cs4w+@andrew.cmu.edu (Charles William Swiger) (05/03/90)
From larrylro@gnh-applesauce.cts.com (Larry LRO): >If you have 120ns chips already installed on the board you probaly should come >as close as possible to the 120ns. The faster the chips the better the >performance. All the ram boards we sell come with 80ns drams installed and our >prices are the best. [ Are they? Glad to hear it. ;-) ] I'm afraid that putting faster RAM chips in a computer will do nothing to make it run faster. Or else I'd be in a long line of //gs owners who're buying 50ns (very fast, but very expensive) chips that would triple the speed of my computer that currently uses150ns both on the motherboard and on the RAM card. The speed of chip refers the time in nanoseconds that the chip requires to perform a memory access. You really don't need chips that run faster than the CPU (and/or the DMA controller) can possibly access them. From what I remember about the timing cycle for memory access in Apple //'s, you have 2/5's of the total time of one (CPU) clock cycle to do the memory access. Working that out shows that you need 143 ns chips to run at 2.8 MHz. Since the chips aren't rated to run quite that fast, ocasionally they'll be forced to cause a wait cycle, which gives them time do finish the access at the cost of taking an extra (CPU) clock cycle. This slows the computer down slightly, so that the apparent speed seems to be somewhere between 2.6 and 2.7 MHz. If you check the manuals, you'll find that Apple claims to run at 2.8MHz only when running in the ROM, for running in RAM they claim a speed of 2.6 MHz. Now then, if you buy 120ns chips, you're guarenteed to not see this problem. And, if you buy 150ns chips, all that means is that the chips work consistantly at 150ns but do not work consistantly at 120ns. There's a good chance that 150ns chips will be able to keep up with the 143ns access time required by the CPU. When I was buying memory, chips were expensive as hell, and the price difference between 150ns and 120ns was $2 a chip. ($64 per MB) Now, you can buy chips much more cheaply (that MB costs $110 instead of $400, and the difference between 150ns and 120ns chips is about $10 per MB). It's worth it to buy 120ns chips now, for the 5% or so speed increase. Of course, all bets are off when you are running a Transwarp or a high speed SCSI/DMA card. Then the faster access time of the 80ns chips will produce a very noticable increase in performance when compared to the slower 150ns chips. -- Charles William Swiger cs4w+@andrew.cmu.edu
gwyn@smoke.BRL.MIL (Doug Gwyn) (05/03/90)
In article <MaDwIcS00W05I5I1l_@andrew.cmu.edu> cs4w+@andrew.cmu.edu (Charles William Swiger) writes: >Of course, all bets are off when you are running a Transwarp or a high >speed SCSI/DMA card. Then the faster access time of the 80ns chips will >produce a very noticable increase in performance when compared to the >slower 150ns chips. The only improvement might be in reliability, certainly not memory access speed, which is controlled by the memory address logic and runs at the standard 2.8MHz rate regardless of the TWGS CPU clock.
toddpw@tybalt.caltech.edu (Todd P. Whitesel) (05/04/90)
cs4w+@andrew.cmu.edu (Charles William Swiger) writes: >The speed of chip refers the time in nanoseconds that the chip requires >to perform a memory access. You really don't need chips that run faster >than the CPU (and/or the DMA controller) can possibly access them. From >what I remember about the timing cycle for memory access in Apple //'s, >you have 2/5's of the total time of one (CPU) clock cycle to do the >memory access. Sorry to nitpick, but that's not true. You have a lot more time than that, more like 4/5. The Apple // bus originally ran slow enough that all memory could be enabled using much simpler logic, but that only left about 2/5 of the clock cycle (which in a 1 mhz system is still 400 ns -- plenty of time). > Working that out shows that you need 143 ns chips to run >at 2.8 MHz. Since the chips aren't rated to run quite that fast, >ocasionally they'll be forced to cause a wait cycle, which gives them >time do finish the access at the cost of taking an extra (CPU) clock >cycle. Running the 150 ns chips at 2.8 mhz is not the cause of the wait cycle. DRAMs use a much smaller memory cell design (essentially a capacitor with a switch) which makes them hold more memory but requires that they be "refreshed" every so often in order to ensure that no data is lost do to capacitor charge leaking (don't laugh, I've actually seen it happen in a project I built!). It is pauses for refresh that cause the wait cycles. When running out of ROM or at 1 mhz the DRAM is refreshed automatically so no speed penalty occurs. The REAL reason why we should be using 120 ns or better is that you can run 120 ns DRAM at 3.58 Mhz (150's can't, for a reason similar to your argument above) and 3.58 Mhz is much nicer to work with because it is much more simply related to the rest of the machine. You can also "page mode" the DRAMs at 7.1 Mhz (double 3.58 mhz), and for cache and DMA (lots of sequential memory accesses) this gets a cheap doubling in speed. Page Mode is not hard to do: the VGC chip in the GS already uses it. There are so many simple things that could be done to the GS to really make it sing. Todd Whitesel toddpw @ tybalt.caltech.edu
