7800 vs.....

[DracIsBack]

For example, in the case of the 5200, it could show 200(V)X320(H) and 256 colors- but it would have to be mentioned that it cannot do both at the same time. For example, what resolution can it show at least 16 colors?

 

 

Not a programmer by any stretch, but I do remember some of these details from working in Basic on the Atari 130XE. I'll let the experienced 8-bit/5200 programmers offer more detail.

 

Out of the box, the highest resolution mode is 320 x 192, if I recall correctly but that it's monochrome (GR. . On the 8-bit/5200, resolution varies ... typically 80, 160 or 320 pixels wide and 96 or 192 pixels deep. The number of colors that can be displayed on screen varies by resolution mode.

 

The GR. 11 mode (IIRC) can display 80 x 192 pixels at 16 colors on screen without any tricks. However, on the 8-bit and 5200, you can use display list interrupts to essentially split the screen into different display zones where sections of the screen displayed could be in different graphics modes. For example, the top part could be in GR.1 which uses big text. The middle could be in GR.11, which displays the most colors. The bottom could use GR.0, which displays regular text, which displays the score and credits.

 

Each of those display zones can have their own colors (reflective of what is available in the mode)

 

If I remember right, you can even split the display to include different zones that contain the same mode, but different colors. Hence, this is why there are demos that have 256 colors on screen. The overall display is split into zones by DLIs that, in turn, display a specific subset of the overall palette but give the impression that the 8-bit/5200 can display all 256 colors at once.

 

However, as with the other topics discussed, DLIs are tricky and have some limitations.

[Gorf]

I stand by my previous statements and offer NO appologies.

 

I do still OWN my MA, BTW.

 

 

Your lack of compassion fills me with immense sorrow. I beg you to soften your heart and cast off the bonds of prejudice and intolerance. Come now, it won't hurt. Look at your Aquarius. Take it up into your arms and gaze upon it without malice. Allow the smile to naturally take form upon your jaded and world weary face. The heretofore unloved Aquarius will win your heart, sir. This I know to be true. You can now be born anew in its splendor. Use it. Code it. Take good care of your Aquarius, and in return, it will give you decades of blissful happiness.

 

 

If one thing is clear after these Aquarius intervention posts, it is that we all may have some serious soul searching to do. Let us endeavor to not let hate permeate within us.

 

Let the healing begin.

 

well if it helps....I like it in a 'feel sorry for it' kinda way. Again, its a cute machine

but it could have been so much more out of the box. Come now....even the VIC-20

allowed for highres graphics.

[Gorf]

The GR. 11 mode (IIRC) can display 80 x 192 pixels at 16 colors on screen without any tricks

 

Yes....I was going to use it for a few games...I just hated the lop-sidedness of the aspect

ratio. I think that robot demo was done using that mode.

 

 

. However, on the 8-bit and 5200, you can use display list interrupts to essentially split the screen into different display zones where sections of the screen displayed could be in different graphics modes. For example, the top part could be in GR.1 which uses big text. The middle could be in GR.11, which displays the most colors. The bottom could use GR.0, which displays regular text, which displays the score and credits.

 

 

The DLI's were wonderful. You can pull similar tricks like this on the astrocade.

It had a software ability to fake DLI's...you simply set the vertical line on which

you wish to interrupt, change a few color regs and update the interupt line.

 

You really cant do much in this time as it is one horizontal blanking period, but

its a nice feature along side the ability to split the screen in half for two color

zones.

 

Each of those display zones can have their own colors (reflective of what is available in the mode)

If I remember right, you can even split the display to include different zones that contain the same mode, but different colors. Hence, this is why there are demos that have 256 colors on screen. The overall display is split into zones by DLIs that, in turn, display a specific subset of the overall palette but give the impression that the 8-bit/5200 can display all 256 colors at once.

However, as with the other topics discussed, DLIs are tricky and have some limitations.

 

Yes they are and I am just now playing with PacManPlus's sample 7800 code. It reminds me a lot of the object

processor on the Jag....speaking of which... I need to ask Bob a few questions on that code...

[Vigo]

Actually, Vigo, you just cannot expalin things properly, and you are just too arrogant to admit when you do not know something.

 

Yeah,exactly as I have proven multiple times in this thread...

 

You can babble all you want, but you just don't know.

