First off, a 2871 .64 is a great turbo for a drifting sr20 setup. I had two seperate 240sx with that setup. The first made 330whp or so at 18 psi I think on 93 octane, and stock cams. The second made 387whp on 272 cams. They were night and day, and the 272 car sucked to drift because the cams were all wrong for the turbo.
So basically in a drift car you want response. You can get that in a number of ways. Displacement, turbo sizing, compression, octane, tune, and design of the turbo system. As you play with those items you can fudge one item to get away with more or less of one of the other items.
1. Increasing the displacement would make a car have more power and response.
2. Turbo sizing. A lot goes into turbo sizing, take the 30r for example. There are so many versions of the garrett 30r, that some would work great on your car, and so would be absolute trash and make your car useless. The easiest way to choose a turbo is get someone experienced in turbos to help you choose, and drive or ride along in as many different turbo setups as possible before you choose yours. Russell and Derrick could help you choose a great turbo setup, but since you are going to run a 93 octane reasonably inexpensive build, the 2871 will most likely be a great choice.
3. Compression. Your not going to mess with this most likely, but there is no need to drop your compression greatly so you can run more boost with this turbo. And since you won't be running a lot of octane, you don't want to raise it. Leave it alone.
4. Octane. The 2871 setup with 104 octane or so would really come alive, and be very responsive, because the car can run more boost and have a lot more timing advance to get the fun going. This would make the car significantly more fun to drift, but isn't necessary. Just keep in mind if the car isn't as responsive as you want it, a great tune and more octane could be exactly what you need.
5. Tune. A great tune is super important. Don't just let anyone tune it, get someone good. But don't pay a grand if you don't have it, someone should be willing to do it for something reasonable. I normally paid 200 bucks a tune on a PFC with my tuner. But you need to build a relationship with them.
6. Design of the turbo system. Eliminating bends, tubing length, getting sizing of all the piping right, wastegate setup, and lots of other things will help make the power you want, along with creating the response you want. I will include cams in here too, since they are part of what is moving the air in the system. You want either stock cams, or at the very biggest, 262s. You need to balance the cam size with the turbo size, so if a cam works well from 3k-7k rpm you choose a turbo that works well in that same range, such as a 262 and the 2871r. If you want a 35r, there is no point running such a small cam since you are never going to make power from 3-4k, so you can just get rid of that idea, and that cam would choke the turbo out since it wants to go to at least 8k. So a 272 would be great for a 35r on an sr20 because they both have very high powerbands, although the 35r would suck balls on any type of non crazy motor for drifting.
Ok, now to answer your questions about power. You might come up a little shy on no cams and the 2871 of your 370whp goal. You would either need to add some more octane, more boost than your should on 93, or cams to make the 370whp number. Depending on the dyno you use, you might make it there. I would say instead of making a power goal with that turbo, just decide on how you are going to set the car up as nicely, and with the best craftsmanship as possible. Concentrate on making everything nice, dependable, and clean. Don't concentrate on power.
What I mean by that is buy a decent manifold, decent external wastegate, nice garrett bb turbo ( which you already are ), and get everything working well together. The power will just come with a nicely setup car. This would mean you probably cut your boost off at 18psi or so on pump gas. Sure people run 20psi on the forums, but that seems a bit aggressive if you want everything to work great a year from now, and those last 2 psi aren't going to do that much on 93, and they are a liability, although they would get you close to your 370whp goal. So on pump gas and smaller cams you will probably hit 330-350whp.
And the end all be all to response with this setup, which I didn't put up top because you can't build it, is driving technique. You need to stay over 4500rpm with this setup to drift happily in a competitive setup. As long as you stay up there, the power is almost as instant as in my LS1 car. The boost will be instant, the ball bearing center section will already be spun up, and the tires will just light right up. If that means you have to pop the clutch or shift to do that constantly, that is what you have to do, and you just get used to it. If that means you need to pop your redline up to 8k, which I always did even on my stock sr20s with stock cams, then thats what you need to do so you have rpm extension to stay up there without having to shift quite as much. Your technique will have to change, and as the turbo on your sr20 gets bigger, the car gets harder and more technical to drive. Any larger than the 2871 .64, and I would almost go as far to say you NEED 100+ octane to enjoy the setup. I ran my full race 3071 setup on 93, and forced it to work through driving style, but most people wouldn't have been happy with it.
Look at the below dyno. That was my gold car and 272 cams on the 2871. Looks pretty retarded doesn't it. Don't put those cams together with that turbo.http://fabricatedmotorsports.com/images/gold%20car%20dyno.jpg
And really quickly, lets take 3 cars and compare and contrast them. Stew has my old built sr20 with a 2871r setup in his s14. Russell has a 3076 on a s15 sr20 which is stock. Tcox runs a 3071 twin scroll setup. All use race gas.
Stew. His powerband is probably 4000-8000, he runs around 23-28psi I would guess, he makes well over 400whp at this boost. His car clearly works and makes lots of smoke.
Russell. His motor is stock, I think he runs a powerband closer to 4500-7800, he runs 18psi, he makes 380whp.
Tcox. His motor is built, he has a power band of 4500-8500, he runs 28 psi, he makes 485whp.
Stew makes significantly more power on a much smaller turbo because he runs way more boost. Russell is running far less boost because he doesn't want to build his motor. Russell is subscribing to the idea that you choose a turbo that is super efficient to get the job done, and run it at 50% duty cycle, while stew is running his setup at 90% duty cycle. Both make enough power to get things done, both are responsive enough to keep their owners happy. Tcox's is running his setup closer to 100% duty cycle, and makes the most power by quite a bit. His setup clearly works too, and is the only one to run 272 cams. Stew's motor is the only one that would probably still work on pump gas though, the others would loose too much response if they went to pump gas, and their drivers would be frustrated. It should also be noted that both of the larger turbo cars have gone to significant lengths to cut down intercooler piping and make their plumbing design as efficient as possible. It should also be noted that the 2871 car is probably the only one that can simply get on the gas at 4k rpm without a clutch kick after a transition and keep drifting smoothly outside of second gear. Both of the larger turbo cars would have to pop the clutch to stay in boost when transitioning 3rd gear stuff. The drivers probably don't even notice they do it anymore, it just becomes habit and part of your driving style.
place holder. I will finish this later.