jhall@amc.UUCP (John Hall) (02/20/89)
OK...I have a few questions about the much ballyhooed A2500UX. (I do want one, really!) 1. Any idea when it is to be released? (I want it NAAAAAAOW!) 2. Will there be a faster A2620 style card available? 14Mhz seems rediculous. We are currently working on a 33Mhz 68020 emulator, probably extendable to 60Mhz! 3. Why have a 68000 in the case at all? I want to run UNIX, although I do still want to run Amiga software. As a side issue: 4. Can the A2620 run standalone (except for I/O?) It would be great to have the 68000 running I/O, graphics, and other I/O while the 68020 does the general processing. Does the A2620 architecture allow this? Does the CPU slot allow multi-master bus control? (I have an itch about this one! I feel silly paying for a 68000 that just heats the inside of the computer. Even if the part is cheap, so much potential wasted! Sinful!) 5. What graphics support will be available initially? None? Libraries available to developers only? Non-(any) standard libraries? Text only? (God forbid!) 6. And only slightly related...Anyone thinking of making a bridgeboard only, with no CPU on it? The CAM and imbedded systems markets have embraced the PC OS (What do you mean doctor? Oh my god! I have MSDOS!) and hardware. Without trying I could dig up at least 30 computer on a card systems in almost any configuration dreamed about. It would be nice to put my own style of system together instead of being forced to buy a PC or AT with no real enhancements. It would also sell better I think, (Yes I know about the limited number of slots, but the amiga itself should provide much of the functionality the PC cards would normally supply) Maybe I am just strange, but the idea of a VGA, 25MHz 80386 AND a 25Mhz 68020 with MMU and Amiga graphics, AND the 68000 to keep the conversation between CPU's and I/O going, in a box that is affordable to REAL people, MAKES ME HOT!!!! Ah well, maybe I've put a bee in someones' bonnet! Happy daze to all....
daveh@cbmvax.UUCP (Dave Haynie) (02/22/89)
in article <839@amc.UUCP>, jhall@amc.UUCP (John Hall) says: > Keywords: Amiga 2500, 2500UX, UNIX, A2620, bridgeboard > Summary: Why have the 68000 if I want to run on the 68020 always? > OK...I have a few questions about the much ballyhooed A2500UX. > 1. Any idea when it is to be released? (I want it NAAAAAAOW!) It's been out since late December or so. > 2. Will there be a faster A2620 style card available? > 14Mhz seems rediculous. We are currently working on a > 33Mhz 68020 emulator, probably extendable to 60Mhz! Hey, Motorola's having a hard enough time building a 40MHz 68020. And we'd have a long wait for the 40MHz MMU. The CPU speed is the least of the issues, really. Priced 60ns DRAM lately? It doesn't make sense to build a faster unit until you can give it something to talk to. If enough folks are willing to pay 2x-3x what the A2620 costs now, maybe we can talk.... > 3. Why have a 68000 in the case at all? I want to run UNIX, > although I do still want to run Amiga software. As a side > issue: Why not have a 68000 in the case. They're about as cheap as sand and plastic. If you ask nicely, I'll tell you how to run the A2620 with the 68000 removed from the box l^) > 4. Can the A2620 run standalone (except for I/O?) It would be > great to have the 68000 running I/O, graphics, and other I/O > while the 68020 does the general processing. Does the A2620 > architecture allow this? No. > Does the CPU slot allow multi-master bus control? Yes. Though in order for it to work, you'd need some clever software allocation of interrupts -- you certainly wouldn't want both CPUs responding to each interrupt. The main reason we decided not to make the A2620 run in parallel with the 68000 on the motherboard was that it's not something that we could support on future 32 bit Amigas. You don't want to see the 68000 a required part of every system, now, do you? > (I have an itch about this one! I feel silly paying for a 68000 that > just heats the inside of the computer. Even if the part is cheap, so > much potential wasted! Sinful!) The 68000 is a great bargain if you're a developer! For the price of a 68000, you can save yourself the cost of an A500 for compatability testing, and still have the fast machine you want for development. For UNIX, it's a loss, but it really is cheap, probably much cheaper than you ever imagined. > 6. And only slightly related...Anyone thinking of making a > bridgeboard only, with no CPU on it? Actually, I've heard of three such projects at various times, but haven't seen one actually materialize. I think it's a good idea. > Maybe I am just strange, but > the idea of a VGA, 25MHz 80386 AND a 25Mhz 68020 with MMU and Amiga > graphics, AND the 68000 to keep the conversation between CPU's > and I/O going, in a box that is affordable to REAL people, > MAKES ME HOT!!!! Hmmmm... The '386 seems to me to be the computer hardware equivalent of a cold shower. But there are cool things on AT cards that'd be nice to get at from the 68020 -- Targa boards, SPARC, 29K, and 88K coprocessors, Weitek floating point boards, etc. Such neat stuff to put on such a stupid bus. > > Ah well, maybe I've put a bee in someones' bonnet! > > Happy daze to all.... -- Dave Haynie "The 32 Bit Guy" Commodore-Amiga "The Crew That Never Rests" {uunet|pyramid|rutgers}!cbmvax!daveh PLINK: D-DAVE H BIX: hazy Amiga -- It's not just a job, it's an obsession
