Its 1993, you're in charge of the Jag, what do you do?

[Gorf]

The jaguar had to be successful at launch - in the window before the PSX/Saturn arrived,

so the quality of the initial titles was way more important than the games that followed.

 

Agreed. This is exactly why an 020 would have been better. An 020 would not haev been a detriment to the main bus

for several reasons:

 

A ) dynamic cache,

B ) 16 bits wider on the data path

C ) twice the clock

D ) more efficient cycles per instruction

E ) AI and game logic now make sense on the host.

The last reason allows for much faster developmentof much better quality games.

 

However, simply utilizing the Jaguar's true color pallete, taking those low color 16 bit ports

in the first month and porting over about 30 titles of classics 16 bit games favorites with

much higer resolution and photoreal graphics would have been much better. A few new

features in those games would not hurt either.

 

Seeing the same exact title on a machine is no where near as exciting to a fan of any particular

16 bit title as would be seeing it now in all its larger sprite, photo real, paralax scrolling backround

glory on a just released Jaguar console. If this were the case, the 68k would have been good enough.

 

Maybe Atari should have saved cash by not having a 68000 at all , and spent the money saved on a proper GPU compiler with main memory execution.

I cant remember the bottlenecks on WTR - but the rendering took most of the time - Lee finished it off on his own.

 

Yes but then you split the main ram to two seperate busses and only the blitter is allowed to acces either.

So now you have the GPU with 1.5 meg(plenty for high depth color graphics back then) and .5 meg on the

DSP and a private bus. The blitter instead of a costly small dual port ram buffer would do all the communicating

between the DSP to and from the GPU and in reverse as well.

 

This would have also required a bit of redesign on the DSP to open up its ability to hit the bus at full speed.

Now the DSP can do sound and input and the matrix math for any 3D with no big 12 cycle hits to a unified

main bus choked by a 68k.

 

All this while the GPU rocks away with the blitter and OPL to render the display. This is probably the most optimal

cost/power scheme you could get with the T&J.

 

 

I think not having the cart port 64 bits wide was not wise either and the fact that it was so slow.

Edited June 12, 2009 by Gorf

[kgenthe]

The jaguar had to be successful at launch - in the window before the PSX/Saturn arrived,

so the quality of the initial titles was way more important than the games that followed.

 

Agreed.

Finally, after 15 pages of posts, lol.

[Gorf]

The jaguar had to be successful at launch - in the window before the PSX/Saturn arrived,

so the quality of the initial titles was way more important than the games that followed.

 

Agreed.

Finally, after 15 pages of posts, lol.

 

No, not exactly. We already agreed on this a while back. The debate is whether or not a better choice of host processor

would have helped that said software to reach the sevles in a more timely manner.

[carmel_andrews]

I think the argument about whatever processor the jaguar should have been using is not really relevent as whatever processor they used would have meant adapting the hardware to the processor and also getting the support of the 3rd parties

 

remembering ofcourse that no gaming systems existed at the time that utilised an 020 and atari wanted using a processor that 3rd party publishers already knew how to code on (i.e the 68000)

 

 

I liken the argument about the processor the jaguar should have used, to an argument i remember back in the day about the lynx and should the lynx have used a 68000 or a 6502

[Christos]

I think the argument about whatever processor the jaguar should have been using is not really relevent as whatever processor they used would have meant adapting the hardware to the processor and also getting the support of the 3rd parties

 

remembering ofcourse that no gaming systems existed at the time that utilised an 020 and atari wanted using a processor that 3rd party publishers already knew how to code on (i.e the 68000)

 

 

I liken the argument about the processor the jaguar should have used, to an argument i remember back in the day about the lynx and should the lynx have used a 68000 or a 6502

 

Everyone who knows how to code the 68000 can code the 68020 or the 68060.

[philipj]

The 68000 chip was probably a cheaper option for Atari.

[mcjakeqcool]

@philipj

To true.

[Gorf]

I think the argument about whatever processor the jaguar should have been using is not really relevent as whatever processor they used would have meant adapting the hardware to the processor and also getting the support of the 3rd parties

 

remembering ofcourse that no gaming systems existed at the time that utilised an 020 and atari wanted using a processor that 3rd party publishers already knew how to code on (i.e the 68000)

 

 

I liken the argument about the processor the jaguar should have used, to an argument i remember back in the day about the lynx and should the lynx have used a 68000 or a 6502

 

You are the first person I ever heard state someone complain about the lynx using a 6502. The Lynx was perfect from the start

hardware wise. It was doing polygon based games even before some of the other consoles. It's tool kit and hardware are just

what you want for a system. If only the Jaguar had followed that model.

