the noise was not the motor.
the noise they talk about is the resonance between the road and the bottom of the vehicle.
changing the stupid diff ratio to 2.7  which is same as rolls royce limousine..helped the situation..
but they needed more..so they produced the sportspack and lifted the vehicle up in the air.
it was staggering putting a new sportspack and normal cooper next to each other..
its one of the manufacture cons nobody retail noticed.

people used to want thier brand new sportspack upgraded to more driveable and fun.
some experts gave big invoices for the upgrades..like john cooper garages and several big names..

we simply did the diff ratio to 3.4...and a custom easy exhaust..cat or non cat..didnt make much difference driving 
when cat was new..noise was more crackly without the cat.some like it..
and nothing else was required..we could do all this in 3 hours labour..we did so many..
and transformed the car to fun. 
cleaning the inlet manifolds like some advise as rover had bumps inside..made for misfires..because said bumps where 
swirl knobbles..which caused the inlet air fuel mixture to swirl..
so best left alone..rover where clever..

PUtting the K-head on the A-Series isn't as straightforward as people would have you think.  Also, from a warranty point of view, it would be disasterous due to the drop gears and gearbox claims.

The K-Series engine swap required monster shell changes (Rovers own prototype widened the whole car by 2" to accomodate it) which would have meant going through type approval and crash testing.

I think we all know how that would have ended.

BMW wanted to spend the minimum amount of money to keep the Mini name alive until the new MINI was ready to go. Putting a new head on would have been much more expensive.

They could have just as easily installed a K-series engine (Rover K series not BMW motorcycle K engine) instead, which I believe had already been done on a experimental basis, and was already in mass production for other Rovers. But that would have been too expensive.

makes you think the BMW motorcycle design folks were not consulted.  Yes, we have a DOHC cyl head and EFI that fits on that engine block...

The change to MPi was done to reduce emissions and lower noise. There were new emissions standards for 1997. To keep the Mini name alive until the MINI was ready, these had to be met. The name of the engineer who devised the system is Mike Theaker, who did it around 1991. BMW pulled Theaker's research off the shelf to give the Mini a few more years of life.

MPi helped meet noise regulations by allowing the fitting of the tall 2.76:1 final drive. The MPi improved the low and midrange torque enough so it could pull this tall final drive ratio. If you really want to feel what a MPi motor can do, put a 3.1 or 3.4 final drive in to compare it to older Minis.

That's also the reason for the front mount radiator - the European noise tests are drive-by tests, so moving the radiator to the front reduced the noise picked up by the side-located microphones on the sound test. The MPi's also got the side door crash beams and airbag to meet new regulations.

Spank, I recall emissions being a driver, but wasn't there also a horse power boost/power band improvement associated to the TPIs, as well?

If I'm remembering correctly, Vizard claims that during testing the A-series engine acts more like Four 1-cylinder engines than two 2-cylinder engines, and that even dual carbs were not very efficient. But he claimed dual carbs could next some slight gains that a single carb could not offer (with the caveat that they be set up "properly", not the easiest thing to do)

I'm going to take a WAG (Wild-A%$ Guess) and say it has to do with improving emissions vs cost to implement. What's the least investment we can implement to get us compliant with the world markets we sell these things in. With a front mounted radiator and an electric fans (VS mechanical fan and electric fan) it is easier and more efficient to control engine temps to maintain the optimum temp (212-degrees) for cleaner combustion and the dual point injection is probably more efficient than an SPI throttle body setup and the best compromise when incorporating manufacturing and development cost vs the potential marginal gains from going to full 4 individual injectors as well as the need to develop a completely new 4-port cylinder head and associated valve train requirements.

Why did Austin-Rover go to the TPI?
Is it more reliable than the SPI? Better fuel economy? More HP? International pressure? There was some expense to this? New Block?
Radiator reposition (this was because of drive by noise).
Inquiring minds want to know?