6x at nf6x.net> wrote: The color on the hi-res screens looks pretty good, but the vertical lines through the blocks on the lo-res screens isn?t quite right. The bottom 4 lines of text having color bleeding is normal, even on an Apple color composite monitor. The monochrome 80 column screen looks pretty good too. Do you happen to have an old CRT TV around with composite input that you can hook up and compare to, just for yourself? I?ve got an Amdek Color I and Apple IIc Color Composite here that I?ll try to take some sample pictures of. -- Follow me on twitter: @FozzTexx Check out my blog: http://insentricity.com

Well, you're right! The text looks like crap on an analog CRT, too. I updated my blog post with a few more pictures. Now I'm even more curious about the reports I've heard about having trouble with video conversion, since the first cheap converter I tried seemed to work OK with an Apple //c. Of course, it still lacks a tuner for the TV-connected computers, but I got the impression that the Apple II series was especially picky about its video converters. Maybe there's a different program I should try running that does weird stuff with video modes, like some game with particularly interesting graphics? I'm also interested in seeing what the converter will do with composite video tapped out from a Color Computer's innards. I seem to recall that it had some screwy video modes that required the user to keep hitting the reset button until the colors weren't swapped. -- Mark J. Blair, NF6X <nf6x at nf6x.net> http://www.nf6x.net/

When you get it, we can compare pictures of their innards. Mine has a PCB with blue soldermask. Most of the functionality appears to be in a single QFP, probably with a central ground/heat slug based on the vias on the bottom side of the board. The top of the chip appears to have been sanded to remove the markings. -- Mark J. Blair, NF6X <nf6x at nf6x.net> http://www.nf6x.net/

On 2015-06-01 20:24, Mark J. Blair wrote: As far as the crazy cat lady project, as I was trying to explain earlier.. my $.02 which maybe redundant to your own thoughts: VIDEO ADC/DECODER You can certainly use an off-the-shelf part as there are 47 from 5 companies on Digikey alone in solder friendly packages. However if the classic machines you are targetting are producing a clean conforming NTSC_M/J, PAL_B/G/N, or SECAM signals, stock video decoders on most converter boxes would work perfectly already. I doubt you will have much luck with the NDA process explaining a hobby cause but worth a try. Even if you are successful, make sure the parts you are looking at are readily available in 10s and 20s quantity through a distributor chain at a reasonable cost. That may be a second hurdle if it's fairly exotic. What SiLabs part are you looking at? I can't find any recent video decoder offerings. GENERIC ADC/DSP ANALYSIS This is the harder path to go down. You will need a symbol rate of at least 54 Mhz from a single or two interleaved ADCs to have a shot at at least 50% phase accuracy (NTSC = 13.5 MHz). There are some relatively inexpensive options, but I suggest getting a lot of input from folks and other reference designs on the best way to build an analog front-end for composite. And only do composite/S-Video. There are already commodity solutions for stepping down broadcast RF frequencies to base-band. After that, you will need something fairly hefty at the start to find the characteristics of the signal and align the sampling. Then you just need to track clock drift and adjust a VCXO. Straight-forward in hardware, but I doubt many people will have the experience to add new software support for <insert eclectic hardware platform with slightly non-standard composite output here> and fix bugs when they occur. So set expectations on your time early with respect to project support. SOC/ZYNQ/MARKET The fact you mention ZynQ throws up a few warning flags for me. It's probably overkill, it's expensive in low quantities through normal distributors, and it's not very hobby friendly. $65 for a '010 alone + DDR3 + BGA assembly would drive the resulting cost for you to make these boards well beyond $200. And your market is small - vintage computing folks w/o a CRT on a limited budget. Considering that, you probably should not try to build the processing core yourself. Run the ADC output or the video decoder Bt.656 output into a stock board like a Parallella, BeagleBone Black, or other EVM that has the processing power, HDMI output, and input interfaces you need; especially to start with. Then go full custom later. -Alan

I just noticed that the Si2137 and Si2177 tuners are now marked as "factory special order" and Mouser, where they were previously just marked "New at Mouser". So my guess that those parts are a leftover tray with questionable future availability might be unpleasantly accurate. I'm still interested in pursuing a cheap tuner design to see if it goes anywhere. Even if it turns out that some existing composite to HDMI converter satisfies everybody, I think there would be interest in a small, inexpensive standalone RF demod to get composite video out of vintage home machines without internal mods, and Crazy Cat Lady could morph into that. Maybe it would involve custom hardware, maybe it would be a TV tuner dongle operated in SDR mode plugged into a Beaglebone with custom software, maybe it would be a general purpose SDR design that happens to be well-suited for video demod... figuring that out is the fun part for me. -- Mark J. Blair, NF6X <nf6x at nf6x.net> http://www.nf6x.net/

On Sat, 23 May 2015, Mark J. Blair wrote: Where are you located? Would you like some of the REAL monitors? They will do all sorts of bizarre artifacts that would be hard to emulate. They are also perfect matches to the size weight, power draw, smell, reliability (or lack thereof), resolution (or lack thereof), etc. Nothing truly duplicates the experience without drifting tuning, picky horizontal hold, a lttle flickering, etc. I've long ago gotten rid of most of them, but there are always more breeding in the cracks of the woodwork.

