464+ Mod

Started by Bryce, 15:34, 08 June 12

Previous topic - Next topic

0 Members and 2 Guests are viewing this topic.

ralferoo

Quote from: Badstarr on 00:23, 16 June 12
I think Bryce is quite right that the AY is producing the signal we hear on a classic CPC while loading.
However, that isn't what's happening. For example, try this (it's writing to tape) clearly with no AY involvement:

10 OUT &F600,&10
20 OUT &F600,&30
30 GOTO 10


The relevant part of the schematic is on the 6128 page entitled "disc control interface", bottom right in the "cassette control circuit" (maybe page 36).

For tape output, pin D is connected to tape pin 5 (MIC), which also goes via R310 and R308 to the sound amplifier circuit above (R304 goes from that point to the op-amp input).

For input, pin E is connected to the opamp that goes to the PIO, but before that, tape pin 4 (EAR) is connected via R311 and R308 to the same circuit above.

On page 20, you can see how pins E and D are connected directly to the PIO, pins 25 (E tape input) and 12 (D tape output) respectively.

Quote
The reason why music played through the system seems quieter and very degraded is most likely due to the CPC only being interested in a Unipolar signal, the CPC only needs to make a distinction between Off and On or Loud and Quiet.
Effectively, yes - it's because it's been squished through the opamp to make it 0 or 1 rather than analogue, but not because the CPC has sampled and replayed it.
Quote
Also the baud rate would be adequate to produce the horrible version of whatever 80's classic tape you happen to try to play through it.  :laugh:
This is unaffected as the op-amp is "practically" instantaneous.

Bryce

Quote from: ralferoo on 15:52, 16 June 12
For tape output, pin D is connected to tape pin 5 (MIC), which also goes via R310 and R308 to the sound amplifier circuit above (R304 goes from that point to the op-amp input).

For input, pin E is connected to the opamp that goes to the PIO, but before that, tape pin 4 (EAR) is connected via R311 and R308 to the same circuit above.

The connection via R308 is the 5V power supply to the Op-Amps, not a signal connection, if the sound passed through here, then the tape deck wouldn't work. C316 (between R308 and R310) is there to guarantee that nothing analogue makes it from R310/R311 to R308.

Bryce.

Badstarr

Hmm... So the loading sound is essentially an accident? If it is then it's a good one to make, I for one like the reassuring whistles and whines while a tape loads as I'm sure many of us here do. Congratulations on solving the mystery Bryce! Having poked around on the 6128 mainboard and not being able to find a definite link to the CPC speaker I felt sure that your theory about the AY was the most logical/likely explanation. Mind you, I see no real reason why the AY theory couldn't also work in practice, you never know maybe a similar system was contemplated at some point in the CPC design. Perhaps its a little too complicated and over engineered in reality. However it may be the sound engineer in me that likes this solution as its sort of sampling, but not, using the signal produced by the cassette to produce a a clean synthesised counterpart using another device. It kind of makes me imagine it as a way of streaming samples, I dunno, my brain comes up with these sorts of ideas occasionally and sometimes they actually work  :laugh:     
Proud owner of 464 GTM64 6128 GTM65, GX4128 and a 464/6128 Plus Hybrid a 20 year long ambition realised! :-)

ralferoo

Quote from: Bryce on 17:59, 16 June 12
The connection via R308 is the 5V power supply to the Op-Amps, not a signal connection
Ooops, good spot!

Bryce

Quote from: Badstarr on 20:28, 16 June 12
Hmm... So the loading sound is essentially an accident? If it is then it's a good one to make, I for one like the reassuring whistles and whines while a tape loads as I'm sure many of us here do. Congratulations on solving the mystery Bryce! Having poked around on the 6128 mainboard and not being able to find a definite link to the CPC speaker I felt sure that your theory about the AY was the most logical/likely explanation. Mind you, I see no real reason why the AY theory couldn't also work in practice, you never know maybe a similar system was contemplated at some point in the CPC design. Perhaps its a little too complicated and over engineered in reality. However it may be the sound engineer in me that likes this solution as its sort of sampling, but not, using the signal produced by the cassette to produce a a clean synthesised counterpart using another device. It kind of makes me imagine it as a way of streaming samples, I dunno, my brain comes up with these sorts of ideas occasionally and sometimes they actually work  :laugh:   

I would guess that the engineers noticed the mistake during the prototype test phase, but then decided it was actually quite cool, so they chose not to solve it. Just a pity they didn't use a similar solution on the plus. For some reason (most likely cost) they decided that the plus wouldn't have an internal speaker. To be honest, it probably did save them a fortune.