[Gorf]

I think the argument about whatever processor the jaguar should have been using is not really relevent as whatever processor they used would have meant adapting the hardware to the processor and also getting the support of the 3rd parties

 

remembering ofcourse that no gaming systems existed at the time that utilised an 020 and atari wanted using a processor that 3rd party publishers already knew how to code on (i.e the 68000)

 

 

I liken the argument about the processor the jaguar should have used, to an argument i remember back in the day about the lynx and should the lynx have used a 68000 or a 6502

 

Everyone who knows how to code the 68000 can code the 68020 or the 68060.

 

The other thing is the 68000 would be overkill for the Lynx anyway. The 68k is 'underkill' in the Jaguar.

[Crazyace]

The jaguar had to be successful at launch - in the window before the PSX/Saturn arrived,

so the quality of the initial titles was way more important than the games that followed.

 

Agreed. This is exactly why an 020 would have been better. An 020 would not haev been a detriment to the main bus

for several reasons:

 

A ) dynamic cache,

B ) 16 bits wider on the data path

C ) twice the clock

D ) more efficient cycles per instruction

E ) AI and game logic now make sense on the host.

The last reason allows for much faster developmentof much better quality games.

 

 

That's not really relevant to my point though - The initial games were disappointing - Raiden for graphics, and Trevor for Gameplay - If Doom had been a launch title things might have been different - and Rayman showed the graphics prowess for 2D. There should have been more games like Super Burnout as well.

 

 

 

 

However, simply utilizing the Jaguar's true color pallete, taking those low color 16 bit ports

in the first month and porting over about 30 titles of classics 16 bit games favorites with

much higer resolution and photoreal graphics would have been much better. A few new

features in those games would not hurt either.

 

Seeing the same exact title on a machine is no where near as exciting to a fan of any particular

16 bit title as would be seeing it now in all its larger sprite, photo real, paralax scrolling backround

glory on a just released Jaguar console. If this were the case, the 68k would have been good enough.

 

It all comes down to cash - If Atari had paid - they would have got better games. 3D0 had Street Fighter - Atari should have spent money on MK2 - launching with that as a pack in could have sold a million jaguars!

 

 

Yes but then you split the main ram to two seperate busses and only the blitter is allowed to acces either.

So now you have the GPU with 1.5 meg(plenty for high depth color graphics back then) and .5 meg on the

DSP and a private bus. The blitter instead of a costly small dual port ram buffer would do all the communicating

between the DSP to and from the GPU and in reverse as well.

 

This would have also required a bit of redesign on the DSP to open up its ability to hit the bus at full speed.

Now the DSP can do sound and input and the matrix math for any 3D with no big 12 cycle hits to a unified

main bus choked by a 68k.

 

All this while the GPU rocks away with the blitter and OPL to render the display. This is probably the most optimal

cost/power scheme you could get with the T&J.

 

 

I think not having the cart port 64 bits wide was not wise either and the fact that it was so slow.

 

I wasn't actually thinking of anything expensive like that - just removing the 68k, fixing the gpu bugs ( and maybe making the dsp run code slowly from main memory as well ) - and spending some cash on software tools so that the gpu/dsp are coded completely in C.

Having a GPU that runs code at 1/2 the bus and a DSP that runs code at 1/4 the bus is still better than the 68k.

[Gorf]

That's not really relevant to my point though - The initial games were disappointing - Raiden for graphics, and Trevor for Gameplay - If Doom had been a launch title things might have been different - and Rayman showed the graphics prowess for 2D. There should have been more games like Super Burnout as well.

 

Understood and no disagreement. The relevance to the point is that the 020 would have made it simpler to make the games better.

The real trouble is those games were ports from what was to be the Panther: CyberMorph and Trevor. Dino Dudes and Raiden were

obviously ports.

 

It all comes down to cash - If Atari had paid - they would have got better games. 3D0 had Street Fighter - Atari should have spent money on MK2 - launching with that as a pack in could have sold a million jaguars!

 

No doubt, but even with the lack of guys there were still a few in house coders(or hire a few) to port over licensed titles, instead of trying to take

over the video game universe before you even have a reputable foot back in the video game universe...Atari 's name was not exactly golden any

more with developers. They depended way to heavily on 3rd party hoping everyone would run to them(at first they did but most soon ran the other way.)

