C clang versus Lisp SBCL

How the programs are written matters! Always look at the source code.

If the fastest programs are flagged * possible hand-written vector instructions or "unsafe" or naked ffi, does the host language matter? You might be more interested in the less optimised programs — more seconds, less gz source code.

fannkuch-redux
	elapsed	Mean or 95% CI
source	secs	cpu secs	mem	gz	make
* C clang #6	2.25	8.88	2,941	1582	5.47
* C clang #4	13.38	13.39	2,351	1190	5.22
Lisp SBCL #5	30.88	30.86	58,196	689	2.50
C clang #5	8.13	32.00—32.15	4,764	917	5.37
C clang #3	34.18	34.19—34.26	2,355	574	5.02
* C clang #2	8.70	34.15—34.33	2,945	1563	5.27
Lisp SBCL #3	36.28	36.26	37,208	833	2.57
C clang #8	36.36	36.44—36.51	2,347	483	5.13
C clang	36.52	36.54—36.63	2,347	514	5.35
Lisp SBCL #4	9.41	36.96	38,060	1531	2.49
Lisp SBCL #2	48.05	48.03	27,795	525	2.47
n-body
	elapsed	Mean or 95% CI
source	secs	cpu secs	mem	gz	make
* C clang #9	2.20	2.20	2,417	1639	6.15
* C clang #5	4.86	4.86	2,470	1436	5.17
C clang #6	5.17	5.17	2,970	1186	5.42
C clang	5.38	5.39	2,978	1179	5.06
C clang #2	5.43	5.43	2,966	1270	5.25
* C clang #4	5.60	5.61	2,953	1398	6.30
C clang #8	5.70	5.56—6.19	2,961	1195	5.27
C clang #3	6.13	6.13	2,966	1214	5.26
C clang #7	6.78	6.80	2,966	1250	5.05
Lisp SBCL #2	7.83	7.82	27,808	1408	2.51
Lisp SBCL	9.28	9.27	27,554	1373	2.64
spectral-norm
	elapsed	Mean or 95% CI
source	secs	cpu secs	mem	gz	make
* C clang #6	0.39	1.55	4,686	1203	6.45
* C clang #4	0.73	2.87	5,382	1145	5.66
* C clang #5	0.73	2.87—2.91	5,403	576	5.55
* C clang #2	0.74	4.30—7.09	5,288	941	6.43
Lisp SBCL #6	5.36	5.35	29,557	512	2.47
Lisp SBCL	5.37	5.35	29,782	639	2.66
C clang #8	5.41	5.42	3,211	389	5.15
Lisp SBCL #9	1.48	5.72	28,799	799	2.56
Lisp SBCL #7	1.47	5.72	29,258	769	2.58
Lisp SBCL #2	1.48	5.72	29,274	920	2.55
Lisp SBCL #3	1.47	5.72	29,598	893	2.50
C clang #3	1.45	5.75—5.79	5,423	470	5.58
* C clang #7	1.84	8.98—10.12	5,439	906	6.66
mandelbrot
	elapsed	Mean or 95% CI
source	secs	cpu secs	mem	gz	make
C clang #9	1.24	4.92	36,811	700	5.32
C clang #5	1.54	6.10	37,044	788	5.35
* C clang #4	3.49	13.89—13.95	36,102	805	5.26
* C clang	3.51	13.95	35,643	828	5.65
* C clang #7	3.54	14.02—14.06	36,491	1000	5.49
* C clang #3	3.56	14.16—14.20	35,418	769	5.06
Lisp SBCL	4.53	16.56—16.61	60,678	2478	3.27
Lisp SBCL #4	4.20	16.68	60,654	2453	3.17
C clang #8	25.51	25.52	2,359	412	5.22
Lisp SBCL #3	9.82	26.17—26.21	60,191	900	2.59
* C clang #6		Empty
fasta
	elapsed	Mean or 95% CI
source	secs	cpu secs	mem	gz	make
C clang #3	0.79	0.79	2,273	1469	4.57
C clang #5	1.20	1.20	2,204	1286	5.64
C clang #2	2.44	2.44	2,163	1299	5.14
C clang #6	1.04	3.05	5,272	1936	5.40
C clang #7	0.83	3.28—3.32	5,861	2276	5.29
C clang	3.47	3.47	2,392	1213	5.28
Lisp SBCL #6	4.51	4.49	28,107	1763	2.75
Lisp SBCL #3	5.32	5.33	27,558	1592	2.59
C clang #8	8.08	8.14—8.25	2,200	839	5.08
Lisp SBCL #2	8.93	8.92	27,910	1628	2.52
Lisp SBCL	9.06	9.08	30,581	1430	2.65
C clang #4		Empty
k-nucleotide
	elapsed	Mean or 95% CI
source	secs	cpu secs	mem	gz	make
C clang	6.27	13.93—14.23	134,038	1512	6.01
Lisp SBCL #4	15.75	15.78—16.30	168,317	2292	2.80
Lisp SBCL #5	15.84	15.85—15.91	169,165	2323	2.88
Lisp SBCL #6	8.79	31.23—32.50	536,506	2485	2.61
Lisp SBCL	10.80	33.59	589,226	2431	2.54
Lisp SBCL #2	44.39	44.57—45.38	528,470	1291	2.57
Lisp SBCL #3	44.68	45.09—45.91	528,470	1297	2.60
reverse-complement
	elapsed	Mean or 95% CI
source	secs	cpu secs	mem	gz	make
* C clang #7	0.45	0.61—0.67	514,220	1899	6.22
C clang #9	0.47	0.72	514,216	1551	5.13
C clang #2	1.17	1.70	1,021,891	757	5.17
C clang #5	1.97	2.03—2.09	1,020,813	654	5.16
* C clang #4	3.77	3.78	514,023	716	4.93
C clang #6	1.05	4.15—4.20	714,764	828	5.90
Lisp SBCL	5.51	5.50	1,538,748	911	2.58
Lisp SBCL #2		Timed Out
binary-trees [ Contentious. Different approaches. ]
	elapsed	Mean or 95% CI
source	secs	cpu secs	mem	gz	make
C clang #2	1.66	5.23—6.05	176,116	816	5.77
C clang #3	1.91	5.90	138,719	842	5.90
Lisp SBCL #3	2.83	7.36—7.54	641,151	938	2.54
Lisp SBCL	9.60	9.62—9.68	665,092	595	2.51
C clang	17.72	17.78—18.29	271,938	660	5.20
C clang #5	9.25	33.42—33.59	361,689	914	4.96
pidigits [ Contentious. Different libraries. ]
	elapsed	Mean or 95% CI
source	secs	cpu secs	mem	gz	make
C clang #2	0.74	0.74	3,990	422	5.19
C clang	0.75	0.75	3,731	459	5.35
Lisp SBCL #4	3.37	3.39	817,934	495	2.62
Lisp SBCL #3	3.38	3.39	817,897	499	2.59
Lisp SBCL #2	6.00	6.01	836,059	655	2.49
regex-redux [ Contentious. Different libraries. ]
	elapsed	Mean or 95% CI
source	secs	cpu secs	mem	gz	make
C clang #5	0.83	2.82	159,736	1403	5.40
C clang #2	3.22	3.22	106,783	1121	5.07
C clang #4	0.99	3.82—3.86	159,539	1237	5.11
C clang #3	1.76	4.34	109,867	1273	5.08
Lisp SBCL #4	13.84	29.66—29.75	1,089,716	1977	3.35
C clang	Ubuntu clang version 18.1.8 (11)
Lisp SBCL	SBCL 2.4.8