CPC on a VGA Monitor

[Bryce]

Hi All,
     during my endless quest to get my CPC to produce a perfect picture on a PC Monitor, I came across a relatively cheap solution that works better than anything else I've seen. If you've read my article "Converting an MP1/2 to composite output" (http://www.cpcwiki.eu/index.php/LCD_And_Plasma_TV_Solution) you'll have seen that you don't need much electronics knowledge to do this. Now I've found a relatively cheap converter called "PremiumBlue V2V" which I picked up for 29,95€ and it converts the composite output to a very impressive and sharp VGA signal. Because the signal is composite->VGA there is no signal drift or other problems associated with a modulated output. The converter is approximately 130mm/80mm/20mm (ie: small) and is available at Pollin Electronics (http://www.pollin.de/shop/dt/OTA4ODcyOTk-/Computer_und_Zubehoer/Hardware/Monitore/Video_zu_VGA_Konverter_Box_PremiumBlue_V2V.html) for anyone who's interested in trying this solution for themselves.

I'd love to tell you more about it, but Ikari Warriors has just finished loading, so I HAVE to go.....

Bryce.

[Gryzor]

Nice find!

There's a small catch from what I see: it says that the resolutions it supports are:
max. 1280x1024 (640x480, 800x600, 1024x768, 1440x900)

This means that if you have a wide monitor, only the 1440x900 will work correctly I guess, BUT because this is not a native resolution for those monitors there'll be some blur...

But I guess it'll look great at 1280x1024 on a 4:3 monitor...

Ikari Warriors, ehhhh? Been playing this on my arcade cabinet lately

[Bryce]

There's an newer version of this unit which can do Widescreen formats, but it cost 79€. I have a 19inch 4:3 1280x1024 which was just waiting to get connected to my CPC. The picture is Scart quality and crystal clear. The main difficulty with getting a CPC connected to a VGA monitor is the refresh rate, which means that the hardware would have to buffer the entire CPC screen to refresh it to the VGA. Possible, but not easy or cheap to build. It would have cost me more than 29€ to build it myself. The V2V has on-screen menus and I can connect other devices (S-Video) to the monitor too. It was definitely a good investment.

Bryce.

[Gryzor]

Nice info... a shame I have a juicey widescreen monitor, or I'd probably buy it!

[OCT]

Reading this thread and http://www.cpcwiki.eu/index.php/LCD_And_Plasma_TV_Solution I realized that of my own devising I had already modded an MP1 circuitry in a similar way decades ago, before there was an Internet let alone digital cameras (each available to mere mortals, that is) to spread the word.

The following line in http://www.dse-faq.elektronik-kompendium.de/dse-faq.txt attests to the fact that the MC1377 remained every bit as dreaded and prone to losing colour as in the 1980s (cf. MP and Amiga 520 modulators) - IIRC it has been given a less finicky successor around the turn of the millennium (but I can't seem to recall whether simply MC1378 was "The Latest And Greatest"):

RGB->Composite PAL-Encoder: MC1377 (schlecht, Trafo, OnSemi), TDA8501+TDA4568 (Philips)

The V2V box, also a favourite with the builders of "DIY beamers" (i.e. TFTs on OHPs etc.), is nice (and even contains an IR remote control receiver according to http://michael.rstudios.de/hardware/bericht-geniatech-v2v-pro-video-converter) in spite of a tendency to hum bars (as is common to analog video technology).

Then again, the "Holy Grail" would be to find one of the current crop of SCART-to-HDMI converters that accepts the CPC's RGB output (or even design one's own custom chip) and hence skip entirely the degrading detour via composite video - making things come full circle (with 1985 catching up) in time for the launch of the much Duke-Nukemed Eee Keyboard bringing back the CPC form factor to the living rooms of this world.

[Bryce]

Hi OCT,
      I'm not suprised that someone else came up with the MP Composite hack before, but I have never seen it documented anywhere, so I thought I'd add it to the wiki. The MC1378 was a 40 pin monster of a chip which included the hardware of the MC1377 plus overlay and other features, ie: it was probably most useful for allowing the integration of cheap OSDs (On-Screen Displays) such as those terrible first attempts (The horrible green volume bar and channel display seen in the last of the CRT TVs), but I'm not sure if the 1377 content was actually improved in this chip or was the same rubbish in a new package. The V2V box is a quick and easy way of getting the low frequency composite signal buffered up to the scan rates required by a modern VGA monitor, yes it shows some ripples as all analogue to digital convertors of this calibre seem to, but it's price convinced me that this solution is cheaper than anything I could build myself.

The quest doesn't stop there though. The V2V also has a VGA input, which if buffered (haven't open the case yet) would be a perfect CPCRGB (ie: scart) to VGA converter, using an LM1881 or similar, the V and H Sync can be cheaply seperated to give this input all the signals it needs and a perfect SCART signal directly to the VGA monitor.