shs@uts.amdahl.com (Steve Schoettler) (02/22/89)
In article <6030@cbmvax.UUCP> daveh@cbmvax.UUCP (Dave Haynie) writes: >in article <839@amc.UUCP>, jhall@amc.UUCP (John Hall) says: > >> 2. Will there be a faster A2620 style card available? >> 14Mhz seems rediculous. We are currently working on a >> 33Mhz 68020 emulator, probably extendable to 60Mhz! > >The CPU speed is the least of the issues, really. Priced 60ns DRAM lately? >It doesn't make sense to build a faster unit until you can give it something >to talk to. If enough folks are willing to pay 2x-3x what the A2620 costs >now, maybe we can talk.... 60ns RAM is not really necessary. There are 20MHz 386 boards that run nearly full speed with 120ns DRAM's. This may not be a fair comparison since I don't know how many cycles/memory reference a 386 does, but... Suppose you could get a 90% hit rate with a fast 128K SRAM cache, then: 90% of your program would run at full processor speed, and 10% would run at half speed, so overall the program would run at about 91% as fast as if the machine used all fast DRAM, and the RAM cost would be much lower. This also would cut bus traffic by a factor of about 10 (minus protocol/shared memory overhead), so if you want to run multiprocessors, this is the only way to go. Of course, you have the added development effort of designing a cache controller and the extra cost of those chips, but cache controllers are well understood these days, and general purpose controllers are starting to appear on the market. Steve -- "The cache guy" -- Steve Schoettler shs@uts.amdahl.com {sun,decwrl,pyramid,ames,uunet}!amdahl!shs Amdahl Corp., M/S 213, 1250 E. Arques Ave, Sunnyvale, CA 94088
daveh@cbmvax.UUCP (Dave Haynie) (02/23/89)
in article <0130z74c1z1010p0ANE@amdahl.uts.amdahl.com>, shs@uts.amdahl.com (Steve Schoettler) says: > In article <6030@cbmvax.UUCP> daveh@cbmvax.UUCP (Dave Haynie) writes: >>in article <839@amc.UUCP>, jhall@amc.UUCP (John Hall) says: >> >>> 2. Will there be a faster A2620 style card available? >>> 14Mhz seems rediculous. We are currently working on a >>> 33Mhz 68020 emulator, probably extendable to 60Mhz! >>The CPU speed is the least of the issues, really. Priced 60ns DRAM lately? >>It doesn't make sense to build a faster unit until you can give it something >>to talk to. If enough folks are willing to pay 2x-3x what the A2620 costs >>now, maybe we can talk.... > 60ns RAM is not really necessary. 60ns is just about right for a full speed 68020. A 25MHz 68020 runs a 120ns cycle. Take 60ns for RAS access, 40ns for RAS precharge, and you have 20ns left over for decoding and refresh arbitration. Note that a 25MHz 68030 runs an 80ns cycle. But anyway, let's assume we've got a 25MHz A2620, which will have an extra wait state for the MMU. That's now a 160ns cycle. With 80ns DRAMs, you get 80ns for RAS access, 60ns for RAS precharge. Again, that leaves just 20ns for decoding and refresh arbitration, which is currently about 1/2 the time it actually takes on the A2620. Assuming I could knock 10ns off the select/arbitration time, I'd still need 70ns DRAMs for a zero wait state system (excluding the MMU wait state, of course). You could get similar performance, most of the time, with 80ns or 100ns going to a RAM bank interleave (lets you hide the precharge time), but that doubles the minimum number of devices you need for a complete system, and adds quite a bit more RAM support logic. No problem to do, and there are even better "clever" memory designs than this, but don't expect the whole thing to fit on an A2620 board without going to much higher integration than PALs to build the thing. > There are 20MHz 386 boards that run nearly full speed with 120ns DRAM's. > This may not be a fair comparison since I don't know how many cycles/memory > reference a 386 does, but... The 386 boards have two advantages. First of all, they almost all use either a bank interleave or a page mode/static column memory design. Using static column memory, for example, I could build a 25MHz '030 system that runs most of the time without wait states using 80ns or possibly even 100ns DRAM. The complexity of the design goes way up, though. And that takes room. The second thing the '386 benefits from, memory speed wise, is a special trick it does to ease memory timing. Rather than working on caches, the Intel folks decided that a deep pipeline was the speedup they wanted on the '386 (Moto went the other way, obviously, but no matter -- they converge next generation). So once the pipeline is full, the '386 drops into a mode that lets it run a slower cycle while still keeping the pipeline full. This is undoubtedly why '386s take about 80%-90% of the bus bandwidth -- they reorder their memory accesses via