 

Yes but then you split the main ram to two seperate busses and only the blitter is allowed to I wasn't actually thinking of anything expensive like that - just removing the 68k, fixing the gpu bugs ( and maybe making the dsp run code slowly from main memory as well ) - and spending some cash on software tools so that the gpu/dsp are coded completely in C.

Having a GPU that runs code at 1/2 the bus and a DSP that runs code at 1/4 the bus is still better than the 68k.

 

Well that was my point exactly. This would not need a host. The unit design by default would probably be much less of a hassle, easier,

cheaper and not to mention need for a less complicated MMU and DMA scheme(less silicon is always cheaper.) The DSP chip already has

all 32 bit signals(even the very ones in the Jaguar, which are simply not connected but are indeed there.). You dont need a dual port ram

between them anymore as the blitter does data transfers between them every frame.

 

Now you do not have any bus choking data paths since the DSP will now operate full width and no longer have to take cycle hits as a slave.

I think the only reason why the DSP needed 12 cycles was the fact that the 68k forced it.( Kskunk, please enlighten if not correct, since I

know you have a well rounded understanding of Jerry. ) I mean the RISC core is the same chip with a few different instructions and I

am sure it could have been opened up to hit the bus at GPU access speeds if it was not already able to as is.

Edited June 14, 2009 by Gorf

[kskunk]

I think the only reason why the DSP needed 12 cycles was the fact that the 68k forced it.( Kskunk, please enlighten if not correct, since I

know you have a well rounded understanding of Jerry. )

 

There's three big bus performance problems in Jerry, but only one is caused by the 68K. Jerry has a 32-bit bus, but with a 68K installed, Jerry must run in 16-bit mode. (This is because Tom sees all bus masters as the same -- so both the 68K and Jerry must use the same bus width.)

 

The next problem is that Jerry's memory pipeline is hard-coded to delay 6 cycles. So with a 68K, Jerry takes 24 cycles to read a 64-bit word. With a 68020, Jerry would use 12 cycles to read a 64-bit word, but that is still much worse than Tom which can read a 64-bit word in 2 cycles.

 

The final problem is that Jerry's memory pipeline has buggy writes, so writing a 16-bit word takes not 6 cycles, but 12.

 

They may have intended Jerry to work primarily as an audio synthesizer (i.e., all computation, little memory access). It's pretty hard to use it for much else due to its slow memory interface.

 

- KS

[Gorf]

There's three big bus performance problems in Jerry, but only one is caused by the 68K. Jerry has a 32-bit bus, but with a 68K installed, Jerry must run in 16-bit mode. (This is because Tom sees all bus masters as the same -- so both the 68K and Jerry must use the same bus width.)

 

The next problem is that Jerry's memory pipeline is hard-coded to delay 6 cycles. So with a 68K, Jerry takes 24 cycles to read a 64-bit word. With a 68020, Jerry would use 12 cycles to read a 64-bit word, but that is still much worse than Tom which can read a 64-bit word in 2 cycles.

 

The final problem is that Jerry's memory pipeline has buggy writes, so writing a 16-bit word takes not 6 cycles, but 12.

 

They may have intended Jerry to work primarily as an audio synthesizer (i.e., all computation, little memory access). It's pretty hard to use it for much else due to its slow memory interface.

 

So in other words it is the memory interface between the core and the external system that is hard coded for delay which is pretty

awful actually. Since we are fantasizng, make the Jerry just like Tom except now Jerry would be simply a Tom with sound and I/O

hardware instead of gfx/vid hardware. A 64 bit 2 cycle external capable Jerry would allow for reading in 4 channels of 16 bit audio

in 2 cycles. Also you can do 8 channels 8 bit audio as well. In fact I say have the I/O bus a completely seperate 8 bit bus altogether.

A sound system and I/O block is probably a lot less silicon than a Blitter OPL and video system, so you can probably still maintain the

8k local.

 

Now you have an undisputable 64 bit system really operating at full bore 64 bits most of the time. I still maintain the 020 would have

been the best overall choice even with the chips as they are. Just the fact that the 020 would actually be able to do work and stay off

the bus most of the time would be a big plus. When it did hit the bus it would be at twice the data width and clock, with a great deal of

already in place tool support, unlike it's J-RISC counter parts.

[Crazyace]

A 68020 lacks a data cache - so it's still going to have stall points, even with the 32 bit advantage. - If there were no 68k at all the DSP would run at 32 bits as well - 6 cycle .