dseah@wpi.wpi.edu (David I Seah) (05/04/90)
In article <MaDwIcS00W05I5I1l_@andrew.cmu.edu> cs4w+@andrew.cmu.edu (Charles William Swiger) writes: >I'm afraid that putting faster RAM chips in a computer will do nothing >to make it run faster. Or else I'd be in a long line of //gs owners >who're buying 50ns (very fast, but very expensive) chips that would >triple the speed of my computer that currently uses150ns both on the >motherboard and on the RAM card. >[DRAM timing stuff deleted] >143ns access time required by the CPU. When I was buying memory, chips The CPU has a 143ns access time? I find this odd... Apple seems to enjoy putting higher-rated parts into their hardware for improved reliability (like the greater-than-1Mhz 6502 in the Apple IIc). >Of course, all bets are off when you are running a Transwarp or a high >speed SCSI/DMA card. Then the faster access time of the 80ns chips will >produce a very noticable increase in performance when compared to the >slower 150ns chips. Hmmm... The TWGS and SCSI/DMA board don't automatically detect how fast your DRAM chips are, do they? They must assume minimum speeds, I thought. The SCSI/DMA does its stuff at 1MHz, and I don't think the TWGS assume memory accesses any faster than a stock GS. -- Dave Seah | O M N I D Y N E S Y S T E M S - M | "Yargh, cats!" | User Friendly Killing Machines | .............................................................................. I-net: dseah@wpi.wpi.edu - America Online: AFC DaveS (Apple II Art & Graphics)
toddpw@tybalt.caltech.edu (Todd P. Whitesel) (05/04/90)
dseah@wpi.wpi.edu (David I Seah) writes: >Apple seems to enjoy putting higher-rated parts into their hardware >for improved reliability (like the greater-than-1Mhz 6502 in the Apple IIc). Not in this case. The 2 mhz 65C02 in the //e and //c is necessary, because the custom chips (IOU/MMU) are so godawful slow that a 1 mhz 65c02 doesn't give them enough time to decode the address. Todd Whitesel toddpw @ tybalt.caltech.edu
skann@gnh-applesauce.cts.com (Steve Kann) (05/06/90)
>There are so many simple things that can be done on the GS to really make it
sing.
Exactly the point that I've been trying to make for a long time!! The old Apple
// spirit is gone, (at least from s/w manufacturers). Instead of using newer
technology to make better products, they use advanced technology so that they
can write their s/w in a high-level language, and rely on the machine to make
up for their laziness. This just can not be done on the GS. There are just so
many neat little things it can do that are simply overlooked by manufacturers.
For instance: Fill-mode anumation; at the penalty of one color per line, we
could have simply dazzling animation, yet I've only seen one peice of software
that used it (tunnels of armageddon). Multiple pallettes; nobody really touches
these either.. rather just port over the 16/64 color graphics from EGA.. what a
waste!. And all the little points that I forgot to mention. Just neglected. We
are neglected not only by Apple, but also by our s/w companies. It really is a
shame.
Proline:skann@gnh-applesauce <--- Preferred
America online: NYSteve
CI$:70265,1141
************************
UUCP: !crash!pnet01!gnh-applesauce!skann This is just my opinion!
INET: skann@gnh-applesauce.cts.com ************************
ARPA: crash!pnet01!gnh-applesauce!skann@nosc.mil "Don't blame me!"
* Sent by AppleSauce BBS at Fri, 4 May 90 22:03:18lhaider@pro-sol.cts.com (Lawrence Haider) (05/07/90)
In-Reply-To: message from gwyn@smoke.BRL.MIL >The only improvement might be in reliability, certainly not memory >access speed, which is controlled by the memory address logic and >runs at the standard 2.8MHz rate regardless of the TWGS CPU clock. I've just upgraded from a AE GS-RAM, 1.5 MB using 150ns 256K DRAMs to an MDIdeas OctoRAM, 8 MB (Fully loaded) using 80ns 1 MB SIMMs. There isn't any noticable improvement in processing speed, but the displays ARE noticiably snappier. BTW, I'm also running a 7MHz TransWarp. lhaider@pro-sol.cts.com
gwyn@smoke.BRL.MIL (Doug Gwyn) (05/08/90)
In article <2545@crash.cts.com> lhaider@pro-sol.cts.com (Lawrence Haider) writes: >I've just upgraded from a AE GS-RAM, 1.5 MB using 150ns 256K DRAMs to >an MDIdeas OctoRAM, 8 MB (Fully loaded) using 80ns 1 MB SIMMs. There >isn't any noticable improvement in processing speed, but the displays >ARE noticiably snappier. BTW, I'm also running a 7MHz TransWarp. Well, with a different configuration there are reasons why things might seem faster, but if you were to switch to 8MB of 120ns SIMMs in the OctoRAM you should see no difference whatsoever compared to 8MB of 80ns SIMMs.