I agree with you. They want to have their cake and eat it too.
Equity is for the people who work for free (or at a less-than-industry-standard salary) when there's no guarantee of success. These two people want none of the risk and all of the reward, and it's a deal I'd never do, myself.
Under these conditions, they should be getting no equity at all, or perhaps a small amount in line with what early employees get, which is what they'd be under their proposition. 0.25% to maybe 5% (if you're feeling particularly generous) is fair.
I'd make choose one of a few counterproposals.
First, I might suggest that they quit and work on the project full time, and that you'd then do the investor's loan thing. That ought to cut the time to market in half (or more, depending on how many hours people actually work) and so it reduces the risk for you. Also, they have to take a risk by quitting their jobs, so they'll be forced to be more serious about it. I'd say that since you're putting in all of the financial risk, even though it's their know-how and technical skills, the original equity split is still fair here. (And I'm a technical guy myself.)
Second, I might propose what they were initially proposing, but give them the 0.25-5% equity that is normal for early employees, and explain to them how little risk they're taking if they're still working their day jobs full time and getting paid for their time in the new startup.
For the third option, I'd split up the $70K into salary vs. everything else, and drop the salary component out entirely. If you're building software, that will be most of the costs. (If you're building hardware or something else physical, it may not change that much.) The two could still work full-time elsewhere and work on the startup part time, and not get paid. Under these conditions, it might be more like splitting the equity equally three ways.
In any of these scenarios, you'd have to work out the specifics. The bottom line is that it's hard to do a startup when some of the cofounders are getting paid salary while others are not (or are paying others).
On a slightly unrelated note, 8 months is quite a long time to be working in the dark. You may already be accounting for this, but you want to bring your product/idea into contact with customers as early as possible, even in some sort of beta or MVP state. No idea survives contact with the customer, so the sooner you can do it and start refining your idea and business plan, the better. Don't go 8 months in the dark unless you're planning on at least another 8 months of refinement after that before getting it right and beginning to make real sales.