Re:Nobody ever ported Doom to run on a Cray 1
by tempodox on Monday December 21, 2020 @12:19PM
(#25497007)
Wouldn't that make Doom much harder? In-game adversaries would take advantage and execute vectorized attack patterns supported by cached-ammunition flow. Would Doomguy be able to survive at all?
[ Parent
| on HN ]
Re:Nobody ever ported Doom to run on a Cray 1
by IncRnd on Monday December 21, 2020 @02:21PM
(#25498362)
You should be okay with your Reciprocal BFG as a force multiplier.
EDIT for those who may not have understood my comment. You would be hard-pressed to find a Cray that performed division and would instead need to perform a reciprocal multiplication in constant time.
Re:Nobody ever ported Doom to run on a Cray 1
by ficklepickle on Monday December 21, 2020 @05:23PM
(#25500350)
How do you compute the reciprocal without division? A precomputed lookup table?
Re:Nobody ever ported Doom to run on a Cray 1
by variaga on Monday December 21, 2020 @07:11PM
(#25501370)
Newton-Rapheson method to calculate X = 1/D
1) Make a guess at the reciprocal X(0) (this can come from a small LUT, but a static "guess" can also work)
2) Calculate an improved reciprocal X(n) iteratively using the relation:
X(i+1) = X(i)*(2-X(i)*D)
Repeat (2) until X(i) is accurate "enough". This is relatively fast, since the # of correct bits in X() doubles for each iteration.There are several other methods: https://en.wikipedia.org/wiki/Division_algorithm#Fast_divisi...
Re:Nobody ever ported Doom to run on a Cray 1
by whoopdedo on Monday December 21, 2020 @01:22PM
(#25497709)
On the other hand, I can't wait to see E1M1 in <4s.
Re:Nobody ever ported Doom to run on a Cray 1
by Something1234 on Monday December 21, 2020 @05:18PM
(#25500295)
What's E1M1?
Re:Nobody ever ported Doom to run on a Cray 1
by FartyMcFarter on Monday December 21, 2020 @05:38PM
(#25500530)
Episode 1, map 1. The first level in Doom.