 

Obviously.

 

And judging how quickly you gave up on 7800 programming, you'll never know.

 

Oh no, he is trying to judge me!

 

I read your threads, and gave up trying to sort useful info from drivel. Note how that programmer I contacted not only knows more than you do, but how to expalin it. So he has you beat on two counts right there.

 

:lol:

 

Note that I had the good taste not to send that pregrammer cyber-copies of your posts. Maybe I should have- he could no doubt use a good laugh.

 

You sort of remember me more of a schoolkid than a grown up. ("Now I get my friend, and HE WILL BEAT YOU, HUH HUH HUH! *snort*")

 

A comparison of specs- along with a brief explanation of what's what- would give one a decent idea of what each can do. For example, in the case of the 5200, it could show 200(V)X320(H) and 256 colors- but it would have to be mentioned that it cannot do both at the same time.

 

Weren't we talking about the CV?

 

For example, what resolution can it show at least 16 colors?

 

Try to hold a magnet in front of your picture tube, and it will suddenly show 16777216 colours... IN ALL MODES! And the best thing is, it works even with the Aquarius!

Edited September 4, 2008 by Vigo

[supercat]

Try to hold a magnet in front of your picture tube, and it will suddenly show 16777216 colours... IN ALL MODES! And the best thing is, it works even with the Aquarius!

 

May dad's Macintosh PowerBook developed a fault which caused a ~1" wide by half-screen-height zone of the screen to be solid yellow. That wouldn't have been so remarkable except that it was a black and white screen. My guess would be that one of the LCD driver connections went bad, causing a substantial DC bias on that part of the display which caused the liquid crystal stuff to behave rather oddly.

[supercat]

2. Be Turing complete.

 

No computer with finite memory is Turing complete. A simple hardwired graphics manipulation system, if it performed logical bitBlt operations on an infinite-resolution screen, would be.

[frogstar_robot]

No computer with finite memory is Turing complete. A simple hardwired graphics manipulation system, if it performed logical bitBlt operations on an infinite-resolution screen, would be.

 

Architectures are colloquially understood to be Turing complete if all they need to complete the definition is unlimited storage. Basically if you can implement the P'' language or equivalent (like say brainf*ck) then the computing device is potentially Turing complete. A true Turing complete machine is an ideal that will most likely never be realized (that infinite storage bit is a real monster) but we CAN build machines that support all of the operations of the simplest Turing complete computer.

 

P'': http://en.wikipedia.org/wiki/P_prime_prime

 

Good behavior requires googling the brainf*ck article as an exercise for the reader .

Edited September 4, 2008 by frogstar_robot

[supercat]

Architectures are colloquially understood to be Turing complete if all they need to complete the definition is unlimited storage.

 

Perhaps, but wouldn't a piece of bitBlt hardware that was hardwired to Conway's "Life" meet that criterion? I would suggest that a more useful requirement for calling something a "processor" would be that it must be able to perform general operations at a speed at least vaguely comparable to the rate it which it performs specialized ones.

[Vigo]

2. Be Turing complete.

 

No computer with finite memory is Turing complete.

 

And thus, no existing computer would be fitting the original definition. I however, supposed that people would understand I was using the informal definition of turing completeness, like almost everyone else. The instruction set of all CPU's would fit the original definition if they supported unlimited memory in theory.

[+MattelAquarius]

I stand by my previous statements and offer NO appologies.

 

I do still OWN my MA, BTW.

 

 

Your lack of compassion fills me with immense sorrow. I beg you to soften your heart and cast off the bonds of prejudice and intolerance. Come now, it won't hurt. Look at your Aquarius. Take it up into your arms and gaze upon it without malice. Allow the smile to naturally take form upon your jaded and world weary face. The heretofore unloved Aquarius will win your heart, sir. This I know to be true. You can now be born anew in its splendor. Use it. Code it. Take good care of your Aquarius, and in return, it will give you decades of blissful happiness.

 

 

If one thing is clear after these Aquarius intervention posts, it is that we all may have some serious soul searching to do. Let us endeavor to not let hate permeate within us.

 

Let the healing begin.

 

well if it helps....I like it in a 'feel sorry for it' kinda way. Again, its a cute machine

but it could have been so much more out of the box. Come now....even the VIC-20

allowed for highres graphics.

