gibson@unc.UUCP (Bill Gibson) (07/17/86)
>In article <737@Navajo.ARPA> rokicki@Navajo.UUCP writes: (CardCo board:) >> >> 512K machine, 1.1 988 dhrystones >> 1536K machine, 1.1 674 dhrystones >> Loss 31.8% >> 512K machine, 1.2 beta II 983 dhrystones >> 1536K machine, 1.2 beta II 664 dhrystones >> Loss 32.5% >> In article <435@oscvax.UUCP> rico@oscvax.UUCP (Rico Mariani) writes: (Comspec board:) > > 512K machine, 1.1 975.04 dhrys/sec = 51.28 secs for 50000 cycles > 2.5M machine, 1.1 967.49 dhrys/sec = 51.68 secs for 50000 cycles > Comparison: 99.23% of top speed = 0.77% Loss > > 512K machine, 1.2 Beta II 976.56 dhrys/sec = 51.20 secs for 50000 cycles > 2.5M machine, 1.2 Beta II 963.58 dhrys/sec = 51.89 secs for 50000 cycles > Comparison: 98.67% of top speed = 1.33% Loss Were these tests run with the benchmarks *located* in the extra memory, or were they just run with the extra memory attached? Of course, I would assume the memory was being used, but I would like to see this explicitly stated. I don't know how the Amiga's wait states are generated; is timing the same for all memory, or can one have slow expander memory and fast factory-installed memory? Bill Gibson gibson@unc ...[akgua,decvax,philabs]!mcnc!unc!gibson
rico@oscvax.UUCP (07/20/86)
In article <25@unc.unc.UUCP> gibson@unc.UUCP writes: >>In article <737@Navajo.ARPA> rokicki@Navajo.UUCP writes: > (CardCo board:) >>> >>> 512K machine, 1.1 988 dhrystones >>> 1536K machine, 1.1 674 dhrystones >>> Loss 31.8% >>> 512K machine, 1.2 beta II 983 dhrystones >>> 1536K machine, 1.2 beta II 664 dhrystones >>> Loss 32.5% >>> >In article <435@oscvax.UUCP> rico@oscvax.UUCP (Rico Mariani) writes: > (Comspec board:) >> >> 512K machine, 1.1 975.04 dhrys/sec = 51.28 secs for 50000 cycles >> 2.5M machine, 1.1 967.49 dhrys/sec = 51.68 secs for 50000 cycles >> Comparison: 99.23% of top speed = 0.77% Loss >> >> 512K machine, 1.2 Beta II 976.56 dhrys/sec = 51.20 secs for 50000 cycles >> 2.5M machine, 1.2 Beta II 963.58 dhrys/sec = 51.89 secs for 50000 cycles >> Comparison: 98.67% of top speed = 1.33% Loss > >Were these tests run with the benchmarks *located* in the extra memory, >or were they just run with the extra memory attached? Of course, I would >assume the memory was being used, but I would like to see this explicitly >stated. > I can't speak for the CARDCO tests, but the tests I did were with the benchmark program actually running in the "fast" ram. That's all -Rico ...{ihnp4|allegra|decvax|watmath|linus}!utzoo!oscvax!rico
rokicki@navajo.UUCP (07/20/86)
[-] Postnews Window ===================================================[=][=] > >Were these tests run with the benchmarks *located* in the extra memory, > >or were they just run with the extra memory attached? Of course, I would > >assume the memory was being used, but I would like to see this explicitly > >stated. > I can't speak for the CARDCO tests, but the tests I did were with the > benchmark program actually running in the "fast" ram. In the CardCo tests, they were run with the benchmarks located in the fast RAM. As soon as I get to my office with my soldering iron, I'll let you know how fast a fixed CardCo board runs. -tom
rokicki@navajo.UUCP (07/21/86)
CardCo, with strapping problem: >> >> 512K machine, 1.1 988 dhrystones >> 1536K machine, 1.1 674 dhrystones >> Loss 31.8% >> 512K machine, 1.2 beta II 983 dhrystones >> 1536K machine, 1.2 beta II 664 dhrystones >> Loss 32.5% >> Comspec board: > 512K machine, 1.1 975.04 dhrys/sec = 51.28 secs for 50000 cycles > 2.5M machine, 1.1 967.49 dhrys/sec = 51.68 secs for 50000 cycles > Comparison: 99.23% of top speed = 0.77% Loss > > 512K machine, 1.2 Beta II 976.56 dhrys/sec = 51.20 secs for 50000 cycles > 2.5M machine, 1.2 Beta II 963.58 dhrys/sec = 51.89 secs for 50000 cycles > Comparison: 98.67% of top speed = 1.33% Loss I just fixed my CardCo board (my soldering iron is still hot), and these are the new updated timings: 512K machine, 1.1 988 dhrystones 1.5M machine, 1.1 938 dhrystones 94.9% of top speed = 5.1% loss 512K machine, 1.2 Beta II 983 dhrystones 1.5M machine, 1.2 Beta II 933 dhrystones 94.9% of top speed = 5.1% loss So, it's taking about 5% away; sounds like refresh delays. I suspect that this 5% will come back with interest on normal applications which access the disk and therefore use the coprocessors, though. I also decided to run the Mandlebrot set and see what improvement I noticed there. Here are the timings for mse (from the Fish disks) on a 640x200x4 screen, when the reset both window and region menu option is selected, for one scan of the window: 512K machine, 1.1 56m 4s 1.5M machine, 1.1 31m13s Speedup: 1.796 (79.6%) Not bad! -tom