Conference hydra::amiga_v1

    Any price for this v2.0 upgrade ?
    
    				Cheers,
    					Erik
    
    
    PS - Is a 8 channel chip better than using an 8 channel software like
    	the octalizer ?

    There's been very little information on the v2.0 operating system
    upgrade here in the States.  Because of the A3000 and beta copies being
    in the hands of many developers, much has been written about features
    and compatibility.  But so far there's been nothing on price, dates, or
    even whether or not the upgrade will require new custom chips.  BTW we
    don't have the A1500 in the US.
    
    Can you provide any details from your letter and Commodore's replies?
    
    Wes

I have heard RUMOURS that a v2.0 upgrade will be 70-80 pounds, but there was
no official price declaration in the letter. I guess the price will depend on
whether your machine already has the ECS Agnus or not, unfortunately I believe
the upgrade will be "dealer-only" because of the need to fit an actual new KS
ROM (when these are avialable!!).

As for the Super Paula ... it is one of the new breed of chips which humble
A500 owners seem to get their hands on last, just like the 2Mb Super Agnus and
the new Denise chip. I have heard (rumours again - groan, why never actual
details?) that the Super Paula will have a small board of it's own to plug
into the socket of the old Paula. I think that the chip will offer new features
as well as the extra channels, for improved sample quality, so it must be
OVERALL better than using, say OCTOMED, with the present chip; however I am not
sure whether the new channels are gleaned via new design or simply by using the 
same methods employed by the present software (multiplexing, using the blitter
to mix channels (?) ). The chips must be at present with some privileged 
developers, but news seems to be annoyingly scarce.

Anyways, keep your ear to the ground, and thanks for the reply!

							- JIM CAD*

    hmm.... could this be the way they are going to add 1.44Meg
    floppy support?  I remember reading that the current Paula
    chip couldn't handle the doubled data rate needed for ms-dos
    compatible 1.44Meg floppy drives.  I can't believe they really
    intend to go with the half speed drives for the high end systems.
    
    Dave

    I now have an 512k chip in my A2000. If I buy now an 'Fatter' Agnus of
    1mb, could I then get now the upgrade, or should wait and buy the upgrade 
    with all the chips.
    
    The new drive, shouldn't be 1.56mb ?
    
    	Erik

Erik,

	I think the only benefit you would gain from the extra 512k chipRAM
	would be if you were doing intensive graphical work such as rendering
	with Sculp-Animate 4D etc... therefore I reckon you would save yourself
	hassle by waiting for the upgrade when you will probably have a Fatter
	Agnus installed at the same time.

								- JIM CAD*

	I don't know much about the new drive.

    Thanks for the answer, I will wait.
    
    I just got another question turning in my 'processor', Are the
    512kb of the Chip memory in the 'Fat' agnus (inside the square chip).
    I have a friend with an 'Fatter' agnus 1mb, but when he checks the
    memory in his A2000, he's got 1mb and not 1.5mb ?
    
    						Erik

	The figures quoted for versions of the Agnus chip represent how much
	chipRAM the chip can address - i.e. how much of your normal memory
	can be used as chipRAM by the custom chips.

	There is not actually any memory ON the Agnus chip itself.

	For example:

	a 512k AMIGA with Fat Agnus(512k) could have 512k chip - all present
		memory would be used as chip.

	a 1Mb AMIGA with Fat Agnus(512k) could still only have 512k chip -
		there would be 512k fast too in this case. However, now
		changing to a Fatter Agnus(1Mb) would allow all 1Mb to be
		used as chip - *but* this does not increase the available
		memory since, as above, the Agnus figure is only how much
		chip RAM maximum may be used - FROM THE AVAILABLE MEMORY.
		This is why your friend still only has 1Mb but he has a
		"fatter" chip.

	incidentally, the 3000 has a Super Agnus (up to 2Mb chip!), also
	you might have to have your 1Mb chipRAM feature enabled if you get
	a new Agnus - hence my reply to wait and get the dealer to do the
	whole v2.0 upgrade (enabling requires cutting a track on the board
	I believe).

						- JIM CAD*

    Paula info required :
    
    I was thinking about building an electronic project called a Hemisync
    which generates two very stable frequencies of around 200 Hz, when it
    occurred to me that this would be an ideal application fro the Amiga
    as long as the paula chip is up to it.
    
    As background; a Hemisync is a machine for stimulating your brain to
    produce alpha and theta waves by audio waves in your ears (where
    else!). You feed 200 hz in one ear, 210 hz in the other ear and this
    stimulates your brain to produce brain waves at 10 hz (alpha wave
    frequency).
    
    The design I was looking at uses data held in an EPROM to program
    two D/A converters to give the required frequencies and the required
    waveforms.
    
    So is it possible to program the Paula chip to produce a frequency on
    each of the stereo channels of varying waveforms? These channels must
    not be multiplexed, so frequency of channel A will not generate
    harmonics from multiplexing when combined with frequency of channel B.
    Also these frequencies must be stable to at least 1 hz and if possible
    maintain a constant phase relationship with each other.
    
    Next question: what software would allow me to use the paula chip to
    produce these different waveforms?
    
    Thanks,
    Barry.

re: -.1

>    So is it possible to program the Paula chip to produce a frequency on
>    each of the stereo channels of varying waveforms? 

	Yep.
    
>    Next question: what software would allow me to use the paula chip to
>    produce these different waveforms?
    
	WIthout trying it myself, the easy answer is to check out the 
	audio.device chapter of the ROM Kernal Reference manual, Libraries
	and Devices. What you need to do is produce a table of "sample"
	numbers representing (I suppose) a sine wave. Then just pass some
	IO data through BeginIO() for the proper frequencies and audio
	channels. 

	Please excuse my not elaborating, it's not complicated, it's just
	that it's very accurately explained in the manual. There's sample
	code listing in C. The code on disk is on Fred Fish disk 344.

	These low frequencies should be no sweat for the audio.device. As
	far as I know, the output waveforms should be rock steady, because
	they're software sync'd to the computer's clock.

    Thanks MKODEV::OSBORNE. It sounds perfectly suited to the hemisync
    application. I'll get hold of the Kernel manuals and give it a try.
    It should save me 60 pounds in electronic pieces and a lot of time in
    constructing the project (although this is half the fun!).
    
    Thanks,
    Barry.