 

 

 

Thank you for the kind gesture. You and you Aquarius have come a long way in a short period of time. Imagine what the future holds for you both...

 

[+MattelAquarius]

Try to hold a magnet in front of your picture tube, and it will suddenly show 16777216 colours... IN ALL MODES! And the best thing is, it works even with the Aquarius!

 

 

It is good to see gleeful laughter in the face of such exciting potential of the Mattel Aquarius computer. You can do these things, and many other wonderful expressions of creativity utilizing the untapped potential of the humble blue-keyed creation.

 

The Aquarius awaits your latest ideas!

[Gorf]

2. Be Turing complete.

 

No computer with finite memory is Turing complete. A simple hardwired graphics manipulation system, if it performed logical bitBlt operations on an infinite-resolution screen, would be.

 

 

Like the Jaguar blitter and OPL for instance.

[DracIsBack]

Yes....I was going to use it for a few games...I just hated the lop-sidedness of the aspect

ratio. I think that robot demo was done using that mode.

 

I used to have an early picture digitizer called ComputerEyes (anyone remember that) on my 130XE. Was really cool back then ... but had the same issue. Everything was done in black and white (Gr. 9 if I recall right) and 16-shades of grey .... in that 80 x 192 mode.

 

The results could be ok, or terrible, depending on the image you were trying to capture.

[Vigo]

Like the Jaguar blitter and OPL for instance.

 

The Jaguar blitter and OPL are not turing complete. They are specialised for moving and manipulating graphics, but they can not execute full programs, with loops, branches, etc.

 

One hybrid architecture which would fit this scheme would be the TMS34010 microprocessor. It's a CPU, which has built-in commands for graphic manipulations. But show me one assembler or C-Code listing which executes ON (NOT the JRISC or 68K) the Blitter or OPL... Show me a simple PONG game which executes just on the OPL & Blitter, without using the JRISC or 68k.... Or even something simpler, like a time counter.

Edited September 4, 2008 by Vigo

[frogstar_robot]

Perhaps, but wouldn't a piece of bitBlt hardware that was hardwired to Conway's "Life" meet that criterion? I would suggest that a more useful requirement for calling something a "processor" would be that it must be able to perform general operations at a speed at least vaguely comparable to the rate it which it performs specialized ones.

 

It depends how "hardwired" it is. A Turing complete machine must be able to solve any computable problem in principle. The definition of Turing completeness doesn't specify a time frame so if it takes the machine 100,000 years to add 100 numbers then so be it. It also doesn't specify niceties like ease of development, architecture, input, output, or any form of usefulness whatsoever.

 

I suspect the definition of "processor" falls prey to the same sort of vagueness that the bitness of a given architecture is. And there are no end of useful digital devices like the video processors in some machines that aren't Turing complete...but the the CPUs controlling them generally are.

[HARMIK]

I just want to say that i have found this topic incredibly interesting

 

But i do have some questions

 

1 How close are the VDP's from the SMS to the Mega Drive sorry Genesis and to the 9938/58.

 

2 I have read that Yamaha did the 9938/58 did they pay TI for a licence or anything.

 

3. If there was a coleco 2 do you think it would be close to the MSX2 or more Master System i would think a MSX2 type system would be logical.

[supercat]

It depends how "hardwired" it is. A Turing complete machine must be able to solve any computable problem in principle.

 

I'm pretty sure Conway's "Life" is Turing-complete, and that any Turing-machine problem can be expressed in a grid whose X dimension is proportional to the size of the grid and whose Y dimension is proportional to the number of states. Further, it would certainly be possible to design a hard-wired sequence of blitter-operations which would be Turing-complete using a far smaller grid than would be required for a "Life"-based approach.

 

I suspect the definition of "processor" falls prey to the same sort of vagueness that the bitness of a given architecture is. And there are no end of useful digital devices like the video processors in some machines that aren't Turing complete...but the the CPUs controlling them generally are.

 

Video processors don't do much of anything by themselves (nor, for that matter, do most microprocessors). The concept of Turing completeness would only become meaningful if there were some circuitry to drive them (code-storage ROM at the very least). The minimum number of steps required to achieve a Turing-complete system would depend in significant measure upon what the steps were. Suppose one had a graphics chip with two 1xn graphics buffer called L and R, and two nxn graphics buffers P and T. The following sequence of steps, if iterated, would be Turing-complete (the 'program' would be in P; the initial state would be in L).