 

A Jaguar with no 68k - and with GPU and DSP working from main memory ( and programmed in C ) would be as easy to code for as a 68000/68020 version.

Of course straight C wouldn't give you the full use - but Atari could supply optimised libraries for 2D and 3D object creation and sound generation.

And ( more importantly for profitability ) the money saved by not having a 68k could go towards incorporating a CD drive rather than cartridge - giving a single machine that would completely outclass both the SNES/Genesis and the 3D0.

 

.. Of all the changes I actually think adding a CD is the most important - as high quality FMV/Animation would have been the 'eye candy' that sold the machine - especially compared to the dodgy Sega CD video. Also having more profit on games due to lower manufacturing costs would be important both for Atari's profit, and also to pull in 3rd parties ( with increased royalties )

[Gorf]

A 68020 lacks a data cache - so it's still going to have stall points, even with the 32 bit advantage.

- If there were no 68k at all the DSP would run at 32 bits as well - 6 cycle .

 

 

Very few stalls in relation to the 68k and certainly a significant improvement. You can't just blow off the 32 bit width and the double clock.

 

 

A Jaguar with no 68k - and with GPU and DSP working from main memory ( and programmed in C ) would be as easy to code for as a 68000/68020 version.

Of course straight C wouldn't give you the full use - but Atari could supply optimised libraries for 2D and 3D object creation and sound generation.

And ( more importantly for profitability ) the money saved by not having a 68k could go towards incorporating a CD drive rather than cartridge - giving a single machine that would completely outclass both the SNES/Genesis and the 3D0.

 

The CD drive back then would make the machine cost twice as much. The 68k would not even come close to covering that.

 

 

.. Of all the changes I actually think adding a CD is the most important - as high quality FMV/Animation would have been the 'eye candy' that sold the machine - especially compared to the dodgy Sega CD video. Also having more profit on games due to lower manufacturing costs would be important both for Atari's profit, and also to pull in 3rd parties ( with increased royalties )

 

 

If content was the issue, they should have just made sure the communications ports actually worked correctly and have

the fist console with downloadable games and content. Do a Skunkboard like cart for this and still make other cart based games.

[Crazyace]

A 68020 lacks a data cache - so it's still going to have stall points, even with the 32 bit advantage.

- If there were no 68k at all the DSP would run at 32 bits as well - 6 cycle .

 

 

Very few stalls in relation to the 68k and certainly a significant improvement. You can't just blow off the 32 bit width and the double clock.

 

Of course not - the 68020 is a more powerful processor than the 68000 - but with a 32 bit DSP and GPU running from main memory you could get great performance without using either.

 

A Jaguar with no 68k - and with GPU and DSP working from main memory ( and programmed in C ) would be as easy to code for as a 68000/68020 version.

Of course straight C wouldn't give you the full use - but Atari could supply optimised libraries for 2D and 3D object creation and sound generation.

And ( more importantly for profitability ) the money saved by not having a 68k could go towards incorporating a CD drive rather than cartridge - giving a single machine that would completely outclass both the SNES/Genesis and the 3D0.

 

The CD drive back then would make the machine cost twice as much. The 68k would not even come close to covering that.

I disagree - the Sega CD had been out for a while afterall , I'd expect a launch price of 299 instead of 249 as a worst case ( Assuming removing the 68k and cart saves $20 , and a CD adds $50 to the component costs ) - After all the Jaguar CD-rom is almost an Audio CD, rather than a computer CD-rom , so it would be cheaper.

 

 

.. Of all the changes I actually think adding a CD is the most important - as high quality FMV/Animation would have been the 'eye candy' that sold the machine - especially compared to the dodgy Sega CD video. Also having more profit on games due to lower manufacturing costs would be important both for Atari's profit, and also to pull in 3rd parties ( with increased royalties )

 

 

If content was the issue, they should have just made sure the communications ports actually worked correctly and have

the fist console with downloadable games and content. Do a Skunkboard like cart for this and still make other cart based games.

 

Infrastructure wasn't there in 1993 - also, download costs would probably be more than CD pressing. Having all the hardware bugs fixed would have been great though.

[Gorf]

Of course not - the 68020 is a more powerful processor than the 68000 - but with a 32 bit DSP and GPU running from main memory you could get great performance without using either.

 

Not without a decent tool set.