I'll let you know how it goes....

Bryce.

[OCT]

not suprised that someone else came up with the MP Composite hack before, but I have never seen it documented anywhere, so I thought I'd add it to the wiki.

Had no intention to make any claim of "me first", this mod and the credit for documenting it are of course entirely yours.

In what I dug up as attached from a distant past (in an even more ancient PC housing), apparently a resistor had been unsoldered at the end carrying the composite video.
BTW, next to it sits a Panasonic JU-257A604P much like at http://honi.hucki.de/panasonic.html which to document I also had no way back in the day, but thankfully for this drive and many more Honi did.

At any rate it's a staggering realization (for someone rejoining these discussions) what has been and most importantly, is being accomplished in forums like this, more skillful and vivid even than comp.sys.amstrad.8bit in its heyday, and certainly testament to how this medium has every bit as much of an impact on this scene (in terms of speed-up and, erm, "matchmaking" ) as it had on correspondence chess.
Seeing code and hardware designs improved by orders of magnitude within just a day or so, as e.g. at http://www.cpc-forum.de/viewtopic.php?t=209 is such a pleasant sight, and on the other hand makes it all the more impressive how much we could get done decades ago in spite of having to resort to printouts, AAs/A100%s and diskmags at best exchanged by international postal mail (even more so in languages not just foreign but fairly new or even "alien" in almost every sense of the word to many to us)...

The MC1378 was a 40 pin monster of a chip which included the hardware of the MC1377 plus overlay and other features, ie: it was probably most useful for allowing the integration of cheap OSDs (On-Screen Displays) such as those terrible first attempts (The horrible green volume bar and channel display seen in the last of the CRT TVs), but I'm not sure if the 1377 content was actually improved in this chip or was the same rubbish in a new package.

Cheap is not an unwelcome term in this context. In my recollection that "monster" had all the oscillators included that would avoid the calibration nightmares (let alone drift) of its predecessor.

The quest doesn't stop there though. The V2V also has a VGA input, which if buffered (haven't open the case yet) would be a perfect CPCRGB (ie: scart) to VGA converter, using an LM1881 or similar, the V and H Sync can be cheaply seperated to give this input all the signals it needs and a perfect SCART signal directly to the VGA monitor.

Which, incidentally, sounds a lot like adapting Amiga Scan-Doublers to the CPC. Amiga though is not a hardware I ever had a chance to get sufficiently acquainted with.

The V2V's VGA-in, however, might just be a simple switched loop-through - looking forward to what you'll find out.

[Bryce]

Hi OCT,
         just puzzling over the attached picture in your mail. Is that an original MP1/2 MC1377 board with some modifications? Why are the R/C pairs added to the RGB inputs? What have you tuned with the variable resistor at the output and the variable inductor hanging off the end (borrowed from old TV/FM radio perhaps)? And what's the array of massively oversized Electrolytic capacitors for? Just asking

As I mentioned on the wiki page, one of the reasons the MP1/2 performed so badly (drift / colour seperation / etc) was the fact that Amstrad decided to build the cheap alternative. The original design notes from Motorola recommend using a bandpass transformer and the associated delay-line, ie: Coupling the Chromo signal properly. Although the circuit works with the cheaper solution, the quality and stability suffer, with these parts the MC1377 performs much better than it's reputation would suggest. Unfortunately both the Filter and delay line were relatively expensive parts which would have made the unit much more expensive, so I assume that this decision was made for commercial reasons. Also other 8-Bit machines using the MC1377 went the same low-cost route, so Amstrad probably thought "Why bother, when people are happy with the cheap solution". These parts are also no longer available, I've stripped down some old CRT TV's in search of suitable parts, but haven't found anything useful up to now. Even if I did, it would still only be enough for one unit and hardly information that would help others. I added the Chromo coupling information to the page anyway, just in case someone out there has a box of old TV parts in the basement.

Bryce.

[OCT]

puzzling over the attached picture in your mail. Is that an original MP1/2 MC1377 board with some modifications? Why are the R/C pairs added to the RGB inputs? What have you tuned with the variable resistor at the output and the variable inductor hanging off the end (borrowed from old TV/FM radio perhaps)?

The details are buried in time (and http://www.homecon.net/index.php/blog/175-tod-eines-desktop-c64 -to name but one, while denying to have ever made a similar machine d'enfer- cautions against taking too much apart ), but in my recollection that's standard MP1 circuitry with the composite video redirected to an RCA terminal (unshielded as just inches away, with a divider pot and diode to ensure the syncs pass unattenuated; cobwebs optional - as proof of antiquity) rather than to the modulator. AMS being the entrepreneur he is (of televised legend), I would not be surprised at all if MP2 had to make do with an even smaller number of components.

And what's the array of massively oversized Electrolytic capacitors for?

Unrelated open-frame PSU at a safer distance than the image suggests (depth-of-field due to close-up mode), cf. lower right of the original image.

