It seems to me that a hacker with a sleep regulator could get more work done in a day than one without. A regulator comes close to quartering the time spent sleeping, but I imagine time for other neccessities is still needed. Does halfing the time for programming tests seem acceptable to you guys?

For that matter, could the same hacker use Agents loaded with Computer Autosofts as assistants for the skill checks? Would Agents even need the Autosoft, or could Edit suffice? I'm not sure if it's legit to load Autosofts on Agents. If not could Drones with Computer Autosofts help?

Hell, could you just load/copy 9 rating 4 agents into a ghosting 2K$ Fairlight, and let them bang away at programs while you go to work? In theory they would never need to break, so that 8 hour day turns into a 24 hour day... would that reduce the time spent on the extended check to 1/3?

An interesting thought, but basically all you're doing is automating the job of programming. As that is becoming how it's done now, a programmer putting together code or even higher level scripts to be compiled down, what's really needed is creative thought. Something Agents lack. Assuming a GM even allows agents to program, sentient input would still be required about as much as the programming times indicate.

yeah, SR3 had the programming plan that had to be drafted before the real meat of the programming could be done. The plan covered the higher level thought (more or less), and allowed you to work with other programmers who were following the plan provided they were of a sufficient skill level. I think back in SR3 Agents could help with programming provided their rating was high enough to follow the plan.

the new SR4 system (at least until the matrix book comes out) doesn't really seem to set any level of skill as being necessary to helping out with a programming task... I would imagine that if a runner buddy with Logic 1 Computer 1 could roll for successes to help you out, then an agent would be able to as well..

I suppose it really depends on how the GM feels about the capabilities of Agents. (how much I is in that AI?).

either way, the reall payoff wouldn't really be in the extra dice added from helpers. It would be in the reduction in the time needed for the extended tests from having someone working around the clock. If any.

How would you handle an hacker that puts in 8 inventive/creative hours a day, and then has his suite of agents continue to crank away the dross during the other 16?

oh.. and I think the I goofed the cost of the Fairlight, it might be more than 2k$...

My thoughts are pretty much that Agents could only do "grunt work" type programming. Like you said, it depends on how intelligent they actually are. It's my view that as much programming as possible would already be automated to cut down on the amount of labor required, in which case Agents would not be able to add anything specific.

Of course, that's entirely a judgement call, not covered even remotely in any material that I know of.

There were rules for it under SR3, the programming suite with upgrade. Can't remember the exact rules, but it could only do half the work or the rating of the program suffered.

Don't agents usie their rating instead of skills? I thought agents always used Agent Rating+Program for whatever they did.