 

 

A Jaguar with I disagree - the Sega CD had been out for a while afterall , I'd expect a launch price of 299 instead of 249 as a worst case ( Assuming removing the 68k and cart saves $20 , and a CD adds $50 to the component costs ) - After all the Jaguar CD-rom is almost an Audio CD, rather than a computer CD-rom , so it would be cheaper.

 

Players were about $100 back then for 2 speeders.

 

 

A Jaguar with I Infrastructure wasn't there in 1993 - also, download costs would probably be more than CD pressing. Having all the hardware bugs fixed would have been great though.

 

 

I'd have been happy with the RISC bugs fixed if I could have anything.

[Crazyace]

If the GPU/DSP hardware bugs were fixed and both processors could run from main memory the tools would have been ok - The C compiler did work for generating GPU code, and GDB would have targeted the gpu instructions set and registers instead of any 68xxx. Tools would have matched what everyone else supplied at product launch.

I'd expect that atari would go for single speed( 2x if it didn't cost much more ) and they would pay far less than $100 as they would just have the mechanism, with their own silicon.

 

The CD32 came out around the same time - and a jaguar with CD would have blown it out of the water , and if I can 'whatif' I'd have loved Atari to get Lucas online with something like X-Wing at launch - the 3D graphics of the PC version would run amazingly well on the jag.

[Pete5125]

A single speed Jag CD would of been a joke, CD-i and 3D0 were double speed, Jag would of came out at $349-399 price point as a CD only unit less than the 3D0 and cdi, so it may of sold just as many, but would Atari of had a larger liabrary....I don't think it would of changed things much, 3D0 had a huge title list due to Electronic Arts having an exclusive deal and deep pockets to get other developers on board, same with CDi, Phillips spent tons to get games on their system, Atari was out of money....now if Atari could add a slot to pay Lynx games and that way come out the gate with a system that has 90+ Lynx games in its liabrary?

[mcjakeqcool]

@Pete5125

If the Jag CD had been single speed, EVERYBODY would be laughing, except Atari.

[Crazyace]

CD-i was single speed.

My CD jaguar would have been $299 ( Atari were very good at 'power without the price' ) ( That's why I think no 68k/Cart slot - so that costs can be cut to help offset the CD unit cost ) - I expect that CD software costs ( to Atari for manufacturing ) would be $5 to $10 cheaper than cart costs - so Atari would save costs when looking at Console+Game sales together, and also when looking at 'pack in' games. ( Atari could also have shipped with a 'sampler' or demo CD - saving royalties lost on a pack in game )

Not having EA sucked - but if the Jag had sold a million units they would have put something on it.

I dont think Lynx games on Jag would have been impressive - 160x102 pixel res wouldn't look that good on a big TV.

[Gorf]

CD-i was single speed.

My CD jaguar would have been $299 ( Atari were very good at 'power without the price' ) ( That's why I think no 68k/Cart slot - so that costs can be cut to help offset the CD unit cost ) - I expect that CD software costs ( to Atari for manufacturing ) would be $5 to $10 cheaper than cart costs - so Atari would save costs when looking at Console+Game sales together, and also when looking at 'pack in' games. ( Atari could also have shipped with a 'sampler' or demo CD - saving royalties lost on a pack in game )

Not having EA sucked - but if the Jag had sold a million units they would have put something on it.

I dont think Lynx games on Jag would have been impressive - 160x102 pixel res wouldn't look that good on a big TV.

 

Was never impressed with the Lynx connected to the Jag Idea. Nor did I think running 160x102 pixel 8 bit games on a 64 bit console would

be of much use either....perhaps later on when you have been sucessful with the console as a real 64 bit power house, but not initially.

 

First of all the 68k and the cart connector probably did not even add up to 25 bucks. Hardly an offset to a 100 dollar 2x drive.

 

250 - 25 = 225 +100 = 325....they would have taken a 25 dollar hit on each unit at $299. With enough hot selling software this

might have been worth it...but that was not the case.

[kool kitty89]

I liken the argument about the processor the jaguar should have used, to an argument i remember back in the day about the lynx and should the lynx have used a 68000 or a 6502

 

This came up in another recent thread:

 

I think they made Lynx tools for the Atari ST latter on during the systems life span. I do know for a fact that the Lynx sound system used a "Sound Programming Language or SPL." According to some documents I found some time ago it was a tool "by musicians for musicians" which was created to somehow take advantage of the Lynx unique sound system, which was very similar to the 2600 sound chip if I'm not mistaken, but a little better then the 2600. I've physically opened a Lynx 2 system and found that the CPU is a nice size chip; I wonder sometimes could a 68000 be put in there, but then that's just my own personal curiocities.

