I'm trying to track down a copy of PCW from November 1978. I am looking for an article on benchmarking BASIC. They introduced an 8th program to the 7-program Rugg/Feldman test and I'd like to read the article. Does anyone have a copy? I cannot find it online.
No online resource found, but to record for posterity, here are some relevant links:
The Centre for Computing History (Cambridge, UK) has a list of covers and brief summaries of contents. As you note in a comment, the Entry for November 1978 confirms that the article you are interested is in that edition.
A blog entry on the PCW Preservation Project here is an attempt to kick-start creating such an online repository. Although it does not seem to have got anywhere, there are comments from 2018/19; at least one from someone who has the relevant physical copy. It's conceivable that you may be able to get in touch with someone who could photocopy the article in question.
A side-entry notes that someone is/was (in 2010) selling a CD of issues June 1995 - March 1997, though neither the writer (nor I) is certain what it contains/contained.
The National Museum of Computing (Bletchley Park, UK) has a News Release in which it announces (emphasis mine):
The archive at TNMOC is available to bona fide researchers. It is growing rapidly already contains the entire 45 year print history (1966-2011) of Computer Weekly, the world's first weekly computer publication, 26 years of Personal Computer World, and many other magazines, the complete ICL manufacturing archive, the entire Digital Equipment Corporation microfiche and much more.
I have no idea of the feasibility, but again it is conceivable that they may be able to photocopy the relevant article.
Brian N Butterworth has compiled The PCW Index covering the magazine from its inception in 1978 until 1989. It contains:
'benchtests' (machine reviews), 'software' (software and 'firmware' reviews), 'checkouts'(reviews of peripherals such as add-in cards and printers). sections also cover 'features' (articles from programming to politics), 'screenplay' (computer games) and 'book reviews'. there are also sections covering the deceased 'tjs workshop' and 'computer answers'. finally, there are sections on 'subset' and 'programs'.
but virtually no capital letters. [Credit to Uli for finding this resource].
I have a full collection of 1978 and 1979 PCW magazines.
The Benchmark Tests are documented in Vol. 1 No. 1 under the article "Direct Addressing: Where to get your Personal computer". The November issue is simply a report of showing the results of benchmark times that were submitted from contributors that ran the programs listed in issue 1.
The following is a partial extraction from the text of the article....
The first seven benchmarks were used in a series of tests carried out in the States and published in an article in the June 1977 issue of Kilobaud. The eighth benchmark has been introduced to test the transcendental functions of the various interpreters. Unusually poor performance on this benchmark is a clear indication of the use of poor algorithms and is more a reflection on the programmer than on the machine.
Here's the BASIC code for the benchmark tests:
300 PRINT "S" 400 FOR K=1 To 1000 500 NEXT K 700 PRINT "E" 800 END
300 PRINT "S" 400 K=0 500 K=K+1 600 IF K < 1000 THEN 500 700 PRINT "E" 800 END
300 PRINT "S" 400 K=0 500 K=K+1 510 A=K/K*K+K-K 600 IF K<1000 THEN 500 700 PRINT "E" 800 END
300 PRINT "S" 400 K=0 500 K=K+1 510 A=K/2*3+4-5 600 IF K<1000 THEN 500 700 PRINT "E" 800 END
300 PRINT "S" 400 K=0 500 K=K+1 510 A=K/2"3+4-5 520 GOSUB 820 600 IF K<1000 THEN 500 700 PRINT "E" 800 END 820 RETURN
300 PRINT "S" 400 K=0 430 DIM M(5) 500 K=K+1 510 A=K/2"3+4-5 520 GOSUB 820 530 FOR L=1 TO 5 540 NEXT L 600 IF K<1000 THEN 500 700 PRINT "E" 800 END 820 RETURN
300 PRINT "S" 400 K=0 430 DIM M(5) 500 K=K+1 510 A=K/2*3+4-5 520 GOSUB 820 530 FOR L=1 TO 5 535 M(L)=A 540 NEXT L 600 IF K<1000 THEN 500 700 PRINT "E" 800 END 820 RETURN
300 PRINT "S" 400 K=0 500 K=K+1 530 A=K^2 540 B=LOG(K) 550 C=SIN(K) 600 IF K<100 THEN 500 700 PRINT "E" 800 END
I have the original 1978 copy volume 1 no 1.
As pointed out by Chromatix there is a typo in BM5 and BM6 which should be the same as BM4 and BM7.
I wrote a program in Liberty Basic to illustrate these benchmarks but changed the loops to K = 100000 so the speeds can be measured on modern pc's. They still run faster than the best machines used in 1978. Running them as originally written takes a few milliseconds in total.
k=100000 k = 1000
BM1 O.358 0.005
BM2 0.632 0.008 BM3 0.850 0.011 BM4 0.906 0.010 BM5 1.212 0.015 BM6 2.929 0.031 BM7 4.049 0.045 BM8 1.200 0.005 Total 12.117 0.121
I get slightly different numbers if I rerun the program These results are in the "compiled" program. The original code runs slightly faster. I had almost all the copies of PCWorld, but had to get rid of most of them when I moved some years ago. However I still have the first three volumes and a few other early ones 1978/1979 plus 10th, 20th, 25th, 30th anniversary copies and the last copy August 2009. The 25th anniversary copy includes the original Bonus double CD