Bryce.

MacDeath

Anyway, nice to see this aspect of the CPC and this particularity of having an internal speaker has just been studied.

and yes, it was a kool "feature".

arnoldemu

Quote from: Bryce on 15:33, 16 June 12
The effect happens because the classic CPC has shared an Op-Amp chip for the tape amplifier and the sound amplification.
So I was correct? ;)
leakage/cross talk.... ;)
My games. My Games
My website with coding examples: Unofficial Amstrad WWW Resource

arnoldemu

There was a story that the original CPC had a design fault whereby the square wave produced when writing to tape was not correct. The firmware was coded to cope with it.

The problem was spotted by an engineer in the east and corrected. When the machine got back they had to modify the software to make it work.  :laugh:

My games. My Games
My website with coding examples: Unofficial Amstrad WWW Resource

arnoldemu

Quote from: Badstarr on 20:28, 16 June 12
Hmm... So the loading sound is essentially an accident? If it is then it's a good one to make, I for one like the reassuring whistles and whines while a tape loads as I'm sure many of us here do. Congratulations on solving the mystery Bryce! Having poked around on the 6128 mainboard and not being able to find a definite link to the CPC speaker I felt sure that your theory about the AY was the most logical/likely explanation. Mind you, I see no real reason why the AY theory couldn't also work in practice, you never know maybe a similar system was contemplated at some point in the CPC design. Perhaps its a little too complicated and over engineered in reality. However it may be the sound engineer in me that likes this solution as its sort of sampling, but not, using the signal produced by the cassette to produce a a clean synthesised counterpart using another device. It kind of makes me imagine it as a way of streaming samples, I dunno, my brain comes up with these sorts of ideas occasionally and sometimes they actually work  :laugh:   
I believe the Enterprise computer uses the sound chip to write cassette data.
Nice way to re-use some electronics.
My games. My Games
My website with coding examples: Unofficial Amstrad WWW Resource

arnoldemu

Quote from: Bryce on 00:38, 17 June 12
I would guess that the engineers noticed the mistake during the prototype test phase, but then decided it was actually quite cool, so they chose not to solve it. Just a pity they didn't use a similar solution on the plus. For some reason (most likely cost) they decided that the plus wouldn't have an internal speaker. To be honest, it probably did save them a fortune.

Bryce.
I was thinking they had cut it out because of the lack of internal speaker. Imagine tape loading noise being amplified through the monitor/television. Ouch!
My games. My Games
My website with coding examples: Unofficial Amstrad WWW Resource

Bryce

Quote from: arnoldemu on 11:16, 18 June 12
So I was correct? ;)
leakage/cross talk.... ;)

Yes you were correct. The reason I questioned it being a leak is because it is so consistant. Leaks normally vary their pitch and volume and can be easily effected by other sources too. But because the leak seems to be inside an IC, everything is pretty consistant.

Bryce.

Gryzor

This is truly an interesting discussion. Imagine that, a mystery just solved in 2012? Wow.


Now waiting for the simpler explanation. I'm barely following Bryce's story, but as soon as I'm done I'm left with only the general idea. Maybe it's the lack of sleep from last night's elections...


Anyhow, didn't other systems have loading sounds?

mr_lou

Quote from: Gryzor on 17:19, 18 June 12Anyhow, didn't other systems have loading sounds?

I remember finding it a bit frustrating there weren't any loading sounds on my mates C64 back then. Asked him how he knew where a game began on the tape. He replied: "You can see it on the colors on the screen".