Am 23.05.15 um 17:24 schrieb Mark J. Blair: What about the analogy of Software Defined Radio: Use only as much electronics as minimal necessary to get the input signal digitized by a high speed ADC. Maybe some cheap DVB-T USB thing can be abused for the ADC part. Do the processing in software on a "normal" computer. (PeeCee, some ARM single board thing.) Advantage: - Converts to whatever video output your PeeCee has. - Works on a laptop. No extra display needed when on the road. - Easy to capture the video to generate AVI, MPEG, ... - Scaling can be done by GPU of host computer. No need to reinvent the wheel. (If the existing GPU wheel fits the wagon...) - Scaling and other processing can be done in easy to change software. That way it is easy to extent for other people. (Different scaling and interpolating algorithms. Adaption to the odd signal from some obscure hardware of other people.) - Most likely cheaper then a stand alone solution. - No obscure FPGA magic needed. Disadvantage: - needs extra PeeCee / ARM-SBC for processing. - No obscure FPGA magic needed. -- tsch??, Jochen

Am 23.05.15 um 18:11 schrieb emanuel stiebler: Depending on your personal preference FPGAs can be an annoying fight in an obscure HDL with proprietary tools running on Window$, or it can be lots of fun and new things to learn. :-) -- tsch??, Jochen

Yup. I'm a radio amateur and an electrical engineer working with GPS stuff, so I understand the pain! But nowadays, a VHF/UHF tuner is a single IC, possibly surrounded by a small handful of fixed inductors. The problems are: * Might not be able to buy it in small quantity. * Might need to read Chinese to understand the datasheet. * Datasheet might simply be unavailable to individuals, even if the part is available. I haven't begun trying to crack that egg yet, but there may a successful path there. With modern parts, even a dumb digital designer like me can successfully design RF front ends operating at 1.5 GHz, and cram an entire GPS receiver onto a fingernail-sized PCB. It's black magic if somebody makes the right chip. And it's still do-able if they don't make the magic chip... it's just a lot more expensive than consumer electronics have conditioned us to expect. I think there used to be a rule of thumb for microwave work to not try for over 10dB of gain per inch of circuit, or something like that. But now, slapping down a 20dB or greater gain LNA circuit the size of a barley corn is no big deal. Being able to make the whole circuit a lot smaller than a wavelength eliminates a lot of the pain. As an example, Mouser carries a single-chip tuner for $2.32 at quantity 1. It's by Silicon Labs (I think I mistakenly states ST earlier). But only a short-form datasheet is available openly, and I need to contact SiLabs to see if I can get a full datasheet. I also don't know whether this analog TV tuner chip will remain in production, vs. Mouser buying some TV manufacturer's leftover parts. The output of a single-chip tuner might also be at IF. The Maxim part (which I will not use) outputs at 36 MHz, I think. Can't tell the output of the SiLabs part without more info. Hopefully it's either baseband or a lower IF frequency that I could sample with a cheaper ADCs for digital down-conversion. Needing to support a 36 MHz IF would probably increase ADC cost vs. using ones that just need to sample baseband or a low IF. SAW filters are also black magic, and nowadays they are TINY! And US NTSC puts the sound carrier at 4.5 MHz, so there's another thing in favor of using SDR techniques for some portion of the demodulation if I can't find a Magic Chip that does the work more cheaply. The chroma subcarriers are also at different frequencies in the various standards. And that's where the fun begins! The plan is to infer what color the vintage computer was trying to display at any given pixel, with knowledge of the dirty tricks it used to get that color cheaply. Then cram that inferred pixel into the frame buffer, and convert the video format on the other side of the frame buffer. -- Mark J. Blair, NF6X <nf6x at nf6x.net> http://www.nf6x.net/

On 2015-05-23 11:24, Mark J. Blair wrote: OMG, stop. Google "Composite to HDMI" and you will find hundreds of low cost boxes that do this. Hopefully one of them will work for you. I was going to do your idea for C64s long ago until I realized there's no way I can do it nearly as cheap as Mr. Wong in his apartment in Shenzhen, Guangdong. However if you really want to pursue your idea "because it is fun for you", I suggest looking at the Lattice XO2 Control Development Kit (novel idea of OpenLDI in/out + LPDDR2), the Digilent Atlys board (Spartan 6 + dual direct TMDS in&out), or NeTV (Marvell SoC + Spartan 6 w/ dual direct TMDS) as starting points. You don't even need a full frame buffer if you need to scale. If the frame rates match, you can just store one line if you need to horizontal stretch and/or line-double, or three lines if you want to perform multi-tap vertical scaling. All HDMI TVs (at least in North America) support 480i however you generally have to pixel double per the CEA-861 standard to have enough clocks in the HBI/VBI to pass audio. -Alan

And then keep on purchasing and evaluating these converters so that when they are updated to use the latest cheapest IC, you know which ones still work with which machines. -tony

If you do end up building a custom solution, I have a feature request :-) It would nice if the device was also a frame grabber that could, under command, snap one or more frames of the legacy video and export it over USB perhaps. This would allow us to document operation of legacy software with high quality frame grabs since persumably you'd have access to the image in a relatively good quality domain before you turned it into DVI/HDMI or whatever. Chris -- Chris Elmquist N?JCF