Amiga questions you've always been too embarrassed to ask

[Nibbler]

Well, then. Everything is fine. "use" is "save"
(i did this way back in 1992 when i got my first Harddisk, because more Ram & shorter boot)

thanks

[Jope]

Also any temporary env variables will become permanent if you store them in the env-archive. You might not experience any problems, but I recommend happyenv in any case.

[Nibbler]

I know, thats no ploblem for me Thank you Jope

[Glen M]

Originally Posted by TroyWilkins

I could be wrong here, but my understanding is that in the 68000 based models, as well as the A1200 and CD32, the system clock is set by one crystal, and then all the other clocks are exact divisions of those clocks.

My understanding is that on PAL machines, the crystal is 28.37516MHz and NTSC it will be 28.63636 MHz. On a 68000 machine, this is divided by 4 to get the CPU clock speed (7.09MHz for PAL, 7.15MHz for NTSC), and on the 1200 and CD32, it's divided by 2 to get the CPU clock speed (14.18 for PAL, 14.31MHz for NTSC). I could be wrong here with the exact numbers, particularly with the 1200 and CD32, but my understanding is this is how it's set, and the video and audio circuitry is also set with these clocks, using a divisor where necessary.

The problem isn't that the CPUs can't be clocked higher, it's that if the CPU was clocked higher (or lower), all the other clocks speeds would be messed up, because they wouldn't be exact divisions. To fix this would be far more hassle than it's worth. At least, that's my understanding from my layman's reading of the schematics for the Amiga models. This isn't the case in A3000 and A4000, nor in any model with an accelerator card, as then the CPU doesn't have to run synchronously with the system bus, and the CPU gets it's clock from a different source.

If I'm wrong here I'm sure someone will correct me.

Would it therefore be possible (on the likes of the A500 with its big ass chip) to lift pin 15 of the 68000 and hook it up to its own crystal to overclock it. Much in the same way the sega mega drive can be overclocked (not that I've ever got mine stable).

I might try this just for the fun of it (I do need to get out more don't I)

[Glen M]

OK my stupid question

With my 030 accelerator card fitted in my A1200 is the 020 processor complete defunct or is it still used for anything? Could programmers make use of both processors in code for more performance?

It also makes me wonder that if it is indeed defunct then is it possible with upgrades galore to make most if not all the main board defunct? Is there a point in upgrading that the amiga becomes not so much an amiga anymore?

[Daedalus]

In theory that's possible, but in reality it won't really help much since the rest of the system is still running at its original speed, so the CPU will have to wait for its turn on the chip RAM bus to be able to load new instructions and data and save the results of previous operations. You really need a cache that can run at the higher speed for it to be worthwhile, otherwise all that happens is that the CPU spends more of its cycles waiting. Some operations will be faster, sure, but not worth the effort IMHO. Accelerators generally have fast RAM available, and the 68020 and above included an on-die cache that meant running asynchronously had the potential for much greater improvements.

Oh, and that's not taking into account any assumptions that are made on the basis of the CPU running synchronously with the chipset - there may be other issues that become apparent once you try to run it like that.

Edit: To answer your second post... No, the on-board CPU cannot be used since the Amiga architecture was never designed to allow it unfortunately. The on-board CPU is suppressed when an accelerator is added. The 680x0 series does actually have provisions for multiple-CPU operations, but the Amiga architecture is set up for exactly one CPU (and up to one FPU).

[demolition]

Originally Posted by Glen M

It also makes me wonder that if it is indeed defunct then is it possible with upgrades galore to make most if not all the main board defunct? Is there a point in upgrading that the amiga becomes not so much an amiga anymore?

Accelerators are designed to take over the original CPU, so I doubt there is any way to use the 020, even if you could accept that it would slow down the 030 due to memory syncronization.

As for making the original system defunct, this is indeed possible as nobody prevents you from implementing all the functions on the accelerator and just using the original machine simply for delivering power. You cannot access many of the original connectors directly though, so you cannot reimplement Paula on the accelerator and send the audio out directly through the standard audio jacks.

I don't know how the final layout of the standalone Vampire will be, but it could be made as an accelerator card so it could both operate in full standalone mode or inside an Amiga to augment desired features on top of the original machine. In the latest prototype Apollo code for example, they have implemented much of the functionality from Lisa so you can use AGA through the HDMI port, but that doesn't mean that you can't still use the original Lisa if you prefer.