Badstarr

It's been a while since I have used one but I seem to remember the Spectrum had loading sounds, I would imagine that, based on the Spectrum (ehem) build quality, that this would have been a simple pass through from the cassette to the modulator.


I maybe being a little unfair to the Spectrums build quality but in my experience they aren't the most robust of machines even the Amstrad versions were a poor comparison to the CPC. I briefly had a Spectrum+2 (sort of anyway, it was given to me and my sisters by my Dad) I remember the horror as I pressed the play key one day and it snapped off. Of course I was told I was too rough with it! Strange then that my 6128 is still fighting fit after 20 years! Another one of those trips down memory lane that these topics tend to inspire.  :laugh:
Proud owner of 464 GTM64 6128 GTM65, GX4128 and a 464/6128 Plus Hybrid a 20 year long ambition realised! :-)

ralferoo

Quote from: Badstarr on 10:40, 19 June 12
It's been a while since I have used one but I seem to remember the Spectrum had loading sounds, I would imagine that, based on the Spectrum (ehem) build quality, that this would have been a simple pass through from the cassette to the modulator.
The Spectrum solution was a stroke of genius. It's actually a single pin on the ULA used for BEEP, EAR and MIC and connected to all 3 differently. It produces different voltages according to which logical bit is fiddled with by the CPU, but pretty much they can be interchangable as they're not massively different.

ralferoo

Incidentally, if you're interested in such things, this book on the reverse engineering of the Spectrum ULA is a fantastic read: The ZX Spectrum ULA: How to design a microcomputer

beaker

Quote from: mr_lou on 17:43, 18 June 12
I remember finding it a bit frustrating there weren't any loading sounds on my mates C64 back then. Asked him how he knew where a game began on the tape. He replied: "You can see it on the colors on the screen".

Tell me about it... I finally got a tape game, The Sentinel, on Monday for the C64 I bought a few weeks ago and never having used a C64, I wasn't sure if the thing was working or not until the tape ran out and I was still looking at a blue screen. Those old Amstrad tape loading sounds were reassuring...

Gryzor

Even by mistake, Amstrad did the right thing it would seem :D

Bryce

#68
This morning the last parts arrived for stage one of the 464+ Mod. The 464+ has finally grown up and become a 6128+.

The µPD765 Floppy Controller, SED9420 Data Separator, Extra RAM, Floppy Logic ICs, Floppy header / power / filter, External Floppy Connector and lots of other passive components have been added. All new ICs are mounted in sockets to make any future repairs easier. I've just tested the additional RAM and even loaded its very first program from disc. All is working fine. Now it's time to look at what additional stuff can fit inside the casing...

Bryce.

P.s. If anyone else is considering doing the same upgrade, I have a small stock of the hard to find µPD765 / SED9420 etc.

mr_lou

Nice!  :)

emilen72

good work Bryce.... :D
I'm interest in the same mod (ram+floppy controller)...
I send you a pm for info...

steve

Details are in amstrad action issue 90 (ram) and 91 (disk controller).

endangermice

Bryce this is shaping up to be a fantastic 464+ mod and really shows your engineering skills in action! I'm looking forward to watching your progress and more of the insights into the internal workings of the machine that you will no doubt discover along the way! Keep up the great work!
For all the latest Starquake remake news check out my website - www.endangermice.co.uk

emilen72

#73
Quote from: steve on 13:05, 24 June 12
Details are in amstrad action issue 90 (ram) and 91 (disk controller).

Where find they? Megaupload link in cpc wiki section is out... of course...


edit: I have find all issue, but the scan quality is not very good

http://archive.org/search.php?query=collection%3Aamstrad-action-magazine&sort=-publicdate


if you have good quality please snd me a pm

Gryzor

Quote from: emilen72 on 14:10, 24 June 12
Where find they? Megaupload link in cpc wiki section is out... of course...


edit: I have find all issue, but the scan quality is not very good

http://archive.org/search.php?query=collection%3Aamstrad-action-magazine&sort=-publicdate


if you have good quality please snd me a pm


Obviously you've missed the Downloads section above ;)

Powered by SMFPacks Menu Editor Mod