TURBO CARD for XE ATARI

[Rybags]

Possible interest here... but I've got 800XL, Ultimate 1 Meg, VBXE and IDE+2 so it's a crowded system.

 

Re colour stuff - the strength there could be just plugging new values into registers more quickly.

Think GTIA paletted mode with much lesser limitations on colour changes, 160x multicolour with all registers changable multiple times per line. If the block move works well then all the better.

Rastaconverter could go on steroids, faster CPU means stuff like PM reuse becomes feasible.

Edited October 29, 2014 by Rybags

[sloopy]

I would be interesting in having one on my shelf...

 

one day, I will be able to play with my A8 stuffs :/

 

sloopy.

[drac030]

In high speed mode, should be close to 4x the processing power of an ST. Since you need about 5 times the processing power to emulate a different processor, you should be able to emulate an ST with an 8 bit.

These estimates look pretty optimistic. Motorola's 68k 32-bit move.l (a0),(a1) takes 20 clock cycles, and 65C816 identical indirect move takes, say, 31 clock cycles:

 

lda [zp1] ;7 clocks
sta [zp2] ;7 clocks
ldy #$0002 ;3 clocks
lda [zp1],y ;7 clocks
sta [zp2],y ;7 clocks

(note that what I am aiming at is the exact equivalent, you surely can save few cycles if the move does not exceed a 64k segment)

 

This still makes a 20 MHz 65C816 faster than an 8 MHz m68k, but certainly not at a factor of 4, just ~1.6.

Edited November 1, 2014 by drac030

[ricortes]

Hi Drac. International forum so I'm not sure we share lexicons.

 

"Kind of a pipe dream though"

 

I believe this came from Lewis Carol's "Through the Looking Glass/Alice in Wonderland." The reference infers: If I had been smoking opium/hashish and was not of my right mind.

 

Still, I think the '816 system as described should compare very favorably with an ST/Amiga/'286 PC/Mac.

 

Most of the sieve programs I recall => ST 3-7 times faster then an 8 bit. Ditto for the other computers I mentioned. 20 MHz/1.79 MHz => 11X the speed of a stock 8 bit plus the improved instruction set should be able execute mixed instructions at a pretty good clip. I'll concede working on and with 16 bit values, the ST and the others mentioned should/would compare favorably since that its strong suit.

[+Larry]

I remember when the C64 Super CPU (20 MHz) came out. A Commodore mag editor speculated that the Super CPU would be faster than a Pentium! I thought, "dream on..." He came back to reality when some folks laid some numbers on him.

 

But the great increase in cpu speed might allow nice programs to be written in a high level language and still perform similarly to ML on a 1.79 system. That should allow for easier development. Perhaps this new cross-compiler language might have come at just the right time. (Can't remember the name right now.)

 

-Larry

[flashjazzcat]

Yeah: Windows is written in C/C++ and runs quite well on a 2GHz processor.

Edited November 1, 2014 by flashjazzcat

[ricortes]

Yeah: Windows is written in C/C++ and runs quite well on a 2GHz processor.

Only if you have at least 2 gigs of memory and don't multitask.

[pirx]

 

This still makes a 20 MHz 65C816 faster than an 8 MHz m68k, but certainly not at a factor of 4, just ~1.6.

 

 

What would make XFormer quality Atari emulator doable (:

 

pirx

[Prosystemsearch]

This is a great thing! Now all we need is an XE cheat device!

[Prosystemsearch]

I really hope it works with the XEGS!! :thumbsup:

[kogden]

I remember when the C64 Super CPU (20 MHz) came out. A Commodore mag editor speculated that the Super CPU would be faster than a Pentium! I thought, "dream on..." He came back to reality when some folks laid some numbers on him.

 

I don't think a 65816 could keep up with a Pentium if you clocked it at 200MHz. A 286 maybe. Maybe a 386SX/16.

 

But the great increase in cpu speed might allow nice programs to be written in a high level language and still perform similarly to ML on a 1.79 system. That should allow for easier development. Perhaps this new cross-compiler language might have come at just the right time. (Can't remember the name right now.)

 

The problem is that the user base would be so small for such software because it would run like crap on a stock Atari.

[+Larry]

 

The problem is that the user base would be so small for such software because it would run like crap on a stock Atari.

 

 

Yes, probably true -- depends if the accelerator idea takes off. It's the "chicken or egg" thing. I wonder if much "special software" was written for the C64 Super CPU? But they've had their own problems since the original developer went out of business. But still, you should see BASIC fly at 14 MHz. The ABC compiler really is snappy, even at 7 MHz. The MyIDE's (soft OS) really shine with an accelerator.

 

-Larry

[visionik]

I'd be in for five of these, thanks.

[+Larry]

This is a compatibility question... Is the SIDE cart 3.3 VDC (or 5 like a regular cart)?

 

-Larry

[flashjazzcat]

This is a compatibility question... Is the SIDE cart 3.3 VDC (or 5 like a regular cart)?

 

 

 

5v at the connector, although the CF card and other bits run at 3.3.

[ricortes]

Just because you can do something doesn't mean you should do it.

http://www.bit-tech.net/news/hardware/2012/03/30/linux-atmel-microcontrollers/1

 

"Curiously enough, once booted, the system is somewhat usable,' Grinberg claims in defense of the design. 'You can type a command and get a reply within a minute."

Emulated an ARM so he could run Linux on an AVR. I have a wide assortment of older/low end hardware. Sometimes running applications I can't tell if the system has crashed or is just slow. I think there was a cautionary tale about something like this. DEC PDP-8<or older?> software that was run via emulation on each succeeding model. The final incarnation ran slower on a VAX then the original did on the original PDP-8.

 

Still, there are some programs that ran OK on a 10 MHz XT that I would probably use if they were available. Pools of Radiance comes to mind. I believe it was originally written in Turbo Pascal on the MS DOS version so Larry's comment about high level languages being available would hold true if there was a way of getting a hold of the original source code. It also came out for the Apple, Nintendo, C64, essentially everything that didn't have Atari on it. Pipe dream.

[Ffin72]

Put me down for one please.

[+Larry]

@ Lotharek-

 

Any update on the project? Have your boards? A short update once a month or so would be welcome.

 

Thanks,

Larry

[+Larry]

Here's sort of an update in another thread... (pics at #42)

 

http://atariage.com/forums/topic/229928-sio2pc-usb-by-lotharek/page-2

 

-Larry

[DyLucke]

I would be interested on this item too.

But, a question, what about the other accelerator Lotharek is advertising on his site, the external one. Is meant to be used on the A8 computers only or maybe it would be able to work on the C64?

[lotharek]

Langsam

 

step by step.... first RAPIDUS as a lead design. then other clones

[pjbramley]

I would be interested in one.

[vazquezrick]

I'll take one.

[+Larry]

I've already got a couple of 7 MHz accelerators, that really "make things happen" and I probably need 20 MHz like another hole in the head. But what the heck, it is neat and I'm eager to see its strengths and weaknesses (at least as I perceive them). Since it is compatible with the Ultimate and IDE+2, that's a big plus, already. None of the external PBI ram upgrades work even at 7 MHz, so I'm not anticipating that they will work with Rapidus. Hopefully the MyIDE-II will run well.

 

-Larry