[Glen M]

Another thing that came to mind last night.

Why is the PCMCIA slot so shallow? On any old laptop that I've seen for example the PCMCIA slot is of sufficient depth that the card sits in its full length with only the end sticking out. On the amiga the card only goes in maybe 30mm. Surely they could of made this deeper by offsetting the card slot above the motherboard to accommodate the depth, or even relocating it completely to the rear of the unit were that mostly pointless trap door exists?

In its current position I'm always fearful that a card will fall out or get damaged etc.

In its current position it almost looks as if it was a last minute need to find space for a card slot, sure that'll do rightly, no-one will ever use it decision!

[idrougge]

I wonder the very same thing everytime I look at my A600.

[Daedalus]

Yep, it's a strange one alright, and I've cursed it myself many times. It could be that the standard was still only in draft when the A600 was being drawn up, or it could've been a later addition as you say. Perhaps they envisioned smaller form factor memory cards being manufactured specifically for the Amiga, I don't know. It would be tricky to relocate it without extensive board reworking since it's related to Gayle and the IDE hardware, with some extra bus buffer chips. It could've been a space/money limitation either - while there's physical space further behind the port, I'm not sure about routing the displaced signals underneath the memory card without a much more expensive screened card carrier.

All speculation though, I'd like to know the definitive answer too!

[demolition]

Espresso cup holder?

Well, the 600 was made before the final PCMCIA specification was in place so I could imagine that perhaps some shorter PCMCIA devices could have been planned. This is just my guess though. There isn't much place in a 600, but the area behind the PCMCIA slot is not really densely populated with tracks so if they wanted to, I am sure they would have been able to design it further in.

At the time, hardly any other PCMCIA devices existed that they needed to support so perhaps they mainly thought about designing it together with their Squirrel SCSI controller.

[idrougge]

But the Squirrel didn't come out until after the A600 was discontinued, and that one also hangs out a fair bit from the slot. The only hardware that actually fits is the Overdrive hard drive.

[Locutus]

Amusingly the PCMCIA socket on other machines from that timeframe *is* properly recessed like on the HP 95LX.

[Glen M]

To be honest the thought never occurred to me that the design might of been completed before the PCMCIA standard was and I suppose its easier to copy and paste that part of the 600 design to the 1200 rather than rethink it at more cost.

Maybe they seen it as a means of quick temporary connection of peripherals and not intended for long term connections to the machine.

[drHirudo]

Originally Posted by slaapliedje

Speaking of this line of questions, I tried getting ScummVM to run through USB to MIDI at one point, but didn't have much luck. Anyone else get this working? (Spider/Mediator/Poseidon).

Yes, I did it on microAmigaOne with AmigaOS 4. It worked pretty well - [ Show youtube player ]
But it was long time ago, so I don't remember the exact details. I think camd.library had to be configured properly to work.

[idrougge]

Originally Posted by Glen M

To be honest the thought never occurred to me that the design might of been completed before the PCMCIA standard was and I suppose its easier to copy and paste that part of the 600 design to the 1200 rather than rethink it at more cost.

While the PCMCIA standard was not finalised, the packaging was well-known from older Japanese memory card standards, and even Commodore had used such cards before on the CDTV.

[Jope]

Originally Posted by Glen M

Why is the PCMCIA slot so shallow?

Perhaps they saved some pennies by using a slot without a mechanical eject button. Having the card stick out gives you ample space to grab it for removal. :-)

[Predseda]

This always brings and idea to have a SHORT Compact Flash - PCMCIA adapter, especially for Amiga 600 / 1200. Could anyone design it and start to provide through Amigakit, please?

[Locutus]

Originally Posted by Jope

Perhaps they saved some pennies by using a slot without a mechanical eject button.

Seeing this is Commodore i think we can be 95% sure that is the answer ;-)

[appiah4]

Originally Posted by Locutus

Seeing this is Commodore i think we can be 95% sure that is the answer ;-)

As cynical as I am when it comes to Commodore we are talking about the company that opted to use a nonstandard 3.5 mounting mechanism inside the A500 case and went about designing, producing and assembling its own eject tab for the floppy drive even though a standard 3.5" mount and face plate would have worked wonderfully well with the case.

The way I look at it, any cost cutting design decisions made by Commodore were most likely coincidental.