[Greg.0]

Hi Bryce, Hi all,

I also looked a way to display my CPC on new screen generation. There is now a new cheap (about 35$) rgb converter able to accept CPC sync timing, the CTV-910. Maybe your can be interested by the test I made.

> Review on my never-finished Joomla website <

I have some trouble with demos that can not be displayed... Software and game are very good compatible.
How about your way to use composite signal and demos ? Have you some deformations ?

[Bryce]

Hi Greg.0,
        had a look at your (unfinished) Joomla page. Your results seem to be around the same as mine, maybe slightly better (difficult to tell with photos). Which demos have you had problems with? Let me know and I'll try them out with my setup.

Bryce.

[PulkoMandy]

You should try Climax, it runs for about 30 seconds without generating a single VBL.
Any demo using vertical screen splitting is also likely to do strange things.

[Greg.0]

Have a look here, This is the
Orion Prime intro from Arkos
I'm working at this moment to put the "from Scratch" of Vanity, working, except for a part.
I will have a look at the Climax too.

Back asap... 

[Greg.0]

I must ask authorisation before publishing the video.

Could you try the Dreamend demo from Gozeur ?
Have you similar deformation on Orion Prime intro ?

Thanks.

[PulkoMandy]

This is to be expected... To do the screen distorsion in this part they are using wrong HBL timings (shorter and longer lines taht are not really at 15KHz). The CTM can handle that, but most other screens and adapters will have trouble. Moreover, the CTM use a PLL to sync, which means the screen will almost never jump like that, but go progressively to the left or right, which makes the nice distorsions in orion prime.

[Greg.0]

Thanks for explanations PulkoMandy.

I added a low-res video of the from scratch demo of Vanity on my website... Still not have authorisation to publish it on youtube, and maybe it will not be necessary, you can easy see when the converter return to focus with this one.

Tomorrow I will try climax... It's a also a good way to practice camcorder ! And i need it

[Bryce]

If that's how they are achieving the effects, then I don't need to test it on the V2V, it will definitely have the same error, because of the way the frequency is being sampled by the converter.

Can your camcorder not be connected directly to an output on the converter? It would produce a much clearer view of the distortion.

Bryce.

[TheRogue]

I have a converter that should be able to convert CPC RGB to VGA flawlessly. There are actually 2 different ones, both with strengths and weaknesses. One is the Ambery Converter ( http://www.ambery.com/rgbcgatovgac.html ) and the other is the GBS-8220 CGA/EGA/YUV to VGA HD-Converter ( http://www.jammaboards.com/store/cga/ega/yuv-to-vga-converter-pcb-gbs-8220/prod_291.html?ccSID285e73f64c04744fc5f33ff68f529742=90b328dca2722ac920aeeafa602b3927 ) both should work well, but especially the second one as it can work flawlessly with frequencies from 14 to 16 KHz instead of just 15KHz. I might also add that my CPC 6128+ works flawlessly on my Commodore 1084S monitor.

[Greg.0]

Hi TheRogue,

I tried the CGA/RGB/... to 2VGA converter, and never manage to get it work with the CPC... I have 9 different converters board at this time. But never tested the Amberry one (too much expansive imho)

[TheRogue]

I have the ambery board for use with my X68000 XVI, and it works great. I'm crossing my fingers and hoping for the same result with the CPC!

[Apollo]

I have the ambery board for use with my X68000 XVI, and it works great. I'm crossing my fingers and hoping for the same result with the CPC!

Keep us posted as there are a lot of people interested in a *good working* solution! Especially demos and tricks like in Orion Prime would be really cool if there are converted in a correct way to be displayed on a LCD screen!

[mahlemiut]

I have the ambery board for use with my X68000 XVI, and it works great. I'm crossing my fingers and hoping for the same result with the CPC!

Hmmmm.... X68000...  you lucky soul, you.

[TheRogue]

Hmmmm.... X68000...  you lucky soul, you.

I take it from your sig that you have an X68k too? What model? How do you like it? If you need any help with it or anything let me know, I've found out quite a bit of things about it that are virtually unknown.  By the way, I also have an MSX Turbo-R ST. Since you like the X68k I figure you might also be into another great Japanese machine:P

[mahlemiut]

I don't have one (I wish I did), but I do like the system lots.  I've written a driver for it for MESS, so I'm fairly familiar with it.

[bleeper]

I had tried to turn my MP-1 into composite a long time ago but never really finished it properly.

The main problems I had was that the image was really wavy. On certain solid colours, you could see diagonal lines moving up the screen. It made most games look psychadelic!

Sometimes, pressing play / FF on the tape player would 'freeze' these lines, but resetting caused them to start again.

I've powered it up again and sure enough, the same issue.

I can only assume it was a grounding issue, but is there a way to actually diagnose this properly?

Living in the UK, it's probably better if I just modify it to RGB SCART