the pipeline, resulting in fewer cycles with no memory access, and a slightly relaxed timing on each access. > Suppose you could get a 90% hit rate with a fast 128K SRAM cache, then: I have no problem with external caches. The A2620 would have been really nice with a 64K or 128K logical bus cache. Any fool can tell you that. Where would I put it? Add the cache, and you loose all 32 bit memory. Plain and simple. Also given that the current operating system doesn't properly support data caches, and would likely run into trouble with large instruction caches, the choice of a decent amount of 32 bit memory over a large external cache was the only choice. > This also would cut bus traffic by a factor of about 10 (minus protocol/shared > memory overhead), so if you want to run multiprocessors, this is the > only way to go. No kidding. Common knowledge. Though without bus snooping, you loose a great deal of the benefit of caching in a multiple CPU system. > Of course, you have the added development effort of designing a cache > controller and the extra cost of those chips, but cache controllers > are well understood these days, and general purpose controllers are starting > to appear on the market. Adding a cache would have been no more trouble than adding the 32 bit memory. Considering all the support chips available today, it would have likely been simpler. Possibly more expensive, but likely simpler. > Steve -- "The cache guy" -- Dave Haynie "The 32 Bit Guy" Commodore-Amiga "The Crew That Never Rests" {uunet|pyramid|rutgers}!cbmvax!daveh PLINK: D-DAVE H BIX: hazy Amiga -- It's not just a job, it's an obsession
jhall@amc.UUCP (John Hall) (02/24/89)
Thanks Dave H. for your reasonable response, I fear my initial posting was a bit intense, but I had just finished arguing with a local Amiga dealer about the existence, availability, pricing, and possible future of the A2500UX. I do agree with you about the PC bus. Also, thanks Steve S. for your moral support. Now for a few more questions: 1. Does the current Amiga 2000 (or 2500) floppy drive controller support a 1.44MB 3.5" floppy? (If they are standard, forgive me.) 2. Will the 2500UX have a tape drive available? What kind? Media size? 3. Whose 80MB hard drive is supplied with the 2500UX? Speed? 4. What packages are supplied with the 2500UX and what will be available? a. Whose C compiler? b. What object file format? c. What kind of graphics, sound, and other device support will it have? d. Any Networking support? TCP? UUCP (whose)? e. Which "standard" (if any) will the windowing library support? f. Thats enough for now... 5. Who should I have my dealer call to get 2500UX info? 6. Finally, could you give me some pricing info? For the base package, and for any options (unix related). Ballpark figures will do. Thanks much, PS: Why don't you email the info to me, if anyone else wants it, I'll forward. ------------------------------------------------------------------------------- | John "Down Ball" Hall | Volleyball isn't just a sport... | Bump, Set, | | Applied Microsystems Corp. | ...it's a religion! | SPIKE! | | PO Box 97002 |------------------------------------------------| | Redmond, WA 98073-9702 | Disclamer: "Did I say that??? My mind must | | Voice: (206) 882-2000 x654 | be out to lunch. I hate when that happens!" | | UUCP: uunet!amc!jhall | Ollie's Disclaimer: "I don't recall..." | -------------------------------------------------------------------------------
jesup@cbmvax.UUCP (Randell Jesup) (02/27/89)
In article <842@amc.UUCP> jhall@amc.UUCP (John Hall) writes: >1. Does the current Amiga 2000 (or 2500) floppy drive controller > support a 1.44MB 3.5" floppy? (If they are standard, forgive me.) No. The drives run at a higher data rate (They're really 2MB unformatted; current Amiga drives and 760K IBM drives are 1 MB unformatted). The current Paula cannot handle 2MB floppies (unless you run them at 150 RPM :-( ). >2. Will the 2500UX have a tape drive available? What kind? Media size? Yes. I don't know what kind. -- Randell Jesup, Commodore Engineering {uunet|rutgers|allegra}!cbmvax!jesup
bill@cbmvax.UUCP (Bill Koester CATS) (03/01/89)
In article <842@amc.UUCP> jhall@amc.UUCP (John Hall) writes: >Now for a few more questions: > >2. Will the 2500UX have a tape drive available? What kind? Media size? It is a 150MB tape backup, don't know the manufacturer. >3. Whose 80MB hard drive is supplied with the 2500UX? Speed? Quantum 80MB, Fast 16-18ms I think? >4. What packages are supplied with the 2500UX and what will be available? > a. Whose C compiler? AMIX is standard System V Release 3 > b. What object file format? > c. What kind of graphics, sound, and other device support will it have? The system comes stock with support for a few graphics terminals. Many packages support the Tek 4014 terminal so an amiga with ATalkIII makes a nice terminal. Comes with GED a graphical editor that works with GPS (Device independent files) and lets you edit graphics on the 4014. > d. Any Networking support? TCP? UUCP (whose)? Honey Dan Ber UUCP. > f. Thats enough for now...