One more little benchmark, this time doing a variety of primitives.
⎕IO←0
∇
[0] TIME X;TS
[1] TS←⎕TS
[2] ⍎X
[3] (⍕(24 60 60 1000⊥¯4↑⎕TS-TS)÷1000),' Seconds.'
∇
∇
[0] Z←SHOW X;I
[1] Z←(I,50)⍴(50×I←⌈(⍴X)÷50)↑X←,X
∇
∇
[0] Z←X TIMES Y;D;I;C
[1] Z←+⌿(-⍳⍴X)⌽Z←((⍴X),(⍴X)+⍴Y)↑Z←(D⍳X←⌽,X)∘.×(D←'0123456789')⍳Y←⌽,Y
[2] I←¯1◊C←0
[3] L1:→((I←I+1)⍴Z)/L2
[4] C←⌊(÷10)×Z[I]←Z[I]+C
[5] Z[I]←10|Z[I]
[6] →L1
[7] L2:Z←((¯1+⍴Z)⌊+/∧\'0'=Z)↓Z←D[⌽Z]
∇
∇
[0] Z←FACTORIAL N;I
[1] Z←,'1'◊I←1
[2] L1:→((I←I+1)>N)/0
[3] Z←Z TIMES⍕I
[4] →L1
∇
[IBM APL2]
TIME 'SHOW FACTORIAL 300'
30605751221644063603537046129726862938858880417357
69994167767412594765331767168674655152914224775733
49939147888701726368864263907759003154226842927906
97455984122547693027195460400801221577625217685425
59653569035067887252643218962642993652045764488303
88909753943489625436053225980776521270822437639449
12012867867536830571229368194364995646049816645022
77165001851765464693401122260347297240663332585835
06870150169794168850353752137554910289126407157154
83028228493795263658014523523315693648223343679925
45940952768206080622328123873838808170496000000000
00000000000000000000000000000000000000000000000000
000000000000000
2.078 Seconds.
[GNU APL]
TIME 'SHOW FACTORIAL 300'
30605751221644063603537046129726862938858880417357
69994167767412594765331767168674655152914224775733
49939147888701726368864263907759003154226842927906
97455984122547693027195460400801221577625217685425
59653569035067887252643218962642993652045764488303
88909753943489625436053225980776521270822437639449
12012867867536830571229368194364995646049816645022
77165001851765464693401122260347297240663332585835
06870150169794168850353752137554910289126407157154
83028228493795263658014523523315693648223343679925
45940952768206080622328123873838808170496000000000
00000000000000000000000000000000000000000000000000
000000000000000
52.547 Seconds.
I see these numbers, ~2 seconds and ~52 seconds a lot, so I will conclude object-oriented APL is about 25 times as slow as fine-tuned machine language APL. Not too bad a penalty for portability across a wide spectrum of CPU architectures and Operating Systems.
With modern commodity processors doing billion of instructions per second, most people probably won't even notice.