• For all i,j in 1..N, R[i,j] = P[i,j] and L
  
• For all i in 1..N, T = 1
  
• For all i,j in 1..N, T[j] = T[j] and R[i,j]
  
• For all i in 1..N, L = L xor T
  

Most graphics coprocessors could be hardwared to run that sequence pretty easily (with the caveat that they only handle finite screen dimensions). That particular implementation isn't just Turing-complete--it can actually perform work in a reasonable number of steps. The biggest problem is that each dimension of the array would be proportional to the required amount of memory).

[supercat]

Yes, you read everywhere that the 65816 has a 16bit ALU, but from an engineer's point of view, it doesn't make any sense, and I'll explain why...

 

If a machine with 6502-style timing fetches multi-byte operands LSB first, an 8-bit ALU will generally be able to perform just as fast as a 16-bit one. In the 6502 instruction set there are two places where a 16-bit ALU would have helped, and one of those could have been dealt with in the instruction sequencer:

• Branch across a page boundary would save a cycle with a 16-bit ALU;
  
• JMP (abs) wouldn't behave oddly if the ALU were 16 bits, though changing the sequencer to use the PC for temp storage would also have worked.
  

If a machine with 6502-style were to fetch operands MSB first, however, a 16-bit ALU or an extra holding latch would be necessary to avoid wasting two cycles on every indexed fetch; even then, one cycle would always be wasted. If I recall, though, the 6805 uses MSB-first addressing. Don't ask me why.

[Vigo]

I just want to say that i have found this topic incredibly interesting

 

But i do have some questions

 

1 How close are the VDP's from the SMS to the Mega Drive sorry Genesis and to the 9938/58.

 

Comparision with 9938/9958:

 

They are not compatible. The 9938 implements 16 & 256 colour bitmapped display whild the SMS/Mega Drive VDP always operate on tiles. Also, the sprites are different, the 9938 can only display 1 colour sprites (which can change colour every line) while the SMS/Mega Drive VDP display 16 colour sprites. The 9938 has only vertical scrolling (horizontal only with a trick) while SMS/MD have full X/Y scrolling. The 9938 can display 512x448 pixel interlaced maximum, while the SMS can only display 224x192 & the MD can display 320x448 interlaced maximum. Plus, since the 9938 operates on bitmaps on the enhanced modes, it has a blitter.

 

Comparision MD/SMS:

 

The MD can display all SMS graphic modes except the old 9918 video modes. The MD has more sprites, more colours (64 vs 32), tiles/sprites with colour attributes, has 2 parallax layers, can switch between 256x224 & 320x224 (the SMS only has 256x192), line scrolling, shadow/highlight modes, DMA transfers to VRAM... Well, much more advanced of course. But SMS backwards compatible.

 

2 I have read that Yamaha did the 9938/58 did they pay TI for a licence or anything.

 

Dunno. If they had to do, then it would be logical that Sega would have, too, since the SMS VDP can display all 9918 modes.

 

3. If there was a coleco 2 do you think it would be close to the MSX2 or more Master System i would think a MSX2 type system would be logical.

 

It would be closer to the SMS, since the SMS has the same sound chip as the Colecovision. Both have the SN76489, while the MSX Systems are AY-3-8910/YM2149 based. For a game console, the SMS VDP makes more sense, the 9938 is not as well suited for games, but more for computer applications.

[HARMIK]

Thank you very much for the reply Vigo I found it all very interesting and helpful, makes me think more of the SMS and Genesis.

 

Interesting you would think that coleco 2 would have been closer to SMS, I would have thought that coleco would not be in a position to design there own custom chip and would buy off someone els, and use the same sound chip or the next model up from the SN76489 if there was any. And to be honest I just thought that 9938 was the next number up so it made sence, but then again at the begining of this topic I would have thought that 9938/58 were buy TI. So now that I know it was not, Yamaha might not have sold it to them any way.

 

How about the 9958 I believe that it had horisontal scrolling, do you think that would make a better console VDP than the SMS one, and would it take the Genesis one to beat it, or would it be better than both.