SPL is a very good language, but it lacks a few features (which I'm sure would have been implemented if hell froze over and Atari was successful) which would have helped musicians out. A 68000 would be detrimental to the Lynx's performance. It's a great chip, but it has no place in the Lynx hardware.

 

 

Gorf, a 1x speed drive on the Jag would be unaccepable? (even if this was only done on early models, with falling component costs allowing for a more compeditive 2x drive later on -obviously early models would be stuck with the slower drives inless the drive was fitted as a module, which would again add to cost)

 

Load times would be longer, but most video encoding at the time catored to 1x speed units (MPEG-1, Cinepak, ect), so that shouldn't be an issue. (and many smaller games, easily fitting on carts, possibly with added CD audio and streaming video, wouldn't have particularly long load times as the programs themselves would be small)

Edited June 15, 2009 by kool kitty89

[kool kitty89]

The CD32 came out around the same time - and a jaguar with CD would have blown it out of the water , and if I can 'whatif' I'd have loved Atari to get Lucas online with something like X-Wing at launch - the 3D graphics of the PC version would run amazingly well on the jag.

 

Yeah, that then the (more popular) Sequel, Tie Fighter in 1994. As well as a lot of other contemporary PC games should have been ported. (technically they probably could have compressed the floppy version to a practical size to fit on a cart. at least for X-Wing)

 

One thing the Jag's often critisized controller was goof for was PC ports, in particular games that required too many actions to do properly on most console controllers. (hence some crappy controlls on a couple flight sims and the like that ended up on some consoles, F-15 Strike Eagle II and Mig 29 on the Genesis are such examples) They screwed this up with Wolfenstein though (doom had the # keys for weapons). The re-mapped buttons on the Pro controllers are still preferable in many circumstances though. (in addition to the rest of the # keys)

 

Wing Commander would be a nice port, with a CD they could probably combine the first 2 games (plus expansions) onto one disc. Then youve got the very impressive (for the time and even still in some ways) Wing Commander III in 1994 (which did end up getting ported to the Playstation and even the 3DO -though I cant immagine gameplay was very smoothe on the latter)

The Jag should have been able to handle it, though it would have to be modified to work with the limitations of the system (mainly the RAM; the PC version requireng a pleintiful 8 MB), of course the PSX version (and 3DO in particular) would have been optimized to work on those systems as well. (with hardware corrections the memory limits and architecture change would really be the only issues, and with the well known architecture of the 020 that should help as well -perhaps being more attractive development wise than contemporary consoles in this respect)

 

It would be promoting both the 3D and streaming video capabilities of the console. (with the port of an unpresidentedly high budget title) Of course this didn't come out until 1994, so that would be a later considderation (and a Jag port probably wouldn't be out until some time in '95)

 

WC IV would be even later (also ported to PSX) and more demanding (same RAM wise though), but still probably one of the PC titels to go after.

 

Along with these, many popular FPS's should have been ported (these in particular would also work with the online/internet angle mentioned a while back), Duke Nukem 3D, adn Quake should be possible with the hardware improvements. (though in reality, a big reason for lacking such games was a lack of development support, and lack of good tools to use the existing hardware -on top of additional hardware flaws -of course with improving the flaws and programming difficulties should also improve intrest from developers -which I beleive had been initialy high in reality)

 

 

That said, upgraded versions of older console/computer/arcade games (the original 2 WC games included), with higher resolution and more colorful graphics, should have been possible even using the 68k in the current system, granted it's more difficult, especially putting some games on carts (those designed to run from CD or HDD), and newer game in this range might be pushing it with the 68k as well. (Wing Commander even had issues on the CD32, though probably in large part due to lack of optimization for the 68k architecture)

So even for the older games, the 020 (firstly, then other possible improvements like MMU, OP bugs, and blitter buffer) would be a big help to make such conversions look fresher and more appealing.

 

Of course a CD drive could add some of this as well, as CDDA, streaming Video intros/cutscenes, and conversions of some games that would be difficult port on cart, and of course the lower cost and lack of storage limits compared to carts would be appealing to developers and could offer a significantly higher profit margin for game sales for Atari as well. (especially where in-house titles are concerned) On top of lhis theres lower cost for distributors/retailers and, ultimately, consumers, as far as individual game pricing is concerned.

Edited June 16, 2009 by kool kitty89

[walter_J64bit]

:? :? just put a game in and play the Jag :? :?