Crazyace, on Fri Dec 23, 2011 2:52 PM, said:
Looking at it more and more it is much more pleasing than the seperate toilet seat addon - If Jaguar had launched like this I think Atari's fortunes may have been slightly different

- Have you got many more pictures , or a video, 10p6?
I'm less convinced about 52MHz though - it would need much faster memory and generate a lot more heat. ( What was Freeze like, it doesn't look like something that a standard jaguar would need overclocking for )
Yeah, and, again, I find it a lot more appealing (aesthetically) than the actual Jag Duo mock-up.
http://media.photobu...R/JagDUO202.jpg
10p6, on Fri Dec 23, 2011 2:49 AM, said:
Memory prices in 1992/1993 would have added about $40 t0 $60 (Less in Bulk) to the cost to double the DRAM, however this with a 52 MHZ (stock TOM and JERRY do work at that speed) would have more than offset the cost of the 68000 and related hardware. Even more so, with the 68000 gone, all the CD electronics could have been placed on the same JAG PCB and saved even more money by removing the cost of the CD PCB.
Yes, RAM would have been a major expense, and as it was, RAM was a huge chunk of the total component cost of the system. (also remember that raw manufacturing cost will be inflated substantially by the time you make retail, so small price additions can mean significant increases in SRP -let alone substantial cost additions like a CD drive or double the RAM)
Plus (in hindsight), with the CD, drive prices fell rapidly in the mid 90s, but RAM prices (especially from the standpoint of bulk, low-end DRAM) stagnated around $3/Mbit from 1992 through early 1996. (actually the average prices increased somewhat in that time -prior to the considerable drop in mid/late '96)
Albeit adding a single 512k or 128k 16-bit DRAM chip in the 2nd DRAM bank might have been more realistic. (and, more than just adding memory, it would be extremely useful for texture mapping, since -with the Jaguar's dual-bank interleaving- source and destination pages could be held open for a maximum rate of 4 cycles per textel rather than 13 -in the 37.5 MHz example- )
As for clock speeds, yes, the existing chips may handle it (at least in some systems), but it would take much more testing to know for sure what sort of yields they were getting.
If all production models are indeed stable at such clock speeds, it's rather odd that they'd underclock them so much, unless maybe it was due to the speeds used on preproduction units. (RAM would be an issue too, and while 52 MHz would have been to much for the existing DRAM used - 52 MHz would need 38 ns PC timing- that 80 ns Toshiba DRAM used is rated for 50 ns minimum page cycle timing, so 37.5 MHz would seem the ideal clock rate to use -allowing a 3-cycle long pulse to provide 80 ns RAC timing, 2 cycles allowing 53.3 ns for page-mode reads/writes, and 6 cycles -160 ns- for random reads/writes -this DRAM has 150 ns minimum RC time)
TOM's DRAM controller is locked at 2 cycles for page-mode cycle time, but other parameters are programmable. (including RAS, RAC, etc)
52 MHz (38 ns PC time -2 cycles) should be possible with FPM DRAM, but they'd need to bump it up a few speed grades. (most 50 ns FPM DRAM seems to be rated at 35 ns PC timing, but with a 52 MHz clock, you'd be underclocking that a bit at 57.7 ns RAC, so PC timing might not remain fast enough in that case -EDO DRAM would easily handle it though, but I'm not sure what sort of modifications the system would need to support that)
Quote
Also with the 68000 gone we would have seen less crap game ports.
From that standpoint, you also may have simply seen fewer games in general . . . though if the bugs were fixed on the RISCs (or at least 1 of them), that shouldn't have been an issue in any case. (source-ports of some older games would have been more difficult, but actual conversions of many newer games would probably be much easier -since you'd have straightforward programming of much more powerful processors, necessary for many newer games -plus, PC games or SNES games wouldn't cater to 68k source ports anyway -and PC ports probably should have been a major effort for the Jaguar, especially for 3D/pseudo 3D stuff -like Wing Commander I and II, Comanche, X-Wing, Tie Fighter, Wacky Wheels, Blake Stone, Duke 3D, -Wolf 3D and Doom already there, and some 2D games like Jazz Jackrabbit and various adventure games -several of those games would effectively have been exclusives since no other consoles featured them)
Quote
From day one I think Atari should have fixed the Chip design flaws (Two different fabrications of V 1.0 and they did not remove the flaw is BS), as for using the 68000 for a system compiler I would not bother with. Instead I would have included a larger system ROM with more features, and I would have made a version of the Falcon with Tom, Jerry and the CD built in to make game development much easier. If Atari would have shared programming tips between developers too that would have helped.
Yes, there's a lot of things they should have done under better circumstances, but a lot of what they did do was out of desperation (like the whole 1993 test market, mainly to drum-up desperately needed investor interest)
Ongoing management issues didn't help (and were exacerbated by the financial state too), but the major issue was Atari's overall situation/stability by 1993. (due to the steady downward spiral from ~1988/1989 onward -in both their computer and game markets, from the consumer standpoint and in terms of internal management and technology)
Had the computers and/or console business continued to do well and expand from ~'88 into the early 90s, Atari could have been in a much better market position with much better funding (and credit -and PR) by 1993. (and that added funding might also have afforded smoother development of the Jaguar in the first place, but at very least afforded the necessary time to address the major bugs -plus, established products on the market already would mean less need to rush out a new system as they did with the Jag)
And, yes, if they were still producing computers as such, it would indeed have been interesting to have newer computers including the Jaguar chipset (or part of it). More so if they'd kept the computer architectures a bit simpler, meaning less redundant hardware for compatibility. (like if they'd focused mainly on faster CPUs, simple DMA sound, and improved bitmap graphics modes -higher color depth, scrolling, chunky pixels, etc)
Edited by kool kitty89, Fri Dec 23, 2011 5:29 PM.