Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.sys.transputer > #68 > unrolled thread

IMSB427-17 vs. IMS B428-12

Back to article view | Back to comp.sys.transputer

Contents

  IMSB427-17 vs. IMS B428-12 "Mike B." <michael_bruestle@yahoo.com> - 2012-04-22 01:26 -0700
    Re: IMSB427-17 vs. IMS B428-12 "Mike B." <michael_bruestle@yahoo.com> - 2012-04-22 01:44 -0700

#68 — IMSB427-17 vs. IMS B428-12

I've done a simple compare between two TRAMs:

* IMSB427-17 (T805d-30, 4K,1+8192K,3)
* IMSB428-12 (T801b-25, 4K,1+2048K,2)

The CPU ratio is 1.20
The MEM ratio is 0.80

Test 1: icc      savage /ta
Test 2: ilink    savage.tco /f startup.lnk /ta
Test 3: icollect savage.lku /t
Test 4: iserver  /sb savage.btl

The /ta has been selected on purpose to compare mainly the CPU and not the FPU, where the T805-30 will be the obvious winner.

icc       savage /ta	2.492	2.440	1.02
ilink11.405	10.766	1.06
icollect 2.214	2.152	1.03
iserver  8.961	9.310	0.96

[toc] | [next] | [standalone]

#69

Sorry for the incomplete post - I know - a web browser ist not a wise editor!

I've measured the overall runtime, including the boot time.

		  B427	  B428	ratio
icc		 2.492	 2.440	1.02
ilink		11.405	10.766	1.06
icollect	 2.214	 2.152	1.03
iserver 	 8.961	 9.310	0.96

Outcome: For me the B427 is the winner. The B428 is slightly faster in memory intensive tests but the fourfold memory size of the B427 is always smashing.

Nevertheless - memory references are far away from imperceptible, even in 1992.

-Mike

[toc] | [prev] | [standalone]

Back to top | Article view | comp.sys.transputer

csiph-web