Edited September 5, 2008 by HARMIK

[CV Gus]

Since I contacted that programmer not long after starting this thread, and he explained the various graphics modes of the CV, along with limitations, scrolling, sprites, etc., and even comparing Z-80 to 6502 processors (roughly), I have a pretty good idea of what the CV is about. Sort of. My thanks to him, again. But, again, he admitted to not knowing enough about the 7800 to compare it to the CV.

 

Could someone here give a detailed listing of 7800 specs? That way, I can try to compare the two.

Edited September 5, 2008 by CV Gus

[Paranoid]

I always find these debates kind of ironic for a retro forum.

 

"Which is the better gaming console, the 7800 or the Xbox 360".

 

I mean, really, the same relevence is there. I prefer the Atari 8 bit/5200 as the "ultimate classic retro gaming platform". Honestly, the 5200 is mecca for me - and an Atari 8 bit is just a less finicky, easier to support and maintain solution to enjoy most of the same titles.

 

But most of the time, I run emulation either on a PC or an Xbox, either of which now have a suitable number of alternative I/O methods to support a decent old school joystick type interface to accurately reflect the control methods of either of these gaming platforms.

 

I also prefer the Coleco and Coleco games to the vast majority of 7800 titles. I normally use the 7800 as a 2600 with the Cuttle Cart 2, and otherwise would find no real compelling reason to own one. There is no doubt that it reflects a jump in graphics capability over the Colecovision, but the entire feel of most of the games on the 7800 just leaves me flat, whereas Colecovision games were compelling.

 

But it all comes down to subjective opinion - or to hard technical specs that don't tell you a *thing* about if gameplay is actually enjoyable and engaging.

[Gorf]

Like the Jaguar blitter and OPL for instance.

 

The Jaguar blitter and OPL are not turing complete. They are specialised for moving and manipulating graphics, but they can not execute full programs, with loops, branches, etc.

 

One hybrid architecture which would fit this scheme would be the TMS34010 microprocessor. It's a CPU, which has built-in commands for graphic manipulations. But show me one assembler or C-Code listing which executes ON (NOT the JRISC or 68K) the Blitter or OPL... Show me a simple PONG game which executes just on the OPL & Blitter, without using the JRISC or 68k.... Or even something simpler, like a time counter.

 

 

Scott Legrand and doug Engel supossedly have done this. The fact is there really is no such thing as turing complete.

This is also NOT the reason the Jaguar is or isn't 64 bits. A study of the net files will show you that jag is indeed a 64

bit SYSTEM...no there are no 64 bit CPU's...in fact there is no CPU in the Jaguar...there are 3 gen purp processors.

None of which have ultimate control....every priority in the Jaguar can be re-aranged. The 68k is the HOST, not the CPU.

[HARMIK]

I find the VS dedates very interesting coleco VS 7800 as this one is has been a great read. I love all retro systems so don't usually take sides but i like to learn about the technical capabilities of each system. When i was a kid and the 7800 was about to come out, I could not wait for it and thought it was going to be the most powerful system ever made. I started calling many many shops in the U.S. asking about it.

 

All had good things to say about it and could not wait for it to come in. None said anything like we have been burned buy ATARI and will never deal with them again as you hear today. I do think the system would have done allot better if it had come out in 84. I also think some of the best games were the first ones Ms Pac Man, Joust, Dig Dug, Galaga, Xevious back in 84 arcade games were still big, and they would have been almost arcade perfect. And they would have been the first gen of games I don't think programing for the 7800 realy got pushed under the new ATARI.

 

Having said that now that Im not an excited little kid making about 100 long distance calls on my parants phone. I think it was built on the cheep and that GCC did a great job with the budget they were given. And that it turned out better than ATARI thought, after all ATARI were going to call it the 3600. I think ATARI's original plan was to get something out better that the 2600 but compatible with it. Thats why they did not pay for a POKEY or POKEY 2 to be put in it. I mean if your intent was it build something better than the 5200 you would make the sound better or at least equal to it.

 

I also think the 5200 was a great system but as i said i like all or at least most retro systems. My favorite system is the Intellivision

Edited September 6, 2008 by HARMIK

[ZylonBane]

Could someone here give a detailed listing of 7800 specs?

Wow, you really have been accessing this thread in write-only mode, haven't you?