jqj@gvax.UUCP (04/13/87)
I just spent more than an hour wading through all the recent messages on this newsgroup about the new IBM product announcements. Noticeably lacking was any (much) content. Except for a couple of messages, I learned nothing that I didn't already know from reading the trade press. Please, does ANYONE who has played with these new machines have any hard performance data apart from the stupid Norton SI benchmark? In particular, can anyone confirm or deny the IBM claims that the new bus structure and disk interleave provide dramatic performance improvements over, say, a carefully tuned clone? If the IBM claims are correct, the new bus technology could be the one major advantage the new machines have over the current crop of clones.
akk2@ur-tut.UUCP (04/17/87)
In article <813@gvax.cs.cornell.edu> jqj@gvax.cs.cornell.edu (J Q Johnson) writes: >Please, does ANYONE who has played with these new machines have any hard >performance data apart from the stupid Norton SI benchmark? In particular, >can anyone confirm or deny the IBM claims that the new bus structure and >disk interleave provide dramatic performance improvements over, say, a >carefully tuned clone? If the IBM claims are correct, the new bus technology >could be the one major advantage the new machines have over the current crop >of clones. Our local IBM office is having a PC Fair on campus here and will be demonstrating the MODEL 30, MODEL 50, MODEL 60. I will be running some benchmarks (Dhrystones, Whetstone, PC Magazine series etc.) on them and will post the results here sometime at the end of next week. -- ----------------------- Atul Kacker UUCP: ...seismo!rochester!ur-tut!akk2
lowey@sask.UUCP (Kevin Lowey) (04/24/87)
Hi there, "jqj@gvax.UUCP" says: > Please, does ANYONE who has played with these new machines have any hard > performance data apart from the stupid Norton SI benchmark? The following is a table of "benchmarks" I used to compare some of the computers I have access to. (Sorry, but I have SI as well :-). Other than the Norton benchmark, all benchmarks were done using Turbo Pascal. The Sieve and Diskwrite benchmarks both called the system clock at the start and end of the programs, so the times should be accurate to 1/30 of a second. The compile benchmark was done with my wristwatch and cat like reflexes (by cat I mean Garfield). The SIEVE benchmark strictly tests CPU power. There is no disk use. It computes prime numbers. The COMPILE benchmarks compiles a 6800 line Turbo Pascal program. Most of the files came from include files, so the disk speed is important in this benchmark, but the CPU speed also contributes. This test was repeated for both floppy and hard disks. The DISKWRITE benchmark timed how long it took to write 160K to a disk, character by character (not block writes). This was repeated for floppy and hard disks. The SI benchmark is the result from the Norton Utility SYSINFO command. It is supposed to return the speed of the computer relative to the IBM-PC, but it greatly exaggerates results from the 80286 based machines. Because of the limited time I had to do this, the table is incomplete in places. ================ Fire Extinguisher (A grain of salt Dept.) =============== I realize that everything but SI and SIEVE test DISK DRIVE speed rather than CPU horsepower, so most of the benchmarks have to be taken with a grain of salt. Please, no flames about my benchmark methodology. I plan later to put together a full battery of tests of things such as screen speed, Floating arithmetic, Integer arithmetic, Memory Access, etc. If there is enough interest, I'll post some renewed benchmarks at my convienience (we are currently moving our offices so you will just have to wait!!). +----------------------------------------------------------------------------+ | Benchmark results for Various Microcomputers (All times in seconds) | |----------------------------------------------------------------------------| | | | Compile | Compile | Diskwrite | Diskwrite | SI | | System | Sieve | Hard | Floppy | Hard | Floppy | | |-----------------+-------+---------+---------+-----------+-----------+------| | IBM-XT(3.5"disk)| 28.78 | -- | 3:17.31 | -- | 73.43 | 1.0 | |-----------------+-------+---------+---------+-----------+-----------+------| | IBM-XT(5.25") | 28.34 | 2:20.67 | 3:15.37 | 40.75 | 72.17 | 1.0 | |-----------------+-------+---------+---------+-----------+-----------+------| | Zenith-158(v20) | 26.92 | 2:18.68 | 3:12.94 | 41.30 | 71.79 | 1.7 | |-----------------+-------+---------+---------+-----------+-----------+------| | IBM Model 30 | 10.49 | 1:03.85 | 2:06.41 | 17.58 | 72.72 | 1.9 | |-----------------+-------+---------+---------+-----------+-----------+------| | IBM AT (6MHz) | 9.66 | 55.81 | 1:48.17 | 18.24 | 60.14 | 5.7 | |-----------------+-------+---------+---------+-----------+-----------+------| | IBM XT286 | 7.03 | -- | -- | 12.53 | 60.09 | 7.1 | |-----------------+-------+---------+---------+-----------+-----------+------| | DEC VAXmate | 7.00 | 43.84 | 1:48.93 | 11.89 | 72.61 | 7.7 | |-----------------+-------+---------+---------+-----------+-----------+------| | IBM Model 50 | 5.55 | 37.98 | 1:44.66 | 13.45 | 72.61 | 10.3 | |-----------------+-------+---------+---------+-----------+-----------+------| | IBM Model 60 | 5.55 | 33.35 | 1:44.43 | 11.36 | 72.61 | 10.3 | |-----------------+-------+---------+---------+-----------+-----------+------| | Zenith-248 | 4.99 | 33.17 | 1:48.74 | 12.19 | 60.36 | 9.2 | +----------------------------------------------------------------------------+ I guess "the quality goes in before the name goes on", Right IBM? (Sorry, I couldn't resist) ______________________________________________________________________________ | Kevin Lowey |The above is the personal opinion of Kevin | | University of Saskatchewan |Lowey. It does not reflect the position of| | Computing Services |the University of Saskatchewan in any way. | | SaskTel: (306) 966-4826 | | | Bitnet:LOWEY@SASK. (preferred) |I am in no way affiliated with any of the | | UUCP: ihnp4!sask!lowey.uucp |above mentioned companies other than U of S| |________________________________|___________________________________________|